This action might not be possible to undo. Are you sure you want to continue?
:
Comparisons between
FRAPCON3.4, FRAPTRAN 1.4, and
MATPRO
Office of Nuclear Regulatory Research
NUREG/CR7024
PNNL19417
Material Property Correlations:
Comparisons between
FRAPCON3.4, FRAPTRAN 1.4, and
MATPRO
Manuscript Completed: August 2010
Date Published: March 2011
Prepared by
W.G. Luscher and K.J. Geelhood
Pacific Northwest National Laboratory
P.O. Box 999
Richland, WA 99352
M.E. Flanagan, NRC Project Manager
NRC Job Code N6609
Office of Nuclear Regulatory Research
NUREG/CR7024
PNNL19417
iii
Abstract
The U.S. Nuclear Regulatory Commission uses the computer codes FRAPCON3 and
FRAPTRAN to model steadystate and transient fuel behavior, respectively, in regulatory
analysis. In this document, material property correlations for oxide fuels and cladding materials
are presented and discussed. Comparisons are made between the material property correlations
used in the most recent versions of the codes, FRAPCON3.4 and FRAPTRAN 1.4.
Comparisons are also made with MATPRO, which is a compilation of material property
correlations with an extensive history of use with various fuel performance and severe accident
codes. In addition to modeltomodel comparisons, modeltodata comparisons and source codes
for the FRAPCON3.4 and FRAPTRAN 1.4 correlations are provided.
iv
v
Foreword
The U.S. Nuclear Regulatory Commission uses the computer codes FRAPCON and FRAPTRAN
to model steadystate and transient fuel behavior, respectively, in regulatory analysis. To
effectively model fuel behavior, material property correlations must be used for a wide range of
operating conditions (e.g., temperature and burnup). In this sense, a “material property” is a
physical characteristic of the material whose quantitative value is necessary in the analysis
process.
The consolidated resource for “material properties” cited most often in the literature is MATPRO
(as documented in Volume 4 of NUREG/CR6150). MATPRO is a compilation of fuel and
cladding material property correlations with an extensive history of use with fuel performance
and severe accident codes. However, consistency among the material property correlations in
FRAPCON3, FRAPTRAN, and MATPRO has never been complete, as the material property
correlations in the fuel performance codes have evolved to accommodate recent fuel design
changes.
The primary purpose of this report is to consolidate the current material property correlations
used in FRAPCON and FRAPTRAN into a single document. In addition to describing the
material property correlations used in the subroutines of latest versions of the codes, FRAPCON
3.4 and FRAPTRAN 1.4, this report also provides a variety of comparisons to the material
property correlations used in MATPRO.
vi
vii
Contents
Abstract .......................................................................................................................................... iii
Foreword .......................................................................................................................................... v
List of Figures ................................................................................................................................. ix
List of Tables ................................................................................................................................... xi
Executive Summary ..................................................................................................................... xiii
Abbreviations ................................................................................................................................. xv
1.0 Introduction ........................................................................................................................... 1.1
1.1 Relation to Other Reports .................................................................................................. 1.2
1.2 References .......................................................................................................................... 1.3
2.0 Oxide Fuel Properties ............................................................................................................ 2.1
2.1 Fuel Melting Temperature (PHYPRP) ............................................................................... 2.1
2.1.1 Model Development and Comparisons ...................................................................... 2.1
2.1.2 Applicability and Uncertainty .................................................................................... 2.3
2.1.3 References .................................................................................................................. 2.3
2.2 Fuel Specific Heat Capacity (FCP) and Fuel Enthalpy (FENTHL) ................................... 2.4
2.2.1 Model Development and Comparisons ...................................................................... 2.4
2.2.2 Applicability and Uncertainty .................................................................................... 2.7
2.2.3 References .................................................................................................................. 2.7
2.3 Fuel Thermal Conductivity (FTHCON)............................................................................. 2.8
2.3.1 Model Development and Comparisons ...................................................................... 2.8
2.3.2 Applicability and Uncertainty .................................................................................. 2.20
2.3.3 References ................................................................................................................ 2.21
2.4 Fuel Emissivity (FEMISS) ............................................................................................... 2.22
2.4.1 Model Development and Comparisons .................................................................... 2.22
2.4.2 Applicability and Uncertainty .................................................................................. 2.23
2.4.3 References ................................................................................................................ 2.23
2.5 Fuel Thermal Expansion (FTHEXP) ............................................................................... 2.24
2.5.1 Model Development and Comparisons .................................................................... 2.24
2.5.2 Applicability and Uncertainty .................................................................................. 2.26
2.5.3 References ................................................................................................................ 2.26
2.6 Fuel Densification (FUDENS) ......................................................................................... 2.27
2.6.1 Model Development and Comparisons .................................................................... 2.27
2.6.2 Applicability and Uncertainty .................................................................................. 2.30
2.6.3 References ................................................................................................................ 2.30
2.7 Fuel Swelling (FSWELL) ................................................................................................ 2.31
2.7.1 Model Development and Comparisons .................................................................... 2.31
2.7.2 Applicability and Uncertainty .................................................................................. 2.34
2.7.3 References ................................................................................................................ 2.34
3.0 Cladding Material Properties ................................................................................................ 3.1
3.1 Cladding Specific Heat (CCP) ........................................................................................... 3.1
3.1.1 Model Development and Comparisons ...................................................................... 3.1
3.1.2 Applicability and Uncertainty .................................................................................... 3.3
3.1.3 References .................................................................................................................. 3.3
3.2 Cladding Thermal Conductivity (CTHCON) ..................................................................... 3.3
3.2.1 Model Development and Comparisons ...................................................................... 3.4
3.2.2 Applicability and Uncertainty .................................................................................... 3.6
3.2.3 References .................................................................................................................. 3.6
3.3 Cladding Oxide Thermal Conductivity (ZOTCON) .......................................................... 3.7
viii
3.3.1 Model Development and Comparisons ...................................................................... 3.8
3.3.2 Applicability and Uncertainty .................................................................................. 3.11
3.3.3 References ................................................................................................................ 3.11
3.4 Cladding Surface Emissivity (ZOEMIS) ......................................................................... 3.11
3.4.1 Model Development and Comparisons .................................................................... 3.12
3.4.2 Applicability and Uncertainty .................................................................................. 3.14
3.4.3 References ................................................................................................................ 3.14
3.5 Cladding Thermal Expansion (CTHEXP) ....................................................................... 3.14
3.5.1 Model Development and Comparisons .................................................................... 3.14
3.5.2 Applicability and Uncertainty .................................................................................. 3.17
3.5.3 References ................................................................................................................ 3.18
3.6 Cladding Elastic Modulus (CELMOD) and Shear Modulus (CSHEAR) ........................ 3.18
3.6.1 Model Development and Comparisons .................................................................... 3.19
3.6.2 Applicability and Uncertainty .................................................................................. 3.24
3.6.3 References ................................................................................................................ 3.24
3.7 Cladding Axial Growth (CAGROW) .............................................................................. 3.25
3.7.1 Model Development and Comparisons .................................................................... 3.25
3.7.2 Applicability and Uncertainty .................................................................................. 3.29
3.7.3 References ................................................................................................................ 3.29
3.8 Creep Rate (CREPR) ....................................................................................................... 3.30
3.8.1 Model Development and Comparisons .................................................................... 3.30
3.8.2 Applicability and Uncertainty .................................................................................. 3.33
3.8.3 References ................................................................................................................ 3.34
3.9 Cladding Meyer Hardness (CMHARD)........................................................................... 3.34
3.9.1 Model Development and Comparisons .................................................................... 3.35
3.9.2 Applicability and Uncertainty .................................................................................. 3.36
3.9.3 References ................................................................................................................ 3.36
4.0 Gas Material Properties ......................................................................................................... 4.1
4.1 Gas Conductivity (GTHCON) ........................................................................................... 4.1
4.1.1 Model Development and Comparisons ...................................................................... 4.1
4.1.2 Applicability and Uncertainty .................................................................................... 4.7
4.1.3 References .................................................................................................................. 4.8
Appendix A Subroutine Source Codes ......................................................................................... A.1
ix
List of Figures
Figure 2.11 A comparison between solidus temperatures calculated by MATPRO using the
FHYPRP subroutine and FRAPCON3/FRAPTRAN (FRAP) using the
PHYPRP subroutine. Results for burnup varying from 0 to 62 GWd/MTU and
Pu contents of 0, 3, and 7 wt% are presented. .................................................. 2.2
Figure 2.12 A comparison between liquidus temperatures calculated by MATPRO using
the FHYPRP subroutine and FRAPCON3/FRAPTRAN (FRAP) using the
PHYPRP subroutine. Results for burnup varying from 0 to 62 GWd/MTU and
Pu contents of 0, 3, and 7 wt% are presented. .................................................. 2.3
Figure 2.21 Specific heat capacity of UO
2
from three experimenters compared with the
FRAPCON3/FRAPTRAN/MATPRO (FRAP) correlation (solid line) for UO
2
.2.6
Figure 2.22 Specific heat capacity of (U
0.8
Pu
0.2
)O
2
from three experimenters compared with
the FRAPCON3/FRAPTRAN/MATPRO (FRAP) correlations (solid line) for
MOX. ............................................................................................................... 2.7
Figure 2.3.31 MATPRO and modified NFI thermal conductivity models for UO
2
as a
function of temperature. Unlike the MATPRO model, the modified NFI model
can account for burnup and is presented at burnup levels of 0 and
30 GWd/MTU for comparison. ...................................................................... 2.14
Figure 2.3.32 MATPRO modeltodata comparison for thermal conductivity of unirradiated
UO
2
. ................................................................................................................ 2.15
Figure 2.3.33 Modified NFI modeltodata comparison for thermal conductivity of
unirradiated UO
2
. ........................................................................................... 2.15
Figure 2.3.34 MATPRO modeltodata comparison for thermal conductivity of irradiated
UO
2
. ................................................................................................................ 2.16
Figure 2.3.35 Modified NFI modeltodata comparison for thermal conductivity of irradiated
UO
2
. ................................................................................................................ 2.17
Figure 2.3.36 MATPRO and Duriez/modified NFI thermal conductivity models as a function
of temperature for MOX fuel with 7 wt% plutonium content. Unlike the
MATPRO model, the Duriez/modified NFI model can account for burnup and
is presented at burnup levels of 0 and 30 GWd/MTU for comparison. ......... 2.18
Figure 2.3.37 MATPRO modeltodata comparison for thermal conductivity of MOX
modeled with 7 wt% of plutonium. ................................................................ 2.19
Figure 2.3.38 Duriez/modified NFI modeltodata comparison for thermal conductivity of
MOX with modeled with 7 wt% of plutonium. ............................................. 2.19
Figure 2.41 Data to model comparison for FEMISS. ........................................................ 2.23
Figure 2.51 Modeltodata comparison for UO
2
correlation used in MATPRO and
FRAPCON3/FRAPTRAN with updated constants. ..................................... 2.25
Figure 2.52 Modeltodata comparison for PuO
2
correlation used in MATPRO and
FRAPCON3/FRAPTRAN. ........................................................................... 2.26
Figure 2.61 Comparison of densification predictions generated by the FUDENS model and
experimental densification measurements determined in reactor. ................. 2.30
Figure 2.71 Data comparison between MATPRO and FRAPCON3.4 versions of
FSWELL. ....................................................................................................... 2.33
Figure 2.72 Comparison between swelling rates determined for various instrumented fuel
assemblies tests at Halden and swelling rates predicted by FRAPCON3.4. . 2.34
Figure 3.21 CTHCON model and data used for development. ........................................... 3.5
Figure 3.22 CTHCON model and data collected since development. ................................. 3.6
x
Figure 3.31 Oxide thermal conductivity measurements as a function of temperature
compared with the MATPRO and FRAPCON3.4/FRAPTRAN 1.4 (FRAP)
cladding oxide thermal conductivity models. ................................................ 3.10
Figure 3.41 Modeltodata comparison for cladding oxide emissivity. ............................. 3.13
Figure 3.51 Modeltodata comparison for FRAPCON3.4/FRAPTRAN 1.4 (FRAP) and
MATPRO axial thermal expansion. Data from cladding tubes..................... 3.17
Figure 3.52 Modeltodata comparison for FRAPCON3.4/FRAPTRAN 1.4 (FRAP) and
MATPRO diametral thermal expansion. Data from cladding tubes. ............ 3.17
Figure 3.61 Predicted vs. measured values for Young’s modulus for various data sets. .. 3.23
Figure 3.75 Modeltodata comparison for SRA Zircaloy4 (PWR). The standard error of
AL/L (%) is 0.11. ............................................................................................ 3.27
Figure 3.76 Modeltodata comparison for RXA Zircaloy2 (BWR). The standard error of
AL/L (%) is 0.07. ............................................................................................ 3.27
Figure 3.77 Modeltodata comparison for M5. M5 data was collected by Gilbon et al.
(2000). The standard error of AL/L (%) is 0.05. ........................................... 3.28
Figure 3.78 Modeltodata comparison for ZIRLO. Vandellos data was collected by Irisa
(2000) while North Anna and BR3 data was collected by Sabol (1994). The
standard error of AL/L (%) is 0.05. ................................................................ 3.28
Figure 3.81 Predicted vs. measured hoop strain for SRA Zircaloy model. ....................... 3.33
Figure 3.82 Predicted vs. measured hoop strain for RXA Zircaloy model. ...................... 3.33
Figure 3.91 CMHARD correlation as a function of temperature with data from Peggs and
Godin. ............................................................................................................. 3.36
Figure 4.11 Modeltodata comparison for He thermal conductivity. ................................. 4.3
Figure 4.12 Modeltodata comparison for Ar thermal conductivity. ................................. 4.4
Figure 4.13 Modeltodata comparison for Kr thermal conductivity. ................................. 4.5
Figure 4.14 Modeltodata comparison for Xe thermal conductivity. ................................. 4.6
Figure 4.15 Modeltodata comparison for H
2
thermal conductivity. ................................. 4.6
Figure 4.16 Modeltodata comparison for N
2
thermal conductivity. ................................. 4.7
Figure 4.17 Modeltodata comparison for steam thermal conductivity. ............................ 4.7
xi
List of Tables
Table 1.1 Roadmap to documentation of models and properties in NRC fuel performance
codes, FRAPCON3.4 and FRAPTRAN 1.4. .................................................. 1.2
Table 2.21 Constants used in UO
2
, Gd
2
O
3
, and PuO
2
heat capacity and enthalpy
correlations. ...................................................................................................... 2.5
Table 2.51 Parameters used in UO
2
and PuO
2
solidphase thermal expansion correlations.2.24
Table 3.11 Specific heat capacity database for Zircaloy2, Zircaloy4, ZIRLO, and M5. . 3.2
Table 3.12 Specific heat capacity database for ZrNb1 alloys. .......................................... 3.2
Table 3.81 Parameters for FRAPCON3.4 creep equation for SRA and RXA cladding. 3.31
Table 4.11 Constants used in gas thermal conductivity correlations in MATPRO. ........... 4.1
Table 4.12 Constants used in gas thermal conductivity correlations in FRAP. ................. 4.2
Table 4.13 Standard deviation of gas thermal conductivity correlations. .......................... 4.8
xii
xiii
Executive Summary
The U.S. Nuclear Regulatory Commission uses the computer codes FRAPCON3 and
FRAPTRAN to model steadystate and transient fuel behavior, respectively, in regulatory
analysis. In this document, material property correlations for oxide fuels and cladding materials
are presented and discussed. Comparisons are made between the material property correlations
used in the most recent versions of the codes, FRAPCON3.4 and FRAPTRAN 1.4.
Comparisons are also made with MATPRO, which is a compilation of material property
correlations with an extensive history of use with various fuel performance and severe accident
codes. In addition to modeltomodel comparisons, modeltodata comparisons and source codes
for the FRAPCON3.4 and FRAPTRAN 1.4 correlations are also provided.
This is one of three documents that describe the FRAPCON3.4 and FRAPTRAN 1.4 fuel
performance codes. The FRAPCON3.4 code structure and behavioral models are described in
the FRAPCON3.4 code description document (Geelhood et al., 2010a). The FRAPTRAN 1.4
code structure and behavioral models are described in the FRAPTRAN 1.4 code description
document (Geelhood et al., 2010b). Additional descriptions of mechanical models and modelto
data comparisons for mechanical property correlations are provided in the mechanical properties
report (Geelhood, 2008).
xiv
xv
Abbreviations
°C degrees Celsius
°F degrees Fahrenheit
Ar argon
BWR boilingwater reactor
cm
2
square centimeter(s)
crud Chalk River Unidentified Deposit (generic term for various residues deposited on
fuel rod surfaces, originally coined by Atomic Energy of Canada, Ltd. to describe
deposits observed on fuel from the test reactor at Chalk River)
Gd
2
O
3
gadolinia
GPa gigapascal(s)
GWd/MTU gigawattdays per metric ton of uranium
H hydrogen
He helium
J joule(s)
K kelvin
kg kilogram(s)
Kr krypton
LWR lightwater reactor
m meter(s)
m
2
square meter(s)
m
3
cubic meter(s)
MeV megaelectron volt(s)
mm millimeter(s)
MOX mixed oxide, (U, Pu)O
2
MPa megapascal(s)
MWd/kgU megawattdays per kilogram of uranium
MWd/MTU megawattdays per metric ton of uranium
MWs/kgU megawattseconds per kilogram of uranium
n neutron(s)
N nitrogen
Nb niobium
NFI Nuclear Fuels Industries
NRC U.S. Nuclear Regulatory Commission
O/M oxygentometal
Pa pascal(s)
PNNL Pacific Northwest National Laboratory
Pu plutonium
PuO
2
plutonium dioxide
PWR pressurizedwater reactor
RXA recrystallized annealed
s second(s)
SRA stress relief annealed
TD theoretical density
tHM tonne(s) of heavy metal
UO
2
uranium dioxide
UO
2
Gd
2
O
3
uraniagadolinia
W watt(s)
wt% percent by weight
xvi
Xe xenon
ZrNb1 Russian zirconium alloy with 1 wt% of niobium
ZrO
2
zirconium dioxide
µm micrometer(s)
1.1
1.0 Introduction
The U.S. Nuclear Regulatory Commission (NRC) uses the computer codes FRAPCON3 and
FRAPTRAN to model steadystate and transient fuel behavior, respectively, in regulatory
analysis. To effectively model fuel behavior, material property correlations must be used for a
wide range of operating conditions (e.g., temperature and burnup). In this sense, a “material
property” is a physical characteristic of the material whose quantitative value is necessary in the
analysis process. Further, the property may be used to compare the benefits of one material with
those of another. Generally speaking, the material properties of interest in regulatory analysis of
nuclear fuel behavior are mechanical or thermodynamic.
The issue of what is and is not a “material property” will never be universally resolved. In this
report, properties such as thermal conductivity are included. Other characteristics of the material
(e.g., fission gas release) are considered “models” rather than properties and are discussed
elsewhere. The material property correlations for the FRAPCON3 and FRAPTRAN computer
codes were partially documented in NUREG/CR6534 and NUREG/CR6739, respectively
(Lanning et al., 2005; Cunningham et al., 2001) as well as in NUREG/CR6150 (Siefken et al.,
2001). Some of these have been modified or updated since the original code documentation was
published.
The primary purpose of this report is to consolidate the current material property correlations
used in FRAPCON3 and FRAPTRAN into one document. Material property correlations for
oxide fuels, including uranium dioxide (UO
2
) and mixed oxide (MOX) fuels, are described in
Section 2. Throughout this document, the term MOX is used to describe fuels that are blends of
uranium and plutonium oxides, (U,Pu)O
2
. The properties for UO
2
with other additives
(e.g., gadolinia) are also discussed. Material property correlations for cladding materials and
gases are described in Sections 3 and 4, respectively.
In addition to describing the material property correlations used in the subroutines of
FRAPCON3 and FRAPTRAN, this report provides a variety of comparisons between material
property correlations and data. Although they are frequently identical, comparisons are made
between the material property correlations used in the FRAPCON3 and FRAPTRAN codes.
Comparisons are also made between the material property correlations used in MATPRO, a
compilation of fuel and cladding material property correlations with an extensive history of use
with various fuel performance and severe accident codes. For several reasons, consistency
among the material property correlations in FRAPCON3, FRAPTRAN, and MATPRO has never
been complete. However, the current versions of FRAPCON3 and FRAPTRAN use a relatively
consistent set of correlations for the properties that are used by both codes. The material property
correlations in the most recent version of MATPRO are documented in Volume 4 of
NUREG/CR6150 (Siefken et al., 2001). In addition to comparison of the various correlations,
correlationtodata comparisons are made with FRAPCON3, FRAPTRAN, and MATPRO
properties.
All comparisons in this report are based on the material property correlations used in the most
recent version of the FRAPCON3 and FRAPTRAN codes, FRAPCON3.4 and FRAPTRAN 1.4.
The source code for each material property correlation discussed is provided for FRAPCON3.4
and FRAPTRAN 1.4 (see Appendix A) as well as a range of applicability and an estimate of
uncertainty, where possible.
1.2
1.1 Relation to Other Reports
The full documentation of the steadystate and transient fuel performance codes is described in
three documents. The basic fuel, cladding, and gas material properties used in FRAPCON3.4
and FRAPTRAN 1.4 are described in the material properties handbook (this report). The
FRAPCON3.4 code structure and behavioral models are described in the FRAPCON3.4 code
description document (Geelhood et al., 2010a). The FRAPTRAN 1.4 code structure and
behavioral models are described in the FRAPTRAN 1.4 code description document (Geelhood et
al., 2010b). Additional descriptions of mechanical models and modeltodata comparisons for
mechanical property correlations are provided in the mechanical properties report (Geelhood,
2008).
Table 1.1 shows where each specific material property and model used in the NRC fuel
performance codes are documented.
Table 1.1 Roadmap to documentation of models and properties in NRC fuel
performance codes, FRAPCON3.4 and FRAPTRAN 1.4.
Model/Property FRAPCON3.4 FRAPTRAN 1.4
Fuel thermal conductivity Material properties handbook Material properties handbook
Fuel thermal expansion Material properties handbook Material properties handbook
Fuel melting temperature Material properties handbook Material properties handbook
Fuel specific heat Material properties handbook Material properties handbook
Fuel enthalpy Material properties handbook Material properties handbook
Fuel emissivity Material properties handbook Material properties handbook
Fuel densification Material properties handbook NA
Fuel swelling Material properties handbook NA
Fission gas release FRAPCON3 code description FRAPTRAN code description
Fuel relocation
FRAPCON3 code description FRAPTRAN code description
Fuel grain growth FRAPCON3 code description NA
High burnup rim model FRAPCON3 code description NA
Nitrogen release FRAPCON3 code description NA
Helium release FRAPCON3 code description NA
Radial power profile FRAPCON3 code description NA (input parameter)
Stored energy FRAPCON3 code description FRAPTRAN code description
Decay heat model NA FRAPTRAN code description
Fuel and cladding temperature
solution
FRAPCON3 code description FRAPTRAN code description
Cladding thermal conductivity Material properties handbook Material properties handbook
Cladding thermal expansion Material properties handbook Material properties handbook
Cladding elastic modulus Material properties handbook Material properties handbook
Cladding creep model Material properties handbook NA
Cladding specific heat Material properties handbook Material properties handbook
1.3
Model/Property FRAPCON3.4 FRAPTRAN 1.4
Cladding emissivity Material properties handbook Material properties handbook
Cladding axial growth Material properties handbook NA
Cladding Meyer hardness Material properties handbook Material properties handbook
Cladding annealing FRAPCON3 code description FRAPTRAN code description
Cladding yield stress, ultimate
stress, and plastic deformation
FRAPCON3 code description FRAPTRAN code description
Cladding failure criteria NA FRAPTRAN code description
Cladding waterside corrosion FRAPCON3 code description NA (input parameter)
Cladding hydrogen pickup FRAPCON3 code description NA (input parameter)
Cladding high temperature
oxidation
NA FRAPTRAN code description
Cladding ballooning model
NA FRAPTRAN code description
Cladding mechanical deformation FRAPCON3 code description FRAPTRAN code description
Oxide thermal conductivity Material properties handbook Material properties handbook
Crud thermal conductivity FRAPCON3 code description NA
Gas conductivity Material properties handbook Material properties handbook
Gap conductance FRAPCON3 code description FRAPTRAN code description
Plenum gas temperature FRAPCON3 code description FRAPTRAN code description
Rod internal pressure FRAPCON3 code description FRAPTRAN code description
Coolant temperature and heat
transfer coefficients
FRAPCON3 code description FRAPTRAN code description
Optional models and properties not developed at PNNL
VVER fuel and cladding models NA NUREG/IA0164
(Shestopalov et al., 1999)
Cladding FEA model VTTR1133706
(Knuttilla, 2006)
VTTR1133706
(Knuttilla, 2006)
NA = not applicable
VVER = watercooled, watermoderated energy reactor
FEA = finite element analysis
1.2 References
Lanning, D.D., C.E. Beyer, and K.J. Geelhood. 2005. FRAPCON3 Updates, Including Mixed
Oxide Fuel Properties, NUREG/CR6534, Vol. 4, PNNL11513, Pacific Northwest National
Laboratory, Richland, WA.
Cunningham, M.E., C.E. Beyer, F.E. Panisko, P.G. Medvedev, G.A. Berna, and H.H. Scott.
2001. FRAPTRAN: Integral Assessment, NUREG/CR6739, Vol. 2, PNNL13576, Pacific
Northwest National Laboratory, Richland, WA.
1.4
Siefken, L.J., E.W. Coryell, E.A. Harvego, and J.K. Hohorst. 2001. SCDAP/RELAP5/MOD 3.3
Code Manual MATPROA Library of Materials Properties for LightWaterReactorAccident
Analysis, NUREG/CR6150, Vol. 4, Rev. 2, INEL96/0422, Idaho National Engineering and
Environmental Laboratory, Idaho Falls, ID.
Geelhood, K.J., W.G. Luscher, and C.E. Beyer. 2010a. FRAPCON3.4: A Computer Code for
the Calculation of SteadyState, ThermalMechanical Behavior of Oxide Fuel Rods for High
Burnup, NUREG/CR7022, Vol. 1, PNNL19418, Vol. 1, Pacific Northwest National Laboratory,
Richland, WA.
Geelhood, K.J., W.G. Luscher, C.E. Beyer, and J.M. Cuta. 2010b. FRAPTRAN 1.4: A Computer
Code for the Transient Analysis of Oxide Fuel Rods, NUREG/CR7023, Vol. 1, PNNL19400,
Vol. 1, Pacific Northwest National Laboratory, Richland, WA.
Geelhood, K.J., W.G. Luscher, and C.E. Beyer. 2008. PNNL Stress/Strain Correlation for
Zircaloy, PNNL17700, Pacific Northwest National Laboratory, Richland, WA.
Shestopalov, A., K. Lioutov, L. Yegorova, G. Abyshov, and K. Mikitiouk. 1999. Modification of
USNRC’s FRAPT6 Fuel Rod Transient Code for High Burnup VVER Fuel, NUREG/IA0164,
U.S. Nuclear Regulatory Commission, Washington, D.C.
Knuutila, A. 2006. Improvements on FRAPCON3/FRAPTRAN Mechanical Modeling, VTTR
1133706, VTT, Finland.
2.1
2.0 Oxide Fuel Properties
Material property correlations for UO
2
and MOX fuel pellets are described in the following
subsections. These correlations include the melting temperature, specific heat capacity and
enthalpy, thermal conductivity, emissivity, thermal expansion, densification, and swelling of the
fuel pellets. When indicated, some correlations account for gadolinia (Gd
2
O
3
) additions in UO
2
fuel pellets. The subroutine containing the material correlation in the FRAPCON3/FRAPTRAN
codes is given in parentheses in the heading of each subsection.
2.1 Fuel Melting Temperature (PHYPRP)
The subroutine PHYPRP is used to calculate the temperature of the first appearance of liquid
phase (solidus) and the temperature of the melting point of the last solid phase (liquidus) of UO
2
and MOX. These temperatures are calculated as a function of burnup and plutonia content. The
same subroutine is used in both FRAPCON3.4 and FRAPTRAN 1.4 and exhibits smaller burnup
dependence than the corresponding MATPRO correlation, FHYPRP.
2.1.1 Model Development and Comparisons
The equations used to calculate the UO
2
and MOX melting points were derived by using
3113.15K as the melting temperature of urania, which was determined experimentally by
Brassfield (1968), and a least squares fit to parabolic equations for the solidus and liquidus
boundaries from the Lyon and Baily (1967) phase diagram for the stoichiometric (U, Pu)O
2
mixed oxide. These correlations are presented in Equations 2.11 through 2.15.
(2.11)
(2.12)
For c = 0
(2.13)
For c > 0
(2.14)
(2.15)
where
c = Pu content (wt%)
sldus (c) = solidus temperature as a function of Pu content (°C)
liqdus (c) = liquidus temperature as a function of Pu content (°C)
ftmelt = fuel melting temperature (K)
FBu = burnup (MWd/MTU)
fdelta = temperature range between solidus and liquidus (K)
2 3
c 10 x 468390 . 7 c 41395 . 5 2840 ) c ( sldus
÷
+ ÷ =
2 2
c 10 x 448518 . 1 c 21860 . 3 2840 ) c ( liqdus
÷
+ ÷ =
10000 / FBu 5 15 . 3113 ftmelt ÷ =
10000 / FBu 5 ) c ( sldus ) c ( liqdus fdelta ÷ ÷ =
10000 / FBu 5 15 . 273 ) c ( sldus ftmelt ÷ + =
2.2
The correlations defined above for the subroutine PHYPRP are almost identical to those used in
the subroutine FHYPRP documented in MATPRO. However, the burnup dependence of the
PHYPRP subroutine differs from the FHYPRP subroutine. A burnup dependence of
3.2 K/GWd/MTU is used in FHYPRP, while a burnup dependence of 0.5 K/GWd/MTU is used in
PHYPRP. This change was suggested by Popov et al. (2000) and is based on data from Adamson
et al. (1985) and Komatsu et al. (1988).
The burnupdependent term in Equations 2.13 through 2.15 reflects this change in burnup
dependence. Also, the term 3.21860 c in Equation 2.12 is slightly different from the term used
in FHYPRP. The coefficient was changed from 3.21660 in FHYPRP to 3.21860 in PHYPRP.
However, this change does not appreciably affect the liquidus temperature.
Comparisons between solidus and liquidus temperature calculated by PHYPRP and FHYPRP are
presented in Figures 2.11 and 2.12. Both subroutines predict solidus and liquidus temperatures
that decrease with increasing burnup and plutonium concentration. However, these figures reveal
a smaller decline in solidus and liquidus temperatures for the FRAPCON3 and FRAPTRAN
subroutine (PHYPRP) than the MATPRO subroutine (FHYPRP). This is a result of reducing the
burnup dependence from 3.2 to 0.5 K/GWd/MTU between the FHYPRP and PHYPRP
subroutines, respectively.
Solidus Temperatures
2825
2875
2925
2975
3025
3075
3125
0 10 20 30 40 50 60
Burnup (GWd/MTU)
T
e
m
p
e
r
a
t
u
r
e
(
K
)
MATPRO 0 wt% MATPRO 3 wt% MATPRO 7 wt%
FRAP 0 wt% FRAP 3 wt% FRAP 7 wt%
Figure 2.11 A comparison between solidus temperatures calculated by MATPRO using
the FHYPRP subroutine and FRAPCON3/FRAPTRAN (FRAP) using the
PHYPRP subroutine. Results for burnup varying from 0 to 62 GWd/MTU
and Pu contents of 0, 3, and 7 wt% are presented.
2.3
Liquidus Temperatures
2825
2875
2925
2975
3025
3075
3125
0 10 20 30 40 50 60
Burnup (GWd/MTU)
T
e
m
p
e
r
a
t
u
r
e
(
K
)
MATPRO 0 wt% MATPRO 3 wt% MATPRO 7 wt%
FRAP 0 wt% FRAP 3 wt% FRAP 7 wt%
Figure 2.12 A comparison between liquidus temperatures calculated by MATPRO using
the FHYPRP subroutine and FRAPCON3/FRAPTRAN (FRAP) using the
PHYPRP subroutine. Results for burnup varying from 0 to 62 GWd/MTU
and Pu contents of 0, 3, and 7 wt% are presented.
2.1.2 Applicability and Uncertainty
The correlation used in PHYPRP is valid through the melting temperature of UO
2
and MOX for
burnups up to 62 GWd/MTU. Currently, there are no models for uraniagadolinia (UO
2
Gd
2
O
3
)
fuel melting. However, there are plans to develop a correlation in a future revision of the code.
Due to the experimental data and data interpretations on which these temperature correlations are
based, it is difficult to express a quantitative measure of uncertainty. A conservative estimate of
uncertainty based on engineering judgment is ± 50K.
2.1.3 References
Brassfield, H.C., et al. 1968. Recommended Property and Reactor Kinetics Data for Use in
Evaluating a LightWaterCoolant Reactor LossofCoolant Incident Involving Zircaloy4 or 304
SSClad UO
2
, GEMP482, General Electric Company, Cincinnati, OH.
Lyon, W.F., and W.E. Baily. 1967. “The Solid Liquid Phase Diagram for the UO
2
PuO
2
System,” Journal of Nuclear Materials, 22:332.
Popov, S.G., J.J. Carbajo, V.K. Ivanov, and G.L. Yoder. 2000. Thermophysical Properties of
MOX and UO
2
Fuel Including the Effects for Irradiation, ORNL/TM 2000/351, Oak Ridge
National Laboratory, Oak Ridge, TN.
Adamson, M.G., E.A. Aitken, and R.W. Caputi. 1985. “Experimental and thermodynamic
evaluation of the melting behavior of irradiated oxide fuels.” Journal of Nuclear Materials,
130:349365.
2.4
Komatsu, J., T. Tachibana, and K. Konashi. 1988. “The Melting Temperature of Irradiated
Oxide Fuel,” Journal of Nuclear Materials, 154:3844.
2.2 Fuel Specific Heat Capacity (FCP) and Fuel Enthalpy (FENTHL)
The subroutines FCP and FENTHL are used to calculate the specific heat capacity and enthalpy
of the fuel pellet, respectively. The specific heat capacity and enthalpy of nuclear fuel are
modeled empirically as functions of four parameters: temperature, composition, molten fraction,
and oxygentometal (O/M) ratio. The same subroutine is used in FRAPCON3.4, FRAPTRAN
1.4, and MATPRO.
2.2.1 Model Development and Comparisons
Equations for the specific heat capacity and enthalpy of solid UO
2
and plutonium dioxide (PuO
2
)
are assumed to have the same form, but with different constants. The basic relationships are
presented in Equations 2.21 and 2.22, respectively.
(2.21)
(2.22)
where
FCP = specific heat capacity (J/kg*K)
FENTHL = fuel enthalpy (J/kg)
T = temperature (K)
Y = oxygentometal ratio
R = universal gas constant = 8.3143 (J/mol*K)
u = the Einstein temperature (K)
E
D
= activation energy for Frenkel defects (J/mol)
and the constants are given in Table 2.21.
 
) RT / E ( exp
RT 2
E YK
T K
1 ) T / ( exp T
) T / ( exp K
FCP
D
2
D 3
2
2 2
2
1
÷ + +
÷ u
u u
=
  ) RT / E ( exp K
2
Y
2
T K
1 ) T / ( exp
K
FENTHL
D 3
2
2 1
÷ + +
÷ u
u
=
2.5
Table 2.21 Constants used in UO
2
, Gd
2
O
3
, and PuO
2
heat capacity and enthalpy
correlations.
Constant UO
2
PuO
2
GdO
2
*
Units
K
1
296.7 347.4 315.86 J/kg*K
K
2
2.43 x 10
2
3.95 x 10
4
4.044 x 10
2
J/kg*K
2
K
3
8.745 x 10
7
3.860 x 10
7
0.0 J/kg
u 535.285 571.000 348.0 K
E
D
1.577 x 10
5
1.967 x 10
5
0.0 J/mol
* Gadolinia additions are only accounted for in FRAPCON3.4.
The constants in Table 2.21 were determined by Kerrisk and Clifton (1972) for UO
2
and Kruger
and Savage (1968) for PuO
2
. The specific heat capacity of UO
2
in the liquid state (see Equation
2.23) was determined by Leibotwitz (1971) and assumed to be valid for PuO
2
in the liquid state.
FCP
Liquid
= 503 J/kg*K (2.23)
Inspection of Equations 2.21 and 2.22 reveals that fuel enthalpy, or stored energy, is calculated
by integrating the fuel specific heat capacity correlation from 0 to T in kelvin. Stored energy is
important in reactor transient analysis because the severity of the transient is significantly
affected by the initial stored energy of the fuel. Since UO
2
and PuO
2
are the principal oxides in
lightwater reactor (LWR) fuels, they are the constituents considered.
The correlations for specific heat and enthalpy in both UO
2
and MOX fuel pellets are identical in
FRAPCON3.4, FRAPTRAN 1.4, and MATPRO. However, FRAPCON3.4 can accommodate
gadolinia additions whereas FRAPTRAN cannot. For a mixture of UO
2
, Gd
2
O
3
, and PuO
2
, the
specific heat capacity of the solid is determined by combining the contribution from each
constituent in proportion to its weight fraction. When the material is partially molten, the heat
capacity is determined similarly with a weighted sum of the solid and molten fractions.
Since the specific heat correlation is only valid above a fuel temperature of about 300K, the fuel
enthalpy correlation is not valid below a temperature of about 300K. In addition, it is necessary
to calculate fuel enthalpy with respect to a reference temperature > 300K. Thus, the fuel enthalpy
at any desired temperature, T, is calculated by evaluating Equation 2.22 at T and a reference
temperature, T
ref
, of 300K and taking the difference [FENTHL(T) – FENTHL(T
ref
)]. For
temperatures greater than 2K below melting, the molten fraction and heat of fusion are used to
interpolate between the enthalpy of unmelted fuel and justmelted fuel at the melting temperature.
Specific heat correlations are compared with independent experimental data in Figure 2.21 and
Figure 2.22 for UO
2
and PuO
2
, respectively. Comparison between the UO
2
model and
experimental data from Hein (1968), Leibowitz (1969), and Gronvold (1970) in Figure 2.21
reveals agreement between predicted and measured values up to high temperatures (a few
hundred kelvin below melting). At this point the data begin to fall lower than the model. This is
attributed to partial melting due to a nonuniform temperature distribution within the sample.
With the exception of the Affortit (1970) data, relatively good agreement is observed in Figure
2.22, which compares the values measured by Gibby (1974), Leibowitz (1972), and Affortit with
predicted heat capacities for a MOX fuel, (U
0.8
, Pu
0.2
)O
2
. Since the Affortit results are known to
be generally low in comparison to results from other investigators, the correlation is considered to
be in good agreement with the experimental data.
2.6
200
300
400
500
600
700
800
300 800 1300 1800 2300 2800 3300
Temperature (K)
S
p
e
c
i
f
i
c
H
e
a
t
C
a
p
a
c
i
t
y
(
J
/
k
g
*
K
)
Hein
Leibowitz
Gronvold
FRAP
Figure 2.21 Specific heat capacity of UO
2
from three experimenters compared with the
FRAPCON3/FRAPTRAN/MATPRO (FRAP) correlation (solid line) for
UO
2
.
2.7
200
250
300
350
400
450
500
550
600
650
700
300 800 1300 1800 2300 2800 3300
Temperature (K)
S
p
e
c
i
f
i
c
H
e
a
t
C
a
p
a
c
i
t
y
(
J
/
k
g
*
K
)
Gibby
Leibowitz
Affortit
FRAP
Figure 2.22 Specific heat capacity of (U
0.8
Pu
0.2
)O
2
from three experimenters compared
with the FRAPCON3/FRAPTRAN/MATPRO (FRAP) correlations (solid
line) for MOX.
2.2.2 Applicability and Uncertainty
The correlations for fuel specific heat and enthalpy are valid for temperatures from 300K to more
than 4000K. The standard error of the UO
2
specific heat capacity correlation is ± 3 J/kg*K.
However, this error increases at temperatures above 2300K, as seen in Figure 2.21, and may lead
to underpredictions and overpredictions of 59 and 66 J/kg*K, respectively. As for the mixed
oxidespecific heat capacity correlation, it is 6 to 10 J/kg*K, depending on the fraction of PuO
2
.
For nonstoichiometric fuels, these uncertainties are approximately double.
2.2.3 References
Kerrisk, J.F., and D.G. Clifton. 1972. “Smoothed Values for the Enthalpy and Heat Capacity of
UO
2,
” Nuclear Technology, 16:531535.
Kruger, O.L., and H. Savage. 1968. “Heat Capacity and Thermodynamic Properties of
Plutonium Dioxide,” Journal of Chemical Physics, 45:45404544.
Leibowitz, L., et al. 1971. “Enthalpy of Liquid Uranium Dioxide to 3500 K,” Journal of Nuclear
Materials, 39:115116.
Hein, R.A., L.H. Sjodahl, and R. Swarc. 1968. “Heat Content of Uranium Dioxide from 1,200 to
3,100 K,” Journal of Nuclear Materials, 25:99102.
2.8
Leibowitz, L., L.W. Mschler, and M.G. Chasanov. 1969. “Enthalpy of Solid Uranium Dioxide
from 2,500 K to its Melting Point,” Journal of Nuclear Materials, 29:356358.
Gronvold, F., et al. 1970. “Thermodynamics of the UO
2+x
Phase I. Heat Capacities of UO
2.017
and UO
2.254
from 300 to 1,000 K and Electronic Contributions,” Journal of Chemical
Thermodynamics, 2:665679.
Affortit, C., and J. Marcon. 1970. “Chaleur Specifique a Haute Temperature des Oxydes
D'Uranium et de Plutonium,” Revue Internationale des Hautes Temperatures et des Refractaires,
7:236241.
Gibby, R.L., et al. 1974. “Analytical Expressions for Enthalpy and Heat Capacity for Uranium
Plutonium Oxide,” Journal of Nuclear Materials, 50:155161.
Leibowitz, L., D.F. Fischer, and M.G. Chasanov. 1972. “Enthalpy of Uranium Plutonium
Oxides (U0.8,Pu0.2) O
1.07
from 2,350 to 3,000 K,” Journal of Nuclear Materials, 42:113116.
2.3 Fuel Thermal Conductivity (FTHCON)
The subroutine FTHCON is used to calculate the thermal conductivity of the fuel pellet. Since
accurate predictions of fuel rod behavior are strongly dependent on temperature, the calculation
of fuel thermal conductivity is critical. The FTHCON subroutine is used in FRAPCON3.4,
FRAPTRAN 1.4, and MATPRO. However, the calculation of fuel thermal conductivity has
evolved from the model originally proposed in MATPRO to the modified Nuclear Fuels
Industries (NFI) and Duriez/modified NFI models used in the FRAPCON3.4 and FRAPTRAN
1.4 codes for UO
2
, UO
2
Gd
2
O
3
, and MOX, respectively. An additional fuel thermal conductivity
model for MOX developed at Halden is included as an option in FRAPCON3.4. Each of the
models is discussed in detail in the following sections along with the evolution of the MATPRO
model and the applicability and uncertainty of the models.
Following the initial description of the MATPRO fuel thermal conductivity model, an overview
of subsequent thermal conductivity models is given. This historical summary of how thermal
conductivity modeling has evolved provides insight to model development and previous
motivations for change. For example, the MATPRO model was followed by a model by Lucuta
that accounted for burnup degradation and was incorporated in FRAPCON3. However, there
were several issues with the Lucuta model, including its nonstandard form and weak burnup
degradation term, that led to its replacement by a modified model developed at NFI. Further
refinements of this model were implemented by Pacific Northwest National Laboratory (PNNL)
to provide a better fit to UO
2
data. In addition, PNNL added refinements to include the Duriez
derived modifications to the NFI model, which permit MOX fuels to be modeled. These updates
were incorporated in FRAPCON3.3 and continue to be used in FRAPCON3.4.
2.3.1 Model Development and Comparisons
The original MATPRO model for fuel thermal conductivity is based on a mechanistic description
of thermal conductivity, including lattice vibration (i.e., phonon) and electronhole pair
(i.e., electronic) contributions (Hagrman et al., 1981). The MATPRO thermal conductivity model
is a function of temperature, density, O/M ratio, and plutonium content. Burnup is also included;
however, it is only used to calculate the melt temperature. Unlike the models used in
2.9
FRAPCON3.4 and FRAPTRAN 1.4, MATPRO does not account for the degradation of thermal
conductivity with increasing burnup.
The dominant contributor to thermal conductivity (i.e., phonon or electron) is dependent on
temperature. Phonon contributions to thermal conductivity are dominant below 1500K, whereas
electronic contributions are dominant above 2000K. In general, thermal conductivity decreases
with increasing porosity, but this effect is significantly decreased above 1600K due to radiation
effects and gas conductivity within the pores at these temperatures. At higher temperatures
(> 3000K), estimates of liquid fuel thermal conductivity are made from physical considerations
because data for molten fuel have not been found.
The MATPRO model used to express thermal conductivity of solid fuel is described with
Equation 2.31.
(2.31)
where
k = thermal conductivity (W/m*K).
D = fraction of theoretical density (unitless).
C
V
= phonon contribution to the specific heat at constant volume (J/kg*K). The first
term of the correlation for fuel specific heat capacity is used for this factor.
e
th
= linear strain caused by thermal expansion when temperature is > 300K (unitless);
the MATPRO correlation for fuel thermal expansion is used for this factor
T = fuel temperature (K).
T ' = porosity correction for temperature, for temperature < 1364, T ' = 6.50 –
T*(4.69x10
3
), for temperature > 1834, T ' = 1, and for temperatures in the range
from 1364 to 1834K, T ' is found by interpolation.
T '' = fuel temperature if < 1800K. For temperatures > 2300K, T'' is equal to 2050K;
for temperature in the range 1800 to 2300K, T'' is found by interpolation.
A = a factor proportional to the point defect contribution to the phonon mean free
path (m*s/kg*K); the correlation used for this factor is 0.339 + 12.6 x absolute
value (2.0 – O/M ratio).
B = a factor proportional to the phononphonon scattering contribution to the phonon
mean free path (m*s/kg*K); the correlation used for this factor is 0.06867 x (1+
0.6238 x plutonium content of the fuel).
The first term of Equation 2.31 represents the phonon contribution to specific heat. The
temperature and density dependence of this term can be seen in Equation 2.31. Although not as
evident, the dependence of the first term on O/M ratio and plutonium content is also included.
The specific heat capacity of the fuel (C
V
) and the phonon scattering factor (B) are dependent on
the plutonium content, while the point defect contribution factor (A) is dependent on the O/M
ratio. Temperature dependence can be observed in the first and second terms of Equation 2.31,
which represents the electronic contribution.
( ) ( ) ( )( )
( )   ( )   { }
2 3
th
V
2 T / 13358 169 . 0 1 T / 13358 exp T 10 x 2997 . 5
e 3 1 ' ' T B A
C
D 1 ' T 00469 . 0 5 . 6 1
D
k
+ + ÷ +


.

\

+ +


.

\

÷ ÷ +
=
÷
2.10
This MATPRO model (Hagrman et al., 1981) was initially replaced in FRAPCON3 and
FRAPTRAN with a model from Lucuta (Lucuta et al., 1996) which accounts for the degradation
of thermal conductivity due to burnup. The Lucuta model for fuel thermal conductivity, K, is
presented in Equation 2.32
(2.32)
where
K
o
= thermal conductivity of unirradiated, fully dense urania
FD = factor for dissolved fission products
FP = factor for precipitated fission products
FM = factor to correct for the Maxwell porosity effect
FR = factor for the radiation effect
Thermal conductivity of unirradiated, fully dense urania and factors included in the Lucuta model
are described by the Equations 2.33 through 2.37.
(2.33)
(2.34)
(2.35)
(2.36)
(2.37)
where
K
o
= conductivity of unirradiated, fully dense urania (W/mK)
T = temperature (K)
B = burnup in atom% (1 atom% = 9.383 GWd/MTU at 200 MeV/fission)
p = porosity fraction (asfabricated plus swelling)
s = shape factor ( = 1.5 for spherical pores)
Although the Lucuta model for thermal conductivity accounted for burnup degradation, this
model tended to overpredict thermal conductivity values at high temperature (> 2200K).
FR FM FP FD K K
o
    =
(
¸
(
¸
÷
(
¸
(
¸
+
+
=
÷
T
16361
exp
T
10 x 715 . 4
T 10 x 165 . 2 0375 . 0
1
K
2
9
4
o
(
(
(
(
¸
(
¸
+
(
¸
(
¸
+ =
T
B
0643 . 0
B
09 . 1
1
arctan T
B
0643 . 0
B
09 . 1
FD
265 . 3
265 . 3
(
(
(
¸
(
¸
÷
÷ +
(
¸
(
¸
÷
+ =
)
100
1200 T
exp( 1
1
B 019 . 0 3
B 019 . 0
1 FP
p ) 1 s ( 1
p 1
FM
÷ +
÷
=
)
80
900 T
exp( 1
2 . 0
1 FR
÷
+
÷ =
2.11
Furthermore, the model had too little burnup degradation, making it nonconservative above
30 GWd/MTU. In addition, the form of the Lucuta model is nonstandard, which does not
facilitate comparison between models. As a result, the Lucuta model was later replaced with a
modified version of a thermal conductivity model proposed by NFI (Ohira and Itagaki, 1997)
(Lanning et al., 2005).
Both FRAPCON3.4 and FRAPTRAN 1.4 currently model urania fuel pellet thermal conductivity
with the modified version of the pellet thermal conductivity model proposed by NFI (Ohira and
Itagaki, 1997). The original NFI model was modified to alter the temperaturedependent portion
of the burnup function in the phonon terms and change the electronic term (Lanning et al., 2005).
The original (unmodified) and modified versions of the NFI thermal conductivity model are
presented in Equations 2.38 and 2.39, respectively.
(2.38)
(2.39)
where
K = thermal conductivity, W/mK
T = temperature, K
Bu = burnup, GWd/MTU
f(Bu) = effect of fission products in crystal matrix (solution)
= 0.00187 * Bu
g(Bu) = effect of irradiation defects
= 0.038 * Bu
0.28
h(T) = temperature dependence of annealing on irradiation defects
=
) T / Q ( exp 396 1
1
÷ +
Q = temperaturedependent parameter (“Q/R”) = 6380K
A = 0.0452 mK/W
B = 2.46 x 10
4
mK/W/K
C = 5.47 x 10
9
W/mK
3
D = 2.29 x 10
14
W/mK
5
E = 3.5 x 10
9
WK/m
F = 16,361K
The NFI model was further modified in FRAPCON3.4 with a data correlation presented by
Massih (Massih et al., 1992) to include gadolinia content. The modified NFI with gadolinia
dependency used in FRAPCON3.4 is presented in Equation 2.310.
(2.310)
4 2
95
DT CT
) T ( h ) Bu ( g ) Bu ( f BT A
1
K + ÷
+ + +
=
) T / F exp(
T
E
) T ( h ) Bu ( g )) Bu 04 . 0 exp( 9 . 0 1 ( ) Bu ( f BT A
1
K
2
95
÷ +
÷ ÷ + + +
=
) T / F exp(
T
E
) T ( h ) Bu ( g )) Bu 04 . 0 exp( 9 . 0 1 ( ) Bu ( f BT gad a A
1
K
2
÷
+
÷ ÷ + + + · +
=
2.12
where
a = constant = 1.1599
gad = weight fraction of gadolinia
As applied in FRACON3.4 and FRAPTRAN 1.4, the above models are adjusted for asfabricated
fuel density (in fraction of theoretical density [TD]) using the Lucuta recommendation for
sphericalshaped pores (Lucuta et al., 1996), as follows:
(2.311)
where
K
d
= thermal conductivity adjusted for asfabricated fuel density, d
K
95
= thermal conductivity for 95 percent dense fuel
d = asfabricated fuel density
The phononterm modification in the modified NFI model reduces defect annealing at low burnup
from that by Ohira and Itagaki (NFI). However, the original temperaturedependent annealing is
restored at higher burnups such that, for burnups greater than 40 GWd/MTU, the phonon term is
equivalent to that in the original NFI model. The electronic terms (which in either case become
significant above 1500K) are altered in the modified NFI model to a more theoretically based
equation. The magnitude is slightly lower than the original model at high temperature. This
adjustment was indicated by the Institute for Transuranium Elements data on unirradiated
pressurizedwater reactor (PWR) pellet material at temperatures approaching fuel melting
(Ronchi et al., 1999).
At low burnups (< 20 GWd/MTU) and low temperatures (< 1000K), the modified model is higher
than the unmodified NFI model and roughly equivalent to the Lucuta model without its radiation
term. At higher burnup (> 30 GWd/MTU), the modified model is equivalent to the original NFI
model with the exception of the small reduction at very high temperatures.
Fuel thermal conductivity modeling was extended to include MOX in FRAPCON3 and
FRAPTRAN by adopting a model that was a combination of the Duriez stoichiometrydependent
correlation, which is derived from diffusivity measurements on unirradiated fuel pellets, and the
modified version of the NFI model, which includes burnup degradation effects. The Duriez data
places greater dependence of MOX thermal conductivity on stoichiometry, but only a minor
dependence on plutonia content (Duriez et al., 2000). The Duriez/modified NFI model is
presented in Equation 2.312.
(2.312)
where
K
95
= conductivity at 95 percent TD, W/mK
x = 2.00 – O/M (i.e., oxygentometal ratio)
T = temperature, K
A(x) = 2.85x + 0.035, mK/W
)}] d 1 ( 5 . 0 0 . 1 /{ d [ * K * 0789 . 1 K
95 d
÷ + =
)
T
D
exp(
T
C
) T ( h ) Bu ( g )) Bu 4 . 0 exp( 9 . 0 1 ( ) Bu ( f T ) x ( B gad a ) x ( A
1
K
2
mod
95
÷ +
÷ ÷ + + + · +
=
2.13
B(x) = (2.86 – 7.15x)*1E4 m/W
C = 1.689x10
9
, WK/m
D = 13,520K
a = 1.1599
gad = weight fraction gadolinia (not expected in MOX)
Bu = burnup in GWd/tHM
f(Bu) = effect of fission products in crystal matrix (solution)
= 0.00187*Bu
g(Bu) = effect of irradiation defects
= 0.038*Bu
0.28
h(T) = temperature dependence of annealing on irradiation defects
=
) T / Q ( exp 396 1
1
÷ +
Q = temperature dependent parameter (“Q/R”) = 6380K
C
mod
= 1.5x10
9
WK/m
FRAPCON 3 provides an option to use a fuel thermal conductivity model derived at Halden. The
Halden correlation does not include dependency on stoichiometry, but does emphasize burnup
effects. The Halden correlation tends to overpredict thermal conductivity data at high
temperatures (> 2500K) (Lanning et al., 2005).
(2.313)
where
K
95
= conductivity at 95 percent TD, W/mK
T = temperature, °C
a = 1.1599
gad = weight fraction gadolinia (not expected in MOX)
B = burnup in MWd/kg UO
2
u = minimum of 1650°C or current temperature in °C
To summarize, the modified NFI model is used for UO
2
, whereas the Duriez/NFI model is used
for MOX. The modified NFI model used in FRAPCON3.4 accounts for gad, but the modified
NFI model in FRAPTRAN 1.4 does not. The Duriez/NFI models account for gad in both codes,
although no gad additions are included in MOX at this time. Finally, FRAPCON3.4 has the
option for Halden thermal conductivity correlation, but FRAPTRAN does not.
A comparison between the MATPRO model and the modified NFI model is presented in Figure
2.3.31. The MATPRO model predicts lower thermal conductivities than the modified NFI
model across the applicable temperature range (i.e., 500 to 3000K) for unirradiated UO
2
.
However, unlike the MATPRO model, the modified NFI model can account for degradation of
thermal conductivity with increasing burnup. The modified NFI model at a burnup level of
30 GWd/MTU is shown for comparison and reveals significant degradation of thermal
conductivity. This represents an improvement relative to the MATPRO model, which does not
account for burnup degradation of thermal conductivity.
) T 00188 . 0 exp( 0132 . 0
) B 00333 . 0 1 ( 10 x 475 . 2 B 0040 . 0 x 1599 . 1 gad a 1148 . 0
92 . 0
K
4
95
+
u ÷ + + + · +
=
÷
2.14
0.00
1.00
2.00
3.00
4.00
5.00
6.00
7.00
8.00
9.00
10.00
0 500 1000 1500 2000 2500 3000
Temperature (K)
T
h
e
r
m
a
l
C
o
n
d
u
c
t
i
v
i
t
y
(
W
/
m
*
K
)
MATPRO
Modified NFI 
0 GWd/MTU
Modified NFI 
30GWdMTU
Figure 2.3.31 MATPRO and modified NFI thermal conductivity models for UO
2
as a
function of temperature. Unlike the MATPRO model, the modified NFI
model can account for burnup and is presented at burnup levels of 0 and
30 GWd/MTU for comparison.
Modeltodata comparisons for the MATPRO and modified NFI models and data collected from
unirradiated UO
2
are made in Figures 2.3.32 through 2.3.33. In addition to Ronchi (1999), the
unirradiated UO
2
data sets from that were used to derive the MATPRO model are also included.
Based on Figure 2.3.32, it can be seen that the MATPRO model tends to underpredict the
thermal conductivity of the unirradiated UO
2
. The average bias in Figure 2.3.32 underpredicts
thermal conductivity by about 9.5 percent, with a standard deviation of about 11 percent. The
modified NFI model also underpredicts the unirradiated data, but only by about 2.4 percent and
with a standard deviation of 8 percent (see Figure 2.3.33). Based on the smaller bias and
standard deviation, it can be seen that the modified NFI model predicts the thermal conductivity
of UO
2
much better than the MATPRO model.
2.15
MATPRO (UO
2
)
1
2
3
4
5
6
7
8
1 2 3 4 5 6 7 8
Measured Thermal Conductivity (W/m*K)
P
r
e
d
i
c
t
e
d
T
h
e
r
m
a
l
C
o
n
d
u
c
t
i
v
i
t
y
(
W
/
m
*
K
)
Predicted = Measured Christensen Godfrey
Bates Gibby Weilbacher
Goldsmith and Douglas Hobson Ronchi
Figure 2.3.32 MATPRO modeltodata comparison for thermal conductivity of unirradiated
UO
2
.
Modified NFI (UO
2
)
1
2
3
4
5
6
7
8
1 2 3 4 5 6 7 8
Measured Thermal Conductivity (W/m*K)
P
r
e
d
i
c
t
e
d
T
h
e
r
m
a
l
C
o
n
d
u
c
t
i
v
i
t
y
(
W
/
m
*
K
)
Predicted = Measured Christensen Godfrey
Bates Gibby Weilbacher
Goldsmith and Douglas Hobson Ronchi
Figure 2.3.33 Modified NFI modeltodata comparison for thermal conductivity of
unirradiated UO
2
.
2.16
Modeltodata comparisons for the MATPRO and modified NFI models and data collected from
irradiated UO
2
are made in Figures 2.3.34 through 2.3.35. Based on Figure 2.3.34, it can be
seen that the MATPRO model predicts significantly higher values for thermal conductivity than
the measured values obtained from irradiated UO
2
. The average bias in Figure 2.3.34 indicates
that the thermal conductivity is overpredicted by about 40 percent, with a standard deviation of
about 31 percent. The modified NFI model also tends to overpredict the small amount of
irradiated data on average, but only by about 5 percent and with a standard deviation of about
8 percent (see Figure 2.3.35). Based on the significantly smaller bias and standard deviation, it
can be seen that the modified NFI model predicts the thermal conductivity of irradiated UO
2
better than the MATPRO model. The disparity between models is attributed to the degradation of
thermal conductivity with increasing burnup, which is accounted for in the modified NFI model
but not the MATPRO model.
1
2
3
4
5
1 2 3 4 5
P
r
e
d
i
c
t
e
d
T
h
e
r
m
a
l
C
o
n
d
u
c
t
i
v
i
t
y
(
W
/
m
*
K
)
Measured Thermal Conductivity (W/m*K)
MATPRO (UO
2
)
Predicted = Measured Ronchi 3035 GWd/MTU Carrol 40 GWd/MTU
Ronchi 5055 GWd/MTU Ronchi 82 GWd/MTU Ronchi 92 GWd/MTU
Ronchi 96 GWd/MTU
Figure 2.3.34 MATPRO modeltodata comparison for thermal conductivity of irradiated
UO
2
.
2.17
Modified NFI (UO
2
)
1
2
3
4
5
1 2 3 4 5
Measured Thermal Conductivity (W/m*K)
P
r
e
d
i
c
t
e
d
T
h
e
r
m
a
l
C
o
n
d
u
c
t
i
v
i
t
y
(
W
/
m
*
K
)
Predicted = Measured Ronch 3035 GWd/MTU Carrol 40 GWd/MTU
Ronchi 5055 GWd/MTU Ronchi 82 GWd/MTU Ronchi 92 GWd/MTU
Ronchi 96 GWd/MTU
Figure 2.3.35 Modified NFI modeltodata comparison for thermal conductivity of
irradiated UO
2
.
A comparison between the MATPRO model and the Duriez/modified NFI model is presented in
Figure 2.3.36 for MOX fuel with a plutonium concentration of 7 wt%. The MATPRO model
predicts lower thermal conductivities than the Duriez/modified NFI model across the applicable
temperature range (i.e., 500 to 3000K). However, unlike the MATPRO model, the
Duriez/modified NFI model can account for degradation of thermal conductivity with increasing
burnup. The Duriez/modified NFI model at a burnup level of 30 GWd/MTU is shown for
comparison and reveals significant degradation of thermal conductivity. This represents an
improvement relative to the MATPRO model, which does not account for burnup degradation of
thermal conductivity.
2.18
0.00
1.00
2.00
3.00
4.00
5.00
6.00
7.00
8.00
9.00
10.00
0 500 1000 1500 2000 2500 3000
Temperature (K)
T
h
e
r
m
a
l
C
o
n
d
u
c
t
i
v
i
t
y
(
W
/
m
*
K
)
MATPROMOX
DuriezModified NFI
DuriezModified NFI
30GWd/MTU
Figure 2.3.36 MATPRO and Duriez/modified NFI thermal conductivity models as a
function of temperature for MOX fuel with 7 wt% plutonium content.
Unlike the MATPRO model, the Duriez/modified NFI model can account for
burnup and is presented at burnup levels of 0 and 30 GWd/MTU for
comparison.
Modeltodata comparisons for the MATPRO and Duriez/modified NFI models and unirradiated
MOX fuel are made in Figures 2.3.37 and 2.3.38. The comparisons utilize the data used to
derive the Duriez/modified NFI model. Data collected from MOX with plutonium concentrations
of 3 and 15 wt% were used (Duriez et al., 2000). Based on Figure 2.3.37, it can be seen that the
MATPRO model tends to underpredict thermal conductivity. The average bias in Figure 2.3.37
underpredicts thermal conductivity by about 9 percent at both 3 and 15 wt% plutonium
concentrations. The standard deviation of the MATPRO predictive bias at both plutonium
concentrations is under 6 percent. As expected, the Duriez/modified NFI model exhibits better
agreement with the limited experimental data from which it was derived. The average bias
indicates that the MOX model overpredicts by less than 2 percent and the standard deviation
between the model and the data is less than 3 percent. Based on the smaller bias and standard
deviation, it can be seen that the Duriez/modified NFI model fits this MOX thermal conductivity
data better than the MATPRO model.
It should be noted that the MOX fuel thermal conductivity is strongly influenced by O/M ratio,
with conductivity decreasing with a deviation from stoichiometry (2.0). Based on Figure 2.3.38,
it can be seen that plutonium content has a relatively minor effect at concentrations less than
15 wt%. The current MOX fuel is fabricated closer to being stoichiometric than fuel typically
fabricated more than 25 years ago. The older data from which the MATPRO model was
developed were not as close to being stoichiometric as the more recent data.
2.19
MATPRO (MOX)
1
2
3
4
5
1 2 3 4 5
Measured Thermal Conductivity (W/m*K)
P
r
e
d
i
c
t
e
d
T
h
e
r
m
a
l
C
o
n
d
u
c
t
i
v
i
t
y
(
W
/
m
*
K
)
Predicted = Measured 3 wt% Pu 15 wt% Pu
Figure 2.3.37 MATPRO modeltodata comparison for thermal conductivity of MOX
modeled with 7 wt% of plutonium.
DuriezModified NFI (MOX)
1
2
3
4
5
1 2 3 4 5
Measured Thermal Conductivity (W/m*K)
P
r
e
d
i
c
t
e
d
T
h
e
r
m
a
l
C
o
n
d
u
c
t
i
v
i
t
y
(
W
/
m
*
K
)
Predicted = Measured 3 wt% Pu 15 wt% Pu
Figure 2.3.38 Duriez/modified NFI modeltodata comparison for thermal conductivity of
MOX with modeled with 7 wt% of plutonium.
2.20
2.3.2 Applicability and Uncertainty
The following are limits that restrict the applicability of the MATPRO fuel thermal conductivity
model (Equation 2.31):
 Theoretical density must range between 90 and 100 percent.
 Temperature must range between 500 (the Debye temperature) and 3000K.
 Above 3000K, it is assumed that the fuel becomes molten and the uncertainties on predictions
made above this temperature are large.
In addition, several assumptions have been made to provide an approximate model for effects of
variation in the plutonium content and O/M ratio of ceramic fuels:
 The effect of variation in theoretical density for MOX fuels has been assumed to be described
by the porosity correction derived with UO
2
data.
 The hightemperature electronic contribution to thermal conductivity has been assumed to be
the same for PuO
2
, UO
2
, UO
2
Gd
2
O
3
, and nonstoichiometric fuels.
 Variations in plutonium content have been assumed to affect only the phononphonon
scattering factor.
 Variation in O/M ratio has been assumed to affect only the defect term.
The bias and relative error of the MATPRO fuel thermal conductivity model was assessed based
on the modeltodata comparisons previously presented for unirradiated UO
2
, irradiated UO
2
, and
MOX. The average and standard deviations of the bias for unirradiated UO
2
are approximately
9.5 and 11 percent, respectively, while the relative error is approximately 19 percent. For
irradiated UO
2
, which has a limited number of data points for comparison, the average and
standard deviations of the bias are 40 and 31 percent, respectively. The relative error for the
limited number of irradiated UO
2
data points is approximately 51 percent. Finally, for MOX, the
average and standard deviations of the bias are 9 and 6 percent, respectively, while the relative
error is approximately 13 percent. For reference, a negative bias indicates that the model
underpredicts the data, whereas a positive bias indicates that the model overpredicts the data
For the modified NFI and the Duriez/modified NFI models, the recommended ranges for
application for modeling the thermal conductivity of UO
2
are as follows:
 Temperature: 300 to 3000K
 Rodaverage burnup: 0 to 62 GWd/MTU
 Asfabricated density: 92 to 97 percent TD
 Gadolinia content: 0 to 10 wt%
For the Duriez/modified NFI model, the thermal conductivity model is applicable for plutonia
particle sizes of less than 20 microns.
The bias and relative error of the modified NFI and the Duriez/modified NFI fuel thermal
conductivity models were assessed based on the modeltodata comparisons previously presented
for unirradiated UO
2
, irradiated UO
2
, and MOX. The average and standard deviations of the bias
for unirradiated UO
2
are approximately 2.4 and 8 percent, respectively, while the relative error is
approximately 9 percent. For irradiated UO
2
, which has a limited number of data points for
2.21
comparison, the average and standard deviations of the bias are approximately 5 and 8 percent,
respectively, while the relative error is approximately 10 percent. Finally, for the
Duriez/modified NFI fuel thermal conductivity model for MOX, the average and standard
deviations of the bias are less than 2 and 3 percent, respectively, while the relative error is
approximately 3 percent. These are values are based on comparisons to the small amount of near
stoichiometric MOX data. For reference, a negative bias indicates that the model underpredicts
the data, whereas a positive bias indicates that the model overpredicts the data
2.3.3 References
Hagrman, D.L., G.A. Reymann, and G.E. Mason. 1981. A Handbook of Materials Properties for
Use in the Analysis of Light Water Reactor Fuel Rod Behavior, MATPRO Version 11 (Revision
2), NUREG/CR0479 (TREE1280), prepared by EG&G Idaho, Inc., Idaho Falls, ID, for the U.S.
Nuclear Regulatory Commission, Washington, D.C.
Lucuta, P.G., H.S. Matzke, and I.J. Hastings. 1996. “A Pragmatic Approach to Modeling
Thermal Conductivity of Irradiated UO
2
Fuel: Review and Recommendations,” Journal of
Nuclear Materials, 232:166180.
Ohira, K., and N. Itagaki. 1997. “Thermal Conductivity Measurements of High Burnup UO
2
Pellet and a Benchmark Calculation of Fuel Center Temperature,” in Proceedings of the ANS
International Topical Meeting on LWR Fuel Performance, Portland, Oregon, March 26, 1997,
pp. 541549.
Lanning, D.D., C.E. Beyer, and K.G. Geelhood. 2005. FRAPCON3 Updates, Including Mixed
Oxide Fuel Properties, NUREG/CR6534, Vol. 4, PNNL11513, Pacific Northwest National
Laboratory, Richland, WA.
Massih, A.R., S. Persson, and Z. Weiss. 1992. “Modeling of (U,Gd)O
2
Fuel Behavior in Boiling
Water Reactors,” in Proceedings of Symposium E on Nuclear Materials for Fission Reactors of
the 1991 EMRS Fall Conference; Journal of Nuclear Materials, 188:323330.
Ronchi, C., M. Sheindlin, M. Musella, and G.J. Hyland. 1999. “Thermal Conductivity of
Uranium Dioxide Up to 2900K from Simultaneous Measurement of the Heat Capacity and
Thermal Diffusivity,” Journal of Applied Physics, 85(2):776789.
Duriez, C., J.P. Allesandri, T. Gervais, and Y. Philipponnea. 2000. “Thermal Conductivity of
Hypostoichiometric Low Pu Content (U,Pu)O
2x
Mixed Oxide,” Journal of Nuclear Materials,
277:143158.
Bates, J.L., C.A. Hinman, and T. Kawada. 1967. “Electrical Conductivity of Uranium Dioxide,”
Journal of the American Ceramic Society, 50:652656.
Goldsmith, L.A., and J.A.M. Douglas. 1973. “Measurements of the Thermal Conductivity of
Uranium Dioxide at 6701270 K,” Journal of Nuclear Materials, 47:3142.
Christensen, J.A. et al. 1964. “Uranium Dioxide Thermal Conductivity,” Transactions of the
American Nuclear Society, 7:391392.
2.22
Gibby, R.L. 1971. “The Effect of Plutonium Content on the Thermal Conductivity of (U,Pu)O
2
Solid Solutions,” Journal of Nuclear Materials, 38:163177.
Hobson, I.C., R. Taylor, and J.B. Ainscough. 1974. “Effect of Porosity and Stoichiometry on the
Thermal Conductivity of Uranium Dioxide,” Journal of Physics, Section D: Applied Physics,
7:10031015.
Godfrey, T.G., et al. 1964. Thermal Conductivity of Uranium Dioxide and Armco Iron by an
Improved Radial Heat Flow Technique, ORNL3556, Oak Ridge National Laboratory, Oak
Ridge, TN.
Weilbacher, J.C. 1972. “Diffusivite Thermique de l’Oxyde d’Uranium et de l’Oxyde de
Thorium a Haute Temperature,” High TemperaturesHigh Pressure, 4:431438.
Ronchi, C., M. Sheindlin, D. Staicu, and M. Kinoshita. 2004. “Effect of Burnup on the Thermal
Conductivity of Uranium Dioxide up to 100,000 MWdt
1
,” Journal of Nuclear Materials, 327:58
76.
2.4 Fuel Emissivity (FEMISS)
The subroutine FEMISS is used to calculate the total hemispherical
fuel emissivity (emissivity
integrated over all wavelengths) as a function of temperature. Fuel emissivity is defined as the
ratio of radiant energy emitted from a material to that emitted by a black body at the same
temperature. The subroutine FEMISS is used to calculate radiant energy transfer from fuel to
cladding in conjunction with thermal conduction. Radiant energy transfer can be a significant
heat transfer mechanism, depending on the gap size, temperature gradient across the gap, and
plenum gas. The FEMISS subroutine used by FRAPCON3 and FRAPTRAN is the same as the
subroutine documented in MATPRO.
2.4.1 Model Development and Comparisons
According to the StefanBoltzmann law, the total radiant power per unit area emitted by a body at
temperature T is
(2.41)
where
P = radiant power per unit area (W/m
2
)
e = total hemispherical emissivity (unitless)
o = the StefanBoltzmann constant (5.672 x 10
8
W/m
2
K)
T = temperature (K)
The expression used in the FEMISS subroutine to describe total emissivity is
(2.42)
The first term of Equation 2.42 is slightly less than the value used in MATPRO, 0.7856. The
standard error of Equation 2.42 with respect to its database is ± 6.796 percent, which is in
4
T e P o =
T 10 x 5263 . 1 78557 . 0 e
5 ÷
+ =
2.23
agreement with the error reported in MATPRO, ± 6.8 percent. These small differences are
attributed to rounding error and do not significantly alter emissivity predictions.
The emissivity data were measured at temperatures up to approximately 2400K, and use of
FEMISS above this temperature is speculative because of possible hightemperature effects that
are not modeled. At the time of model development, there were no data to develop a MOX
emissivity equation, so Equation 2.42 is also recommended for MOX.
The hemispherical spectral data of Held and Wilder (1969) and the emissivity data of Cabannes
(1967) and J.M. Jones and D.G. Murchison (1965) were used in developing the FEMISS model.
The emissivity data used to model comparison for FEMISS are presented in Figure 2.41.
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
300 800 1300 1800 2300 2800
Temperature (K)
E
m
i
s
s
i
v
i
t
y
(
u
n
i
t
l
e
s
s
)
Held and Wilder
Cabannes et al.
FEMISS
Standard Error
Figure 2.41 Data to model comparison for FEMISS.
2.4.2 Applicability and Uncertainty
The correlation used in the FEMISS subroutine to calculate fuel emissivity is applicable at
temperatures up to 2400K. The standard error of this correlation with respect to the database
from which it was derived is approximately ± 6.8 percent
2.4.3 References
Held, P.C., and D.R. Wilder. 1969. “High Temperature Hemispherical Spectral Emittance of
Uranium Oxides at 0.65 and 0.70 µm,” Journal of the American Ceramic Society, 52(4):182185.
M.M.F. Cabannes, J.P. Stora, and J. Tsakiris. 1967. “ORTIQUEMOLECULAIREFracteurs de
re’flexion et d’e’mission de UO
2
a’ haute Temperature,” C.R. Acad. SC. Paris, 264.
J.M. Jones and D.G. Murchison. 1965. “Optical Properties of Uranium Oxides,” Nature,
205(4972):663665.
2.24
2.5 Fuel Thermal Expansion (FTHEXP)
The subroutine FTHEXP models dimensional changes in unirradiated fuel pellets caused by
thermal expansion. It can handle any combination of UO
2
, UO
2
Gd
2
O
3
, or PuO
2
in solid, liquid,
or solidliquid states and includes expansion due to the solidliquid phase change. Dimensional
changes in the fuel affect the pellettocladding gap size, which is a major factor in determining
gap heat transfer and thus the stored energy, an important quantity for safety analysis. The
subroutine FTHEXP is used in FRAPCON3.4, FRAPTRAN 1.4, and MATPRO. However, the
FRAPCON3 and FRAPTRAN codes use updated fitting coefficients based on more recent data.
2.5.1 Model Development and Comparisons
The FTHEXP subcode models fuel thermal expansion as a function of temperature, fraction of
PuO
2
, and the fraction of fuel which is molten. The O/M ratio is not included. When the
departure from stoichiometry (O/M – 2.0) is greater than 0.2, there is clearly an effect. However,
this effect is ignored in modeling thermal expansion, since typical reactor fuels only deviate less
than a tenth this much from the stoichiometric composition.
The equations for thermal expansion of UO
2
, UO
2
Gd
2
O
3
, and PuO
2
have the same form. In the
solid phase, Equation 2.51 is used.
(2.51)
where
AL/L = linear strain caused by thermal expansion (equal to zero at 300K) (unitless)
T = temperature (K)
E
D
= energy of formation of a defect (J)
k = Boltzmann’s constant (1.38 x 10
23
J/K)
and K
1
, K
2
, and K
3
are constants to be determined.
The correlation given in Equation 2.51 was originally used in MATPRO and was incorporated
into both the FRAPCON3 and FRAPTRAN codes. Data collected by Baldock et al. (1966),
Burdick and Paker (1956), Gronvold (1955), Christensen (1963), and Kempeter and Elliott (1958)
were used to determine the correlation constants for UO
2
used in MATPRO. However, newer
data provided by Martin (1988) and Momin et al. (1991) required the constants to be updated to
improve the fit between the correlation and hightemperature data. These updated constants are
included in FRAPCON3.4/FRAPTRAN 1.4 (FRAP). Table 2.51 presents the original
MATPRO constants and the updated constants for UO
2
, as well as the constants for PuO
2
.
Table 2.51 Parameters used in UO
2
and PuO
2
solidphase thermal expansion
correlations.
Constant
MATPRO
UO
2
FRAP
UO
2
PuO
2
Units
K
1
1.0 x 10
5
9.80 x 10
6
9.0 x 10
6
K
1
K
2
3.0 x 10
3
2.61 x 10
3
2.7 x 10
3
Unitless
K
3
4.0 x 10
2
3.16 x 10
1
7.0 x 10
2
Unitless
E
D
6.9 x 10
20
1.32 x 10
19
7.0 x 10
20
J
) kT / E exp( K K T K L / L
D 3 2 1 0
÷ + ÷ = A
2.25
For mixed UO
2
and PuO
2
, the thermal expansion of the solid is found by combining the
contribution from each constituent in proportion to its weight fraction.
During melting, an expansion equal to a linear strain of 0.043 occurs. If the fuel is partially
molten, the strain due to thermal expansion is given by Equation 2.52.
(2.52)
where
AL/L
0
(T
m
) = thermal expansion strain of solid fuel from equations with T = T
m
T
m
= melting temperature of the fuel (K)
FACMOT = fraction of the fuel which is molten (unitless)
If FACMOT = 0.0, the fuel is all solid;
If FACMOT = 1.0, the fuel is all molten
The correlation used to describe the expansion of entirely molten fuel is given by Equation 2.53.
(2.53)
The solidtoliquid phase transition is isothermal only for pure UO
2
or pure PuO
2
. For MOX, the
transition occurs over a finite temperature range, denoted in Equation 2.53 by AT
m
.
Comparisons between the databases for UO
2
and PuO
2
with their respective correlations are
presented in Figures 2.51 and 2.52, respectively. In Figure 2.51, it can be seen that the updated
constants used in FRAPCON3.4/FRAPTRAN 1.4 (FRAP) for UO
2
provide a better fit to the
hightemperature data than the previous constants used in MATPRO. The comparison in Figure
2.52 reveals good agreement between the correlation and the database.
0.000
0.005
0.010
0.015
0.020
0.025
0.030
0.035
0.040
0.045
0.050
0 500 1000 1500 2000 2500 3000
Temperature (K)
F
r
a
c
t
i
o
n
a
l
L
e
n
g
t
h
C
h
a
n
g
e
(
A
L
/
L
)
Baldock Burdick Gronvold
Other MATPRO References Martin References MATPRO UO2
FRAP UO2
Figure 2.51 Modeltodata comparison for UO
2
correlation used in MATPRO and
FRAPCON3/FRAPTRAN with updated constants.
FACMOT 043 . 0 ) T ( L / L L / L
m 0 0
· + A = A
( )  
m m
5
m 0 0
T T T 10 x 6 . 3 043 . 0 ) T ( L / L L / L A + ÷ + + A = A
÷
2.26
0.000
0.002
0.004
0.006
0.008
0.010
0.012
0.014
0.016
0 500 1000 1500 2000
Temperature (K)
F
r
a
c
t
i
o
n
a
l
L
e
n
g
t
h
C
h
a
n
g
e
(
A
L
/
L
)
FRAP and MATPRO PuO2 Tokar and Brett References
Figure 2.52 Modeltodata comparison for PuO
2
correlation used in MATPRO and
FRAPCON3/FRAPTRAN.
2.5.2 Applicability and Uncertainty
The fuel thermal expansion correlation used in the FTHEXP subroutine is applicable through the
melting temperature of the fuel (3111.15K for UO
2
). The uncertainty of the pooled data was
found to be temperature dependent, increasing approximately linearly with temperature.
Therefore, a percentage error is given rather than a fixed number. The ± o limits were found to
be within ±10 percent of the calculated value.
2.5.3 References
Baldock, P.J., et al. 1966. “The Xray Thermal Expansion of NearStoichiometric UO
2
,” Journal
of Nuclear Materials, 18:305313.
Burdick, M.D., and H.S. Parker. 1956. “Effect of Particle Size on Bulk Density and Strength
Properties of Uranium Dioxide Specimens,” Journal of the American Ceramic Society, 39:181
187.
Gronvold, F. 1955. “High Temperature Xray Study of Uranium Oxides in the UO
2
U
3
O
8
Region,” Journal of Inorganic and Nuclear Chemistry, 1:357370.
J.A. Christensen. 1963. “Thermal Expansion and Change in Volume of Uranium Dioxide on
Melting” Journal of the American Ceramic Society, 46:607608.
Kempter, C.P., and R.E. Elliott. 1958. “Thermal Expansion of UN, UO
2
, UO
2
ThO
2
, and ThO
2
,”
Journal of Chemical Physics, 30:15241526.
2.27
Martin, D.G. 1988. “The Thermal Expansion of Solid UO
2
and (U,Pu) Mixed Oxides – A
Review and Recommendations,” Journal of Nuclear Materials, 152:94101.
Momin, A.C., E.B. Mirza, and M.D. Mathews. 1991. “High Temperature XRay
Diffacractometric Studies on the Lattice Thermal Expansion Behavior of UO
2
, ThO
2
, and
(U
0.2
Th
0.8
)O
2
Doped with Fission Products,” Journal of Nuclear Materials, 185:308310.
Tokar, M., and A.W. Nutt. 1972. “Thermal Expansion of PuO
2
from 25 to 1420°C,”
Transactions of the American Nuclear Society, 10:210211.
Brett, N.H., and L.E. Russel. 1960. “The Thermal Expansion of PuO
2
and Some Other Actinide
Oxides Between Room Temperature an 1000°C,” in Proceedings of the Second International
Conference on Plutonium Metallurgy, Grenoble, France, April 1922, 1960, pp. 397410.
2.6 Fuel Densification (FUDENS)
The subroutine FUDENS calculates fuel dimensional changes due to densification of UO
2
and
MOX during the first few thousand hours of water reactor operation. If available, data from
resintering tests should be input by the user to determine densification (Regulatory Guide 1.126
[NRC 1978]). Otherwise, the subcode FUDENS calculates densification as a function of fuel
burnup, temperature, and initial density.
The data used to develop FUDENS were taken from irradiated fuel which also included swelling
(Rolstad, 1974). If fuel densification is much greater than swelling during the first 1000 hours of
irradiation, then, to a first approximation, swelling can be neglected during this period. This was
done during the development of the FUDENS model.
The FUDENS subroutine is used in FRAPCON3.4 and is similar to the correlation described by
MATPRO. However, FUDENS is not included in FRAPTRAN because FRAPTRAN is intended
for transient events occurring over a short time scale. Since densification occurs over longer time
scales, it is not included in FRAPTRAN.
2.6.1 Model Development and Comparisons
The subroutine FUDENS uses one of two methods to calculate the maximum density change
during irradiation. The RSNTR method uses the density change observed during resintering tests
(1973K for 24 hours based on Regulatory Guide 1.126 [NRC 1978]) in a laboratory furnace and
is the preferred input for the calculation. When RSNTR > 0, MATPRO uses either
Equation 2.61 or 2.62 and 2.66 while FRAPCON3.4 uses Equations 2.63 and 2.66. If a
resintering density change is not input (i.e., RSNTR = 0), the TSINT method uses the initial
unirradiated density of the fuel and the fuel fabrication sintering temperature and burnup for
density calculations. For TSNTR calculations, both MATPRO and FRAPCON3.4 use either
Equation 2.64 or 2.65 and 2.66. Both the RSNTR and TSNTR methods for MATPRO and
FRAPCON3.4 are described by the following equations to calculate the maximum densification
length change during irradiation.
In the MATPRO version of FUDENS, if a nonzero value for the resintering density change is
input, then either Equation 2.61 or Equation 2.62 is used, depending on the fuel temperature,
FTEMP (Siefken et al., 2001).
2.28
For FTEMP < 1000K (2.61)
For FTEMP > 1000K (2.62)
In FRAPCON3.4, if the resintering density is a nonzero input, then the following equation is
used to calculate the maximum dimensional change.
(2.63)
If zero is input for the resintering density change, then the MATPRO and FRAPCON3.4 models
use either Equation 2.64 or Equation 2.65, depending on the fuel temperature, FTEMP (Rolstad
et al., 1974).
For FTEMP < 1000K (2.64)
For FTEMP > 1000K (2.65)
where
m
L
L

.

\
 A
= maximum possible dimension change of fuel due to irradiation (percent)
RSNTR = resintered fuel density change (kg/m
3
)
FTEMP = fuel temperature (K)
DENS = initial density (percent theoretical)
TSINT = sintering temperature (K)
FDENS = starting density (kg/m
3
)
Densification as a function of burnup is calculated using
(2.66)
where

.

\
 A
L
L
= dimension change (percent)
FBU = fuel burnup (MWd/kgU)
( )RSNTR 0015 . 0
L
L
m
÷ =

.

\
 A
( )RSNTR 00285 . 0
L
L
m
÷ =

.

\
 A
( )
( ) 1453 TSINT
DENS 100 2 . 22
L
L
m
÷
÷ ÷
=

.

\
 A
( )
( ) 1453 TSINT
DENS 100 6 . 66
L
L
m
÷
÷ ÷
=

.

\
 A
( )   ( )  
( )
B FBU 35 B FBU 3
m
e 0 . 2 e
L
L
L
L
+ ÷ + ÷
+ +

.

\
 A
=
A
) FDENS * 0 . 3 /( RSNTR * 100
L
L
m
=

.

\
 A
2.29
B = a constant determined by the subcode to fit the boundary condition: AL/L
= 0 when FBU = 0
The FUDENS subcode presented in Equation 2.66 is used to calculate total densification, and
then the densification from the previous time step is subtracted to obtain the incremental
densification. The incremental densification for the time step being considered is the output of
the subcode FUDENS and is determined by Newton’s method (Hamming, 1971).
The relationship between densification and burnup in Equation 2.66 was originally suggested by
Rolstad et al. (1974). This relationship was adopted for use in the FUDENS subroutine because it
successfully describes the burnup dependence of both the original data (Rolstad et al., 1974) and
EPRI data (Freshley et al., 1976) that was collected shortly thereafter. Since these initial studies,
there has been additional inreactor densification data collected for UO
2
and MOX fuels
irradiated in thermal reactors.
Banks (1974) reported densification data on solid and annular UO
2
pellets irradiated in the
Winfrith Steam Generating Heavy Water Reactor and the Halden Boiling Water Reactor. The
solid pellets experienced centerline temperatures of 733 to 2443K over the burnup range 0.37 to
23.7 MWd/kgU. They concluded that pellets with centerline temperatures less than 1423K
densified during irradiation, but at higher temperatures dimensional change was affected by
intergranular fission gas swelling. Therefore, only data for temperatures less than 1423K were
included in comparison with the FUDENS model (Equations 2.64, 2.65, and 2.66). The
mechanism for densification was determined to be the loss of pores with initial diameters less
than 2 µm.
Freshley et al. (1978) reported densification data on MOX pellets irradiated in the radially
adjustable facility tubes of the General Electric Test Reactor. Two pins were irradiated with fuel
centerline temperatures up to 2048K and burnup to 4.6 MWd/kgM. The pellets were fabricated
with sintering temperatures from 1748 to 2008K and had starting density values ranging from 91
to 95 percent TD. They concluded that dilute additions of PuO
2
in UO
2
(up to 6 wt% PuO
2
) do
not change the densification behavior of the fuel. The mechanism of densification was observed
to be sintering of pores with initial diameters of 0.5 µm or smaller.
Small (1987) irradiated four types of UO
2
pellets in the Harwell Materials Test Reactor. Pellet
centerline temperatures ranged from 623 to 1823K and burnup ranged from 0.0025 to
2.6 MWd/kgU. Three of the four types of pellets were fabricated with pores smaller than 2 µm in
diameter to more carefully examine the mechanisms behind densification. For reference, fuel
currently fabricated has very few pores less than 2 µm in diameter. Densification was determined
by measuring porosity before and after irradiation. At centerline temperatures above 1273K,
densification was caused by sintering of pores less than 1 µm in diameter.
Figure 2.61 compares measured and predicted values of densification using the TSNTR
correlation, which is used in both FRAPCON3.4 and MATPRO and incorporates Equations
2.64, 2.65, and 2.66.
2.30
1.00
0.00
1.00
2.00
3.00
4.00
5.00
6.00
7.00
8.00
1.00 0.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00 8.00
Experimental Densification (% TD)
P
r
e
d
i
c
t
e
d
D
e
n
s
i
f
i
c
a
t
i
o
n
(
%
T
D
)
Banks
Freshley
Small
Freshley 1976
Figure 2.61 Comparison of densification predictions generated by the FUDENS model
and experimental densification measurements determined in reactor.
This data comparison demonstrates that basing densification on the sintering temperature
provides a large degree of uncertainty. However, it should be noted that the Small (1987) data,
which exhibited very high densification at very low burnup (less than 0.07 MWd/kgU), possessed
a nonprototypic pore size distribution that was heavily weighted toward smaller diameters.
Three of the four fuel types had 100 percent of their porosity in pores smaller than 2 µm diameter,
which is atypical of current fuel fabrication.
Since inreactor densification is strongly influenced by fuel fabrication, exreactor resintering
tests are the preferred method for predicting inreactor fuel densification (NRC, 1978).
Exreactor resintering tests conducted by Freshly et al. (1976 and 1978) at 1600 and 1700°C
provided reasonable estimate of inreactor densification tests with standard deviations of 0.5 to
0.6 percent of the theoretical density.
2.6.2 Applicability and Uncertainty
The densification correlation used in the subroutine FUDENS is applicable to fuels with pore size
distributions that are similar to those included in the Freshley et al. (1976) study. Due to the
scatter in the experimental data, it is difficult to establish a meaningful measure of uncertainty.
2.6.3 References
Rolstad, E., et al. 1974. “Measurements of the Length Changes of UO
2
Fuel Pellets During
Irradiation,” Enlarged HPG Meeting on Computer Control and Fuel Research, June 47, 1974.
2.31
NRC – U.S. Nuclear Regulatory Commission. 1978. An Acceptable Model and Related
Statistical Methods for the Analysis of Fuel Densification. Regulatory Guide 1.126, Rev.1, U.S.
Nuclear Regulatory Commission, Washington, D.C.
Siefken, L.J., E.W. Coryell, E.A. Harvego, J.K. Hohorst. 2001. SCDAP/RELAP5/MOD 3.3 Code
Manual MATPROA Library of Materials Properties for LightWaterReactorAccident Analysis,
NUREG/CR6150, Vol. 4, Rev. 2, INEL96/0422, Idaho National Engineering and
Environmental Laboratory, Idaho Falls, ID.
Hamming, R.W. 1971. Introduction to Applied Numerical Analysis, New York: McGrawHill
Book Company, Inc.
Freshley, M.D., D.W. Brite, J.L. Daniel, and P.E. Hart. 1976. “IrradiationInduced Densification
of UO
2
Pellet Fuel,” Journal of Nuclear Materials, 62:138166.
Banks, D.A. 1974. “Some Observations of Density and Porosity Changes in UO
2
Fuel Irradiated
in WaterCooled Reactors,” Journal of Nuclear Materials, 54:97107.
Freshley, M.D., D.W. Brite, J.L. Daniel, and P.E. Hart. 1978. “IrradiationInduced Densification
and PuO
2
Particle Behavior in MixedOxide Pellet Fuel,” Journal of Nuclear Materials, 81:63
92.
Small, G.J. 1987. “Densification of Uranium Dioxide at Low BurnUp,” Journal of Nuclear
Materials, 148:302315.
2.7 Fuel Swelling (FSWELL)
The subroutine FSWELL calculates fuel swelling, which is caused by the buildup of solid and
gaseous fission products during irradiation. Fuel swelling (FSWELL) is combined with
creepinduced elongation (FCREEP), densification due to pressure sintering (FHOTPS), and
irradiation (FUDENS) to calculate the overall dimensional changes in fuel.
The gaseous swelling correlation in FSWELL is not used in FRAPCON3 because it significantly
overpredicts swelling and its effects on cladding deformation. The FSWELL correlation for solid
swelling is used in both MATRPO and FRAPCON3. However, the FSWELL solid swelling
correlation used in FRAPCON3.4 has evolved from the original model described in MATPRO.
The FSWELL subroutine is not used in FRAPTRAN 1.4 because FRAPTRAN 1.4 is intended for
shorter time scales than those that result in significant solid swelling.
2.7.1 Model Development and Comparisons
In MATPRO, the correlation used by FSWELL to calculate swelling due to the buildup of solid
fission products is give by Equation 2.7.1.
(2.71)
where
S
S
= fractional volume change due to solid fission products
(m
3
volume change/m
3
fuel)
S
29
S
B 10 x 5 . 2 S
÷
=
2.32
B
S
= burnup during a time step (fissions/m
3
)
At temperatures below 2800K, swelling due to the buildup of gaseous fission products is given by
Equation 2.72.
(2.72)
where
S
g
= fractional volume change due to gas fission products (m
3
volume change/m
3
fuel)
T = temperature (K)
B = total burnup of fuel (fissions/m
3
)
For temperatures greater than 2800K, S
g
is zero because the gas that causes swelling is assumed
to have released. The correlations presented in Equations 2.71 and 2.72 are used to determine
swelling due to solid and gaseous fission product buildup, respectively, in the FSWELL subcode
described by MATPRO.
Unlike the MATPRO version of FSWELL, FRAPCON3.4 does not include a correlation for
swelling due to the buildup of gaseous fission products. Results of an NRCsponsored test rod
were compared with predictions made by FRAPCON3 and revealed that swelling models that
included gaseous fission product buildup tend to overpredict test results at burnup levels up to
10 GWd/MTU (Lanning and Bradley, 1984; Lanning et al., 1997). It was determined that
swelling could be modeled better by eliminating the contribution from gaseous fission products
and increasing the contribution of solid fission products from 0.699 percent (MATPRO and
FRAPCON2) to 0.77 percent (FRAPCON3) per 10 GWd/MTU.
In FRAPCON3.4, the swelling due to solid fission products was decreased to 0.62 percent per
10 GWd/MTU for burnup less than 80 GWd/MTU and increased to 0.86 percent per
10 GWd/MTU for burnup greater than 80 GWd/MTU. The correlations used to calculate fuel
swelling under these conditions are given in Equations 2.73 and 2.74, respectively.
For burnup < 80 GWd/MTU
(2.73)
For burnup > 80 GWd/MTU
(2.74)
where
soldsw = fractional volume change due to solid fission products (m
3
volume change/m
3
fuel)
sigswell = userdefined parameter that incrementally changes error
bus = fuel burnup during time step (analogous to B
S
in MATPRO model)
fdens = initial pellet density (kg/m
3
)
bu = burnup at end of time step (MWs/kgU)
bul = burnup at end of previous time step (MWs/kgU)
and
( )
( )    
S
B 10 x 0 . 8 T 2800 0162 . 0 73 . 11 56
g
B e e T 2800 10 x 8 . 8 S
27 ÷
÷ ÷ ÷ ÷
÷ =
) 10 x 315 . 2 * sigswell 10 x 315 . 2 ( * bus soldsw
24 23 ÷ ÷
+ =
) 10 x 211 . 3 * sigswell 10 x 211 . 3 ( * bus soldsw
24 23 ÷ ÷
+ =
2.33
The variable “bus” in Equations 2.73 and 2.74 is analogous to the variable “B
S
” in Equation
2.71. However, “bus” permits the fractional volume change due to solid fission products to be
determined as a function of burnup in MWs/kgU as opposed to fissions/m
3
.
A comparison between the MATPRO and the FRAPCON3.4 versions of FSWELL is presented
in Figure 2.71 along with experimental data from immersion density measurements. Based on
this figure, it can be seen that the FRAPCON3.4 model predicts less swelling than the MATPRO
model, which tends to overpredict the experimentally determined data points.
0.00
0.50
1.00
1.50
2.00
2.50
3.00
3.50
4.00
4.50
0 10 20 30 40 50 60 70 80
Burnup (GWd/MTU)
A
V
/
V
(
%
)
MATPRO FRAPCON3.4 Garde (1986)
Newman (1986) Smith (1994) Dideon and Bain (1983)
Figure 2.71 Data comparison between MATPRO and FRAPCON3.4 versions of
FSWELL.
For burnup > 80 GWd/MTU, a higher swelling rate is used in the FRAPCON3.4 version of
FSWELL. Data comparisons between FRAPCON3.4 predictions and data from instrumented
fuel assemblies tested at Halden are presented in Figure 2.72 for varying levels of burnup. These
Halden swelling data are based on measurements of fuel stack length and cladding diameter
change with burnup and converted to volume swelling assuming isotropic swelling. The
diametral cladding change data versus burnup is particularly valuable because it implicitly
includes the effects of cladding restraint on fuel swelling.
( ) bul bu * 10 x 974 . 2 * fdens bus
10
÷ =
2.34
0
0.2
0.4
0.6
0.8
1
1.2
1.4
I
F
A

6
3
3
I
F
A

6
1
0
I
F
A

5
3
4
.
1
4
I
F
A

6
2
9
.
4
I
F
A

6
2
9
.
3
I
F
A

5
9
7
.
3
I
F
A

5
1
5
.
1
0
I
F
A

5
1
5
.
1
0
I
F
A
5
1
9
.
9
I
F
A

5
0
4
I
F
A

6
5
5
R
o
d
s
1
&
3
I
F
A

6
5
5
R
o
d
s
2
&
4
S
w
e
l
l
i
n
g
R
a
t
e
,
%
D
V
/
V
p
e
r
1
0
G
W
d
/
M
T
M
9
9
G
W
d
/
M
T
U

9
9
G
W
d
/
M
T
U

1
7
G
W
d
/
M
T
U

6
2
G
W
d
/
M
T
U

6
8
G
W
d
/
M
T
U

6
5
G
W
d
/
M
T
U

6
2
G
W
d
/
M
T
U

9
0
G
W
d
/
M
T
U

7
5
G
W
d
/
M
T
U

7
8
G
W
d
/
M
T
U

9
0
G
W
d
/
M
T
U

7
0
G
W
d
/
M
T
U

Figure 2.72 Comparison between swelling rates determined for various instrumented fuel
assemblies tests at Halden and swelling rates predicted by FRAPCON3.4.
2.7.2 Applicability and Uncertainty
Based on the data from which it is derived, the correlation used in FSWELL is applicable to fuels
with burnups up to 100 GWd/MTU. Below 80 GWd/MTU, a 1sigma uncertainty of 0.08 percent
AV/V per 10 GWd/MTU has been calculated. Above 80 GWd/MTU, a 1 sigma uncertainty of
0.16 percent AV/V per 10 GWd/MTU has been calculated.
2.7.3 References
Lanning, D.D., and E.R. Bradley. 1984. Irradiation History and Interim Postirradiation Data
for IFA432, NUREG/CR3071, Pacific Northwest Laboratory for the U.S. Nuclear Regulatory
Commission, Washington, D.C.
Lanning, D.D. C.E. Beyer, C.L. Painter. 1997. FRAPCON3: Modifications to Fuel Rod
Material Properties and Performance Models for HighBurnup Application, NUREG/CR6534,
Vol. 1, Pacific Northwest National Laboratory for the U.S. Nuclear Regulatory Commission,
Washington, D.C.
Garde, A.M. 1986. Hot Cell Examination of Extended Burnup Fuel from Fort Calhoun,
DOE/ET3403011 (CEND427), Combustion Engineering, Windsor, CT.
Newman, L.W. 1986. The Hot Cell Examination of Oconee 1 Fuel Rods after Five Cycles of
Irradiation, DOE/ET3421236, Babcock and Wilcox, Lynchburg, VA.
2.35
Smith, G.P., R.C. Pirek, H.R. Freeburn, and D. Schrire. 1994. The Evaluation and
Demonstration of Methods for Improved Nuclear Fuel Utilization, DOE/ET3401315,
Combustion Engineering, Windsor, CT.
Dideon, C.G., and G.M. Bain. 1983. Fuel Performance under Extended Burnup for the B&W
15x15 Design, DOE/ET3421238, Babcock and Wilcox, Lynchburg, VA.
3.1
3.0 Cladding Material Properties
Material property correlations for fuel claddings are described in the following subsections.
Unless otherwise specified, the correlations below are applicable to Zircaloy2, 4, ZIRLO, and
M5. In addition, several FRAPTRAN 1.4 subroutines include additional correlations to describe
a Russian zirconium alloy containing 1 wt% niobium (ZrNb1). The subroutine used to
determine the material correlation in either the FRAPCON3 or FRAPTRAN codes is given in
parentheses.
3.1 Cladding Specific Heat (CCP)
The specific heat subcode, CCP, determines the true specific heat at constant pressures for
cladding. Specific heat calculations are based on interpolation of measured data. The correlation
in FRAPCON3.4, FRAPTRAN 1.4, and MATPRO is applicable to Zircaloy2, Zircaloy4,
ZIRLO, and M5 alloys. However, FRAPTRAN 1.4 contains correlations to calculate the specific
heat of ZrNb1. This calculation can be based on data collected at either fast or slow heating
rates and the user can designate the appropriate data set.
3.1.1 Model Development and Comparisons
The CCP subcode requires temperature as an input to calculate specific heat. For the alpha phase
of the Zircaloy alloys (temperature less than 1090K), CCP returns linear interpolations for the
points listed in Table 3.11. These data points are based on precise data taken by Brooks and
Stansbury (1966) with a Zircaloy2 sample that had been vacuumannealed at 1075K to remove
hydrogen, which would have otherwise affected the measurement.
The standard errors associated with this interpolation technique differ between MATPRO and
FRAPCON3.4/FRAPTRAN 1.4. In MATPRO, the standard error of the CCP interpolation was
based on 90 points in the Brooks and Stansbury (1966) database and was found to be temperature
dependent. For the 57 data points between 300 and 800K, the standard error is 1.1 J/kg*K.
Between 800 and 1090K, it is 2.8 J/kg*K. The FRAPCON3.4/FRAPTRAN 1.4 codes assume a
standard error of 10 J/kg*K below 1090K.
For temperatures from 1090 to 1300K (where Brooks and Stansbury do not report results), values
of specific heat proposed by Deem and Eldridge (1967) are adopted. The Deem and Eldridge
(1967) values are based on measurements of enthalpy and temperature which provide
considerably less precise specific heat data than the results of Brooks and Stansbury (1966).
As a result, the MATPRO standard error estimated from the Deem and Eldridge (1967) data in
the region of 1090 through 1310K is 10.7 J/kg*K. This standard error is a measure only of the
precision of the fit, since only a single data source is used. The standard error in
FRAPCON3.4/FRAPTRAN is assumed to be 25 J/kg*K between 1090 and 1300K.
Above the alpha + beta to beta transformation temperature (about 1250K) and up to about 1320K,
a constant value of 355.7 J/kg*K was reported by Deem and Eldridge (1967). This value agrees
well with a value of 365.3 reported by Coughlin and King (1950) for pure beta zirconium. The
standard error of specific heat calculations made above 1300K is assumed to be 100 J/kg*K in the
FRAPCON3.4/FRAPTRAN 1.4 codes.
3.2
In addition to Zircaloy alloys, FRAPTRAN 1.4 has been modified to include specific heat
calculations for ZrNb1. These calculations are based on data collected at two different heating
rates. Depending on the user input, specific heat calculations can be based on either the fast or
the slow heating rate data. The data used to interpolate specific heat values for the ZrNb1 alloy
are presented in Table 3.12. There is no standard error described for specific heat calculations
based on these data sets.
Table 3.11 Specific heat capacity database for Zircaloy2, Zircaloy4, ZIRLO, and M5.
Standard Error
Temperature
(K)
Specific Heat
Capacity
(J/kg*K)
Source MATPRO
(J/kg*K)
FRAPCON3 /
FRAPTRAN
(J/kg*K)
Alpha Phase
300 281 Brooks and Stansbury 1.1 10
400 302 Brooks and Stansbury 1.1 10
640 331 Brooks and Stansbury 1.1 10
1090 375 Brooks and Stansbury 2.8 10
Beta Phase
1093 502 Deem and Eldridge 10.7 25
1113 590 Deem and Eldridge 10.7 25
1133 615 Deem and Eldridge 10.7 25
1153 719 Deem and Eldridge 10.7 25
1173 816 Deem and Eldridge 10.7 25
1193 770 Deem and Eldridge 10.7 25
1213 619 Deem and Eldridge 10.7 25
1233 469 Deem and Eldridge 10.7 25
1248 356 Deem and Eldridge 10.7 25
2098 356 Coughlin and King 100 100
2099 356 Coughlin and King 100 100
Table 3.12 Specific heat capacity database for ZrNb1 alloys.
Temperature
(K)
Specific Heat Capacity
(J/kg*K)
Source
 Low Heating Rate ( < 1000 K/s) 
280 345 Volkov B. Yu et al.
473 360 Volkov B. Yu et al.
573 370 Volkov B. Yu et al.
673 380 Volkov B. Yu et al.
773 383 Volkov B. Yu et al.
873 385 Volkov B. Yu et al.
883 448 Volkov B. Yu et al.
973 680 Volkov B. Yu et al.
1025 816 Volkov B. Yu et al.
1073 770 Volkov B. Yu et al.
3.3
1153 400 Volkov B. Yu et al.
1173 392 Volkov B. Yu et al.
1248 356 Volkov B. Yu et al.
 High Heating Rate ( > 1000 K/s) 
1100 412 Ljusternik V.E. et al.
1110 420 Ljusternik V.E. et al.
1120 480 Ljusternik V.E. et al.
1134 600 Ljusternik V.E. et al.
1142 1000 Ljusternik V.E. et al.
1150 1400 Ljusternik V.E. et al.
1155 1600 Ljusternik V.E. et al.
1161 1400 Ljusternik V.E. et al.
3.1.2 Applicability and Uncertainty
The CCP subroutine is valid over the temperature range from which the data sets have been
collected on zirconium alloys (300 to 2099K). This correlation is applied to Zircaloy2, 4,
ZIRLO, and M5. However the standard error increases with increasing temperature as indicated
in Table 3.11.
3.1.3 References
Brooks, C.R., and E.E. Stansbury. 1966. “The Specific Heat of Zircaloy2 from 50 to 700°C,”
Journal of Nuclear Materials, 18:223.
Deem, H.W., and E.A. Eldridge. 1967. Specific Heats of Transformation of Zirclaoy2 and Low
Nickel Zircaloy2, USAEC BM11803, Battelle Memorial Institute, Columbus, OH.
Coughlin, J.P., and E.G. King. 1950. “HighTemperature Heat Contents of Some Zirconium
Containing Substances,” Journal for the American Chemical Society, 72:2262.
Volkov, B. Yu., et al. 1989. Library of Subprograms on Physical and Mechanical Properties of
the N1Alloy Fuel Rod Cladding Material, Preprint IAE4941/11, Institute of Automation and
Electrometry, Moscow, Russia.
Ljusternik, V.E., et al. 1993. “Experimental research of zirconium reactor materials thermal
properties: 110alloy,” M.J. Higher Temperature Thermal Physics, 31(4).
3.2 Cladding Thermal Conductivity (CTHCON)
The subroutine CTHCON is used to calculate cladding thermal conductivity, which is required
for accurate predictions of fuel temperature. The thermal conductivity of the cladding is
primarily a function of temperature. Other characteristics, such as residual stress levels, crystal
orientation, and minor composition differences, may have secondary effects on thermal
conductivity. The correlation used in CTCHCON to calculate cladding thermal conductivity is
the same in MATPRO, FRAPCON3.4, and FRAPTRAN 1.4. This correlation is applied to
Zircaloy2, 4, ZIRLO, and M5. In addition, FRAPTRAN 1.4 includes a correlation to calculate
the cladding thermal conductivity of ZrNb1.
3.4
3.2.1 Model Development and Comparisons
Considering only temperature as the defining parameter, the thermal conductivity of Zircaloy for
temperatures less than 2098K is described by CTHCON in Equation 3.21. The uncertainty of
the conductivity calculations is given in Equation 3.22.
(3.21)
(3.22)
For temperatures greater than or equal to 2098K, the thermal conductivity and uncertainty are
given in Equations 3.23 and 3.24, respectively.
(3.23)
(3.24)
where
k = thermal conductivity of Zircaloy (W/m* K)
T = temperature (K)
o
k
= standard deviation (W/m*K)
Equation 3.21 predicts k very well from room temperature to the data limit of about 1800K
(Figure 3.21) and may be extrapolated with some confidence to the melting point. The standard
deviation (o
k
) of the data with respect to this correlation appears to be temperature independent
over the data range.
The standard deviation of the constants in Equation 3.21 is 20 to 30 percent of the value of the
constant. Jensen (1969) performed a parametric analysis of several variables involved in the
estimation of fuel and cladding temperatures. Both steadystate and transient analysis showed
that variations of ± 20 percent resulted in calculated cladding temperature variations of about
2.8K. Fuel centerline temperatures are more sensitive to cladding thermal conductivity and
showed variations of 28K. Similar findings were reported by Korber and Unger (1974).
For temperatures less than 2133K, the thermal conductivity of the ZrNb1 alloy is calculated by
FRAPTRAN 1.4 with the correlation presented in Equation 3.25.
(3.25)
For temperatures above 2133K, the thermal conductivity of the ZrNb1 cladding is assumed to be
36 W/m*K.
Figure 3.21 compares the CTHCON model as incorporated in MATPRO, FRAPCON3.4, and
FRAPTRAN1.4 with the data on which it is based, extrapolated to the melting point of Zircaloy
(2098K).
Figure 3.22 compares the CTHCON model with the Zircaloy thermal conductivity and
3 9 2 5 2
T 10 x 67 . 7 T 10 x 45 . 1 T 10 x 09 . 2 51 . 7 k
÷ ÷ ÷
+ ÷ + =
01 . 1
k
= o
36 k =
5
k
= o
T * 000461843 . 0
e 0636 . 15 k =
3.5
diffusivity data that were not included in development of the model. Thermal diffusivity data
were converted to thermal conductivity using the recommended models in Fink (2000) for
specific heat and density, respectively. Although different alloys were examined, there were no
discernable differences between the alloys. The scatter in the data due to differences in alloy,
form, or measurement direction is comparable to the variation between individual data sets. The
scatter in the data is distributed relatively uniformly around the curve representing the CTHCON
model, except for anomalously high Peggs data below 1000K and the anomalously low Bunnell
et al. (1983) data above 1300K.
0
10
20
30
40
50
60
70
0 500 1000 1500 2000 2500
Temperature (K)
T
h
e
r
m
a
l
C
o
n
d
u
c
t
i
v
i
t
y
(
W
/
m

K
)
MATPRO
Anderson et al. (1962)
Chirigos et al. (1961)
Feith (1966)
Lucks and Deem (1958)
Powers (1961)
Scott (1965)
Figure 3.21 CTHCON model and data used for development.
3.6
0
10
20
30
40
50
60
70
0 500 1000 1500 2000 2500
Temperature (K)
T
h
e
r
m
a
l
C
o
n
d
u
c
t
i
v
i
t
y
(
W
/
m

K
)
MATPRO
Krett and Cleveland (1997)  China
Krett and Cleveland (1997)  AECL
Gilchrist (1976)
Bunnell et al. (1983)
Murabayashi et al. (1975)
Peggs et al. (1976)
Maglic et al. (1994)
Figure 3.22 CTHCON model and data collected since development.
3.2.2 Applicability and Uncertainty
The correlation used in CTHCON to calculate the cladding thermal conductivity may be applied
to temperatures up to the melting temperature of Zircaloy (2098K). The uncertainty through this
range of temperature is given by Equation 3.22. Above the melting temperature, the uncertainty
increases to the value given by Equation 3.24.
3.2.3 References
Jensen, S.E. 1969. Parametric Studies of Fuel Pin Temperature Response, IDO17295, Philips
Petroleum Company, Atomic Energy Division, Idaho Falls, ID.
Korber, H., and H.E. Unger. 1974. “Sensitivity Study on Core Heatup and Meltdown by
Variation of Heat Conductivity and Thermal Emissivity,” Transactions for the American Nuclear
Society, 18:234235.
Fink, J.K. 2000. “Zircaloy Thermal Conductivity,” International Nuclear Safety Center Material
Properties Database, http://www.insc.anl.gov/matprop/zircaloy/index.php, Argonne National
Laboratory, Argonne, IL.
Bunnell, L.R., J.L. Bates, and G.B. Mellinger. 1983. “Some HighTemperature Properties of
ZircaloyOxygen Alloys,” Journal of Nuclear Materials, 116(23):219232.
Anderson, W.K., C.J. Beck, A.R. Kephar, and J.S. Theilacker. 1962. “Zirconium Alloys,”
Reactor Structural Materials: Engineering Properties as Affected by Nuclear Reactor Service,
3.7
ASTMSTP314, pp. 6293, American Society for Testing and Materials, West Conshohocken,
PA.
Chirigos, J.N., et al. 1961. “Development of Zircaloy4,” Fuel Element Fabrication, New York:
Academic Press, pp. 1955.
Feith, A.D. 1966. Thermal Conductivity and Electrical Resistivity of Zircaloy4, GEMP669,
October 1966, General Electric Company, Missile and Space Division, Cincinnati, OH.
Lucks, C.F., and H.W. Deem. 1958. Progress Relating to Civilian Applications during June,
1958, R.W. Dayton and C.R. Tipton, Jr., eds., BMI1273, pp. 79, Battelle Memorial Institute,
Columbus, OH.
Powers, A.E. 1961. Application of the Ewing Equation for Calculating Thermal Conductivity
from Electrical Conductivity from Electrical Conductivity, KAPL2146, Knolls Atomic Power
Laboratory, General Electric Company, Schenectady, NY.
Scott, D.B. 1965. Physical and Mechanical Properties of Zircaloy 2 and 4, WCAP326941,
pp. 5, 9, Westinghouse Electric Corporation, Atomic Power Division, Pittsburgh, PA.
Krett, V., and J. Cleveland. 1997. Thermophysical Properties of Materials for Water Cooled
Reactors, IAEATECDOC949, pp. 6784, International Atomic Energy Agency, Vienna,
Austria.
Gilchrist, K.E. 1976. “Thermal Property Measurements on Zircaloy2 and Associated Oxide
Layers up to 1200°C,” Journal of Nuclear Materials, 62:257264.
Murabayashi, M., S. Tanaka, and Y. Takahashi. 1975. “Thermal Conductivity and Heat
Capacity of Zircaloy2, 4 and Unalloyed Zirconium,” Journal of Nuclear Science and
Technology, 12(10):661662.
Peggs, I.D., A.M. Stadnyk, and D.P. Godin. 1976. “Thermophysical Properties of Zirconium
Alloy Fuel Channel Components,” High TemperaturesHigh Pressures, 8(4):441450.
Maglic, K.D., N.L.j. Perovic, and A.M. Stanimirovic. 1994. “Calorimetric and Transport
Properties of Zircalloy 2, Zircalloy 4, and Inconel 625,” International Journal of Thermophysics,
15(4):741755.
3.3 Cladding Oxide Thermal Conductivity (ZOTCON)
The ZOTCON subroutine calculates the thermal conductivity of the zirconium oxide layer that
forms on zirconium alloys. Cladding temperature is the only parameter used to calculate
zirconium oxide thermal conductivity. The correlation used in the ZOTCON subroutine differs
between MATPRO and the FRAPCON3.4/FRAPTRAN 1.4 codes. The MATPRO correlation is
based on an arbitrary correction to an earlier correlation that was fit to a large database of oxide
thermal conductivity measurements. This database contained measurements from a variety of
zirconium oxide materials and test methods. The correction was applied to better fit data that was
more representative of the oxide layer that forms on zirconium alloys. The correlation used in
FRAPCON3.4 and FRAPTRAN 1.4 is based on measurements obtained by Kingery (1954) from
fully dense and porous (87 percent TD) zirconium oxides.
3.8
3.3.1 Model Development and Comparisons
Oxide thermal conductivity models in MATPRO and FRAPCON3.4/FRAPTRAN 1.4 are only a
function of temperature. After the cladding temperature is established, the thermal conductivity
of the oxide layer can be calculated using the correlation determined from either the MATPRO or
the FRAPCON3.4/FRAPTRAN 1.4 databases.
The MATPRO (2001) correlation is derived from a database that includes data collected by
Adams (1954), Maki (1973), Lapshov and Bashkatov (1973), and Gilchrist (1976). The
correlation used in FRAPCON3.4/FRAPTRAN 1.4 is derived from data collected by Kingery
(1954). The FRAPCON3.4/FRAPTRAN correlation and database was adopted from an earlier
version of MATPRO (Hagrman et al., 1981).
The database for the MATPRO (2001) model includes data from a variety of zirconium oxide
materials and test conditions. The Maki (1973) data include two samples oxidized in steam over
a small temperature range and show a sharp increase in conductivity between 400 and 500K. The
principal recommendation for the data is that they were taken with black oxide from Zircaloy
tubes. Two sets of data attributed to Waldman by Maki are also shown in the table. The data of
Lapshov and Bashkatov (1973) are from films formed by plasma sputtering of zirconium dioxide
on tungsten substrates. However, these data may not be representative of Zircaloy cladding oxide
thermal conductivity because sputtered coatings are quite porous. In addition, these films do not
have the same O/M ratio as cladding oxide and may not adhere well to the substrate. The
Gilchrist (1976) data includes two types of oxide films. One is a nodular oxide and the other a
black oxide characteristic of the kinds of layers usually reported in hightemperature tests with
cladding. The nodular oxide thermal conductivities are much lower than the black oxide thermal
conductivities, and both kinds of oxide have conductivities that are significantly lower than the
stabilized Zircaloy dioxide conductivities reported by Adams (1954). Considerable uncertainty is
reported by Gilchrist (1976) because of difficulty in measuring oxide film thickness.
The database for the FRAPCON3.4 and FRAPTRAN 1.4 model is based on data from Kingery
(1954). The data are from two different samples; one was a bulk sample of 100 percent dense
ZrO
2
and the other with approximately 13 percent porosity.
Each database was fit with a different model to relate the thermal conductivity of the cladding
oxide layer to temperature. The equations used to describe these models are presented in
Equations 3.31 and 3.32 for MATPRO and FRAPCON3.4/FRAPTRAN 1.4 (i.e., FRAP),
respectively. These models relate the thermal conductivity of the cladding oxide layer, K
(W/m*K), to cladding temperature, T (K).
(3.31)
(3.32)
Both databases are presented in Figure 3.31 along with both cladding oxide thermal conductivity
models. In addition, a more recent data set from Gilchrist (1979) is included. Figure 3.31
reveals that the FRAP model fits the Kingery data from which it was derived and agrees well with
the measurements performed by Adams (1954). The FRAP model tends to overestimate both the
Gilchrist data sets and the data from Lapshov and Bashkatov (1973). However, as stated earlier,
the earlier Gilchrist (1976) data is associated with considerable uncertainty and the data from
T 10 x 81 . 1 835 . 0 K
4
MATPRO
÷
+ =
)) 10 x 946 . 1 * T 10 x 43 . 6 ( * T 10 x 41 . 2 ( * T 9599 . 1 K
10 7 4
FRAP
÷ ÷ ÷
÷ ÷ ÷ =
3.9
Lapshov and Bashkatov (1973) is likely to be low due to different O/M ratios and high porosity in
the sputtered films.
The MATPRO model was initially fit to the data collected by Adams (1954), but it was adjusted
to the form presented in Equation 3.31 to better fit the Gilchrist (1976) data for black oxide. The
rationale for this correction was that the black oxide was more prototypic of the cladding oxide
layer and was achieved by dividing the original equation by two.
3.10
Figure 3.31 Oxide thermal conductivity measurements as a function of temperature compared with the MATPRO and FRAPCON
3.4/FRAPTRAN 1.4 (FRAP) cladding oxide thermal conductivity models.
0
1
2
3
4
5
6
7
8
9
300 500 700 900 1100 1300 1500 1700 1900
Temperature (K)
T
h
e
r
m
a
l
C
o
n
d
u
c
t
i
v
i
t
y
(
W
/
m
*
K
)
Adams Maki Maki (Waldman data)
Lapshov and Bashkatov Gilchrist (black oxide) Gilchrist (nodular oxide)
Kingery (13% Porosity) Kingery (100% Dense) Gilchrist (1979)
MATPRO
F
3.11
3.3.2 Applicability and Uncertainty
The subroutine ZOTCON is applicable over the temperature range for which the data was
collected, 375 to 1673K. Since the relatively high density of the bulk ZrO
2
materials bias the
thermal conductivity measurements high, it is difficult to establish a meaningful measurement of
uncertainty.
The MATPRO model is applicable from 300K to the melting temperature (2973K). Upon
melting the thermal conductivity, K
Liquid
(W/m*K), is assumed to be described by Equation 3.33
(Hagrman et al., 1981).
(3.33)
The standard error for the MATPRO model is approximately ±0.75 W/m*K, but is assumed to be
within 10 percent for materials that are known to be dense (>87%TD) ZrO
2
.
3.3.3 References
Adams, M. 1954. “Thermal Conductivity: III, Prolate Spheroidal Envelope Method,” Journal of
the American Ceramic Society, 37:7479.
Maki, H. 1973. “Heat Transfer Characteristics of Zircaloy2 Oxide Film,” Journal of Nuclear
Science and Technology, 10:107175.
Lapshov, V. N., and A. V. Bashkatov. 1973. “Thermal Conductivity of Coatings of Zirconium
Dioxide Applied by the Plasma Sputtering Method,” Heat TransferSoviet Research, 5:1922.
Gilchrist, K. E. 1976. “Thermal Property Measurements on Zircaloy2 and Associated Oxide
Layers,” Journal of Nuclear Materials, 62:257264.
Kingery, W.D., et al. 1954. “Thermal Conductivity: X, Data for Several Pure Oxide Materials
Corrected to Zero Porosity,” Journal of the American Ceramic Society, 37:107100.
Hagrman, D.L., et al. 1981. MATPROVersion 11 (Revision 2). A Handbook of Materials
Properties for Use in the Analysis of Light Water Reactor Fuel Rod Behavior, NUREG/CR0479
(TREE1280, Rev. 2), EG&G Idaho, Inc., Idaho Falls, ID.
Gilchrist, K.E. 1979. “Thermal Conductivity of Oxide Deposited on Zircaloy Fuel Tube
Material – A Continuation of Previous Work,” Journal of Nuclear Materials, 82:193194.
3.4 Cladding Surface Emissivity (ZOEMIS)
The subcode ZOEMIS returns the cladding surface emissivity, which is directly proportional to
the radiant heat transfer from the cladding surface during an abnormal transient. The ZOEMIS
model described in MATPRO is the same as the model used in FRAPCON3.4 and FRAPTRAN
1.4.
4 . 1 K
Liquid
=
3.12
3.4.1 Model Development and Comparisons
Surface emissivities are significantly affected by surface layers on the cladding. For cladding
with thin oxide coatings, the oxide surface thickness is only a few wavelengths of near infrared
radiation and is partly transparent. Oxide thickness is an important parameter for these thin
coatings. Thicker oxide layers are opaque, so the oxide thickness is not as important as the nature
of the outer oxide surface, which is affected by temperature and by chemical environment. The
effect of temperature has been modeled, but variations in crud on the external cladding surface
and chemical reaction products on the inside surface are not modeled explicitly.
The model for emissivity was constructed by considering measured emissivities reported by
several investigators. Expressions used to predict the emissivity of Zircaloy cladding surfaces are
summarized below.
When the cladding surface temperature has not exceeded 1500K, emissivities are modeled by
Equations 3.41 and 3.42. Equation 3.41 is used for oxide layers less than 3.88x10
6
m thick
and Equation 3.42 is used for oxide layers equal to or greater than 3.88x10
6
m thick. Both
equations relate the hemispherical emissivity, c
1
(unitless), to the oxide layer thickness, d (m).
(3.41)
(3.42)
When the maximum cladding temperature has exceeded 1500K, emissivity is taken to be the
larger of 0.325 and the result of Equation 3.43. Equation 3.43 relates the emissivity above
1500K, c
2
(unitless), to c
1
and the maximum cladding temperature, T (K).
(3.43)
Comparisons between data and the high and low temperature models are presented in Figure
3.41.
d 10 x 1246 . 0 325 . 0
6
1
+ = c
d 0 . 50 808642 . 0
1
÷ = c
(
¸
(
¸
÷
c = c
300
T 1500
exp
1 2
3.13
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
0 20 40 60 80 100 120 140
Surface Layer Thickness (mm)
E
m
i
s
s
i
v
i
t
y
(
u
n
i
t
l
e
s
s
)
Murphy and Havelock Juenke and Sjodahl
Burgoyne and Garlick Uniform Oxide Burgoyne and Garlick Nodular Oxide
Burgoyne and Garlick with Crud Model Prediction < 1500 K
Model Prediction > 1500K
Figure 3.41 Modeltodata comparison for cladding oxide emissivity.
3.14
3.4.2 Applicability and Uncertainty
The absolute standard error, o
1
, expected from the use for Equation 3.41 to predict emissivity in
a reactor when cladding surface temperature has never exceeded 1500K is given by Equation
3.44.
(3.44)
When the maximum cladding temperature exceeds 1500K, the expected standard error is
estimated by o
2
, which is described by Equation 3.45.
(3.45)
If Equations 3.43 and 3.45 predict values of c
2
and o
2
that fall inside the range of physically
possible values of emissivity (0.0 – 1.0), the value o
2
is returned as the expected standard error.
If the prediction c
2
+ o
2
is greater than 1 or if c
2
 o
2
is less than 0, the standard error of Equation
3.45 is modified to limit c
2
+ o
2
at 1 and/or c
2
 o
2
at 0.
3.4.3 References
Juenke, E.F., and L.H. Sjodahl. 1968. “Physical and Mechanical Properties: Emittance
Measurements,” Advanced Pressure Vessel Materials, GEMP1008, pp. 239242, General
Electric Co., Missile and Space Division, Cincinnati, OH.
Murphy, E.V., and F. Havelock. 1976. “Emissivity of Zirconium Alloys in Air in the
Temperature Range 100400°C,” Journal of Nuclear Materials, 60:167176.
Burgoyne, T.B. and A. Garlick. 1976. “The effect of oxidation and crud deposition on the
emissivity of Zircaloy2 cladding.” Paper presented to the OECDCSNI Meeting on the Behavior
of Water Reactor Fuel Elements under Accident Conditions, Spåtnid, Norway, Sept. 1316, 1976.
3.5 Cladding Thermal Expansion (CTHEXP)
The subroutine CTHEXP returns the axial and diametral components of thermal expansion in the
cladding as a function of temperature. The MATPRO model calculates these strain components
for single crystal Zircaloy and applies these to polycrystalline cladding materials through the use
of pole figures collected over a cladding section. The FRAPCON3.4/FRAPTRAN 1.4 models
use the correlation from Mehan and Wiesinger (1961) for temperatures below 1073K and a
constant value of thermal expansion above 1273K. Values of thermal expansion between 1073
and 1273K are determined in the FRAPCON3.4/FRAPTRAN 1.4 codes by linear interpolation.
In addition, the FRAPTAN 1.4 model includes a calculation of the thermal expansion for ZrNb1.
3.5.1 Model Development and Comparisons
In the MATPRO model, a total of six correlations that are functions of temperature only are used
to find single crystal thermal strains. In addition, basal plane symmetry (c
11
= c
22
) is assumed.
1 . 0
1
± = o
(
¸
(
¸
÷
± = o
300
1500 T
exp 1 . 0
2
3.15
The model was developed for as fabricated Zircaloy4, but comparisons with Zircaloy2 and
zirconium data also show good agreement for these materials.
The correlations for single crystal thermal strains are:
For 300 < T < 1083K
(3.51)
(3.52)
where
c
11
= circumferential thermal expansion (m/m)
c
33
= axial thermal expansion (m/m)
For 1083 s T < 1244K
(3.53)
(3.54)
where the arguments of the cosines are in radians.
For 1244 s T < 2098K
(3.55)
(3.56)
For temperatures > 2098K, consideration of the volume change associated with melting is
required. Since no data were found for the MATPRO model, a typical 2 percent volume increase
at melt was assumed (Siefken et al., 2001). The expressions used for the thermal strain in liquid
zirconium (temperatures > 2098K) are thus:
(3.57)
c
p
= thermal expansion strain in liquid Zircaloy (m/m)
c
11
= circumferential thermal expansion strain of a single crystal of Zircaloy at 2098K
(m/m)
c
33
= axial thermal expansion strain of a single crystal of Zircaloy at 2098K (m/m)
where the c
11
and c
33
are calculated by Equations 3.53 and 3.54, respectively.
3 6
11
10 x 485 . 1 T 10 x 95 . 4
÷ ÷
÷ = c
3 5
33
10 x 78 . 3 T 10 x 26 . 1
÷ ÷
÷ = c
3
11
10 x
161
1083 T
cos 09822 . 1 77763 . 2
÷
(
¸
(
¸

.

\

t
÷
+ = c
3
33
10 x
161
1083 T
cos 09822 . 1 77763 . 2
÷
(
¸
(
¸

.

\

t
÷
+ = c
2 6
11
10 x 04 . 1 T 10 x 7 . 9
÷ ÷
÷ = c
3 6
33
10 x 4 . 4 T 10 x 76 . 9
÷ ÷
÷ = c
0067 . 0
3
1
3
2
33 11 p
+ c + c = c
3.16
To obtain cladding strains from these single crystal strains, it is necessary to do a volume
weighted averaging of crystalline orientation over the entire cladding section. Such an averaging
requires the collection of pole figures and is described in greater detail in the MATPRO (2001)
handbook. The input parameters for typical cladding are not readily available.
The model for cladding thermal expansion in the FRAPCON3.4/FRAPTRAN (FRAP) codes is
different from the MATPRO model but provides similar predictions of expansion (< 6 percent).
The data for the FRAP correlation used from room temperature to 1273K was taken from Mehan
and Wiesinger (1961), Scott (1965), and Kearns (1965). Above 1273K, the coefficient of thermal
expansion is the constant value of 9.7x10
6
, as recommended by Lustman and Kerze (1955).
Between 1073 and 1273K (approximately the alphabeta transition range for Zircaloy), the
thermal expansion components are determined by linear interpolation.
The correlations used to calculate the axial and diametral components of thermal expansion in the
cladding between room temperature and 1073K are presented in Equations 3.58 and 3.59,
respectively. Strain is given a function of temperature, T (°C).
(3.58)
(3.59)
The correlations used to calculate the axial and diametral components of thermal expansion in the
cladding above 1273K are presented in Equations 3.510 and 3.511, respectively. Strain is given
a function of temperature, T (°C).
(3.510)
(3.511)
Modeltodata comparisons for the MATPRO and FRAPCON3.4/FRAPTRAN 1.4 models are
presented in Figures 3.51 and 3.52 for axial and diametral thermal expansion, respectively.
These comparisons illustrate that the axial and diametral components of cladding thermal
expansions are being underpredicted. These correlations will be examined further and possibly
revised in a future revision of FRAPCON3.4/FRAPTRAN 1.4.
T
axial
6 5
10 4410 . 4 10 5060 . 2
÷ ÷
× + × ÷ = c
T
diametral
6 5
10 7210 . 6 10 3730 . 2
÷ ÷
× + × ÷ = c
T
axial
6 3
10 70 . 9 10 300 . 8
÷ ÷
× + × ÷ = c
T
diametral
6 3
10 70 . 9 10 800 . 6
÷ ÷
× + × ÷ = c
3.17
0.0E+00
1.0E03
2.0E03
3.0E03
4.0E03
5.0E03
6.0E03
7.0E03
8.0E03
9.0E03
1.0E02
200 400 600 800 1000 1200 1400 1600 1800
Temperature (K)
T
h
e
r
m
a
l
E
x
p
a
n
s
i
o
n
(
m
/
m
)
FRAP Axial
MATPRO Axial
Bunnel Axial
Kearns, Scott, and Mehan Axial
Figure 3.51 Modeltodata comparison for FRAPCON3.4/FRAPTRAN 1.4 (FRAP) and
MATPRO axial thermal expansion. Data from cladding tubes.
0.0E+00
1.0E03
2.0E03
3.0E03
4.0E03
5.0E03
6.0E03
7.0E03
8.0E03
9.0E03
1.0E02
200 400 600 800 1000 1200 1400 1600 1800
Temperature (K)
T
h
e
r
m
a
l
E
x
p
a
n
s
i
o
n
(
m
/
m
)
FRAP Diametral
MATPRO Diametral
Bunnel Diametral
Kearns, Scott, and Mehan Diametral
Figure 3.52 Modeltodata comparison for FRAPCON3.4/FRAPTRAN 1.4 (FRAP) and
MATPRO diametral thermal expansion. Data from cladding tubes.
3.5.2 Applicability and Uncertainty
The subroutine CTHEXP is applicable over the temperature range for which the experimental
data was collected, which included temperatures up to 1100K. The uncertainty of the axial and
diametral thermal expansion correlations has not been quantified.
3.18
3.5.3 References
Mehan, R.L., and F.W. Wiesinger. 1961. Mechanical Properties of Zircaloy2, KAPL2110,
General Electric Company, Knolls Atomic Power Laboratory, Schenectady, NY.
Siefken, L.J., E.W. Coryell, E.A. Harvego, and J.K. Hohorst. 2001. SCDAP/RELAP5/MOD 3.3
Code Manual MATPROA Library of Materials Properties for LightWaterReactorAccident
Analysis, NUREG/CR6150, Vol. 4, Rev. 2, INEL96/0422, Idaho National Engineering and
Environmental Laboratory, Idaho Falls, ID.
Scott, P.B. 1965. Physical and Mechanical Properties of Zircaloy2 and 4, WCAP326941,
Westinghouse Electric Corporation, Atomic Power Division, Pittsburgh, PA.
Kearns, J.J. 1965. Thermal Expansion and Preferred Orientation in Zircaloy, WAPDTM472,
pp. 1718, Bettis Atomic Power Laboratory, Pittsburgh, PA.
Lustman and F. Kerze. 1955. The Metallurgy of Zirconium, McGrawHill Book Company, New
York, p. 355.
Bunnell, L.R., et al. 1977. High Temperature Properties of Zircaloy Oxygen Alloys, EPRI NP
524, Battelle Pacific Northwest Laboratory, Richland, WA.
3.6 Cladding Elastic Modulus (CELMOD) and Shear Modulus
(CSHEAR)
Elastic moduli are required to relate stresses to strains. The elastic moduli are defined by the
generalized form of Hooke’s law as elements of the fourth rank tensor that relates the second rank
stress and strain tensors below the yield point. In practice, cladding is frequently assumed to be
an isotropic material. In such a case, only two independent elastic moduli are needed to describe
the relation between elastic stress and strain: the Young’s modulus and the shear modulus.
The subcodes CELMOD and CSHEAR are used in the MATPRO and the
FRAPCON3.4/FRAPTRAN 1.4 (FRAP) codes to determine the Young’s modulus and the shear
modulus. The tensor from which these moduli are derived is calculated by CELAST, which is
included in MATPRO and determines the compliance matrix for isotropic cladding. However,
CELAST is not used in the FRAP codes. The subcode CELAST would only be required if the
Young’s modulus and shear modulus for an anisotropic cladding is desired and, as stated
previously, the cladding is assumed to be isotropic.
Elastic moduli are affected primarily by temperature and oxygen content. Fast neutron fluence,
cold work, and texture effects are also included in the models described herein, but they are not as
important as temperature and oxygen content for typical LWR fuel rod cladding. The models are
based primarily on data published by Bunnell et al. (1977), Fisher and Renken (1964), Armstrong
and Brown (1964), and Padel and Groff (1976), since these data include the best description of
texture for the temperature range in which they were used. Data from several other sources are
used to evaluate the expected standard error of the CELMOD and CSHEAR codes and to
estimate the effect of fast neutron fluence (Shober, 1957; Whitmarsh, 1962). To calculate
Zircaloy elastic moduli at temperatures greater than the melting temperature of Zircaloy (2098K),
the moduli are set to zero (actually, 1.0x10
10
is used to avoid dividing by zero).
3.19
3.6.1 Model Development and Comparisons
CELMOD
The CELMOD subcode used in MATPRO is the same equations written in a different formas the
one used in FRAPCON3.4/FRAPTRAN 1.4 (FRAP). The differences in form between the codes
will be discussed in this subsection.
MATPRO uses Equations 3.61 and 3.62 to describe the Young’s modulus in the alpha phase
and the beta phase, respectively.
(3.61)
(3.62)
where
Y = Young’s modulus for Zircaloy2 and 4 with random texture (Pa)
T = cladding temperature (K)
K
1
= modification to account for the effect of oxidation (Pa)
K
2
= modification to account for the effect of cold work (Pa)
K
3
= modification to account for the effect of fast neutron fluence (unitless)
In the alpha plus beta phase, MATPRO determines Y by linearly interpolating between values
calculated at the alpha to alpha plus beta and the alpha plus beta to beta phase boundaries.
The MATPRO expressions used to model the effects of oxidation, cold work, and fast neutron
fluence are presented in Equations 3.63 through 3.65.
(3.63)
(3.64)
(3.65)
where
A = average oxygen concentration minus oxygen concentration of asreceived
cladding (kg oxygen/kg Zircaloy). Asreceived oxygen concentrations are so
small (0.0012 kg oxygen/kg Zircaloy) that the exact magnitude of the asreceived
concentration will not affect the correlation predictions.
C = cold work (unitless ratio of areas).
u = fast neutron fluence (n/m
2
).
The standard error of CELMOD is 6.4x10
9
Pa. Although FRAPCON3.4 is coded to include this
error, this value is not used. This standard error is not included in FRAPTRAN 1.4.
3 2 1
7 11
K / ) K K T 10 x 475 . 5 10 x 088 . 1 ( Y + + ÷ =
T 10 x 05 . 4 10 x 21 . 9 Y
7 10
÷ =
A + = ) T 10 x 912 . 5 10 x 61 . 6 ( K
8 11
1
C 10 x 6 . 2 K
10
2
÷ =

.

\
 u
÷ + =
25 3
10
exp 12 . 0 88 . 0 K
3.20
The CELMOD subcode included in the FRAP codes differs only in formy from the MATPRO
version. For instance, the expression for Young’s modulus in the alpha phase is presented in
Equation 3.66.
(3.66)
where
celmod = Young’s modulus (Pa)
ctemp = cladding temperature (K)
deloxy = input average oxygen concentration excluding oxide layer (kg oxygen/kg
Zircaloy)(hardwired to zero in FRAPCON3.4/FRAPTRAN 1.4)
cwkf = input effective cold work (unitless ratio of areas)
In Equation 3.66, c
1
, c
3
, and c
2
are expressions that account for oxygen content, cold work, and
fast neutron fluence respectively.
Equations 3.67 through 3.69 describe the constants c
1
, c
2
, and c
3
, respectively.
(3.67)
(3.68)
(3.69)
For neutron fluences greater than 1x10
22
, c
2
is given by Equation 3.610.
(3.610)
where
fnck = input effective fast fluence (n/m
2
)
Both FRAPCON3.4 and FRAPTRAN 1.4 use the same correlations to determine Young’s
modulus in the subcode CELMOD. Although FRAPCON3.4 includes the coding to return the
standard error of the model, neither FRAPCON3.4 nor FRAPTRAN 1.4 uses these errors.
FRAPTRAN 1.4 includes additional coding to determine Young’s modulus for ZrNb1.
CSHEAR
In MATPRO, the expressions used to determine shear modulus in the alpha and beta phases are
given in Equations 3.611 and 3.612, respectively.
(3.611)
(3.612)
In the alpha plus beta phase, MATPRO determines G by linearly interpolating between values
calculated at the alpha to alpha plus beta and the alpha plus beta to beta phase boundaries.
2 3 1
7 11
c / ) cwkf * c deloxy * c ctemp * 10 x 475 . 5 10 x 088 . 1 ( mod cel + + ÷ =
( ) 7015 . 5 * 10 x 037 . 1 * ctemp 10 x 16 . 1 c
8 11
1
+ =
0 . 1 c
2
=
10
3
10 x 6 . 2 c ÷ =
( ) ( ) ( )
25 25
2
10 x 1 / fnck exp 10 x 1 / fnck exp 1 * 88 . 0 c ÷ + ÷ ÷ =
( )
3
2 1
7 10
K
K K T 10 x 168 . 2 10 x 04 . 4
G
+ + ÷
=
T 10 x 66 . 1 10 x 49 . 3 G
7 10
÷ =
3.21
The constants K
1
, K
2
, and K
3
have the same definitions as stated before in CELMOD. However,
the expression used to model the effect of oxidation for shear modulus is given in Equation
3.613.
(3.613)
The standard error of the CSHEAR code is 9x10
9
Pa.
The CSHEAR subcode included in the FRAP codes differs slightly from the MATPRO version.
For instance, the expression for shear modulus in the alpha phase is presented in Equation 3.614.
(3.614)
where
cshear = shear modulus (Pa)
ctemp = cladding temperature (K)
deloxy = input average oxygen concentration excluding oxide layer
cwkf = input effective cold work (unitless ratio of areas)
In Equation 3.614, c
1
, c
3
, and c
2
are expressions that account for oxygen content, cold work, and
fast neutron fluence, respectively.
Equations 3.615 through 3.617 describe the constants c
1
, c
2
, and c
3
, respectively.
(3.615)
(3.616)
(3.617)
For neutron fluences greater than 1x10
22
, c
2
is given by Equation 3.618.
(3.618)
where
fnck = input effective fast fluence (n/m
2
)
Both FRAPCON3.4 and FRAPTRAN 1.4 use the same correlations to determine Young’s
modulus in the subcode CSHEAR. Although FRAPCON3.4 includes the coding to return the
standard error of the model, neither FRAPCON3.4 nor FRAPTRAN 1.4 returns the error.
FRAPTRAN 1.4 includes additional coding to determine Young’s modulus, shear modulus, and
Poisson’s ratio as a function of temperature for ZrNb1.
Since there is limited data available from shear modulus measurements, a modeltodata
comparison for Young’s modulus is presented. Figure 3.61 presents predicted vs. measured
Young’s modulus values for data sets used in the development of the CELMOD models as well
as more recent data.
A + = ) T 10 x 315 . 2 10 x 07 . 7 ( K
8 11
1
2 3 1
7 10
c / ) cwkf * c deloxy * c ctemp * 10 x 168 . 2 10 x 04 . 4 ( cshear + + ÷ =
8 11
1
10 x 315 . 2 * ctemp 10 x 07 . 7 c ÷ =
0 . 1 c
2
=
10
3
10 x 867 . 0 c ÷ =
( ) ( ) ( )
25 25
2
10 x 1 / fnck exp 10 x 1 / fnck exp 1 * 88 . 0 c ÷ + ÷ ÷ =
3.22
The upper and lower bounds of Figure 3.61 represent an interval of ± 6.4 GPa. It may be noted
that the data fit is slightly better below 80 GPa ( ~ < 4 percent) than above 80 GPa
(~ > 5 percent). This is attributed to the relatively low temperatures that were used to collect
these data (e.g., a quarter of Bolmaro’s data is collected at < 250°F). Nevertheless the CELMOD
model, in general, provides a satisfactory fit to the data within the temperature range it will be
applied for predicting inreactor performance (> 250°F).
3.23
0
40
80
120
0 40 80 120
Measured Young's Modulus, GPa
P
r
e
d
i
c
t
e
d
Y
o
u
n
g
'
s
M
o
d
u
l
u
s
,
G
P
a
Lower Bound Upper Bound Armstrong and Brown
Padel and Groff Busby Spasic
Mehan Northwood Bolmaro and Povolo
Measured = Predicted
Figure 3.61 Predicted vs. measured values for Young’s modulus for various data sets.
3.24
3.6.2 Applicability and Uncertainty
The CELMOD and CSHEAR correlations are applicable below the melting temperature of the
cladding (2098K). Above this temperature, the moduli are assumed to be zero (but actually set at
1x10
10
to avoid dividing by zero). The CELMOD and CSHEAR correlations are similar between
MATPRO and the FRAPCON3.4/FRAPTRAN 1.4. The uncertainty of the predictions for
CELMOD are within 6.4 GPa (the upper and lower bounds), and typically < 5 percent below
80 GWd/MTU. The uncertainty of the CSHEAR predictions is estimated to be within 9 GPa.
3.6.3 References
Bunnell, L.R., et al. 1977. High Temperature Properties of Zircaloy Oxygen Alloys, EPRI NP
524, Battelle Pacific Northwest Laboratory, Richland, WA.
Fisher, E.S., and C.J. Renken. 1964. “Single – Crystal Elastic Moduli and the CHPBCC
Transformation in Ti, Zr, and Hf,” Physical Review, 135(2A):A482494.
Armstrong, P.E., and H.L. Brown. 1964. “Dynamic Young’s Modulus Measurement above
1000°C on Some Pure Polycrystalline Metals and Commercial Graphites,” Transactions of the
Metallurgical Society of AIME 230, August 1964, pp. 962966.
Padel, A., and A. Groff. 1976. “Variation du Module de Young du Zirconium b en Function de
la Temperature,” Journal of Nuclear Materials 59:325326.
Shober, F.B., et al. 1957. The Mechanical Properties of Zirconium and Zircaloy2, BMI1168,
Battelle Memorial Institute, Columbus, OH.
Whitmarsh, C.L. 1962. Review of Zircaloy2 and Zircaloy4 Properties Relevant to N.S.
Savannah Reactor Design, ORNL3281, Oak Ridge National Laboratory, Oak Ridge, TN.
Mehan, R.L. 1958. Modulus of Elasticity of Zircaloy2 Between Room Temperature and
1,000°F, KAPLMRLM16, General Electric Company, Knolls Atomic Power Laboratory,
Schenectady, NY.
Mehan, R.L., and F.W. Wiesinger. 1961. Mechanical Properties of Zricaloy2, KAPL2110, pp.
1112, General Electric Company, Knolls Atomic Power Laboratory, Schenectady, NY.
Busby, C.C. 1966. Properties of Zircaloy4 Tubing, WAPDTM585, p. 65, Bettis Atomic
Power Laboratory, Pittsburgh, PA.
Northwood, D.O., et al. 1975. “Elastic Constants of Zirconium Alloys,” Journal of Nuclear
Materials, 55:299310.
Spasic, Z., et al. 1968. Conference on the Use of Zirconium Alloys in Nuclear Reactors,
Marlanske Lanze, Czechoslovakia, USAEC CONF681086, 1968, pp. 277284.
Bolmaro, R.E., and F. Povolo. 1988. “Anelastic behaviour of materials under multiaxial strains,”
J. Materials Science, 23:371.
3.25
3.7 Cladding Axial Growth (CAGROW)
The subcode CAGROW calculates the factional change in length of Zircaloy tubes due to
irradiationinduced growth. The change in length of commercial fuel rods due to irradiation
growth is small. However, growth can be a significant fraction of the clearance between the rod
and the top and bottom assembly nozzles. Contact with the nozzles can cause rods to bow and
possibly fail at points where the rods contact each other. In addition, rod growth increases the
internal void volume that impacts the rod internal pressure calculation.
The original CAGROW model used in MATPRO was modified in FRAPCON3.4 to better
describe the cladding axial growth at higher fluence. The MATPRO version of CAGROW was
applied to both Zircaloy2 and 4. The FRAPCON3.4 version includes correlations for
Zircaloy2 and 4, M5, and ZIRLO. These correlations may be applied to either PWR or boiling
water reactor (BWR) conditions. The CAGROW subroutine is not used in FRAPTRAN 1.4
because significant irradiation axial growth is not expected under the short time scales considered
by FRAPTRAN 1.4.
3.7.1 Model Development and Comparisons
The MATPRO correlation used to describe cladding axial growth is presented in Equation 3.71.
This equation has been developed to model the irradiation growth of Zircaloy tubes at
temperature between 40 and 360°C (the normal range of cladding temperatures in LWRs).
(3.71)
where
AL/L = fractional change in length due to growth.
A = 1.407x10
16
(n/m
2
)
1/2
.
T = cladding temperature (K).
u = fast neutron flux (n/m
2
s) (E > 1.0 MeV).
t = time (s).
f
z
= texture factor for the tubing. The variable f
z
is the effective fraction of cells
aligned with their <0001> axis parallel to the tubing axis, as determined by X
ray diffraction analysis. A value of f
z
= 0.05 is typical.
CW = cold work (fraction of crosssectional area reduction).
Axial growth for temperature below 40°C is approximated by using T = 40°C in Equation 3.71
and growth above 360°C is approximated by using T = 360°C.
The MATPRO subroutine CAGROW was based on lowburnup, lowfluence data found in
Harbottle (1970), Kreyns (1960), and Daniel (1971, 1972). Consequently, the CAGROW model
was updated in FRAPCON3.4 to better model axial growth under greater fluences ( > 1x10
25
n/m
2
). A model proposed by Franklin (1982) that was based on high fluence PWR data was
adopted for FRAPCON3.4. This model is presented in Equation 3.72 and relates the axial
growth, ax (m/m), of stress relief annealed (SRA) Zircaloy4 in a PWR to fluence (n/cm
2
).
(3.72)
( )  ( ) ( )( ) CW 02 . 0 1 f 1 t T / 8 . 240 exp A L / L
z
2 / 1
+ ÷ u = A
845 . 0 21
* 10 x 18 . 2 ax u =
÷
3.26
where
ax = axial growth increment (m/m)
u = fast neutron fluence (n/cm
2
) (E > 1.0 MeV)
The expression in Equation 3.72 is used to determine the axial growth at the beginning and the
end of a time increment. The difference is used to determine the axial growth for the time
increment, which is the value returned by the CAGROW subroutine. The expression in Equation
3.72 can be used to describe the axial growth of fully recrystallized annealed (RXA) Zircaloy2
cladding in a BWR by multiplying by a factor of 0.5.
In addition to updating the cladding axial growth model for SRA Zircaloy4 and RXA Zircaloy2,
new correlations have been added to describe the cladding axial growth of M5 and ZIRLO. The
cladding axial growth correlations for M5 and ZIRLO are presented in Equations 3.73 and 3.74,
respectively. These equations relate the cladding axial growth, ax (m/m), to fluence (n/cm
2
). The
variables have the same definitions as those described in Equation 3.72 and the time difference is
used to determine the axial growth during the time increment, which is the value returned by the
CAGROW subroutine.
For M5 cladding:
(3.73)
For ZIRLO cladding:
(3.74)
Modeltodata comparisons are presented in Figures 3.75 through 3.78 for SRA Zircaloy4,
RXA Zircaloy2, M5, and ZIRLO, respectively. Based on these figures, it appears that the
cladding axial growth of these alloys can be predicted well with the CAGROW subroutine.
81787 . 0 21
10 013 . 7 ax u × =
÷
98239 . 0 25
10 7893 . 9 ax u × =
÷
3.27
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
0 2E+21 4E+21 6E+21 8E+21 1E+22 1.2E+22
Fast Neutron Fluence, n/cm²s
A
L
/
L
,
%
Newman Franklin Gilbon Zr4 PWR Model
Figure 3.75 Modeltodata comparison for SRA Zircaloy4 (PWR). The standard error of
AL/L (%) is 0.11.
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
0 2E+21 4E+21 6E+21 8E+21 1E+22 1.2E+22
Fast Neutron Fluence, n/cm²s
A
L
/
L
,
%
FRAPCON Data Gilbon Zr2 BWR Model
Figure 3.76 Modeltodata comparison for RXA Zircaloy2 (BWR). The standard error
of AL/L (%) is 0.07.
3.28
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
0 2E+21 4E+21 6E+21 8E+21 1E+22 1.2E+22
Fast neutron fluence, n/cm², E>1MeV
F
u
e
l
r
o
d
g
r
o
w
t
h
,
%
M5 Data PWR M5 Model
Figure 3.77 Modeltodata comparison for M5. M5 data was collected by Gilbon et al.
(2000). The standard error of AL/L (%) is 0.05.
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
0 2E+21 4E+21 6E+21 8E+21 1E+22 1.2E+22
Fast neutron fluence, n/cm², E>1MeV
F
u
e
l
r
o
d
g
r
o
w
t
h
,
%
Vandellos ZIRLO Data North Anna ZIRLO Data
BR3 ZIRLO Data PWR ZIRLO Model
Figure 3.78 Modeltodata comparison for ZIRLO. Vandellos data was collected by Irisa
(2000) while North Anna and BR3 data was collected by Sabol (1994). The
standard error of AL/L (%) is 0.05.
3.29
3.7.2 Applicability and Uncertainty
The CAGROW model in FRAPCON3.4 is applicable over the temperature range from 700 to
900K for local burnups up to 65 GWd/MTU (or 12x10
25
n/m
2
fluence, E > 1MeV). The absolute
uncertainties for the growth strains predicted for stressrelief annealed Zr4, M5, and ZIRLO are
0.11, 0.05, and 0.05, respectively. These three alloys are typically used in PWRs, whereas
recrystallized Zr2 is used in BWRs. The absolute uncertainty in growth strains for Zr2 is 0.07.
3.7.3 References
Harbottle, J.E. 1970. “The Temperature and Neutron Dose Dependence of Irradiaton Growth in
Zircaloy2,” Irradiation Effect on Structural Alloys for Nuclear Reactor Applications, ASTM
STP485, American Society of Testing and Materials, West Conshohocken, PA, pp. 287299.
Kreyns, P.H., quoted by E. Duncombe et al. 1996. Comparison with Experiment of Calculated
Changes and Failure Analysis of Irradiated Bulk Oxide Fuel Test Rods Using the CYGRO01
Computer Program, WAPDTM583, Bettis Atomic Power Laboratory, Pittsburgh, PA.
Daniel, R.C. 1971. InPile Dimensional Changes of Zircaloy4 Tubing Having Low Hoop
Stresses (LWBR Development Program), WAPDTM973, Bettis Atomic Power Laboratory,
Pittsburgh, PA.
Daniel, R.C. 1972. “InPile Dimensional Changes of Zircaloy4 Tubing Having Stresses (Light
Water Breeder Reactor Development Program),” Nuclear Technology, 14:171186.
Franklin, D.G. 1982. “Zircaloy Cladding Deformation during Power Reactor Irradiation.” In
Proceedings of the Fifth International Symposium on Zirconium in the Nuclear Industry, ASTM
STP754, American Society of Testing and Materials, West Conshohocken, PA, pp. 235267.
Newman, L.W. 1986. The Hot Cell Examination of Oconee 1 Fuel Rods After Five Cycles of
Irradiation, DOE/ET/3421250, Babcock and Wilcox Company, Lynchburg, VA.
Gilbon, D., A. Soniak, S. Doriot, and J.P. Mardon. 2000. “Irradiation Creep and Growth
Behavior, and Microstructural Evolution of Advanced ZrBase Alloys,” Zirconium in the Nuclear
Industry: Twelfth International Symposium, ASTMSTP1354, GP Sabol and GD Moan eds.,
American Society of Testing and Materials, West Conshohocken, PA, pp. 5173.
Irisa, Y., Murai, K., Goto, K., Suzuki, S., Sabate, R., Alonso, J.M. 2000. “Segmented Fuel Rod
Irradiation Programme on Advanced Materials for High Burnup,” in Proceedings of the 2000
Light Water Fuel Performance Meeting, Park City, UT, April 1013, 2000.
Sabol, G.P., R.J. Comstock, R.A. Weiner, P. Larouere, and R.N. Stanutz. 1994. “InReactor
Corrosion Performance of ZIRLO and Zircaloy4,” Zirconium in the Nuclear Industry: Tenth
International Symposium, ASTMSTP1425, A.M. Garde and E.R. Bradley, eds. American
Society for Testing and Materials, Philadelphia, PA, pp. 724744.
3.30
3.8 Creep Rate (CREPR)
The subcode CREPR is used by FRAPCON3.4 to determine the cladding creep rate, accounting
for both thermal and irradiation creep. The CREPR subcode was not used in MATPRO. Also,
since creep is a timedependent deformation process, CREPR is not used in FRAPTRAN 1.4,
because creep occurs over much longer time intervals than those considered in FRAPTRAN 1.4
of a minute or less. The FRAPTRAN calculation of cladding ballooning as a result of a LOCA is
handled with a separate subroutine named BALON2
3.8.1 Model Development and Comparisons
The creep model used by FRAPCON3.4 is based on the model given by Limbäck and Andersson
(1996). This model uses a thermal creep model described by Matsuo (1987) and an empirical
irradiation creep rate with tuned model parameters that were fit to data by Franklin et al. (1983).
The Limbäck model was further modified by PNNL to use effective stress rather than hoop stress
as an input so that the difference in creep behavior during tensile and compressive creep would be
modeled correctly. Several of the fitting coefficients were consequently changed to
accommodate this modification. In addition, a temperaturedependent term was added to the
formula for irradiation creep strain rate. This model has different parameters for SRA and RXA
cladding types, and provides much more reasonable creep strains in the LWR range of
temperatures and cladding hoop stresses than the previous model used in FRAPCON3.3.
The steadystate thermal and irradiation creep rates are given by Equations 3.81 and 3.82,
respectively.
(3.81)
(3.82)
where
th
c = thermal strain rate (in./in./hr)
irr
c = irradiation strain rate (in./in./hr)
T = temperature (K)
o
eff
= effective stress (MPa)
R = universal gas constant (0.008314 kJ/molK)
 = fast neutron flux (n/m
2
s)
The variables A, E, a
i
, n, Q, C
o
, C
1
, C
2
, and the function f(T) have different values depending on
temperature, flux, and cladding type (SRA or RXA). The values for these variables under
different conditions are presented in Table 3.81.

.

\
 ÷

.

\
 o
= c
RT
Q
exp
E
a
sinh
T
E
A
n
eff i
th
) T ( f C
2 1
C
eff
C
o irr
· o ·  · = c
3.31
Table 3.81 Parameters for FRAPCON3.4 creep equation for SRA and RXA cladding.
Parameter Units SRA Cladding RXA Cladding
A* K/MPa/hr 1.08x10
9
* 5.47x10
8
*
E MPa 1.149x10
5
59.9*T
a
i
MPa
1
650{10.56[1exp(1.4x10
27
*u
1.3
)]}
u=fast neutron fluence (n/cm
2
)
n unitless 2.0 3.5
Q kJ/mole 201
C
o
(n/m
2
s)
Cl
MPa
C2
4.0985x10
24
1.87473x10
24
C
1
unitless 0.85
C
2
unitless 1.0
f(T)* unitless
T < 570 0.7283
570< T< 625 7.0237+0.0136T
T > 625 1.4763
0.7994
3.18562+0.00699132T
1.1840
*Parameters changed from original Limbäck equation.
The thermal and irradiation creep rates may be added together as shown in Equation 3.83 and
used to calculate the saturated primary hoop strain, which is presented in Equation 3.84.
(3.83)
(3.84)
The total thermal strain can then be calculated as a function of time, t (hours), as shown in
Equation 3.85. However, in FRAPCON3.4 the strain rate is used, which is obtained by taking
the derivative of Equation 3.85. This derivative is presented in Equation 3.86, which relates the
total thermal strain rate to the saturated primary hoop strain, the combined thermal and irradiation
creep rates, and time, t, in hours.
(3.85)
(3.86)
The effective stress in the cladding is found using the principal stresses at the midwall radius
using the thick wall formula. The principle stresses can be determined with Equations 3.87
through 3.89.
(3.87)
(3.88)
irr th irr th
c + c = c
+
( ) ( )
05 . 2
irr th
109 . 0
irr th
s
p
35500 tanh 2 0216 . 0
÷
+ +
c · ÷ c · = c
( ) ( ) t t 52 exp 1
irr th irr th
s
p H
· c + · c · ÷ ÷ c = c
+ +
( )
irr th irr th
2 / 1
2 / 1
irr th
s
p
H
t 52 exp
t 2
52
+ +
+
c + · c · ÷
·
c · c ·
= c
( )
2
i
2
o
2
i o
2
o
2
i
2
o o
2
i i
r
r r
r
P P r r
r P r P
÷
÷
+ ÷
= o
( )
2
i
2
o
2
i o
2
o
2
i 2
o o
2
i i
t
r r
r
P P r r
r P r P
÷
÷
÷ ÷
= o
3.32
(3.89)
where
P
i
= inner pressure
P
o
= outer pressure
r
i
= inner radius
r
o
= outer radius
r = radius within tube
o
r
= radial stress
o
t
= tangential stress
o
l
= longitudinal stress
The effective stress can then be calculated by Equation 3.810.
(3.810)
The subcode CREPR can be used for newer alloys like M5 and ZIRLO. It has been found that
the Zircaloy RXA model adequately describes the creep behavior of M5 (Gilbon et al., 2000;
Soniak et al., 2002). The Zircaloy SRA model is used for ZIRLO with a reduction factor of 0.8.
The reduction factor is the result of studies that have shown that ZIRLO exhibits about 80 percent
of the Zircaloy4 creepdown (Sabol et al., 1994).
Modeltodata comparisons are presented in Figures 3.81 and 3.82 for SRA and RXA Zircaloy
cladding. Creep data was obtained from irradiated SRA and RXA tube from Franklin (1983),
Soniak et al. (2002), and Gilbon et al. (2000). The relative standard error for the predictions is
14.5 percent for SRA and 21.6 percent for RXA cladding.
2
i
2
o
2
o o
2
i i
l
r r
r P r P
÷
÷
= o
( ) ( ) ( ) ( )
2
l r
2
r t
2
t l eff
5 . 0 o ÷ o + o ÷ o + o ÷ o = o
3.33
0
0.005
0.01
0.015
0.02
0.025
0 0.005 0.01 0.015 0.02 0.025
Measured hoop strain, in/in
P
r
e
d
i
c
t
e
d
h
o
o
p
s
t
r
a
i
n
,
i
n
/
i
n
Franklin Soniak 350°C Soniak 320°C Gilbon Predicted=Measured
Figure 3.81 Predicted vs. measured hoop strain for SRA Zircaloy model.
0
0.002
0.004
0.006
0.008
0 0.002 0.004 0.006 0.008
Measured hoop strain, in/in
P
r
e
d
i
c
t
e
d
h
o
o
p
s
t
r
a
i
n
,
i
n
/
i
n
Franklin Soniak 350°C Soniak 320°C Gilbon Predicted=Measured
Figure 3.82 Predicted vs. measured hoop strain for RXA Zircaloy model.
3.8.2 Applicability and Uncertainty
The subroutine CREPR is applicable over the range of conditions for which the data was
collected. The database spans a temperature range of 570 to 625K, an effective stress range of 40
to 130 MPa, and a fast neutron flux range of 1x10
17
to 2x10
18
n/m
2
s. This covers the nominal
3.34
operating conditions for LWRs. The relative standard error for the predictions is 14.5 percent for
SRA and 21.6 percent for RXA cladding. The uncertainty for predicting cladding creep in a fuel
rod will most likely be greater than these values because there may be greater uncertainty in the
estimate of fast neutron flux and cladding temperature than those from the creep data used to
determine the creep model uncertainties.
3.8.3 References
Limbäck, M., and T. Andersson. 1996. “A Model for Analysis of the Effect of Final Annealing
on the In and OutofReactor Creep Behavior of Zircaloy Cladding,” Zirconium in the Nuclear
Industry: Eleventh International Symposium, ASTM STP 1295, E.R. Bradley and G.P. Sabol,
eds., American Society for Testing and Materials, West Conshohocken, PA, pp. 448468.
Matsuo, Y. 1987. “Thermal Creep of Zircaloy4 Cladding under Internal Pressure,” Journal of
Nuclear Science and Technology, 24(2):111119.
Franklin, D.G., G.E. Lucas, A.L. Bement. 1983. “Creep of Zirconium Alloys in Nuclear
Reactors,” ASTM STP 815, American Society for Testing and Materials, West Conshohocken,
PA.
Gilbon, D., A. Soniak, S. Doriot, and J.P. Mardon. 2000. “Irradiation Creep and Growth
Behavior, and Microstructural Evolution of Advanced ZrBase Alloys,” Zirconium in the Nuclear
Industry: Twelfth International Symposium, ASTMSTP1354, G.P. Sabol and G.D. Moan, eds.,
American Society of Testing and Materials, West Conshohocken, PA, pp. 5173.
Soniak, A., N. L’Hullier, JP. Mardon, V. Rebeyrolle, P. Boffioux, C. Bernaudat. 2002.
“Irradiation Creep Behavior of ZrBase Alloys,” Zirconium in the Nuclear Industry: Thirteenth
International Symposium, ASTM STP 1423, G.D. Moan and P. Rudling, eds., ASTM
International, West Conshohocken, PA, pp.837862.
Sabol, G.P., Comstock, R.J., Weiner, R.A., Larouere, P., Stanutz, R.N. 1994. “InReactor
Corrosion Performance of ZIRLO™ and Zircaloy4,” Zirconium in the Nuclear Industry: Tenth
International Symposium, ASTM STP 1245, A.M. Garde and E.R. Bradley, eds. American
Society for Testing and Materials, Philadelphia, PA, pp. 724744.
3.9 Cladding Meyer Hardness (CMHARD)
The subroutine CMHARD calculates Meyer hardness as a function of cladding temperature.
Hardness is one of the parameters required for calculating fueltocladding contact conductance.
As the contact pressure between the two surfaces increases, the points of contact enlarge due to
localized plastic deformation and the solidtosolid thermal conductance is improved. The Meyer
hardness is used by Ross and Stoute (1962) in their heat transfer correlation as an indication of
the hardness of resistance to deformation of the softer (Zircaloy) material.
The same CMHARD subroutine is used in the MATPRO, FRAPCON3.4, and FRAPTRAN 1.4
codes. However, FRAPTRAN 1.4 includes additional coding that ensures that the minimum
hardness returned is 1.94x10
8
N/m
2
(the highest temperature data point) and includes provisions
for ZrNb1.
3.35
3.9.1 Model Development and Comparisons
In MATPRO, the Meyer hardness number is a measure of indentation hardness and is defined in
conjunction with Meyer’s law, which is presented in Equation 3.91.
(3.91)
where
L = load
d = the diameter of impression at the surface of a specimen in a static ball test
n = the Meyer work hardening coefficient
a = a material constant
The Meyer hardness number (MH) is defined as 4L/td
2
. Other hardness numbers are available
(Brinell, Rockwell, etc.), and conversion from one to another is possible. However, the routine
CMHARD was created to provide information required by the Ross and Stoute gap conductance
model that includes a dependence on Meyer hardness.
Meyer hardness numbers for temperatures from 298 to 877K were taken from Peggs and Godin
(1975). A regression analysis of the reciprocal of the Meyer hardness values versus the log of
temperature was used to obtain the analytical expression used in CMHARD. The correlation
used is given by Equation 3.92.
(3.92)
where
MH = Meyer hardness (N/m
2
)
T = temperature (K)
The Meyer hardness decreased rapidly with increasing temperature, beginning at 2x10
9
MPa at
room temperature and decreasing to 2x10
8
MPa at 875K. The hardness is presumed to continue
its rapid rate of decrease at temperatures above 875K. The minimum Meyer hardness number of
Zircaloy cladding is 1.0x10
5
N/m
2
.
A comparison between the CMHARD correlation and the data from which it was derived is
presented in Figure 3.91.
n
ad L =
( )   { } { }
8 5 2 1
10 x 5621 . 2 T 10 x 3502 . 4 T 10 x 6394 . 2 T 10 x 6034 . 2 exp MH
÷ ÷ ÷
+ + ÷ + =
3.36
0
200
400
600
800
1000
1200
1400
1600
1800
2000
200 400 600 800 1000 1200 1400 1600 1800 2000 2200
Temperature (K)
M
e
y
e
r
H
a
r
d
n
e
s
s
(
M
P
a
)
Figure 3.91 CMHARD correlation as a function of temperature with data from Peggs and
Godin.
3.9.2 Applicability and Uncertainty
The subroutine CMHARD is applicable for temperatures up to 875K. An estimate of the
uncertainty has not been established for this correlation due to the limited data.
3.9.3 References
Ross, A.M., and R.L. Stoute. 1962. Heat Transfer Coefficient Between UO
2
and Zircaloy2,
AECL1552, Atomic Energy of Canada, Ltd., Chalk River, Ontario.
Peggs, I.D., and D.P. Godin. 1975. “The Yield StrengthHot Hardness Relationship of Zircaloy
4,” Journal of Nuclear Materials, 57:246248.
4.1
4.0 Gas Material Properties
This section describes material property correlations for gap gases. The subroutine used to
determine the material correlation in either the FRAPCON3/FRAPTRAN codes is given in
parentheses in the heading of each subsection.
4.1 Gas Conductivity (GTHCON)
The subroutine GTHCON calculates the gas thermal conductivity as a function of temperature
and gas fraction for seven gases: helium (He), argon (Ar), krypton (Kr), xenon (Xe), hydrogen
(H), nitrogen (N), and water vapor (steam). The MATPRO, FRAPCON3.4, and FRAPTRAN
1.4 codes use similar correlations to determine the gas thermal conductivity. However,
FRAPCON3.4 and FRAPTRAN 1.4 use updated fitting parameters to better estimate gas thermal
conductivity at higher temperatures.
4.1.1 Model Development and Comparisons
The heat conductance of gasfilled gaps or pores is dependent on the thermal conductivity of the
gas mixture when the dimensions of the gasfilled regions are large compared to the mean
distance between gas molecule collisions (mean free path of the gas molecules). This section
presents data and correlations for the thermal conductivities of the gases of interest in fuel rod
analysis.
The correlations used for the gases of interest are all of the form presented in Equation 4.11.
(4.11)
where
K = thermal conductivity (W/mK)
T = gas temperature (K)
The constants A and B are fitting parameters. The values for A and B used in MATPRO for each
gas are given in Table 4.11 and the values used in FRAPCON3.4 and FRAPTRAN 1.4 (FRAP)
are given in Table 4.12.
Table 4.11 Constants used in gas thermal conductivity correlations in MATPRO.
Gas A B
He 2.639x10
3
0.7085
Ar 2.986x10
4
0.7224
Kr 8.247x10
5
0.8363
Xe 4.351x10
5
0.8616
H
2
1.097x10
3
0.8785
N
2
5.314x10
4
0.6898
B
AT K =
4.2
Table 4.12 Constants used in gas thermal conductivity correlations in FRAP.
Gas A B
He 2.531x10
3
0.7146
Ar 4.092x10
4
0.6748
Kr 1.966x10
4
0.7006
Xe 9.825x10
5
0.7334
H
2
1.349x10
3
0.8408
N
2
2.984x10
4
0.7799
The MATPRO steam correlation is also used in FRAPCON3.4/FRAPTRAN 1.4 codes. For
temperatures less than or equal to 973.15K, the correlation presented in Equation 4.12 is used.
For temperatures greater than 973.15K, the correlation presented in Equation 4.13 is used to
determine the thermal conductivity of steam. Temperature, T, is in kelvin and pressure, P, is in
pascals.
(4.12)
(4.13)
The thermal conductivity of gas mixtures is calculated with Equation 4.14.
(4.14)
where +
ij
is given in Equation 4.15
(4.15)
and 
ij
is given in Equation 4.16
(4.16)
and
o
ij
= Kronecker delta = 1 for i = j, 0 otherwise (unitless)
n = number of components in mixture (unitless)
M
i
= molecular weight of component i (kg)
( )
( ) ( )
( ) ( ) ] 15 . 273 T * 10 x 51 . 4 15 . 273 T * 10 x 08 . 1
15 . 273 T * 10 x 87 . 5 10 x 6 . 17 [ 15 . 273 T / T / P * 009 . 1
T * 10 x 005 . 6 T * 10 x 424 . 9 10 x 8516 . 2 * T / P k
3 11 2 7
5 4 2 . 4 2 2
2 14 10 8
÷ ÷ ÷
+ ÷ + + ÷
+ ÷ + ÷ =
÷ ÷
÷ ÷
÷ ÷ ÷
3 . 1 9 5 45 . 1 6
) T / P * 10 x 1668 . 2 ( * 10 x 45 . 9 T * 10 x 44 . 4 k
÷ ÷ ÷
+ =
( )
¿
¿





.

\

+ o ÷ +
=
=
n
i
n
1 j
j ij ij i
i i
mix
x 1 x
x k
k
( )( )
( )


.

\

+
÷ ÷
+  = +
2
j i
j i j i
ij ij
M M
M 142 . 0 M M M
41 . 2 1
2 / 1
j
i 2 / 3
2
4 / 1
j
i
2 / 1
j
i
ij
M
M
1 2
M
M
k
k
1


.

\

+
(
(
¸
(
¸


.

\



.

\

+
= 
4.3
x
i
= mole fraction of component i (unitless)
k
i
= thermal conductivity of the component i (W/mK)
Modeltodata comparisons for the gas conductivity models are presented in Figures 4.11
through 4.17.
0.00
0.10
0.20
0.30
0.40
0.50
0.60
0.70
0.80
0 500 1000 1500 2000 2500 3000
Temperature, K
T
h
e
r
m
a
l
C
o
n
d
u
c
t
i
v
i
t
y
,
W
/
m

K
MATPRO FRAP
Faubert and Springer (1973) Jain and Saxena (1975)
Jody et al. (1977)
Figure 4.11 Modeltodata comparison for He thermal conductivity.
4.4
0.00
0.01
0.02
0.03
0.04
0.05
0.06
0.07
0.08
0.09
0.10
0 500 1000 1500 2000 2500 3000
Temperature, K
T
h
e
r
m
a
l
C
o
n
d
u
c
t
i
v
i
t
y
,
W
/
m

K
MATPRO FRAP
Timrot and Umanski (1966) Saxena and Saxena (1968)
Faubert and Springer (1972) Springer and Wingeier (1973)
Stefanov et al. (1976)
Figure 4.12 Modeltodata comparison for Ar thermal conductivity.
4.5
0.00
0.01
0.02
0.03
0.04
0.05
0.06
0.07
0 500 1000 1500 2000 2500 3000
Temperature, K
T
h
e
r
m
a
l
C
o
n
d
u
c
t
i
v
i
t
y
,
W
/
m

K
MATPRO FRAP
Saxena and Saxena (1969) Stefanov et al. (1976)
Vargaftik and Yakush (1971)
Figure 4.13 Modeltodata comparison for Kr thermal conductivity.
4.6
0.00
0.01
0.01
0.02
0.02
0.03
0.03
0.04
0.04
0.05
0 500 1000 1500 2000 2500 3000
T
h
e
r
m
a
l
C
o
n
d
u
c
t
i
v
i
t
y
,
W
/
m

K
Temperature, K
MATPRO FRAP
Stefanov (1976) Springer and Wingeier (1973)
Saxena and Saxena (1969) Vargaftik and Yakush (1971)
Figure 4.14 Modeltodata comparison for Xe thermal conductivity.
0.00
0.20
0.40
0.60
0.80
1.00
1.20
0 500 1000 1500 2000 2500 3000
Temperature, K
T
h
e
r
m
a
l
C
o
n
d
u
c
t
i
v
i
t
y
,
W
/
m

K
MATPRO FRAP Saxena and Saxena (1970)
Figure 4.15 Modeltodata comparison for H
2
thermal conductivity.
4.7
0.00
0.02
0.04
0.06
0.08
0.10
0.12
0.14
0.16
0 500 1000 1500 2000 2500 3000
Temperature, K
T
h
e
r
m
a
l
C
o
n
d
u
c
t
i
v
i
t
y
,
W
/
m

K
MATPRO FRAP
Vargaftik and Zimina (1964) Faubert and Springer (1972)
Chen and Saxena (1973)
Figure 4.16 Modeltodata comparison for N
2
thermal conductivity.
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
0.16
600 700 800 900 1000
Temperature (K)
T
h
e
r
m
a
l
C
o
n
d
u
c
t
i
v
i
t
y
(
W
/
m

K
) MATPRO/FRAP
Data
Figure 4.17 Modeltodata comparison for steam thermal conductivity.
4.1.2 Applicability and Uncertainty
The gas thermal conductivity correlations for used in GTHCON are applicable over the
temperature range for which data was collected. The standard error for each correlation is given
in Table 4.13.
4.8
Table 4.13 Standard deviation of gas thermal conductivity correlations.
Gas Standard Deviation
He 8.99x10
3
Ar 9.66x10
4
Kr 8.86x10
4
Xe 5.34x10
4
H
2
1.67x10
2
N
2
1.97x10
4
Steam 1.69x10
3
4.1.3 References
Faubert, F.M., and G.S. Springer. 1973. “Measurement of the Thermal Conductivity of Helium
up to 2100°K by the Column Method,” Journal of Chemical Physics, 58(10):40804083.
Jain, P.C., and S.C. Saxena. 1975. “Transport Properties of Helium in the Temperature Range
4002300K,” Chemical Physics Letters, 36(4):489491.
Jody, B.J., S.C. Saxena, V.P.S. Nain, and R.A. Aziz. 1977. “Thermal Conductivity of Helium:
A Problem for the Repulsive Wall of the Interatomic Potential,” Chemical Physics, 22:5358.
Timrot, D.L., and A.S. Umanskii. 1966. “Thermal Conductivity of Hydrogen and Argon,” High
Temperature, 3:285287.
Saxena, V.K., and S.C. Saxena. 1968b. “Measurement of the Thermal Conductivity of Argon
Using HotWire Type Thermal Diffusion Columns,” Chemical Physics Letters, 2(1):4446.
Faubert, F.M., and G.S. Springer. 1972. “Measurement of the Thermal Conductivity of Argon,
Krypton, and Nitrogen in the Range 8002000°K,” Journal of Chemical Physics, 57(6):2333
2340.
Springer, G.S., and E.W. Wingeier. 1973. “Thermal Conductivity of Neon, Argon, and Xenon at
High Temperatures,” Journal of Chemical Physics, 59(5):27472750.
Stefanov, B., L. Zarkova, and D. Oliver. 1976. “Measurement of the ThermalConductivity
Coefficient of Gases and Vapors up to 2500°K,” High Temperature, 14(1):4857.
Saxena, V.K., and S.C. Saxena. 1969. “Thermal Conductivity of Krypton and Xenon in the
Temperature Range 3501500K,” Journal of Chemical Physics, 51(8):33613368.
Vargaftik, N.B., and L.V. Yakush. 1971. “Measurement of the Thermal Conductivities of Neon,
Krypton, and Xenon over a Wide Range of Temperatures,” Journal of Engineering Physics,
21(3):11561161.
Saxena, S.C., and V.K. Saxena. 1970. “Thermal Conductivity Data for Hydrogen and Deuterium
in the Range 1001100°C,” Journal of Physics A, 3(3):309320.
Vargaftik, N.B., and N.Kh. Zimina. 1964. “Thermal Conductivity of Nitrogen at High
Temperatures,” High Temperature, 2(6):782790.
4.9
Chen, S.H.P., and S.C. Saxena. 1973. “Experimental Determination of Thermal Conductivity of
Nitrogen in the Temperature Range 1002200°C,” High Temperature Science, 5(3):206233.
Appendix A
Subroutine Source Codes
A.1
Appendix A Subroutine Source Codes
A.1 Uranium Dioxide/Mixed Oxide Properties
A.1.1 Fuel Melting Temperature (PHYPRP)
FRAPCON3.4: PHYPRP
*deck phyprp
subroutine phyprp
c
c phyprp returns uo2, (u,pu)o2, and zircaloy melting points
c and heats of fusion, and zirconium and zircaloy alpha to beta
c transition temperatures.
common / phypro / ftmelt,fhefus,ctmelt,chefus,ctranb,
+ ctrane,ctranz,fdelta,bu,comp,deloxy
c ftmelt = output uo2 or mixed oxide fuel melting points (k)
c fhefus = output uo2 or mixed oxide fuel heat of fusion (j/kg)
c ctmelt = output zr clad melting point (k)
c chefus = output zr clad heat of fusion (j/kg).
c ctranb = output start of zr4 alphabeta transus temperature (k)
c ctrane = output end of zr4 alphabeta transus temperature (k)
c ctranz = output zr isothermal alphabeta transus temperature (k)
c fdelta = output liquidsolid coexistence temperature range (k)
c bu = input burnup (mws/kgu)
c comp = input puo2 content (wt%)
c deloxy = input oxygen concentration  oxygen concentration
c of as received cladding (kg oxygen/kg zircaloy)
c the uo2 and zr cladding melting points and heats of fusion are
c taken from the following references. uo2 fuel melting point of
c 3113k from h.c.brassfield et al gemp482. uo2 heat of fusion
c of 17.7 kcal/mole from l.leibowitz et al, j.nuc.mat. 39 p 115
c (1971). cladding melting point of 2098 k from m.l. picklesimer
c private communication. cladding heat of fusion of 4.9 kcal/mol
c from brassfield et al, gemp482.
c beginning and end of alphabeta transus are from data in
c figure iii.33 of anl7649
c isothermal zirconium alphabeta transus temperature is 1135 k
c taken from b.lustman & f.kerze "the metallurgy of zirconium"
c mcgrawhill book co., new york, 1955
c mixed oxide melting point was obtained from lyon et al, j. nuc.
c mat., 22 (1967) p 332
c phypro was coded by v.f. baston in may 1974
c modified by c.s. olsen in feb. 1975
c modified by b.w. burnham in nov. 1977
c modified by d.l. hagrman in june 1979
c burnup dependance for fuel melting temperature changed from 3.2K/GWd/tHM
c to 0.5 K/GWd/tHM by K.J. Geelhood in June 2004 based on
c S. G. Popov, et al., "Thermophysical Properties of MOX and UO2 Fuels
c including the Effects of Irradiation," ORNL/TM2000/351.
include 'lacmdl.h'
data on / 1 /,
+ off / 2 /,
+ locidx / 2 /
sldus(c) = 2840.05.41395*c+7.468390e3*c*c
liqdus(c) = 2840.03.21860*c1.448518e2*c*c
fbu = bu/86.4
if (comp.gt.0.0) go to 100
ftmelt = 3113.155.0*fbu/10000.0
A.2
fdelta = 1.0e10
go to 110
100 c1 = comp
ftmelt = sldus(c1)+273.155.0*fbu/10000.0
fdelta = liqdus(c1)sldus(c1)5.0*fbu/10000.0
110 fhefus = 27.4e+4
ctmelt = 2098.15
chefus = 22.5e+04
wfox = deloxy+0.0012
ctranb = 1094.+wfox*(1.289e+03+wfox*7.914e+05)
if (wfox.lt.0.025) go to 120
ctranb = 1556.4+3.8281e+04*(wfox0.025)
120 ctrane = 392.46*((100.*deloxy+0.1242807)**2+3.1417)
if (deloxy.lt.4.7308937e03) go to 130
ctrane = (100.*deloxy+0.12)*491.157+1081.7413
130 continue
ctranz = 1135.15
return
end
FRAPTRAN 1.4: PHYPRP
*deck phyprp
c
subroutine phyprp (icm)
c
implicit real (ah,oz)
c
real liqdus
c
c This is a MATPRO11, Rev. 2 routine modified by PNNL for use in FRAPT
c
c phyprp returns uo2, (u,pu)o2, and zircaloy melting points
c and heats of fusion, and zirconium and zircaloy alpha to beta
c transition temperatures.
c
include 'phypro.h'
c
c ftmelt = output uo2 or mixed oxide fuel melting points (k)
c fhefus = output uo2 or mixed oxide fuel heat of fusion (j/kg)
c ctmelt = output zr clad melting point (k)
c chefus = output zr clad heat of fusion (j/kg).
c ctranb = output start of zr4 alphabeta transus temperature (k)
c ctrane = output end of zr4 alphabeta transus temperature (k)
c ctranz = output zr isothermal alphabeta transus temperature (k)
c fdelta = output liquidsolid coexistence temperature range (k)
c
c bumtp = input burnup (mws/kgu)
c compmt = input puo2 content (wt%)
c deloxy = input oxygen concentration  oxygen concentration
c of as received cladding (kg oxygen/kg zircaloy)
c
c the uo2 and zr cladding melting points and heats of fusion are
c taken from the following references. uo2 fuel melting point of
c 3113k from h.c.brassfield et al gemp482. uo2 heat of fusion
c of 17.7 kcal/mole from l.leibowitz et al, j.nuc.mat. 39 p 115
c (1971). cladding melting point of 2098 k from m.l. picklesimer
c private communication. cladding heat of fusion of 4.9 kcal/mol
c from brassfield et al, gemp482.
c
c beginning and end of alphabeta transus are from data in
c figure iii.33 of anl7649
c isothermal zirconium alphabeta transus temperature is 1135 k
A.3
c taken from b.lustman & f.kerze "the metallurgy of zirconium"
c mcgrawhill book co., new york, 1955
c
c mixed oxide melting point was obtained from lyon et al, j. nuc.
c mat., 22 (1967) p 332
c
c phypro was coded by v.f. baston in may 1974
c modified by c.s. olsen in feb. 1975
c modified by b.w. burnham in nov. 1977
c modified by d.l. hagrman in june 1979
c Modified by PNNL, January 1997, to clean up coding and delete
c sensitivity uncertainty analysis coding
c burnup dependance for fuel melting temperature changed from
c 3.2 K/GWd/tHM to 0.5 K/GWd/tHM by K.J.Geelhood in June 2004 based on
c S.G. Popov, et al., "Thermophysical Properties of MOX and UO2 Fuels
c including the Effects of Irradiation," ORNL/TM200/351
c
c
sldus(c) = 2840.0d0  5.41395d0*c + 7.468390d3*c*c
liqdus(c) = 2840.0d0  3.21860d0*c  1.448518d2*c*c
fbu = bumtp/86.4d0
if (compmt .gt. 0.0d0) go to 10
c
ftmelt = 3113.15d0  5.0d0*fbu/10000.0d0
fdelta = 1.0d10
go to 20
c
10 continue
c1 = compmt
ftmelt = sldus(c1) + 273.15d0  5.0d0*fbu/10000.0d0
fdelta = liqdus(c1)  sldus(c1) 5.0d0*fbu/10000.0d0
c
20 continue
fhefus = 27.4d+4
c
go to (400,400,400,400,400,600,400,600,400,400,400)icm
c
400 continue
c generic Zry properties
ctmelt = 2098.15d0
chefus = 22.5d+04
go to 50
600 continue
c Zr1%Nb properties from RRCKI
ctmelt = 2133.d0
chefus = 21.0d+4
50 continue
c
c 5/7/03: ctranb, ctrane, ctranz phase transition temperatures
c not currently used in FRAPTRAN; RRCKI does have different
c transition temperatures for Zr1%Nb
wfox = deloxy + 0.0012d0
ctranb = 1094.d0 + wfox*(1.289d+03 + wfox*7.914d+05)
if(wfox .lt. 0.025d0) go to 30
ctranb = 1556.4d0 + 3.8281d+04*(wfox  0.025d0)
30 continue
ctrane = 392.46d0*((100.d0*deloxy + 0.1242807d0)**2 + 3.1417d0)
if(deloxy .lt. 4.7308937d03) go to 40
ctrane = (100.d0*deloxy + 0.12d0)*491.157d0 + 1081.7413d0
40 continue
ctranz = 1135.15d0
c
return
A.4
c
A.1.2 Fuel Specific Heat Capacity (FCP) and Fuel Enthalpy (FENTHL)
FRAPCON3.4: FCP
*deck fcp
function fcp (ftemp,facmot,fotmtl,gadoln)
c
c the function fcp is used to calculate the specific heat capacity
c of uo2, puo2, and (u,pu)o2 fuels as a function of temperature,
c fraction of fuel which is molten, puo2 content, and oxygento
c metal ratio.
c fcp = output fuel specific heat capacity (j/(kg*K))
c ftemp = input fuel meshpoint temperature (K)
c facmot = input fuel fraction molten (unitless)
c facmot = 1.0  fuel is all molten
c facmot = 0.0  fuel is all solid
c fotmtl = input oxygen to metal ratio (unitless)
c fotmtl = (atoms oxygen)/(atoms metal) if not known,
c enter 2.0
c gadoln = weight fraction of gadolinia in the fuel
c ufcp = estimated standard error for uo2 (j/(kg*K))
c (not currently returned)
c pufcp = estimated standard error for puo2 (j/(kg*K))
c (not currently returned)
c the equations used in this function are based on data from;
c (1) j.kerrisk and d.clifton, nuclear technology,16 (1972).
c (2) o.kruger and h.savage, journal of chemical physics,45
c (1968).
c the effect of oxygentometal ratio was estimated from
c equations published by;
c (3) c.affortit and j.marcon, revue internationale des hautes
c temperatures et des refractaires,7 (1970).
c the specific heat capacity of molten uo2 was taken from
c (4) l.leibowitz et al, journal of nuclear materials,39 (1970)
c fcp was originally coded by v.f.baston in march 1974.
c last modified by g.a.reymann in may 1978.
c Modification for gadolinia additions were made by DD Lanning
c in 1996.
common / phypro / ftmelt,fhefus,ctmelt,chefus,ctranb,
+ ctrane,ctranz,fdelta,bu,comp,deloxy
c the phypro common block is set by call to the subroutine
c phypro which is part of the materials properties package
c used at the inel.
c quantities contained in it used in this subcode are;
c ftmelt = fuel melting temperature (K)
c fdelta = liquidsolid coexistence temperature (K)
c comp = puo2 content (wt.%)
c when the function fcp is used alone, not as part of the
c inel code, values for ftmelt, fdelta, and comp must be
c input.
data c1u ,c2u ,c3u ,thu ,edu / 296.7 , 2.43e02, 8.745e07,
+ 535.285, 1.577e05 /
data c1pu,c2pu,c3pu,thpu,edpu/ 347.4 , 3.95e04, 3.860e07,
+ 571.0 , 1.967e05 /
data c1gd,c2gd,c3gd,thgd,edgd/ 315.86 , 4.044e02 , 0.0,
+ 348.0 , 0.0 /
cp(c1,c2,c3,th,ed,t,fotmtl) = c1*(th**2)*exp(th/t)/((t**2)*((exp(t
+h/t)1.)**2))+c2*t+(fotmtl/2.)*(c3*ed/(8.314*(t**2)))*exp(ed/
+(8.314*t))
t = ftemp
A.5
r = facmot
tm = ftmelt
fcomp = comp/100.0
fcpmol = 503.0
c fcpmol = specific heat capacity of molten fuel (j/(kg*K))
if (t.gt.(tm+fdelta)) go to 100
fcp = cp(c1u,c2u,c3u,thu,edu,t,fotmtl)*(1.0fcomp)+cp(c1pu,c2pu,c3
+pu,thpu,edpu,t,fotmtl)*fcomp
fcp = fcp*(1.0gadoln) + gadoln*cp(c1gd,c2gd,c3gd,thgd,edgd,t,
+ fotmtl)
if (t.lt.(tm0.1)) go to 110
fcp = (1.0r)*fcp+r*fcpmol
go to 110
100 fcp = fcpmol
ufcp = 2.0
pufcp = 5.6
110 continue
return
end
FRAPTRAN 1.4: FCP
*deck fcp
function fcp (ftemp,facmot,fotmtl,frden)
implicit real (ah,oz)
c
c This is a MATPRO11, Rev. 2 routine modified by PNNL for use in FRAPT
c
c the function fcp is used to calculate the specific heat capacity
c of uo2, puo2, and (u,pu)o2 fuels as a function of temperature,
c fraction of fuel which is molten, puo2 content, and oxygento
c metal ratio.
c
c fcp = output fuel specific heat capacity (j/(kg*k))
c
c ftemp = input fuel meshpoint temperature (k)
c facmot = input fuel fraction molten (unitless)
c facmot = 1.0  fuel is all molten
c facmot = 0.0  fuel is all solid
c fotmtl = input oxygen to metal ratio (unitless)
c fotmtl = (atoms oxygen)/(atoms metal) if not known,
c enter 2.0
c ufcp = estimated standard error for uo2 (j/(kg*k))
c (not currently returned)
c pufcp = estimated standard error for puo2 (j/(kg*k))
c (not currently returned)
c
c the equations used in this function are based on data from;
c (1) j.kerrisk and d.clifton, nuclear technology,16 (1972).
c (2) o.kruger and h.savage, journal of chemical physics,45
c (1968).
c
c the effect of oxygentometal ratio was estimated from
c equations published by;
c (3) c.affortit and j.marcon, revue internationale des hautes
c temperatures et des refractaires,7 (1970).
c the specific heat capacity of molten uo2 was taken from
c (4) l.leibowitz et al, journal of nuclear materials,39 (1970)
c
c fcp was originally coded by v.f.baston in march 1974.
c last modified by g.a.reymann in may 1978.
c Modified by PNNL, January 1997, to clean up coding and delete lic
A.6
c analysis and sensitivity uncertainty coding
c
c
include 'phypro.h'
c
c the phypro common block is set by call to the subroutine
c phypro which is part of the materials properties package
c used at the inel.
c quantities contained in it used in this subcode are;
c ftmelt = fuel melting temperature (k)
c fdelta = liquidsolid coexistence temperature (k)
c compmt = puo2 content (wt.%)
c when the function fcp is used alone, not as part of the
c inel code, values for ftmelt, fdelta, and compmt must be
c input.
c
data c1u ,c2u ,c3u ,thu ,edu / 296.7d0 , 2.43d02, 8.745d07,
& 535.285d0, 1.577d05 /
data c1pu,c2pu,c3pu,thpu,edpu/ 347.4d0 , 3.95d04, 3.860d07,
& 571.0d0 , 1.967d05 /
c
cp(c1,c2,c3,th,ed,t,fotmtl) =
& c1*(th**2)*exp(th/t)/((t**2)*((exp(th/t)1.d0)**2)) + c2*t
& + (fotmtl/2.d0)*(c3*ed/(8.314d0*(t**2)))*exp(ed/(8.314d0*t))
c
t = ftemp
r = facmot
tm = ftmelt
fcomp = compmt/100.0d0
fcpmol = 503.0d0
c fcpmol = specific heat capacity of molten fuel (j/(kg*k))
c
if (t .gt. (tm+fdelta)) go to 50
c
fcp = cp(c1u ,c2u ,c3u ,thu ,edu ,t,fotmtl)*(1.0d0  fcomp) +
& cp(c1pu,c2pu,c3pu,thpu,edpu,t,fotmtl)*fcomp
c
if (t .lt. (tm0.1d0)) go to 100
fcp = (1.0d0  r)*fcp + r*fcpmol
go to 100
c
50 continue
fcp = fcpmol
ufcp = 2.0d0
pufcp = 5.6d0
c
100 continue
c
return
c
end
FRAPCON3.4: FENTHL
*deck fenthl
c
function fenthl (temp,fcomp,fotmtl,ftmelt,facmot,fhefus,gadoln)
c implicit real * 8 (ah,oz)
c ********************************************************************
c this function is called by subroutine energy and computes the
c enthalpy of fuel at a point relative to zero degrees absolute
c temperature
A.7
c ********************************************************************
c input arguments
c ********************************************************************
c facmot  fraction of molten fuel
c fcomp  puo2 fraction of the fuel
c fhefus  heat of fussion of the fuel (j/kg)
c fotmtl  fuel oxygen to metal ratio
c ftmelt  fuel melting temperature (K)
c gadoln  weight fraction of gadolinia in the fuel
c temp  local temperature (K)
c ********************************************************************
c output
c ********************************************************************
c fenthl  local fuel enthalpy relative to zero degreesK (j/kg)
c ********************************************************************
c fcpmol = specific heat capacity of molten fuel ( j/(kg*K) )
data fcpmol / 503.e0 /
c the following data statements contain constants from matpro11 fcp
data c1u, c2u, c3u, thu, edu / 296.7 , 2.43e02, 8.745e07,
+ 535.285, 1.577e05 /
data c1pu,c2pu,c3pu,thpu,edpu / 347.4 , 3.95e04, 3.860e07,
+ 571.0 , 1.967e05 /
data c1gd,c2gd,c3gd,thgd,edgd / 315.86 , 4.044e2, 0.0 ,
+ 348.0 , 0.0 /
c the following equation is the integral of the fuel specific heat
c with respect to temperature
cpdt(c1,th,c2,otm,ed,t,c3) = c1*th*(1./(exp(th/t)1.))+c2*t*t/2.e0
++c3*otm*exp(ed/(t*8.314e0))/2.e0
tx = temp
if (tx.gt.ftmelt) tx = ftmelt
fenthl = cpdt(c1u,thu,c2u,fotmtl,edu,tx,c3u)*(1.e0fcomp)
fenthl = fenthl+cpdt(c1pu,thpu,c2pu,fotmtl,edpu,tx,c3pu)*fcomp
fenthl = fenthl*(1gadoln)+
&gadoln*cpdt(c1gd,thgd,c2gd,fotmtl,edgd,tx,c3gd)
if (temp.le.ftmelt2.e0) go to 100
fenthl = fenthl+fhefus*facmot
if (temp.le.(ftmelt+2.)) go to 100
fenthl = fenthl+(tempftmelt)*fcpmol
100 continue
return
end
FRAPTRAN 1.4: FENTHL
*deck fenthl
c
function fenthl (temp, fcomp, fotmtl, ftmelt, facmot, fhefus)
c
implicit real (ah,oz)
c
c
c *********************************************************************
c
c this function is called by subroutine energy and computes the
c enthalpy of fuel at a point relative to zero degrees absolute
c temperature
c
c *********************************************************************
c input arguments
c *********************************************************************
c
c facmot  fraction of molten fuel
A.8
c fcomp  puo2 fraction of the fuel
c fhefus  heat of fussion of the fuel (j/kg)
c fotmtl  fuel oxygen to metal ratio
c ftmelt  fuel melting temperature (K)
c temp  local temperature (K)
c
c *********************************************************************
c output
c *********************************************************************
c
c fenthl  local fuel enthalpy relative to zero degreesK (j/kg)
c
c *********************************************************************
c
c fcpmol = specific heat capacity of molten fuel ( j/(kg*K) )
c
data fcpmol / 503.d0 /
c
c the following data statements contain constants from matpro11 fcp
c
data c1u, c2u, c3u, thu, edu / 296.7d0 , 2.43d02, 8.745d07,
& 535.285d0, 1.577d05 /
data c1pu,c2pu,c3pu,thpu,edpu / 347.4d0 , 3.95d04, 3.860d07,
& 571.0d0 , 1.967d05 /
c
c the following equation is the integral of the fuel specific heat
c with respect to temperature
c
cpdt(c1,th,c2,otm,ed,t,c3) =
& c1*th * ( 1.d0 / (exp(th/t)1.d0))
& + c2 * t * t / 2.d0
& + c3 * otm * exp(ed /(t * 8.314d0) )/2.d0
c
tx = temp
if(tx .gt. ftmelt) tx = ftmelt
c
fenthl = cpdt(c1u ,thu ,c2u ,fotmtl ,edu
& ,tx ,c3u ) * ( 1.d0  fcomp )
c
if( fcomp .gt. 0.d0 ) fenthl = fenthl + cpdt(c1pu ,thpu ,c2pu
& ,fotmtl ,edpu ,tx ,c3pu) * fcomp
c
if( temp .le. ftmelt  2.d0 ) go to 100
c
c
c
fenthl = fenthl + fhefus * facmot
c
if(temp.le.(ftmelt+2.d0))go to 100
c
c
fenthl = fenthl + (temp  ftmelt) * fcpmol
100 continue
c
return
end
A.1.3 Fuel Thermal Conductivity (FTHCON)
FRAPCON3.4: FTHCON
*deck fthcon
A.9
subroutine fthcon (ftemp,fraden,fotmtl,con,burnup
+ ,gadoln,imox)
c
c fthcon calculates the fuel thermal conductivity and its
c derivative with respect to temperature as a function of
c temperature, density, composition and burnup.
c
c UO2 Fuel (IMOX = 0)
c
c The equation used in this subroutine is that proposed by
c staff at NFI, Japan, at the May 1997 ANS Topical Meeting on
c Light Water Reactor Fuel performance in Portland, OR: (Ohira,
c K., and N.Itagaki, 1997. "Thermal Conductivity Measurements
c of High Burnup UO2 Pellet and a Benchmark Calculation of Fuel
c Center Temperature", proceedings pp. 541549. Applies to UO2.
c
c MOX:
c
c Option number 1 (IMOX = 1)
c
c The 100% dense solid MOX fuel thermal conductivity formulation is based
c on a combination of the Duriez stoichiometrydependent correlation,
c derived from diffusivity measurements on unirradiated fuel pellets
c (C.Duriez, et al, J.Nuclear Materials 277, 143158 2000) and the burnup
c degradation conatined in a modified version of the NFI fuel thermal
c conductivity model
c
c Option number 2 (IMOX = 2)
c
c The MOX fuel thermal conductivity formulation is based
c on the OECD Halden Reactor Project report "Thermal Performance of
c of High Burnup Fuel Inpile Temperature Data and Analysis"
c W.Wiesnack, T. Tverberg, Proceedings of the 2000 International
c Topical Meeting on LWR Fuel Performance
c
c
c burnup = current local burnup (MWd/MTU)
c con = output fuel thermal conductivity (W/(m*K))
c ftemp = current fuel ring temperature (K)
c fraden = input fuel density (ratio of actual density to
c theoretical density)
c fotmtl = input oxygen to metal ratio of fuel (atoms oxygen/
c atoms metal)
c gadoln = input weight fraction of gadolinia in the fuel
c
c the following inputs are by common block
c comp = input puo2 content of fuel (percent puo2
c in total fuel weight)
c bu = input burnup (mws/kgu)
c emflag(12) = input switch for evaluation model. if this
c variable is equal to 1.0, the matpro model for
c fuel thermal conductivity is replaced by the
c subcode emfton
c
common /uncert/sigftc, sigftex, sigfgr
& ,sigswell ,sigcreep ,siggro ,sigcor ,sigh2
common / phypro / ftmelt,fhefus,ctmelt,chefus,ctranb,
+ ctrane,ctranz,fdelta,bu,comp,deloxy
c
include 'lacmdl.h'
data on / 1 /,
+ off / 2 /,
+ locidx / 12 /
A.10
c
c find constants
c
frpu = comp/100.
t = ftemp
c
c Burnup in GWd/MTU
c
bug = burnup/1000.0
c
if(imox.eq.0) then
c
c NFI formula (Ohira & Itagaki, ANS LWR Fuel perf. Topical mtg. 1997)
c MODIFIED in January 2002 to raise lowburnup thermal conductivity
c at low temperature and lower thermal conductivity at very high temp.
c
h = 1/(1.0+396.0*exp(6380.0/t))
rphonon= 1.0/(0.0452+0.000246*t + 1.0*0.00187*bug+1.1599*gadoln
& + (1.00.9*exp(0.04*bug))*0.038*bug**0.28*h)
elect = (3.50e9/t**2)*exp(16361/t)
base = rphonon + elect
c
c fm is the Lucuta porosity correction factor(applied to 100% TD fuel)
c
fm = fraden/(1.0 + 0.5*(1.0fraden))
c
c NFI base equation is for 95% TD fuel, so multiply by 1.079 to
c raise to 100% TD fuel conductivity, then multiply by fm
c
con = base*fm*1.079
c
else if(imox.eq.1) then
c
c Using the Duriez/NFI Mod correlation combination
c
c base term for MOX
c where X = deviation from stoichiometry (2O/M)
fm = 1.0789*fraden/(1.0+0.5*(1.0fraden))
c fm is multiplied by 1.0789 to account for 95% TD
c Porosity correction is Lucuta correction, not MaxwellEuken
c as proposed by Duriez et al.
x = 2.0fotmtl
ax=2.85*x+0.035
cx=(2.867.15*x)*1.0e4
c
h = 1/(1.0+396.0*exp(6380.0/t))
rphonon = 1.0/(ax + cx*t + 0.00187*bug+1.1599*gadoln
&+ (1.00.9*exp(0.04*bug))*0.038*bug**0.28*h)
elect = (1.50e9/t**2)*exp(13520/t)
base = rphonon + elect
con = base*fm
c
else if(imox.eq.2) then
c
c Using the Halden correlation
c
tc=t273.15
tco=min(1650.0,tc)
buguo2=bug*0.8815
fm = 1.0789*fraden/(1.0+0.5*(1.0fraden))
base=0.92/(0.1148+0.004*buguo2+1.1599*gadoln+
& 2.475e4*(1.00.00333*buguo2)*tco)+
& 0.0132*exp(0.00188*tc)
A.11
con=base*fm
c
c If IMOX.ne.0,1,2 then stop the calculations
c
else
stop 'fthcon  IMOX not within bounds'
end if
c
c find uncertainty
if(sigftc.gt.0.0) con = con*(1.0+sigftc*0.088)
if(sigftc.lt.0.0) con = con/(1.0sigftc*0.088)
if(imox.eq.0) then
if(t.lt.ftmelt) then
ucon = 0.2*(1.0+abs(2.0fotmtl)*10.)
else
ucon = con/2.0
endif
else
if(t.le.1800.0) then
ucon = 0.07*con
else
frac=(t1800.0)/(3100.01800.0)*(0.200.07)+0.07
ucon=frac*con
endif
endif
if (emflag(locidx).eq.on) call emfton (ftemp,fraden,ftmelt,con)
return
end
FRAPTRAN 1.4: FTHCON
*deck fthcon
c
subroutine fthcon (ftemp, fraden, fotmtl, con, dkdt, burnup,
& gadoln, compmt)
c
implicit real (ah, oz)
c
c FTHCON calculates the UO2 fuel thermal conductivity
c as a function of temperature, density, composition and burnup.
c
c The equation used in this subroutine is that proposed by
c staff at NFI, Japan, at the May 1997 ANS Topical Meeting on
c Light Water Reactor Fuel Performance in Portland, Oregon.
c reference: Ohira and Itakaki, 1997. "Thermal Conductivity
c Measurements of High Burnup UO2 Pellet and a Benchmark
c Calculation of Fuel Center Temperature," proceedings pg. 541549.
c
c The NFI formula was modified in January 2002 by PNNL to raise
c lowburnup thermal conductivity at low temperature and lower thermal
c conductivity at very high temperature.
c The PNNL reference is: Lanning and Beyer. 2002. "Revised UO2 Thermal
c Conductivity for FRAPCON3 NRC Fuel Performance Codes," in Proceedings of
c ANS Annual Meeting, June 913, 2002, Hollywood, Florida.
c
c This is the same formulation used in FRAPCON3.2 (February 2003),
c minus the formulations for MOX fuel.
c
c burnup = current local burnup (MWd/MTU)
c con = output fue thermal conductivity (W/(m*K))
A.12
c ftemp = current fuel ring temperature (K)
c fraden = input fuel density (ratio of actual to theoretical)
c fotmtl = input oxygen to metal ratio of fuel
c (atoms oxygen/atoms metal)
c gadoln = input weight fraction of gadolinia in the fuel
c compmt = input weight % of Pu in Pu and U
c
c convert burnup to GWd/MTU
bug = burnup/1.d3
if(bug .lt. 1.d10) bug = 0.001
c use modified NFI model for UO2 and Duriez/NFI Mod model for MOX
if(compmt.gt.0.0) go to 100
c UO2 Model
c
c NFI formula (Ohira & Itagaki, ANS LWR Fuel perf. Topical mtg. 1997)
c MODIFIED in January 2002 to raise lowburnup thermal conductivity
c at low temperature and lower thermal conductivity at very high temp.
c
c calculate terms
c temperature dependence of annealing of irradiation effects
h = 1.0/(1.0 + 396.0*exp(6380.0/ftemp))
c phonon term
rphonon = 1.0/(0.04520 + 0.000246*ftemp + 0.00187*bug
& + (1.0  0.9*exp(0.04*bug))*(0.038*bug**0.28)*h)
c electronic term
elect = (3.50d9/ftemp**2)*exp(16361./ftemp)
c base is conductivity before apply porosity correction
base = rphonon + elect
c apply lucuta porosity correction factor (applied to 100% TD fuel)
fm = fraden/(1.0 + 0.5*(1.0fraden))
c NFI base equation is for 95% TD fuel, so multiply by 1.079 to
c raise to 100% TD fuel conductivity, then multiply by fm
con = base*fm*1.079
go to 200
c
c MOX Model
c
100 continue
c Using the Duriez/NFI Mod correlation combination
c
c base term for MOX
c where X = deviation from stoichiometry (2O/M)
fm = 1.0789*fraden/(1.0+0.5*(1.0fraden))
c fm is multiplied by 1.0789 to account for 95% TD
c Porosity correction is Lucuta correction, not MaxwellEuken
c as proposed by Duriez et al.
x = 2.0fotmtl
ax=2.85*x+0.035
cx=(2.867.15*x)*1.0e4
c
h = 1/(1.0+396.0*exp(6380.0/ftemp))
rphonon = 1.0/(ax + cx*ftemp + 0.00187*bug+1.1599*gadoln
&+ (1.00.9*exp(0.04*bug))*0.038*bug**0.28*h)
elect = (1.50e9/ftemp**2)*exp(13520/ftemp)
base = rphonon + elect
con = base*fm
c
200 continue
return
end
A.1.4 Fuel Emissivity (FEMISS)
A.13
FRAPCON3.4: FEMISS
*deck femiss
function femiss (ftemp)
c
c femiss calculates fuel emissivity as a function of temperature.
c ftemp = input fuel temperature (k)
c femiss = output fuel emissivity (unitless)
c data used to develop the model are
c held and wilder, journ. amer. ceram. soc.
c vol. 52, (1969)
c cabannes, et at, c. r. acad. sci., paris, ser. b
c (1967)
c femiss was coded by r. e. mason in october 1978.
femiss = 0.78557+1.5263e05*ftemp
c the following calculations provide the upper and lower bounds.
c the upper and lower bounds are not an output unless the user
c desires to modify the subcode appropriately.
c femisu is the upper bound
femisu = femiss*(1.+0.06796)
c femisl is the lower bound
femisl = femiss*(1.0.06796)
100 continue
return
end
FRAPTRAN 1.4: FEMISS
*deck femiss
function femiss (ftemp)
implicit real (ah,oz)
c
c This is a MATPRO11, Rev. 2 routine modified by PNNL for use in FRAPT
c
c femiss calculates fuel emissivity as a function of temperature.
c
c ftemp = input fuel temperature (k)
c femiss = output fuel emissivity (unitless)
c
c data used to develop the model are
c held and wilder, journ. amer. ceram. soc.
c vol. 52, (1969)
c cabannes, et at, c. r. acad. sci., paris, ser. b
c (1967)
c
c femiss was coded by r. e. mason in october 1978.
c Modified by PNNL, January 1997, to clean up coding and delete lic
c analysis and sensitivity uncertainty coding
c
femiss = 0.78557d0 + 1.5263d05 * ftemp
c
c the following calculations provide the upper and lower bounds.
c the upper and lower bounds are not an output unless the user
c desires to modify the subcode appropriately.
c femisu is the upper bound
femisu = femiss * (1.d0 + 0.06796d0)
c femisl is the lower bound
femisl = femiss * (1.d0  0.06796d0)
c
return
c
A.14
end
A.1.5 Fuel Thermal Expansion (FTHEXP)
FRAPCON3.4: FTHEXP
*deck fthexp
function fthexp (ftemp,facmot)
c
c the subprogram fthexp is used to calculate the strain of the
c fuel caused by thermal expansion.
c thermal expansion strain is calculated for uo2,pu02, or (u,pu)02
c as a function of temperature, puo2 content, and fraction of
c the fuel which is molten.
c the correlations were derived with the assumption that the
c thermal expansion strain is zero at 300k.
c fthexp = output fuel strain due to thermal expansion (unitless)
c ufthex = estimated standard error of fthexp (per cent)
c (not currently returned).
c ftemp = input fuel temperature (k)
c facmot = input fuel fraction which is molten (unitless)
c facmot = 1.0  fuel is all molten
c facmot = 0.0  fuel is all solid
c the correlations used in this subroutine are based on data from;
c (1) p.j.baldock et al, journal of nuclear materials, 18 (1966)
c (2) n.h.brett and l.e.russel, proceedings of the 2nd
c international conference on plutonium metallurgy, grenoble,
c france (1960) pp 397410.
c (3) m.d.burdock and h.s.parker, journal of the american
c ceramic society,39 (1956) pp 181187.
c (4) j.a.christensen, journal of the american ceramic society,
c 46 (1963) pp 607608.
c (5) j.b.conway et al, transactions of the american nuclear
c society,6 (1963).
c (6) f.gronvold, journal of inorganic and nuclear chemistry,1
c (1955) pp 357370.
c (7) m.hoch and a.c.momin, high temperatureshigh pressures,1
c (1969) pp 401407.
c (8) c.p.kempter and r.o.elliott, the journal of chemical
c physics,30 (1958) pp 15241526.
c (9) w.a.lambertson and j.h.handwerk, anl5053 (1956).
c (10)m.tokar et al, nuclear technology, 17 (1973) pp 147152.
c fthexp was originally coded by v.f.baston in march 1974
c modified by c.s.olsen in feb. 1975
c last modified by g.a.reymann in july 1978
common /uncert/sigftc, sigftex, sigfgr
& ,sigswell ,sigcreep ,siggro ,sigcor ,sigh2
common / phypro / ftmelt,fhefus,ctmelt,chefus,ctranb,
+ ctrane,ctranz,fdelta,bu,comp,deloxy
c the phypro common block is set by call to the subroutine
c phypro which is part of the materials properties package
c used at the inel.
c quantities contained in it used in this subcode are;
c comp = puo2 content (wt. :)
c ftmelt = fuel melting temperature (k)
c fdelta = liquidsolid coexistence temperature range (k)
c when the function fthexp is used alone, not as part of
c the inel code, values for comp, ftmelt, and fdelta must
c be input.
c
c c1u, c2u, c3u, edu parameters updated by W.G. Luscher 101707
c
A.15
data c1u ,c2u ,c3u ,edu / 9.8e06, 2.61e03, 3.16e01, 1.32e19 /
data c1pu,c2pu,c3pu,edpu/ 9.0e06, 2.7e03, 7.0e02, 7.0e20 /
ftx(c1,c2,c3,ed,bk,t) = c1*tc2+c3*exp(ed/(bk*t))
t = ftemp
bk = 1.38e23
c bk = boltzmann's constant (j/k)
fcomp = comp/100.0
if (t.gt.ftmelt) go to 100
fthexp = ftx(c1u,c2u,c3u,edu,bk,t)*(1.0fcomp)+ftx(c1pu,c2pu,c3pu,
+edpu,bk,t)*fcomp
go to 120
100 fthexm = ftx(c1u,c2u,c3u,edu,bk,ftmelt)*(1.0fcomp)+ftx(c1pu,c2pu,
+c3pu,edpu,bk,ftmelt)*(fcomp)
if (t.ge.(ftmelt+fdelta)) go to 110
fthexp = fthexm+.043*facmot
go to 120
110 fthexp = fthexm+.043+3.6e05*(t(ftmelt+fdelta))
120 continue
c add on uncertainty
if(sigftex.gt.0.0) fthexp = fthexp*(1.0+sigftex*0.103)
if(sigftex.lt.0.0) fthexp = fthexp/(1.0sigftex*0.103)
ufthex = 10.0
return
end
FRAPTRAN 1.4: FTHEXP
*deck fthexp
c
function fthexp (ftemp,facmot)
c
implicit real (ah,oz)
c
c This is a MATPRO11, Rev. 2 routine modified by PNNL for use in FRAPT
c
c the subprogram fthexp is used to calculate the strain of the
c fuel caused by thermal expansion.
c thermal expansion strain is calculated for uo2,pu02, or (u,pu)02
c as a function of temperature, puo2 content, and fraction of
c the fuel which is molten.
c the correlations were derived with the assumption that the
c thermal expansion strain is zero at 300k.
c
c fthexp = output fuel strain due to thermal expansion (unitless)
c ftemp = input fuel temperature (k)
c facmot = input fuel fraction which is molten (unitless)
c facmot = 1.0  fuel is all molten
c facmot = 0.0  fuel is all solid
c
c the correlations used in this subroutine are based on data from;
c (1) p.j.baldock et al, journal of nuclear materials, 18 (1966)
c (2) n.h.brett and l.e.russel, proceedings of the 2nd
c international conference on plutonium metallurgy, grenoble,
c france (1960) pp 397410.
c (3) m.d.burdock and h.s.parker, journal of the american
c ceramic society,39 (1956) pp 181187.
c (4) j.a.christensen, journal of the american ceramic society,
c 46 (1963) pp 607608.
c (5) j.b.conway et al, transactions of the american nuclear
c society,6 (1963).
c (6) f.gronvold, journal of inorganic and nuclear chemistry,1
c (1955) pp 357370.
c (7) m.hoch and a.c.momin, high temperatureshigh pressures,1
A.16
c (1969) pp 401407.
c (8) c.p.kempter and r.o.elliott, the journal of chemical
c physics,30 (1958) pp 15241526.
c (9) w.a.lambertson and j.h.handwerk, anl5053 (1956).
c (10)m.tokar et al, nuclear technology, 17 (1973) pp 147152.
c
c fthexp was originally coded by v.f.baston in march 1974
c modified by c.s.olsen in feb. 1975
c last modified by g.a.reymann in july 1978
c Modified by PNNL, January 1997, to clean up coding and delete lic
c analysis and sensitivity uncertainty coding
c
include 'phypro.h'
c
c the phypro common block is set by call to the subroutine
c phypro which is part of the materials properties package
c used at the inel.
c quantities contained in it used in this subcode are;
c compmt = puo2 content (wt. :)
c ftmelt = fuel melting temperature (k)
c fdelta = liquidsolid coexistence temperature range (k)
c when the function fthexp is used alone, not as part of
c the inel code, values for compmt, ftmelt, and fdelta must
c be input.
c
c c1u, c2u, c3u, and edu parameters updated by W.G. Luscher 101707
c
data c1u ,c2u ,c3u ,edu / 9.8d06, 2.61d03, 3.16d01, 1.32d19 /
data c1pu,c2pu,c3pu,edpu/ 9.0d06, 2.7d03, 7.0d02, 7.0d20 /
c bk is Boltzmann's constant (J/k)
data bk / 1.38d23 /
c
ftx (c1,c2,c3,ed,bk,t) = c1 * t  c2 + c3 * exp(ed/(bk * t))
c
t = ftemp
fcomp = compmt/100.0d0
c
if(t .gt. ftmelt) go to 25
c
fthexp = ftx(c1u ,c2u ,c3u ,edu ,bk,t) * (1.0d0  fcomp)
& + ftx(c1pu,c2pu,c3pu,edpu,bk,t) * fcomp
go to 100
c
25 continue
fthexm = ftx(c1u ,c2u ,c3u ,edu ,bk,ftmelt) * (1.0d0  fcomp)
& + ftx(c1pu,c2pu,c3pu,edpu,bk,ftmelt) * fcomp
c
if(t .ge. (ftmelt+fdelta)) go to 75
c
fthexp = fthexm + 0.043d0 * facmot
go to 100
c
75 continue
fthexp = fthexm + 0.043d0 + 3.6d05 * (t (ftmelt + fdelta))
c
100 continue
c
return
end
A.1.6 Fuel Densification (FUDENS)
A.17
FRAPCON3.4: FUDENS
*deck fudens
function fudens (ftemp,bu,fdens,rsntr,tsint,comp,prvden)
c
c fudens calculates irradiationinduced densification.
c fudens = output fuel dimensional change (%)
c ftemp = input fuel temperature (k)
c bu = input burnup (mws/kgu)
c fdens = input fuel density (kg/m**3)
c rsntr = input maximum density change determined by a resintering
c test of 1973 k for 24 hours (kg/m**3)
c tsint = input fuel sintering temperature (k)
c comp = input plutonia content (weight percent)
c prvden = input total densification from previous time step (%)
c fudens was developed and programmed by c. s. olsen (january 1975).
c updated and corrected by b. w. burnham (october 1975).
c fudens was modified by r. e. mason (november 1978).
dimension c(2),b(5)
data c / 10.96, 11.46/
data b / 3.0, 1.00, 3.0, 2.00, 35.00 /
dlen2(alen,bu,abu) = b(1)+alen+b(2)*exp(b(3)*(bu+abu))+b(4)*exp(
+b(5)*(bu+abu))
dlen3(bu) = b(2)*b(3)*exp(b(3)*bu)b(4)*b(5)*exp(b(5)*bu)
fbu = bu*1.157e05
ts = tsint2.7315e02
c if rsntr or tsint is not defined by user, the default value is
c tsint = 1873 k.
if (tsint.le.0.0) ts = 1600.0
roth = c(1)*c(2)/(0.01*comp*c(1)+(1.00.01*comp)*c(2))
de = fdens/(roth*10.0)
if (rsntr .gt. 0.0) dlen1 = 100.0 * rsntr/(3.0 * fdens)
if ((ftemp.ge.1000.).and.(rsntr.le.0.)) dlen1 = 66.6*(100.0de)/(t
+s1180.0)
if ((ftemp.lt.1000.).and.(rsntr.le.0.)) dlen1 = 22.2*(100.0de)/(t
+s1180.0)
x3 = 0.0
x4 = 1.0
al1 = dlen1
al3 = 3.0al1
al4 = 0.0
if (al3.le.4.27e03) go to 130
do 100 i=1,6
y2 = dlen2(al3,x4,al4)
y1 = dlen2(al3,x3,al4)
if (y1*y2.le.0.) go to 110
x3 = x4
x4 = x4+1.0
if (i.eq.6) go to 160
100 continue
110 continue
x1 = x3
do 120 j=1,50
x = x1dlen2(al3,x1,al4)/dlen3(x1)
err = abs((xx1)*100.0/x)
if (err.le.2.0e04) go to 140
x1 = x
120 continue
130 al3 = 2.996
al2 = 5.384
go to 150
140 al2 = x
150 continue
A.18
fuden = dlen2(al3,fbu,al2)
if (bu.lt.1728) fuden = 0.0
go to 170
160 print 180
fuden = 0.0
170 continue
if (abs(fuden).le.abs(prvden)) fudens = 0.0
if (abs(fuden).gt.abs(prvden)) fudens = fudenprvden
return
180 format (1x,/' no roots found between 0 and 6000 mws/mt uo2')
end
A.1.7 Fuel Swelling (FSWELL)
FRAPCON3.4: FSWELL
*deck fswell
subroutine fswell (fdens,comp,bu,bul,ftemp,soldsw)
c
c comp = input plutonia content (weight percent)
c fdens = input initial density of the fuel (kg/m**3)
c ftemp = input temperature of the fuel ring (k)
c bu = input burnup to end of time step (mws/kgu)
c bul = input burnup to end of last time step (mws/kgu)
c soldsw = output fuel swelling due to solid fission products
c (fraction)
c fswell was developed and programmed by r. e. mason  june 1978.
c fswell was revised July 1995 by G. A. Berna per D. D. Lanning
c recommendations in ltr to L. Siefken, May 4, 1995
c fswell was revised October 2008 by K.J. Geelhood
c  program 
common /uncert/sigftc, sigftex, sigfgr
& ,sigswell ,sigcreep ,siggro ,sigcor ,sigh2
bus = fdens*2.974e+10*(bubul)
soldsw = bus*(2.315e23+sigswell*2.315e24)
if(bu.ge.6912000.0) soldsw = bus*(3.211e23+sigswell*3.211e24)
return
end
A.2 Cladding Material Properties
A.2.1 Cladding Specific Heat (CCP)
FRAPCON3.4: CCP
*deck ccp
function ccp (ctemp)
c
c ccp calculates the specific heat at constant pressure
c for zircaloys
c ccp = output cladding specific heat at constant
c pressure (j/kgk)
c ctemp = input cladding temperature (k)
c this code is based on data from
c (1) c. r. brooks and e. e. stansbury, "the specific heat 0f
c of zircaloy2 from 50 to 700 c", journal of nuclear
c materials 18 (1966) p 233
c (2) e. a. eldridge and h. w. deem, specific heats and heats
c of transformation of zircaloy2 and low nickel zircaloy2
A.19
c bmi1803 (may 31, 1967)
c conversion from j/(kg*k) to cal/(gm*c) is
c 2.39006d4 (cal/(gm*c))/(j/(kg*k))
c estimated standard error of the prediction of
c ccp for the specific heat of zircaloy cladding
c samples  (sum of squared residuals/(number of
c residualsdegrees of freedom))**0.5  is
c (1) for temperature less than 1090 k, 10j/kgk)
c (2) for temperature 1090 k to 1300 k, 25j/kgk)
c (3) for temperature above 1300 k , 100j/kgk)
c ccp coded by r. l. miller october 1974
c modified by d. l. hagrman may 1976
include 'lacmdl.h'
dimension cpdata(26)
data cpdata/ 281., 300., 302., 400., 331.,640.,
+ 375.,1090., 502.,1093., 590.,1113., 615.,1133.,
+ 719.,1153., 816.,1173., 770.,1193., 619.,1213.,
+ 469.,1233., 356.,1248. /
data npcp, iu / 13,1/
data on / 1 /,
+ off / 2 /,
+ locidx / 4 /
if (emflag(locidx).eq.on) go to 110
if (ctemp.ge.1248.0) go to 100
ccp = polate(cpdata,ctemp,npcp,iu)
go to 120
100 ccp = 356.
go to 120
110 ccp = emccp(ctemp)
120 continue
return
end
FRAPTRAN 1.4: CCP
*deck ccp
function ccp (ctemp, icm)
implicit real (ah,oz)
c
c This is a MATPRO11, Rev. 2 routine modified by PNNL for use in FRAP
c
c ccp calculates the specific heat at constant pressure
c for zircaloys
c
c ccp = output cladding specific heat at constant
c pressure (j/kgk)
c ctemp = input cladding temperature (k)
c
c this code is based on data from
c (1) c. r. brooks and e. e. stansbury, "the specific heat 0f
c of zircaloy2 from 50 to 700 c", journal of nuclear
c materials 18 (1966) p 233
c (2) e. a. eldridge and h. w. deem, specific heats and heats
c of transformation of zircaloy2 and low nickel zircaloy2
c bmi1803 (may 31, 1967)
c
c conversion from j/(kg*k) to cal/(gm*c) is
c 2.39006d4 (cal/(gm*c))/(j/(kg*k))
c
c estimated standard error of the prediction of
c ccp for the specific heat of zircaloy cladding
c samples  (sum of squared residuals/(number of
A.20
c residualsdegrees of freedom))**0.5  is
c (1) for temperature less than 1090 k, 10j/kgk)
c (2) for temperature 1090 k to 1300 k, 25j/kgk)
c (3) for temperature above 1300 k , 100j/kgk)
c
c ccp coded by r. l. miller october 1974
c modified by d. l. hagrman may 1976
c Modified by PNNL, January 1997, to clean up coding, including
c removed licensing assistance coding and sensitivity uncertainty
c coding
c
c Zry properties
dimension cpdata(28)
data cpdata/279.d0,290.d0,
& 281.d0, 300.d0, 302.d0, 400.d0, 331.d0, 640.d0,
& 375.d0,1090.d0, 502.d0,1093.d0, 590.d0,1113.d0, 615.d0,1133.d0,
& 719.d0,1153.d0, 816.d0,1173.d0, 770.d0,1193.d0, 619.d0,1213.d0,
& 469.d0,1233.d0, 356.d0,1248.d0 /
data npcp, iu / 14 , 1 /
c
c Zr1%Nb properties
c low heat rate (<1000K/s)
dimension cpwwer(24)
data cpwwer/345.0d0,280.d0,
& 360.d0, 473.d0, 370.d0, 573.d0, 380.d0, 673.d0,
& 383.d0, 773.d0, 385.d0, 873.d0, 448.d0, 883.d0, 680.d0, 973.d0,
& 816.d0,1025.d0, 770.d0,1073.d0, 400.d0,1153.d0, 392.d0,1173.d0/
data npww / 12 /
c high heat rate (>1000K/s)
dimension cpwwr2(26)
data cpwwr2/412.6d0,1100.d0,
& 420.d0,1110.d0, 480.d0,1120.d0, 600.d0,1134.d0,
& 1000.d0,1142.d0, 1400.d0,1150.d0,1600.d0,1155.d0,1400.d0,1161.d0,
& 1000.d0,1168.d0, 600.d0,1177.d0, 400.d0,1180.d0, 360.d0,1190.d0,
& 348.d0,1200.d0/
data npw2 / 13 /
c
go to (400,400,400,400,400,600,400,600,400,400,400)icm
c
400 continue
c generic Zry
ccp = 279.d0
if (ctemp.ge.290.d0 .and. ctemp.le.1248.d0)
& ccp = polate (cpdata, ctemp, npcp, iu)
if (ctemp .ge. 1248.d0) ccp = 356.d0
return
c
600 continue
c RRCKI Zr1%Nb property
c
c This block calculates the specific heat at constant pressure for
c H1alloy cladding. If iwwer = 1 then data is derived from
c Volkov B.Yu. et. al."Material property library for H1alloy
c cladding", Preprint IAE4941/11, 1989. This property are induced
c by low heating rate. If iwwer = 2 then data is derived from
c Ljusternik V.E. et. al."Experimental research of zirconium reactor
c materials thermal properties:110alloy". M.,J. High Temperature
c Thermal Physics." v.31 n.4, 1993. This property are induced by
c highrate heating>=1000 K/s.
c
c 5/5/03: default to iwwer=1 until can set this flag up
iwwer = 1
c low heat rate
A.21
if(iwwer.eq.2) go to 601
if(ctemp .lt. 1173.d0) ccp = polate(cpwwer,ctemp,npww,iu)
if(ctemp.ge.1173.d0) ccp = 392.d0
return
c
601 continue
c high heat rate
ccp = 237.5d0+15.91d2*ctemp
if(ctemp.ge.1100.d0 .and. ctemp.le.1200.d0)
& ccp = polate(cpwwr2,ctemp,npw2,iu)
if(ctemp.gt.1200.d0) ccp = 199.7d0+12.364d2*ctemp
return
c
end
A.2.2 Cladding Thermal Conductivity (CTHCON)
FRAPCON3.4: (CTHCON)
*deck cthcon
subroutine cthcon (ctemp,time,flux,coldw,ccon)
c
c cthcon calculates cladding thermal conductivity as a function
c of temperature, time, flux, and cold work
c ccon = output thermal conductivity of zircaloy4 (w/mk)
c ctemp = input cladding meshpoint temperature (k)
c time = input time at temperature and flux (s)
c flux = input fast neutron flux ((neutrons/m**2)/s)
c coldw = input cold work (unitless ratio of areas)
c the equation used in this subcode is based on data from
c w.k. anderson, c.j. beck, a.r. kephart and j.s. theilacker
c astmstp314, 1962, pp 62  93
c j.n. chirigos, c. kass, w.s. kirk and g.j. salvaggio
c fuel element fabrication, academic press, 1961, pp 19  55
c a.d. feith, gemp669 (oct 1966)
c c.f. lucks and h.w. deem, bmi1273 (1958) pp 79
c a.e. powers, kapl2146 (1961)
c d.b. scott, wcap326941 (1965) pp 59
c this version of cthcon does not use time, flux or coldwork
c as parameters in calculation of zircaloy thermal conductivity
c one standard deviation of this function = 1.01 w/mk
c this version of cthcon was developed by r.l. miller, dec 1975
c last modified by e.r. carlson in june 1978.
include 'lacmdl.h'
data on / 1 /,
+ off / 2 /,
+ locidx / 7 /
if (emflag(locidx).eq.on) go to 100
ccon = 7.511+ctemp*(2.088e2+ctemp*(1.450e5+ctemp*7.668e09))
go to 110
100 call emcton (ctemp,time,flux,coldw,ccon)
110 continue
return
end
FRAPTRAN 1.4: (CTHCON)
*deck cthcon
subroutine cthcon (ctemp, time, flux, coldw, ccon, cdkdt, icm)
implicit real (ah,oz)
c
A.22
c This is a MATPRO11, Rev. 2 routine modified by PNNL for use in FRAPT
c
c cthcon calculates cladding thermal conductivity as a function
c of temperature, time, flux, and cold work
c
c ccon = output thermal conductivity of zircaloy4 (w/mk)
c cdkdt = output derivative of thermal conductivity with
c respect to temperature
c
c ctemp = input cladding meshpoint temperature (k)
c time = input time at temperature and flux (s)
c flux = input fast neutron flux ((neutrons/m**2)/s)
c coldw = input cold work (unitless ratio of areas)
c
c go to appropriate model for cladding type
go to (400,400,400,400,400,600,400,600,400,400,400) icm
c
400 continue
c Generic Zircaloy property from MATPRO
c
c the equation used in this subcode is based on data from
c w.k. anderson, c.j. beck, a.r. kephart and j.s. theilacker
c astmstp314, 1962, pp 62  93
c j.n. chirigos, c. kass, w.s. kirk and g.j. salvaggio
c fuel element fabrication, academic press, 1961, pp 19  55
c a.d. feith, gemp669 (oct 1966)
c c.f. lucks and h.w. deem, bmi1273 (1958) pp 79
c a.e. powers, kapl2146 (1961)
c d.b. scott, wcap326941 (1965) pp 59
c
c this version of cthcon does not use time, flux or coldwork
c as parameters in calculation of zircaloy thermal conductivity
c
c one standard deviation of this function = 1.01 w/mk
c
c this version of cthcon was developed by r.l. miller, dec 1975
c last modified by e.r. carlson in june 1978.
c Modified by PNNL, January 1997, to clean up coding and delete
c licening analysis and sensitivity uncertainty coding
c
ccon = 7.511d0 + ctemp*(2.088d2 + ctemp * (1.450d5 +
& ctemp*7.668d09))
cdkdt = 2.088d2 + ctemp*(2.9d5 + ctemp*2.3d8)
return
c
600 continue
c E110 property as coded by RRCKI and provided to PNNL
c
c This code block calculates thermal conductivity (w/(m*k)) and
c derivative of thermal conductivity with respect to temperature
c (w/(m*k*k)) of H1alloy. Data is derived from:
c Volkov B.Yu. et. al."Material property library for H1alloy
c cladding", Preprint IAE4941/11, 1989.
c
if(ctemp.le.2133.0d0) then
ccon = dexp(0.000461843d0*ctemp)*15.0636d0
cdkdt = 0.006957018215d0*dexp(0.000461843d0*ctemp)
else
ccon = 36.d0
cdkdt = 0.d0
endif
return
c
A.23
end
A.2.3 Cladding Oxide Thermal Conductivity (ZOTCON)
FRAPCON3.4: ZOTCON
*deck zotcon
function zotcon (ctemp)
c
c zotcon calculates zirconium dioxide thermal conductivity as a
c function of temperature.
c zotcon = output thermal conductivity of zro2 (w/mk)
c ctemp = input cladding meshpoint temperature (k)
c the data used to generate this correlation were taken from:
c w.d. kingery, j. francl, r.l. coble and t. vasilos
c j. amer. ceram. soc., 37 no.i (1954) pp 107110
c zotcon was coded by r.l.miller in march 1974.
c modified by r.l. miller may 1974
c use caution above 1700k
t = ctemp
zotcon = 1.9599t*(2.41e4t*(6.43e7t*1.946e10))
return
end
FRAPTRAN 1.4: ZOTCON
*deck zotcon
function zotcon (ctemp)
implicit real (ah,oz)
c
c zotcon calculates zirconium dioxide thermal conductivity as a
c function of temperature.
c
c zotcon = output thermal conductivity of zro2 (W/mK)
c ctemp = input cladding meshpoint temperature (K)
c
c the data used to generate this correlation were taken from:
c w.d. kingery, j. francl, r.l. coble and t. vasilos
c j. amer. ceram. soc., 37 no.i (1954) pp 107110
c
c zotcon was coded by r.l.miller in march 1974.
c modified by r.l. miller may 1974
c
c use caution above 1700K
c
t = ctemp
zotcon = 1.9599d0  t*(2.41d4  t*(6.43d7  t*1.946d10))
c
return
end
A.2.4 Cladding Oxide Emissivity (ZOEMIS)
FRAPCON3.4: ZOEMIS
*deck zoemis
subroutine zoemis (ctmax,zroxid,emissv)
c
c zoemis calculates the emissivity of the cladding surface
A.24
c as a function of maximum cladding temperature and
c oxide thickness. expected standard error of the
c calculated emissivity compared to inreactor
c data is also returned.
c emissv = output cladding surface emissivity (unitless)
c puemis = output positive standard error expected in emissv
c when compared to inreactor data (not currently returned)
c uuemis = output negative standard error expected in emissv
c when compared to inreactor data (not currently returned)
c ctmax = input maximum cladding temperature (k)
c zroxid = input oxide layer thickness (m)
c the equations used in this subroutine are based on data from
c (1) aec fuels and materials development program progress
c report no. 76, usaec report gemp  1008 (1968). section by
c e. f. juenke and s. j. sjodahl, p 239
c (2) t. b. burgoyne and a. garlick, paper presented at
c specialists meeting on the behaviour of water reactor
c fuel elements under accident conditions, spatind norway
c (september 1976)
c (3) e. v. murphy and f. havelock, emissivity of zirconium
c alloys in air in the temperature range 100  400 c
c j. nuc. mat., 60 (1976) pp 167176
c zoemis coded by r. l. miller sept 1974
c modified by d. l. hagrman october 1976
c model for temperatures below 1500k follows
if (zroxid.ge.3.88e06) go to 100
emissv = 3.25e01+1.246e05*zroxid
go to 110
100 emissv = 8.08642e015.00e01*zroxid
110 puemis = 0.1
uuemis = 0.1
if (ctmax.le.1500.) go to 120
c modification for maximum temperatures above 1500 k follows
emissv = emissv*exp((1.50e03ctmax)/3.00e02)
if (emissv.lt.0.325) emissv = 0.325
puemis = puemis/exp((1.50e03ctmax)/3.00e02)
uuemis = puemis
c standard error cut off at impossible values follows
if (puemis.gt.(1.00emissv)) puemis = 1.00emissv
if (uuemis.gt.emissv) uuemis = emissv
120 continue
return
end
FRAPTRAN 1.4: ZOEMIS
*deck zoemis
subroutine zoemis (ctmax, zroxid, emissv)
implicit real (ah,oz)
c
c This is a MATPRO11, Rev. 2 routine modified by PNNL for use in FRAPT
c
c zoemis calculates the emissivity of the cladding surface
c as a function of maximum cladding temperature and
c oxide thickness. expected standard error of the
c calculated emissivity compared to inreactor
c data is also returned.
c
c emissv = output cladding surface emissivity (unitless)
c puemis = output positive standard error expected in emissv
c when compared to inreactor data (not currently returned)
c uuemis = output negative standard error expected in emissv
A.25
c when compared to inreactor data (not currently returned)
c
c ctmax = input maximum cladding temperature (k)
c zroxid = input oxide layer thickness (m)
c
c the equations used in this subroutine are based on data from
c (1) aec fuels and materials development program progress
c report no. 76, usaec report gemp  1008 (1968). section by
c e. f. juenke and s. j. sjodahl, p 239
c (2) t. b. burgoyne and a. garlick, paper presented at
c specialists meeting on the behaviour of water reactor
c fuel elements under accident conditions, spatind norway
c (september 1976)
c (3) e. v. murphy and f. havelock, emissivity of zirconium
c alloys in air in the temperature range 100  400 c
c j. nuc. mat., 60 (1976) pp 167176
c
c zoemis coded by r. l. miller sept 1974
c modified by d. l. hagrman october 1976
c Modified by PNNL, January 1997, to clean up coding, delete sensiti
c uncertainty coding, and correct error in PUEMIS for T > 1500K
c
c model for temperatures below 1500K:
emissv = 3.25d01 + 1.246d05*zroxid
if (zroxid .ge. 3.88d06) emissv = 8.08642d01  5.0d+01*zroxid
puemis = 0.1d0
uuemis = 0.1d0
if(ctmax .le. 1500.d0) go to 50
c
c modification for maximum temperatures above 1500K:
emissv = emissv * exp((1.50d03  ctmax)/3.00d02)
if(emissv .lt. 0.325d0) emissv = 0.325d0
puemis = puemis*exp((1.50d03  ctmax)/3.00d02)
uuemis = puemis
c
c standard error cut off at impossible values follows
if(puemis .gt. (1.00d0emissv)) puemis = 1.00d0  emissv
if(uuemis .gt. emissv) uuemis = emissv
c
50 continue
c
return
c
end
A.2.5 Cladding Thermal Expansion (CTHEXP)
FRAPCON3.4: CTHEXP
*deck cthexp
subroutine cthexp (ctemp,cathex,cdthex)
c
c cthexp calculates axial and diametral thermal expansion of
c zircaloy lwr cladding.
c cathex = output axial thermal expansion of zircaloy (m/m)
c cdthex = output diametral thermal expansion of zircaloy (m/m)
c ctemp = input cladding meshpoint temperature (k)
c data for the correlation from room temperature to 1273k are
c taken from r.l. mehan and f.w. wiesinger, "mechanical properties
c of zircaloy2", usaec report kapl2110 (1961), d.b. scott,
c "physical and mechanical properties of zircaloy2 and 4",
c wcap326941 (1965), and j.j. kearns, "thermal expansion
A.26
c and preferred orientation in zircaloy", wapdtm472 (1965)
c above 1273k the coefficient of thermal expansion used is the
c constant value 9.7e06/k recommended by b. lustman and f. kerze,
c "the metallurgy of zirconium", mc grawhill book company, new
c york (1955) page 355.
c between 1073 and 1273k (approximately the alphabeta transition
c range for zircaloy) cthexp uses the linear interpolation routine
c polate to find the thermal expansion.
c cthexp was originally coded by r.l. miller in nov. 1974.
c last updated by g.a. reymann in june 1978.
dimension cathxd (44)
dimension dthexp (44)
data cathxd/ 0.0 , 293.15 ,
+ 3.52774e03, 1073.15 , 3.53000e03, 1083.15 ,
+ 3.50000e03, 1093.15 , 3.46000e03, 1103.15 ,
+ 3.41000e03, 1113.15 , 3.33000e03, 1123.15 ,
+ 3.21000e03, 1133.15 , 3.07000e03, 1143.15 ,
+ 2.80000e03, 1153.15 , 2.50000e03, 1163.15 ,
+ 2.00000e03, 1173.15 , 1.50000e03, 1183.15 ,
+ 1.30000e03, 1193.15 , 1.16000e03, 1203.15 ,
+ 1.13000e03, 1213.15 , 1.10000e03, 1223.15 ,
+ 1.11000e03, 1233.15 , 1.13000e03, 1243.15 ,
+ 1.20000e03, 1253.15 , 1.30000e03, 1263.15 ,
+ 1.40000e03, 1273.15 /
data npcatx, iu / 22,1 /
data dthexp/ 4.3480e04, 373.15,
+ 5.1395e03, 1073.15, 5.2200e03, 1083.15, 5.2500e03, 1093.15,
+ 5.2800e03, 1103.15, 5.2800e03, 1113.15, 5.2400e03, 1123.15,
+ 5.2200e03, 1133.15, 5.1500e03, 1143.15, 5.0800e03, 1153.15,
+ 4.9000e03, 1163.15, 4.7000e03, 1173.15, 4.4500e03, 1183.15,
+ 4.1000e03, 1193.15, 3.5000e03, 1203.15, 3.1300e03, 1213.15,
+ 2.9700e03, 1223.15, 2.9200e03, 1233.15, 2.8700e03, 1243.15,
+ 2.8600e03, 1253.15, 2.8800e03, 1263.15, 2.9000e03, 1273.15/
data npdexp / 22 /
if (ctemp.le.1073.15) go to 100
if (ctemp.ge.1273.15) go to 110
cathex = polate(cathxd,ctemp,npcatx,iu)
cdthex = polate(dthexp,ctemp,npdexp,iu)
go to 120
100 cathex = 2.5060e05+(ctemp273.15)*4.4410e06
cdthex = 2.3730e04+(ctemp273.15)*6.7210e06
go to 120
110 cathex = 8.300e03+(ctemp273.15)*9.70e06
cdthex = 6.800e03+(ctemp273.15)*9.70e06
120 continue
return
end
FRAPTRAN 1.4: CTHEXP
*deck cthexp
subroutine cthexp (ctemp, cathex, cdthex, icm)
implicit real (ah,oz)
c
c cthexp calculates axial and diametral thermal expansion of
c zircaloy lwr cladding.
c
c cathex = output axial thermal expansion of zircaloy (m/m)
c cdthex = output diametral thermal expansion of zircaloy (m/m)
c
c ctemp = input cladding meshpoint temperature (k)
c
A.27
c data for the correlation from room temperature to 1273k are
c taken from r.l. mehan and f.w. wiesinger, "mechanical properties
c of zircaloy2", usaec report kapl2110 (1961), d.b. scott,
c "physical and mechanical properties of zircaloy2 and 4",
c wcap326941 (1965), and j.j. kearns, "thermal expansion
c and preferred orientation in zircaloy", wapdtm472 (1965)
c
c above 1273k the coefficient of thermal expansion used is the
c constant value 9.7e06/k recommended by b. lustman and f. kerze,
c "the metallurgy of zirconium", mc grawhill book company, new
c york (1955) page 355.
c
c between 1073 and 1273k (approximately the alphabeta transition
c range for zircaloy) cthexp uses the linear interpolation routine
c polate to find the thermal expansion.
c
c cthexp was originally coded by r.l. miller in nov. 1974.
c last updated by g.a. reymann in june 1978.
c Licensing analysis coding deleted 9/23/97 by ME Cunningham, PNNL
c
c generic Zry
dimension cathxd (44)
dimension dthexp (44)
data cathxd/ 0.0d0,293.15d0, 3.52774d03,1073.15d0,
& 3.5300d03,1083.15d0, 3.5000d03,1093.15d0, 3.460d03,1103.15d0,
& 3.4100d03,1113.15d0, 3.3300d03,1123.15d0, 3.210d03,1133.15d0,
& 3.0700d03,1143.15d0, 2.8000d03,1153.15d0, 2.500d03,1163.15d0,
& 2.0000d03,1173.15d0, 1.5000d03,1183.15d0, 1.300d03,1193.15d0,
& 1.1600d03,1203.15d0, 1.1300d03,1213.15d0, 1.100d03,1223.15d0,
& 1.1100d03,1233.15d0, 1.1300d03,1243.15d0, 1.200d03,1253.15d0,
& 1.3000d03,1263.15d0, 1.4000d03,1273.15d0/
data npcatx /22/
data dthexp/ 4.3480d04,373.15d0,
& 5.1395d03,1073.15d0, 5.2200d03,1083.15d0, 5.2500d03,1093.15d0,
& 5.2800d03,1103.15d0, 5.2800d03,1113.15d0, 5.2400d03,1123.15d0,
& 5.2200d03,1133.15d0, 5.1500d03,1143.15d0, 5.0800d03,1153.15d0,
& 4.9000d03,1163.15d0, 4.7000d03,1173.15d0, 4.4500d03,1183.15d0,
& 4.1000d03,1193.15d0, 3.5000d03,1203.15d0, 3.1300d03,1213.15d0,
& 2.9700d03,1223.15d0, 2.9200d03,1233.15d0, 2.8700d03,1243.15d0,
& 2.8600d03,1253.15d0, 2.8800d03,1263.15d0, 2.9000d03,1273.15d0/
data npdexp, iu /22, 1/
c
go to (400,400,400,400,400,600,400,600,400,400,400)icm
c
400 continue
c generic Zircaloy thermal expansion
if(ctemp .le. 1073.15d0) go to 401
if(ctemp .ge. 1273.15d0) go to 402
c
cathex = polate(cathxd,ctemp,npcatx,iu)
cdthex = polate(dthexp,ctemp,npdexp,iu)
return
c
401 continue
cathex = 2.5060d05 + (ctemp273.15d0)*4.4410d06
cdthex = 2.3730d04 + (ctemp273.15d0)*6.7210d06
return
c
402 continue
cathex = 8.300d03 + (ctemp273.15d0)*9.70d06
cdthex = 6.800d03 + (ctemp273.15d0)*9.70d06
return
c
A.28
600 continue
c RRCKI thermal expansion
c
if(ctemp .ge. 2133.d0) then
cathex=1.0582459d2
cdthex=1.3133600d2
c
else if(ctemp .gt.1153.d0) then
cathex=1.076459d3+9.7d6*(ctemp1153.d0)
cdthex=3.627600d3+9.7d6*(ctemp1153.d0)
c
else if(ctemp .gt. 883.d0)then
cathex=3.0465577d3+2.312d8*(ctemp883.d0)
& 7.358d8*(ctemp883.d0)**2+1.7211d10*(ctemp883.d0)**3
cdthex=5.5977000d3+2.312d8*(ctemp883.d0)
& 7.358d8*(ctemp883.d0)**2+1.7211d10*(ctemp883.d0)**3
c
else if(ctemp .gt. 573.d0) then
cathex=0.13725577d2+5.4d6*(ctemp573.d0)
cdthex=0.3336985d8*ctemp**2+5.65390d6*ctemp0.199649865d2
c
else
cathex=0.1338985d8*ctemp**2+3.85875d6*ctemp0.127813365d2
cdthex=0.3336985d8*ctemp**2+5.65390d6*ctemp0.199649865d2
endif
c
return
end
A.2.6 Cladding Elastic Modulus (CELMOD) and Shear Modulus (CSHEAR)
FRAPCON3.4: CELMOD
*deck celmod
function celmod (ctemp,fnck,cwkf,deloxy)
c
c celmod calculates cladding young's modulus as a function
c of temperature, fast neutron fluence, cold work, and average
c oxygen concentration. grain orientation is assumed random.
c the expected standard error is computed but not returned
c (ucelmd).
c celmod = output young's modulus for zircaloy 2 and 4
c with random texture (pa)
c ucelmd = output standard error expected in celmod (pa)
c (not currently returned)
c ctemp = input cladding temperature (k)
c fnck = input effective fast fluence (neutrons/(m**2))
c cwkf = input effective cold work (unitless ratio of areas)
c deloxy = input average oxygen concentration excluding
c oxide layer  average oxygen concentration of
c asreceived cladding (kg oxygen/kg zircaloy)
c the young's modulus calculated by this function is based on
c data from the following references
c (1) l. r. bunnell, g. b. mellinger and j. l. bates, high
c temperature properties of zircaloy  oxygen alloys, epri
c report np  524 (1977).
c (2) e. s. fisher and c. j. renken, physical review 135 2a
c (20 july 1964) pp a482  494.
c (3) p. e. armstrong and h. l. brown, transactions of the
c metallurgical society of aime 230 (august 1964)
c pp 962  966
c (4) a. padel and a. groff, journal of nuclear materials 59
A.29
c (1976) pp 325326
c (5) w. r. smalley, saxton core ii fuel performance evaluation.
c part i: materials, wcap338556 (september 1971)
c the expected standard error was derived using additional
c data from the following references
c (1) c. c. busby and c.r. woods (eds.) "properties of zircaloy4
c tubing", usaec report wapdtm585 (december 1966) p 65
c (2) z. spasic, m. pavlovic and g. simis, conference on
c the use of zirconium alloys in nuclear reactors, marianske
c lanze, czech. conf681086 (1968) pp 277  284
c (3) r. l. mehan, modulus of elasticity of zircaloy2 between
c room temperature and 1000 f, kaplmrlm16 (july 1958)
c (4) d. o. northwood, i. m london, and l. e. bahen, journal
c of nuclear materials 55 (1975) pp 299310
c (5) f. r. shober, j. a. van echo, l. l. marsh jr. and
c j. r. keeler, the mechanical properties of zirconium and
c zircaloy2, bmi1168 (1957)
c celmod was coded by r. l. miller in march 1974.
c updated and corrected by b. w. burnham in october 1975
c modified by b. w. burnham october 1977
c modified by d. l. hagrman december 1977
include 'lacmdl.h'
data on / 1 /,
+ off / 2 /,
+ locidx / 5 /
if (emflag(locidx).eq.on) go to 130
c best estimate model young's modulus
c1 = (1.16e+11+ctemp*1.037e+08)*5.7015
c2 = 1.0
if (fnck.gt.1.0e+22) c2 = 0.88*(1.0exp(fnck/1.0e+25))+exp(fnck/
+1.0e+25)
c3 = 2.6e+10
celmod = (1.088e+115.475e+07*ctemp+c1*deloxy+c3*cwkf)/c2
if (ctemp.lt.1090.) go to 140
c calculate a to a + b and a + b to b boundaries
wfox = deloxy+0.0012
taab = 1094.+wfox*(1.289e+03+wfox*7.914e+05)
if (wfox.lt.0.025) go to 100
taab = 1556.4+3.8281e+04*(wfox0.025)
100 tabb = 392.46*((100*deloxy+0.1242807)**2+3.1417)
if (deloxy.lt.4.7308937e03) go to 110
tabb = (100*deloxy+0.12)*491.157+1081.7413
110 continue
if (ctemp.lt.taab) go to 140
if (ctemp.gt.tabb) go to 120
amodl = (1.088e+115.475e+07*taab+c1*deloxy+c3*cwkf)/c2
amodr = 9.21e+10tabb*4.05e+07
celmod = amodl+(ctemptaab)*(amodramodl)/(tabbtaab)
go to 140
120 celmod = 9.21e+10ctemp*4.05e+07
go to 140
c em model young's modulus
130 celmod = emclem(ctemp)
140 continue
if (celmod.lt.1.0) celmod = 1.0
ucelmd = 6.4e+09
return
end
FRAPTRAN1.4: CELMOD
*deck celmod
A.30
c
function celmod (ctemp, fnck, cwkf, deloxy, icm)
c
implicit real (ah,oz)
c
c This is a MATPRO11, Rev. 2 routine modified by PNNL for use in FRAPT
c
c celmod calculates cladding young's modulus as a function
c of temperature, fast neutron fluence, cold work, and average
c oxygen concentration. grain orientation is assumed random.
c celmod = output young's modulus for zircaloy 2 and 4
c with random texture (pa)
c ctemp = input cladding temperature (k)
c fnck = input effective fast fluence (neutrons/(m**2))
c cwkf = input effective cold work (unitless ratio of areas)
c deloxy = input average oxygen concentration excluding
c oxide layer  average oxygen concentration of
c asreceived cladding (kg oxygen/kg zircaloy)
c
c the young's modulus calculated by this function is based on
c data from the following references
c (1) l. r. bunnell, g. b. mellinger and j. l. bates, high
c temperature properties of zircaloy  oxygen alloys, epri
c report np  524 (1977).
c (2) e. s. fisher and c. j. renken, physical review 135 2a
c (20 july 1964) pp a482  494.
c (3) p. e. armstrong and h. l. brown, transactions of the
c metallurgical society of aime 230 (august 1964)
c pp 962  966
c (4) a. padel and a. groff, journal of nuclear materials 59
c (1976) pp 325326
c (5) w. r. smalley, saxton core ii fuel performance evaluation.
c part i: materials, wcap338556 (september 1971)
c
c the expected standard error was derived using additional
c data from the following references
c (1) c. c. busby and c.r. woods (eds.) "properties of zircaloy4
c tubing", usaec report wapdtm585 (december 1966) p 65
c (2) z. spasic, m. pavlovic and g. simis, conference on
c the use of zirconium alloys in nuclear reactors, marianske
c lanze, czech. conf681086 (1968) pp 277  284
c (3) r. l. mehan, modulus of elasticity of zircaloy2 between
c room temperature and 1000 f, kaplmrlm16 (july 1958)
c (4) d. o. northwood, i. m london, and l. e. bahen, journal
c of nuclear materials 55 (1975) pp 299310
c (5) f. r. shober, j. a. van echo, l. l. marsh jr. and
c j. r. keeler, the mechanical properties of zirconium and
c zircaloy2, bmi1168 (1957)
c
c celmod was coded by r. l. miller in march 1974.
c updated and corrected by b. w. burnham in october 1975
c modified by b. w. burnham october 1977
c modified by d. l. hagrman december 1977
c Modified by PNNL, January 1997, to clean up coding, delete licensi
c analysis and sensitivity analysis coding
c
go to (400,400,400,400,400,600,400,600,400,400,400)icm
c
400 continue
c generic Zry model
c best estimate model young's modulus
c1 = (1.16d+11 + ctemp * 1.037d+08) * 5.7015d0
c2 = 1.0d0
A.31
if(fnck .gt. 1.0d+22) c2 = 0.88d0 * (1.0d0  exp(fnck/1.0d+25))
& + exp(fnck/1.0d+25)
c3 = 2.6d+10
celmod = (1.088d+11  5.475d+07 * ctemp + c1 * deloxy + c3 * cwkf)
& /c2
if(ctemp .lt. 1090.d0) go to 401
c
c calculate a to a + b and a + b to b boundaries
wfox = deloxy + 0.0012d0
taab = 1094.d0 + wfox * (1.289d+03 + wfox * 7.914d+05)
if (wfox .lt. 0.025d0) go to 415
taab = 1556.4d0 + 3.8281d+04 * (wfox  0.025d0)
415 continue
tabb = 392.46d0 * ((100 * deloxy + 0.1242807d0)**2 + 3.1417d0)
if (deloxy .lt. 4.7308937d03) go to 425
tabb = (100 * deloxy + 0.12d0) * 491.157d0 + 1081.7413d0
425 continue
c
if (ctemp .lt. taab) go to 401
if (ctemp .gt. tabb) go to 435
amodl = (1.088d+11  5.475d+07 * taab + c1 * deloxy + c3 * cwkf)
& /c2
amodr = 9.21d+10  tabb * 4.05d+07
celmod = amodl + (ctemp  taab) * (amodr  amodl)/(tabb  taab)
go to 401
c
435 continue
celmod = 9.21d+10  ctemp * 4.05d+07
c
401 continue
if(celmod .lt. 1.0d0) celmod = 1.0d0
return
c
600 continue
c Zr1%Nb model from RRCKI
c
c This code block calculates cladding Young's modulus as
c a function of temperature ; data is derived from
c Volkov B.Yu. et.al."Material property library for
c H1alloy cladding", Preprint IAE4941/11, 1989.
c Fedorov et.al."Influence of oxygen content and gamma
c irradiation on the elastic module and internal
c friction pure zirconium" Metal Physics vol.32(3)
c p.621, 1971
c
if(ctemp .gt. 1073.0d0) go to 601
celmod = 1.121d11  6.438d07*ctemp
celmod = celmod + 3.021d12*deloxy
go to 602
601 continue
celmod = 9.129d10  4.5d07*ctemp
602 continue
if(celmod .lt. 1.0d0) celmod = 1.0d0
return
c
end
FRAPCON3.4: CSHEAR
*deck cshear
function cshear (ctemp,fnck,cwkf,deloxy)
c
c cshear calculates the shear modulus of zircaloy2 and 4
A.32
c as a function of temperature, fast neutron fluence,
c cold work, and average oxygen concentration. grain
c orientation is assumed random.
c the expected standard error is computed but not returned
c (ucsher)
c cshear = output shear modulus for zircaloy2 and 4
c with random texture (pa)
c ucsher = output standard error expected in cshear (pa)
c (not currently returned)
c ctemp = input cladding temperature (k)
c fnck = input effective fast fluence (neutrons/(m**2))
c cwkf = input effective cold work (unitless ratio of areas)
c deloxy = input average oxygen concentration excluding
c oxide layer  average oxygen concentration of
c asreceived cladding (kg oxygen/kg zircaloy)
c the shear modulus calculated by this function is based on
c data from the following references
c (1) l. r. bunnell, g. b. mellinger and j. l. bates, high
c temperature properties of zircaloy  oxygen alloys, epri
c report np  524 (1977)
c (2) e. s. fisher and c. j. renken, physical review 135 2a
c (20 july 1964) pp a482  494.
c (3) p. e. armstrong and h. l. brown, transactions of the
c metallurgical society of aime 230 (august 1964)
c pp 962  966
c (4) a. padel and a. groff, journal of nuclear materials 59
c (1976) pp 325  326.
c (5) w. r. smalley, saxton core ii fuel performance evaluation.
c part i: materials, wcap338556 (september 1971)
c cshear was coded by r. l. miller, june 1974
c modified by d. l. hagrman december 1977
c1 = 7.07e+11ctemp*2.315e+08
c2 = 1.0
if (fnck.gt.1.0e+22) c2 = 0.88*(1.0exp(fnck/1.0e+25))+exp(fnck/
+1.0e+25)
c3 = .867e+10
cshear = (4.04e+102.168e+07*ctemp+c1*deloxy+c3*cwkf)/c2
if (ctemp.lt.1090.) go to 130
c calculate a to a + b and a + b to b boundaries
wfox = deloxy+0.0012
taab = 1094.+wfox*(1.289e+03+wfox*7.914e+05)
if (wfox.lt.0.025) go to 100
taab = 1556.4+3.8281e+04*(wfox0.025)
100 tabb = 392.46*((100*deloxy+0.1242807)**2+3.1417)
if (deloxy.lt.4.7308937e03) go to 110
tabb = (100*deloxy+0.12)*491.157+1081.7413
110 continue
if (ctemp.lt.taab) go to 130
if (ctemp.gt.tabb) go to 120
amodl = (4.04e+102.168e+07*taab+c1*deloxy+c3*cwkf)/c2
amodr = 3.49e+10tabb*1.66e+07
cshear = amodl+(ctemptaab)*(amodramodl)/(tabbtaab)
go to 130
120 cshear = 3.49e+10ctemp*1.66e+07
130 continue
if (cshear.lt.1.0) cshear = 1.0
ucsher = 9.0e+09
return
end
FRAPTRAN 1.4: CSHEAR
A.33
*deck cshear
c
function cshear (ctemp, fnck, cwkf, deloxy, icm)
c
implicit real (ah,oz)
c
c This is a MATPRO11, Rev. 1 routine modified by PNNL for use in FRAPT
c
c cshear calculates the shear modulus of zircaloy2 and 4
c as a function of temperature, fast neutron fluence,
c cold work, and average oxygen concentration. grain
c orientation is assumed random.
c cshear = output shear modulus for zircaloy2 and 4
c with random texture (pa)
c ctemp = input cladding temperature (k)
c fnck = input effective fast fluence (neutrons/(m**2))
c cwkf = input effective cold work (unitless ratio of areas)
c deloxy = input average oxygen concentration excluding
c oxide layer  average oxygen concentration of
c asreceived cladding (kg oxygen/kg zircaloy)
c
c the shear modulus calculated by this function is based on
c data from the following references
c
c (1) l. r. bunnell, g. b. mellinger and j. l. bates, high
c temperature properties of zircaloy  oxygen alloys, epri
c report np  524 (1977)
c (2) e. s. fisher and c. j. renken, physical review 135 2a
c (20 july 1964) pp a482  494.
c (3) p. e. armstrong and h. l. brown, transactions of the
c metallurgical society of aime 230 (august 1964)
c pp 962  966
c (4) a. padel and a. groff, journal of nuclear materials 59
c (1976) pp 325  326.
c (5) w. r. smalley, saxton core ii fuel performance evaluation.
c part i: materials, wcap338556 (september 1971)
c
c cshear was coded by r. l. miller, june 1974
c modified by d. l. hagrman december 1977
c Modified by PNNL, January 1997, to clean up coding and delete
c sensitivity uncertainty analysis coding
c
go to (400,400,400,400,400,600,400,600,400,400,400)icm
c
400 continue
c generic Zry properties
c1 = 7.07d+11  ctemp * 2.315d+08
c2 = 1.0d0
if(fnck .gt. 1.0d+22) c2 = 0.88d0 * (1.0d0  exp(fnck/1.0d+25))
& + exp(fnck/1.0d+25)
c3 = 0.867d+10
cshear = (4.04d+10  2.168d+07 * ctemp + c1 * deloxy + c3*cwkf)/c2
if(ctemp .lt. 1090.d0) go to 401
c
c calculate a to a + b and a + b to b boundaries
wfox = deloxy + 0.0012d0
taab = 1094.d0 + wfox * (1.289d+03 + wfox * 7.914d+05)
if(wfox .lt. 0.025d0) go to 415
taab = 1556.4d0 + 3.8281d+04 * (wfox  0.025d0)
415 continue
tabb = 392.46d0 * ((100 * deloxy + 0.1242807d0)**2 + 3.1417d0)
if(deloxy .lt. 4.7308937d03) go to 425
tabb = (100 * deloxy + 0.12d0) * 491.157d0 + 1081.7413d0
A.34
425 continue
c
if(ctemp .lt. taab) go to 401
if(ctemp .gt. tabb) go to 435
amodl = (4.04d+10  2.168d+07 * taab + c1 * deloxy + c3*cwkf)/c2
amodr = 3.49d+10  tabb * 1.66d+07
cshear = amodl + (ctemp  taab ) * (amodr  amodl )/(tabb  taab)
go to 401
435 continue
cshear = 3.49d+10  ctemp * 1.66d+07
401 continue
if(cshear .lt. 1.0d0) cshear = 1.0d0
return
c
600 continue
c Zr1%Nb from RRCKI
c
c This code block calculates cladding young's modulus and
c poisson's ratio as a function of temperature ; data is
c derived from Volkov B.Yu. et.al."Material property library
c for H1alloy cladding", Preprint IAE4941/11, 1989.
c
c elastic modulus
ey = celmod (ctemp, fnck, cwkf, deloxy, CladType)
c
c poisson's ratio:
xnu = 0.42628d0  5.556d5*ctemp
c
c shear modulus
cshear = 0.5d0*ey/(1.d0+xnu)
if(cshear .lt. 1.0d0) cshear = 1.0d0
return
c
end
A.2.7 Cladding Axial Growth (CAGROW)
FRAPCON3.4: CAGROW
*deck cagrow
function cagrow (flux,fluenc,icm,dtime)
common /uncert/sigftc, sigftex, sigfgr
& ,sigswell ,sigcreep ,siggro ,sigcor ,sigh2
c
c cagrow calculates axial cladding growth strain increment as a functi
c flux = input fast neutron flux ((neutrons/m**2)/s)
c fluenc = input fast neutron fluence (neutrons/m**2)
c dtime = input time increment at temperature (s)
c icm = input cladding material index:4  Zircaloy 4
c 2  Zircaloy 2
c cagrow = output axial cladding growth straini increment (m/m)
c cagrow coded by D. L. Hagrman, August 1975
c revised April, 1995 by D. D. Lanning and K. J. Geelhood
c to conform to EPRI model by D. G. Franklin, "Zircaloy4
c Cladding Deformation During Power Reactor Irradiation,"
c ASTM STP 754, 1982, pp.235267.
f2=fluenc/10000.
f1=(fluencflux*dtime)/10000.
ax1=2.18e21*f1**0.845
ax2=2.18e21*f2**0.845
if(icm.eq.5) then
ax1=7.013e21*f1**0.81787
A.35
ax2=7.013e21*f2**0.81787
elseif(icm.eq.6) then
ax1=9.7893e25*f1**0.98239
ax2=9.7893e25*f2**0.98239
endif
cagrow = ax2ax1
if( icm .lt. 4.0 ) cagrow = cagrow * 0.5
c add on uncertainty
c PWR cladding
if(siggro.gt.0.0.and.icm.ge.4)
&cagrow = cagrow*(1.0+siggro*0.223)
if(siggro.lt.0.0.and.icm.ge.4)
&cagrow = cagrow/(1.0siggro*0.223)
c BWR cladding
if(siggro.gt.0.0.and.icm.lt.4)
&cagrow = cagrow*(1.0+siggro*0.203)
if(siggro.lt.0.0.and.icm.lt.4)
&cagrow = cagrow/(1.0siggro*0.203)
return
end
A.2.8 Cladding Creep Rate (CREPR)
FRAPCON3.4: CREPR
*deck crepr
c
subroutine crepr(sig,edot,phi,fluence,cldwks,CladAveTemp
&,GapPress,CoolantPress,rci,rco, ProblemTime)
c
common /cladtype/ icm
common /uncert/sigftc, sigftex, sigfgr
& ,sigswell ,sigcreep ,siggro ,sigcor ,sigh2
c implicit real*8 (ah,oz)
c ********************************************************************
c crepr is called from cladf.
c crepr computes (transverse) cladding creep strain rate
c as a function of effective stress,temperature, fast
c neutron flux and accumulated total creep strain
c ********************************************************************
c input arguments
c ********************************************************************
c CreepStrain  total accumulated creep strain (dimensionless)
c phi  flux (neutrons/m**2/sec)
c ProblemTime  time to end of power step (hr)
c CladAveTemp  cladding average temperature (F)
c sig  cladding hoop stress per node (psi)
c ********************************************************************
c output arguments
c ********************************************************************
c edot  creep strain rate (1/hr)
c ********************************************************************
c
c NEW MODEL
c New model uses a modified version of the Limback and Andersson equation
c ASTM STP 1295 pp. 448468
c calculate stress in each direction
sr=(GapPress*rci**2CoolantPress*rco**2
&+rci**2*rco**2*(CoolantPressGapPress)/((rco+rci)/2.0)**2)
&/(rco**2rci**2)
st=(GapPress*rci**2CoolantPress*rco**2
&rci**2*rco**2*(CoolantPressGapPress)/((rco+rci)/2.0)**2)
A.36
&/(rco**2rci**2)
sl=(GapPress*rci**2CoolantPress*rco**2)/(rco**2rci**2)
c sag is effective stress in MPa
c
c sag = sig*6894.7573/1.0E6
sag=0.707*sqrt((slst)**2+(stsr)**2+(srsl)**2)*6894.7573/1.0E6
tcak = (CladAveTemp+459.67)/1.8
fluenceLimback = fluence/10000.0
c
c Thermal Creep Rate
c
A = 1.08E9
if(cldwks.eq.0.0.or.icm.eq.5) A = 5.47E8
an = 2.0
if(cldwks.eq.0.0.or.icm.eq.5) an = 3.5
E = 1.148E5  59.9*tcak
Q = 201000.0
R = 8.314
ai = 650.0*(1.00.56*(1.0exp(1.4e27*fluenceLimback**1.3)))
edottherm = A*(E/tcak)*(sinh(ai*abs(sag)/E))**an*exp(Q/(R*tcak))
c
c Irradiation Creep Rate
c
c0SR = 4.0985E24
c0RXA = 1.87473E24
c0 = C0SR
if(cldwks.eq.0.0) c0 = c0RXA
c1 = 0.85
c2 = 1.0
tmod=tcak
if(tcak.lt.570.0) tmod=570.0
if(tcak.gt.625.0) tmod=625.0
ctemp=7.0237+0.0136*tmod
if(cldwks.eq.0.0) ctemp=3.18562+0.00699132*tmod
edotirr = c0*phi**c1*(abs(sag))**c2*ctemp
c
c Primary Creep Rate
c
if(ProblemTime.ge.0.1) then
esat=0.0216*(edottherm+edotirr)**0.109*
&(2.0tanh(3.55e4*(edottherm+edotirr)))**2.05
edotprimary=
&0.5*esat*52.0*(edottherm+edotirr)**0.5/((ProblemTime)**0.5)*
&exp(52.0*(edottherm+edotirr)**0.5*(ProblemTime)**0.5)
edottherm=edottherm+edotprimary
endif
c
edot = edottherm + edotirr
edot = edot*1.1547
if(icm.eq.6) edot = edot*0.8
100 continue
c add on uncertainty
c SRA cladding
if(sigcreep.gt.0.0.and.cldwks.ne.0.0)
&edot = edot*(1.0+sigcreep*0.145)
if(sigcreep.lt.0.0.and.cldwks.ne.0.0)
&edot = edot/(1.0sigcreep*0.145)
c RXA cladding
if(sigcreep.gt.0.0.and.cldwks.eq.0.0)
&edot = edot*(1.0+sigcreep*0.216)
if(sigcreep.lt.0.0.and.cldwks.eq.0.0)
&edot = edot/(1.0sigcreep*0.216)
return
A.37
end
A.2.9 Cladding Meyer Hardness (CMHARD)
FRAPCON3.4: CMHARD
*deck cmhard
function cmhard (ctemp)
c
c the routine cmhard calculates meyer hardness as a function of
c cladding temperature.
c cmhard = output meyer hardness of zircaloy cladding (n/m**2)
c ctemp = input cladding temperature (k)
c the equations used in this function are based on data from
c (1) a. m. ross and r. l. stoute, heat transfer coefficient
c between uo2 and zircaloy  2, aecl  1552 (june 1962)
c (2) i. d. peggs and d. p. godin, the yield strength  hot
c hardness relationship of zircaloy  4, journal of nuclear
c materials 57 pp 246  248 (1975)
c cmhard was coded by v.f.baston in may 1974.
c modified by m. a. morgan june 1978
t = ctemp
cmhard = exp(2.6034e012.6394e02*t+4.3502e05*t**22.5621e08*t**
+3)
return
end
FRAPTRAN 1.4: CMHARD
*deck cmhard
function cmhard (ctemp, icm)
implicit real (ah,oz)
c
c This is a MATPRO11, Rev. 2 routine modified by PNNL for use in FRAPT
c
c the routine cmhard calculates meyer hardness as a function of
c cladding temperature.
c
c cmhard = output meyer hardness of zircaloy cladding (n/m**2)
c
c ctemp = input cladding temperature (k)
c
c the equations used in this function are based on data from
c (1) a. m. ross and r. l. stoute, heat transfer coefficient
c between uo2 and zircaloy  2, aecl  1552 (june 1962)
c (2) i. d. peggs and d. p. godin, the yield strength  hot
c hardness relationship of zircaloy  4, journal of nuclear
c materials 57 pp 246  248 (1975)
c
c cmhard was coded by v.f.baston in may 1974.
c modified by m. a. morgan june 1978
c Modified by PNNL, January 1997, to clean up coding, delete sensiti
c uncertainty coding
c
go to (400,400,400,400,400,600,400,600,400,400,400)icm
c
400 continue
c generic Zry property
cmhard = exp(2.6034d+01  2.6394d02*ctemp + 4.3502d05*ctemp**2
&  2.5621d08*ctemp**3)
if (cmhard .lt. 1.94d+08) cmhard = 1.94d+08
A.38
return
c
600 continue
c Zr1%Nb from RRCKI
if(ctemp .lt. 800.d0) cmhard=1.d6*(2172.1d0  10.7055d0*ctemp
& + 0.02765d0*ctemp**2  3.278d5*ctemp**3 + 1.423d8*ctemp**4)
if(ctemp .ge. 800.d0) cmhard=exp(26.034d0  2.6394d2*ctemp
& + 4.3502d5*ctemp**2  2.5621d8*ctemp**3)
if (cmhard .lt. 1.0d+05) cmhard = 1.0d+05
c
end
A.3 Gas Material Properties
A.3.1 Gas Thermal Conductivity (GTHCON)
FRAPCON3.4
*deck gthcon
function gthcon (gmix,gtemp,gpres,gpthk)
c
c gthcon calculates gas thermal conductivity as a function of
c temperature and gas fraction for seven gases:
c gthcon = output gas thermal conductivity (w/mk).
c gmix(i) = input mole fractions of the gas mixture
c the seven elements of gmix must sum to 1.0
c constituent gas number key
c 1 helium
c 2 argon
c 3 krypton
c 4 xenon
c 5 hydrogen
c 6 nitrogen
c 7 water vapor
c gtemp = input gas temperature (k)
c gpres = input gas pressure (pa)
c used for knudsen domain correction and for steam
c not used if 0.
c gpthk = input effective gap thickness for knudsen domain (m)
c (maximum of gap dimension or surface roughness)
c not used if 0.
c formula for gas mixtures is from r.s.brokaw, report nasa tr r81
c (1960). conductivity of rare gases is based on j.m.gandhi and
c s.c.saxena, jour. chem. and eng. data, vol.13, no.3 (1968)
c also: wisconsin electric power co., docket no. 50301 (jan 1973)
c the accommodation factor is from r.a.dean, cvna127 (1962)
c steam equation is from meyer, et.al., "thermodynamic and
c transport properties of steam", the american society of
c mechanical engineers (1967)
c gthcon coded by r.c.young march 1975
c adapted from routine cmix by p.e.macdonald
include 'lacmdl.h'
dimension gmix(7), a(7), c(7), r(7), aa(6), bb(6)
data a /4.003 , 39.944 ,83.80 ,131.30 , 2.016 ,28.8 ,
+ 18.016 /
c r = sqrt(a)
data r /2.00075 , 6.3201 , 9.1542 ,11.4586 , 1.41986 ,
+ 5.3666 , 4.2445 /
c
c The "aa" and "bb" parameters are from MATPRO11 Rev. 1, 1981
c "aa" and "bb" parameters updated by W.G. Luscher 101707
A.39
c
data aa/2.531e3,4.092e4,1.966e4,9.825e5,1.349e3,2.984e4/
data bb/0.7146, 0.6748, 0.7006, 0.7334, 0.8408, 0.7799/
c
c
data eps/1.e9 /
data on / 1 /,
+ off / 2 /,
+ locidx / 16 /
if (emflag(locidx).eq.on) gthcon = emgton(gmix,gtemp,gpres,gpthk)
if (emflag(locidx).eq.on) go to 170
if (gtemp.gt.5555.) go to 100
if (gtemp.gt.0.0000) go to 110
100 write (6,180) gtemp,gmix
stop
110 l = 0
c
do 120 i=1,6
if (gmix(i).le.eps) go to 120
l = i
c(i) = aa(i)*gtemp**bb(i)
120 continue
if (gmix(7).le.eps) go to 140
l = 7
c steam
c c(7) defination updated by W.G. Luscher 101707
tc = gtemp  273.15d0
if (gtemp.le.973.15d0)
& c(7) = gpres/gtemp*(2.8516d08 + 9.424d10*gtemp
& 6.005d14*gtemp**2) + 1.009d0*gpres**2/gtemp**2
& /(tc)**4.2d0 + (17.6d4 + 5.87d5*(tc)
& + 1.08d7*(tc)**2  4.51d11*(tc)**3)
if (gtemp.gt.973.15d0)
& c(7) = 4.44d6*gtemp**1.45d0 + 9.45d005*(2.1668d09*gpres
& /gtemp)**1.3d0
140 if (l.le.0) go to 100
gthcon = 0.
do 160 i=1,l
if (gmix(i).le.eps) go to 160
sum = 0.
do 150 j=1,l
if (j.eq.i) go to 150
if (gmix(j).le.eps) go to 150
rc = c(i)/c(j)
ra = a(i)/a(j)
fij = 1.+2.41*(ra1.)*(ra.142)/(1.+ra)**2
gij = (1.+sqrt(rc*r(i)/r(j)))**2/sqrt(8.*(1.+ra))
sum = sum+fij*gij*gmix(j)
150 continue
gthcon = gthcon+c(i)*gmix(i)/(gmix(i)+sum)
160 continue
170 continue
return
180 format ('1gthcon input bad. gtemp=',1pe12.4,' k',/'0fractions',(
+10e12.4))
end
FRAPTRAN 1.4
*deck gthcon
function gthcon (gmix,gtemp,gpres,gpthk)
implicit real (ah, oz)
A.40
c
c gthcon calculates gas thermal conductivity as a function of
c temperature and gas fraction for seven gases:
c
c gthcon = output gas thermal conductivity (w/mK).
c gmix(i) = input mole fractions of the gas mixture
c the seven elements of gmix must sum to 1.0
c constituent gas number key
c 1 helium
c 2 argon
c 3 krypton
c 4 xenon
c 5 hydrogen
c 6 nitrogen
c 7 water vapor
c
c gtemp = input gas temperature (K)
c gpres = input gas pressure (pa)
c used for knudsen domain correction and for steam
c not used if 0.
c gpthk = input effective gap thickness for knudsen domain (m)
c (maximum of gap dimension or surface roughness)
c not used if 0.
c
c formula for gas mixtures is from r.s.brokaw, report nasa tr r81
c (1960). conductivity of rare gases is based on j.m.gandhi and
c s.c.saxena, jour. chem. and eng. data, vol.13, no.3 (1968)
c also: wisconsin electric power co., docket no. 50301 (jan 1973)
c the accommodation factor is from r.a.dean, cvna127 (1962)
c steam equation is from meyer, et.al., "thermodynamic and
c transport properties of steam", the american society of
c mechanical engineers (1967)
c
c gthcon coded by r.c.young march 1975
c adapted from routine cmix by p.e.macdonald
c
dimension gmix(7), a(7), c(7), r(7), aa(6), bb(6)
data a /4.003d0, 39.944d0, 83.80d0, 131.30d0, 2.016d0,
& 28.02d0, 18.02d0/
c r = sqrt(a)
data r /2.00075d0, 6.3201d0, 9.1542d0, 11.4586d0, 1.41986d0,
& 5.3666d0, 4.2445d0 /
c
c The "aa" and "bb" parameters are from MATPRO11 Rev. 2, 8/1981
c "aa" and "bb" parameters updated by W.G. Luscher 101707
c
data aa/2.531d3,4.092d4,1.966d4,9.825d5,1.349d3,2.984d4/
data bb/0.7146d0,0.6748d0,0.7006d0,0.7334d0,0.8408d0,0.7799d0/
data eps/1.d9 /
c
if (gtemp.gt.5555.d0) go to 100
if (gtemp.gt.0.0000d0) go to 110
100 continue
write (6,180) gtemp,gmix
stop
c
110 continue
l = 0
c
do 120 i=1,6
if (gmix(i).le.eps) go to 120
l = i
c(i) = aa(i)*gtemp**bb(i)
A.41
120 continue
c
if (gmix(7).le.eps) go to 140
l = 7
c steam (from MATPRO11, Rev. 2)
c c(7) definition updated by W.G. Luscher 101707
tc = gtemp 273.15d0
if (gtemp .le. 973.15d0)
& c(7) = gpres/gtemp*(2.8516d08 + 9.424d10*gtemp
&  6.005d14*gtemp**2) + 1.009d0*gpres**2/gtemp**2
& /(gtemp273.15d0)**4.2d0 + (17.6d4 + 5.87d5*(tc)
& + 1.08d7*(tc)**2  4.51d11*(tc)**3)
if (gtemp .gt. 973.15d0)
& c(7) = 4.44d06*gtemp**1.45d0 + 9.45d05*(2.1668d09*gpres
& /gtemp)**1.3d0
c c(7) = 17.6e3+tc*(5.87e5+tc*(1.04e74.51e11*tc))
140 continue
if (l.le.0) go to 100
gthcon = 0.d0
c
do 160 i=1,l
if (gmix(i).le.eps) go to 160
sum = 0.d0
do 150 j=1,l
if (j.eq.i) go to 150
if (gmix(j).le.eps) go to 150
rc = c(i)/c(j)
ra = a(i)/a(j)
fij = 1.d0+2.41d0*(ra1.d0)*(ra.142d0)/(1.d0+ra)**2
gij = (1.d0+sqrt(rc*r(i)/r(j)))**2/sqrt(8.d0*(1.d0+ra))
sum = sum+fij*gij*gmix(j)
150 continue
gthcon = gthcon+c(i)*gmix(i)/(gmix(i)+sum)
160 continue
c
return
c
180 format ('gthcon input bad. gtemp= ',1pe12.4,' K',/' fractions',
& (10e12.4))
c
end
NUREG/CR7024 PNNL19417
Material Property Correlations: Comparisons between FRAPCON3.4, FRAPTRAN 1.4, and MATPRO
Manuscript Completed: August 2010 Date Published: March 2011 Prepared by W.G. Luscher and K.J. Geelhood Pacific Northwest National Laboratory P.O. Box 999 Richland, WA 99352 M.E. Flanagan, NRC Project Manager NRC Job Code N6609
Office of Nuclear Regulatory Research
Abstract
The U.S. Nuclear Regulatory Commission uses the computer codes FRAPCON3 and FRAPTRAN to model steadystate and transient fuel behavior, respectively, in regulatory analysis. In this document, material property correlations for oxide fuels and cladding materials are presented and discussed. Comparisons are made between the material property correlations used in the most recent versions of the codes, FRAPCON3.4 and FRAPTRAN 1.4. Comparisons are also made with MATPRO, which is a compilation of material property correlations with an extensive history of use with various fuel performance and severe accident codes. In addition to modeltomodel comparisons, modeltodata comparisons and source codes for the FRAPCON3.4 and FRAPTRAN 1.4 correlations are provided.
iii
iv .
FRAPTRAN. a “material property” is a physical characteristic of the material whose quantitative value is necessary in the analysis process.Foreword The U..g. The consolidated resource for “material properties” cited most often in the literature is MATPRO (as documented in Volume 4 of NUREG/CR6150). this report also provides a variety of comparisons to the material property correlations used in MATPRO. In addition to describing the material property correlations used in the subroutines of latest versions of the codes. as the material property correlations in the fuel performance codes have evolved to accommodate recent fuel design changes. respectively.S. and MATPRO has never been complete. v .4. In this sense. However. temperature and burnup). consistency among the material property correlations in FRAPCON3. Nuclear Regulatory Commission uses the computer codes FRAPCON and FRAPTRAN to model steadystate and transient fuel behavior. MATPRO is a compilation of fuel and cladding material property correlations with an extensive history of use with fuel performance and severe accident codes. FRAPCON3. To effectively model fuel behavior. The primary purpose of this report is to consolidate the current material property correlations used in FRAPCON and FRAPTRAN into a single document.4 and FRAPTRAN 1. material property correlations must be used for a wide range of operating conditions (e. in regulatory analysis.
vi .
.2 Applicability and Uncertainty ............................5..........3 2...............3............................................1 Model Development and Comparisons .1 Model Development and Comparisons ....................2 Fuel Specific Heat Capacity (FCP) and Fuel Enthalpy (FENTHL) .........27 2...................... 3.......................................26 2.........2 1..4...1 2...............................4 2.............................. 2....................27 2........... 2..........................................................1.........................1 3... xiii Abbreviations .....1 Model Development and Comparisons ..........................2...............................................2.6... 2.......1 Cladding Specific Heat (CCP) .................7 2...................................................1.................................................................3 References ......................3 3..............................................................3 References .........................7 2....Contents Abstract .............................7..............................................................1.......................6......................................22 2.....3 References .....................3 References ...................................2.......34 3.......................... 3............................................2 Applicability and Uncertainty ........... 3...................................... 2............................................ 2............5 Fuel Thermal Expansion (FTHEXP) ........................................3 2....1 2............................. 2................................................................ 2...2 Applicability and Uncertainty ................................................3 Fuel Thermal Conductivity (FTHCON).............1 Relation to Other Reports ................................... 2..................2.............................. 1...... 3....................2 Cladding Thermal Conductivity (CTHCON)......................................2 Applicability and Uncertainty ................. 2................6 3............................................................................1 Model Development and Comparisons .... 2................................2 Applicability and Uncertainty ..................1 Fuel Melting Temperature (PHYPRP) ....................0 Oxide Fuel Properties ...........................................1 Model Development and Comparisons .............................................................................. 1................................................................................7.........3 References ............. 3...............................................................................................................2 Applicability and Uncertainty ..................................................................................... 2..........................30 2..2 Applicability and Uncertainty ............................................... xv 1.................................... 2............................ 2........ v List of Figures ................................................... 2................................ 3.................. 2. 2...........................................23 2...........................................20 2........................... 2......................................3 References ............................26 2.2 Applicability and Uncertainty .....................................3 References ...................3 References .......................... iii Foreword ..................1 2.......................................................................23 2...............2 Applicability and Uncertainty .................... 3.........6 Fuel Densification (FUDENS) ...........................................................................24 2..................................................1...............................................3 3....................................................................................................3.........8 2..1..............7 vii ... 2..................1 Model Development and Comparisons ...........................................................................................3..8 2... 3................................. 3..........................4 Fuel Emissivity (FEMISS) ..3 References ............................................3 3...........2 References ..............22 2........1 3.............7 Fuel Swelling (FSWELL) ....1..... 2.......................................................................1 Model Development and Comparisons ........................1 Model Development and Comparisons ........................................ix List of Tables ......34 2................4 3................................................................... 2.......................................................................................................... 2.............................. 1..1 1.................................................... 3.....................0 Cladding Material Properties ..................................................................................................................................................................2.............................. 2.................................................................... 2..................31 2....................30 2....5....... 2................31 2.....................................24 2..........................6............. 2....................5.6 3....................................................................................2...........................................................3 2.................21 2......................................... 2...........4................................ 2....................................7....... 2...xi Executive Summary ...........................3 Cladding Oxide Thermal Conductivity (ZOTCON) ...................................0 Introduction ..............................4 2....1 Model Development and Comparisons .................. 2.......4......1 3........................................
....................4..8................................................8..........14 3................1 Model Development and Comparisons ..........................2 Applicability and Uncertainty ...............................1 4....................................9.... 3.........1..........................29 3...............1 4.........................25 3........................9............3 References ...................... 3.........................................0 Gas Material Properties ..................... 3..........................33 3........................14 3........................................................................ 3. 3.......7....................8... 4.................3...................................................... 3.................................................................3 References ............................................. 3...................8 Appendix A Subroutine Source Codes ................. 4..18 3........1 Model Development and Comparisons ..........1 Model Development and Comparisons ........... 3......................1 Model Development and Comparisons .................................14 3.1 Model Development and Comparisons .........5..............................................17 3.................................... 4....1 Gas Conductivity (GTHCON) .............3 References . 3......... 3.....................................................8 Creep Rate (CREPR) .................... 3.........1....................34 3..14 3......... 3...........................................................................24 3.................... 4..11 3..................................................9 Cladding Meyer Hardness (CMHARD).............................................9.........................................................1 viii .....36 3........25 3...6............................2 Applicability and Uncertainty .......................................................5................................................................................................ 3................................3 References ............... 3..12 3........................................ 3..........................................................................................................................35 3.......................24 3.....2 Applicability and Uncertainty ... 3...................... 4......11 3.......................................................7 4.........................................................................................3..18 3.....30 3....................................4 Cladding Surface Emissivity (ZOEMIS) .................4...........................7.......... A.......11 3......6 Cladding Elastic Modulus (CELMOD) and Shear Modulus (CSHEAR) .....1 Model Development and Comparisons ......2 Applicability and Uncertainty .....................................19 3.......................................29 3.................................................... 3............6..... 3.................4..36 4.............. 3....2 Applicability and Uncertainty ........... 3... 3.......................1.8 3.1 Model Development and Comparisons ..............3 References ..................5 Cladding Thermal Expansion (CTHEXP) .................. 3................................. 3...............................5...2 Applicability and Uncertainty ........................7.............................3 References ..........................................................7 Cladding Axial Growth (CAGROW) ...................................................................30 3.....................................................34 3..................................................................6............2 Applicability and Uncertainty .....3 References ................3 References ......1 Model Development and Comparisons .......... 3..........................................2 Applicability and Uncertainty ....1 4..............3........................ 3................ 3............... 3......................3...........................
........35 Figure 2.................................... 2....3...... and 7 wt% are presented..........38 Figure 2...... .......................51 Figure 2............. 2.15 Modified NFI modeltodata comparison for thermal conductivity of unirradiated UO2........3. 2............................................................33 Comparison between swelling rates determined for various instrumented fuel assemblies tests at Halden and swelling rates predicted by FRAPCON3..................22 Figure 2.................... 2...... 3..12 Figure 2..................................15 MATPRO modeltodata comparison for thermal conductivity of irradiated UO2..................................19 Duriez/modified NFI modeltodata comparison for thermal conductivity of MOX with modeled with 7 wt% of plutonium... 2.11 A comparison between solidus temperatures calculated by MATPRO using the FHYPRP subroutine and FRAPCON3/FRAPTRAN (FRAP) using the PHYPRP subroutine... 2.................5 CTHCON model and data collected since development................................31 Figure 2...21 Figure 3.. ...17 MATPRO and Duriez/modified NFI thermal conductivity models as a function of temperature for MOX fuel with 7 wt% plutonium content.............19 Data to model comparison for FEMISS... 2... 2...................36 Figure 2.......................23 Modeltodata comparison for UO2 correlation used in MATPRO and FRAPCON3/FRAPTRAN with updated constants............................................................ ... Unlike the MATPRO model............ ...............72 Figure 3.......................2)O2 from three experimenters compared with the FRAPCON3/FRAPTRAN/MATPRO (FRAP) correlations (solid line) for MOX.30 Data comparison between MATPRO and FRAPCON3....33 Figure 2..... ..........26 Comparison of densification predictions generated by the FUDENS model and experimental densification measurements determined in reactor...................... 2......22 ix ...... 2.. 3..........................18 MATPRO modeltodata comparison for thermal conductivity of MOX modeled with 7 wt% of plutonium..... . 2...3....... Results for burnup varying from 0 to 62 GWd/MTU and Pu contents of 0............34 CTHCON model and data used for development..............34 Figure 2...41 Figure 2..16 Modified NFI modeltodata comparison for thermal conductivity of irradiated UO2........ Results for burnup varying from 0 to 62 GWd/MTU and Pu contents of 0.... the modified NFI model can account for burnup and is presented at burnup levels of 0 and 30 GWd/MTU for comparison...25 Modeltodata comparison for PuO2 correlation used in MATPRO and FRAPCON3/FRAPTRAN... 2.......... 2.......61 Figure 2......................................3................. and 7 wt% are presented.......................... .................................. ...32 Figure 2.......... 2.............3..... 3............ the Duriez/modified NFI model can account for burnup and is presented at burnup levels of 0 and 30 GWd/MTU for comparison.............. .........3 Specific heat capacity of UO2 from three experimenters compared with the FRAPCON3/FRAPTRAN/MATPRO (FRAP) correlation (solid line) for UO2.... Unlike the MATPRO model... 2.7 MATPRO and modified NFI thermal conductivity models for UO2 as a function of temperature..3. 2.....................................................................21 Figure 2.........List of Figures Figure 2...... ............ 3......14 MATPRO modeltodata comparison for thermal conductivity of unirradiated UO2.....4 versions of FSWELL............................................................ ........37 Figure 2.....52 Figure 2...................8Pu0...3.4.......... 2.......................2 A comparison between liquidus temperatures calculated by MATPRO using the FHYPRP subroutine and FRAPCON3/FRAPTRAN (FRAP) using the PHYPRP subroutine...........................6 Figure 2.................71 Figure 2.........6 Specific heat capacity of (U0..............3......... ................................. .....................2........................
... ......36 Modeltodata comparison for He thermal conductivity.......................4/FRAPTRAN 1.......15 Figure 4..4 (FRAP) and MATPRO diametral thermal expansion....... Data from cladding tubes.... ........... measured hoop strain for SRA Zircaloy model. The standard error of L/L (%) is 0............... 3.... ... Vandellos data was collected by Irisa (2000) while North Anna and BR3 data was collected by Sabol (1994)...27 Modeltodata comparison for RXA Zircaloy2 (BWR)...05.. ............ 3..91 Figure 4............................... 3.......17 Modeltodata comparison for FRAPCON3....10 Modeltodata comparison for cladding oxide emissivity............07... ..27 Modeltodata comparison for M5....23 Modeltodata comparison for SRA Zircaloy4 (PWR).76 Figure 3.............17 Predicted vs......75 Figure 3...12 Figure 4...................................................................... 3.....14 Figure 4.........61 Figure 3...41 Figure 3........... Data from cladding tubes..... ........... ..................................... 3.............. The standard error of L/L (%) is 0.........31 Figure 3....... 4..3 Modeltodata comparison for Ar thermal conductivity......... 4. 3.....28 Modeltodata comparison for ZIRLO..............4 (FRAP) cladding oxide thermal conductivity models............ 4........ The standard error of L/L (%) is 0.............. 4..........5 Modeltodata comparison for Xe thermal conductivity....11..........77 Figure 3....................51 Figure 3..81 Figure 3...33 Predicted vs..6 Modeltodata comparison for H2 thermal conductivity... The standard error of L/L (%) is 0. 3................... ......... measured hoop strain for RXA Zircaloy model...... .... .. ...11 Figure 4...............4/FRAPTRAN 1.. ... 4.. 4............................82 Figure 3..17 Oxide thermal conductivity measurements as a function of temperature compared with the MATPRO and FRAPCON3.. 3...4/FRAPTRAN 1.... ..4 Modeltodata comparison for Kr thermal conductivity..05..... measured values for Young’s modulus for various data sets..............Figure 3...............13 Modeltodata comparison for FRAPCON3...........6 Modeltodata comparison for N2 thermal conductivity........13 Figure 4..............7 x ..... 3......33 CMHARD correlation as a function of temperature with data from Peggs and Godin....... 3.............16 Figure 4.................... 4.78 Figure 3................52 Figure 3............ 3. 3.........7 Modeltodata comparison for steam thermal conductivity.........28 Predicted vs..4 (FRAP) and MATPRO axial thermal expansion..................... (2000).................................................................. M5 data was collected by Gilbon et al... .
4.............24 Specific heat capacity database for Zircaloy2..2 Constants used in UO2.....51 Table 3.... 3.....2 Specific heat capacity database for ZrNb1 alloys......... FRAPCON3........................ and PuO2 heat capacity and enthalpy correlations.8 xi ..............4..........12 Table 3.. .....List of Tables Table 1..... 2...1 Table 2....21 Table 2.......11 Table 3.1 Constants used in gas thermal conductivity correlations in FRAP..2 Standard deviation of gas thermal conductivity correlations........31 Constants used in gas thermal conductivity correlations in MATPRO... Zircaloy4................... 4..........5 Parameters used in UO2 and PuO2 solidphase thermal expansion correlations................. 1.......13 Roadmap to documentation of models and properties in NRC fuel performance codes.12 Table 4............2.4 and FRAPTRAN 1...... 3.. .... . 4..... ZIRLO.......81 Table 4.... Gd2O3..........4 creep equation for SRA and RXA cladding....2 Parameters for FRAPCON3.... 3..........11 Table 4............... .......... and M5...... ....
xii .
S. xiii .4 code description document (Geelhood et al.. 2008). modeltodata comparisons and source codes for the FRAPCON3. respectively. which is a compilation of material property correlations with an extensive history of use with various fuel performance and severe accident codes. 2010a).Executive Summary The U.4. This is one of three documents that describe the FRAPCON3. The FRAPCON3. The FRAPTRAN 1.4 fuel performance codes.4 code structure and behavioral models are described in the FRAPTRAN 1.. In addition to modeltomodel comparisons. Additional descriptions of mechanical models and modeltodata comparisons for mechanical property correlations are provided in the mechanical properties report (Geelhood.4 correlations are also provided. 2010b).4 and FRAPTRAN 1.4 and FRAPTRAN 1. Comparisons are also made with MATPRO. in regulatory analysis.4 code structure and behavioral models are described in the FRAPCON3. FRAPCON3. In this document.4 and FRAPTRAN 1. Nuclear Regulatory Commission uses the computer codes FRAPCON3 and FRAPTRAN to model steadystate and transient fuel behavior.4 code description document (Geelhood et al. material property correlations for oxide fuels and cladding materials are presented and discussed. Comparisons are made between the material property correlations used in the most recent versions of the codes.
xiv .
to describe deposits observed on fuel from the test reactor at Chalk River) gadolinia gigapascal(s) gigawattdays per metric ton of uranium hydrogen helium joule(s) kelvin kilogram(s) krypton lightwater reactor meter(s) square meter(s) cubic meter(s) megaelectron volt(s) millimeter(s) mixed oxide.S. Pu)O2 megapascal(s) megawattdays per kilogram of uranium megawattdays per metric ton of uranium megawattseconds per kilogram of uranium neutron(s) nitrogen niobium Nuclear Fuels Industries U. originally coined by Atomic Energy of Canada.Abbreviations °C °F Ar BWR cm2 crud degrees Celsius degrees Fahrenheit argon boilingwater reactor square centimeter(s) Chalk River Unidentified Deposit (generic term for various residues deposited on fuel rod surfaces. Nuclear Regulatory Commission oxygentometal pascal(s) Pacific Northwest National Laboratory plutonium plutonium dioxide pressurizedwater reactor recrystallized annealed second(s) stress relief annealed theoretical density tonne(s) of heavy metal uranium dioxide uraniagadolinia watt(s) percent by weight Gd2O3 GPa GWd/MTU H He J K kg Kr LWR m m2 m3 MeV mm MOX MPa MWd/kgU MWd/MTU MWs/kgU n N Nb NFI NRC O/M Pa PNNL Pu PuO2 PWR RXA s SRA TD tHM UO2 UO2Gd2O3 W wt% xv . (U. Ltd.
Xe ZrNb1 ZrO2 m
xenon Russian zirconium alloy with 1 wt% of niobium zirconium dioxide micrometer(s)
xvi
1.0 Introduction
The U.S. Nuclear Regulatory Commission (NRC) uses the computer codes FRAPCON3 and FRAPTRAN to model steadystate and transient fuel behavior, respectively, in regulatory analysis. To effectively model fuel behavior, material property correlations must be used for a wide range of operating conditions (e.g., temperature and burnup). In this sense, a “material property” is a physical characteristic of the material whose quantitative value is necessary in the analysis process. Further, the property may be used to compare the benefits of one material with those of another. Generally speaking, the material properties of interest in regulatory analysis of nuclear fuel behavior are mechanical or thermodynamic. The issue of what is and is not a “material property” will never be universally resolved. In this report, properties such as thermal conductivity are included. Other characteristics of the material (e.g., fission gas release) are considered “models” rather than properties and are discussed elsewhere. The material property correlations for the FRAPCON3 and FRAPTRAN computer codes were partially documented in NUREG/CR6534 and NUREG/CR6739, respectively (Lanning et al., 2005; Cunningham et al., 2001) as well as in NUREG/CR6150 (Siefken et al., 2001). Some of these have been modified or updated since the original code documentation was published. The primary purpose of this report is to consolidate the current material property correlations used in FRAPCON3 and FRAPTRAN into one document. Material property correlations for oxide fuels, including uranium dioxide (UO2) and mixed oxide (MOX) fuels, are described in Section 2. Throughout this document, the term MOX is used to describe fuels that are blends of uranium and plutonium oxides, (U,Pu)O2. The properties for UO2 with other additives (e.g., gadolinia) are also discussed. Material property correlations for cladding materials and gases are described in Sections 3 and 4, respectively. In addition to describing the material property correlations used in the subroutines of FRAPCON3 and FRAPTRAN, this report provides a variety of comparisons between material property correlations and data. Although they are frequently identical, comparisons are made between the material property correlations used in the FRAPCON3 and FRAPTRAN codes. Comparisons are also made between the material property correlations used in MATPRO, a compilation of fuel and cladding material property correlations with an extensive history of use with various fuel performance and severe accident codes. For several reasons, consistency among the material property correlations in FRAPCON3, FRAPTRAN, and MATPRO has never been complete. However, the current versions of FRAPCON3 and FRAPTRAN use a relatively consistent set of correlations for the properties that are used by both codes. The material property correlations in the most recent version of MATPRO are documented in Volume 4 of NUREG/CR6150 (Siefken et al., 2001). In addition to comparison of the various correlations, correlationtodata comparisons are made with FRAPCON3, FRAPTRAN, and MATPRO properties. All comparisons in this report are based on the material property correlations used in the most recent version of the FRAPCON3 and FRAPTRAN codes, FRAPCON3.4 and FRAPTRAN 1.4. The source code for each material property correlation discussed is provided for FRAPCON3.4 and FRAPTRAN 1.4 (see Appendix A) as well as a range of applicability and an estimate of uncertainty, where possible.
1.1
1.1
Relation to Other Reports
The full documentation of the steadystate and transient fuel performance codes is described in three documents. The basic fuel, cladding, and gas material properties used in FRAPCON3.4 and FRAPTRAN 1.4 are described in the material properties handbook (this report). The FRAPCON3.4 code structure and behavioral models are described in the FRAPCON3.4 code description document (Geelhood et al., 2010a). The FRAPTRAN 1.4 code structure and behavioral models are described in the FRAPTRAN 1.4 code description document (Geelhood et al., 2010b). Additional descriptions of mechanical models and modeltodata comparisons for mechanical property correlations are provided in the mechanical properties report (Geelhood, 2008). Table 1.1 shows where each specific material property and model used in the NRC fuel performance codes are documented. Table 1.1
Model/Property Fuel thermal conductivity Fuel thermal expansion Fuel melting temperature Fuel specific heat Fuel enthalpy Fuel emissivity Fuel densification Fuel swelling Fission gas release Fuel relocation
Roadmap to documentation of models and properties in NRC fuel performance codes, FRAPCON3.4 and FRAPTRAN 1.4.
FRAPCON3.4 Material properties handbook Material properties handbook Material properties handbook Material properties handbook Material properties handbook Material properties handbook Material properties handbook Material properties handbook FRAPCON3 code description FRAPCON3 code description FRAPTRAN 1.4 Material properties handbook Material properties handbook Material properties handbook Material properties handbook Material properties handbook Material properties handbook NA NA FRAPTRAN code description FRAPTRAN code description
Fuel grain growth High burnup rim model Nitrogen release Helium release Radial power profile Stored energy Decay heat model Fuel and cladding temperature solution Cladding thermal conductivity Cladding thermal expansion Cladding elastic modulus Cladding creep model Cladding specific heat
FRAPCON3 code description FRAPCON3 code description FRAPCON3 code description FRAPCON3 code description FRAPCON3 code description FRAPCON3 code description NA FRAPCON3 code description Material properties handbook Material properties handbook Material properties handbook Material properties handbook Material properties handbook
NA NA NA NA NA (input parameter) FRAPTRAN code description FRAPTRAN code description FRAPTRAN code description Material properties handbook Material properties handbook Material properties handbook NA Material properties handbook
1.2
C. 2. C. D. watermoderated energy reactor FEA = finite element analysis 1. and K. Richland. P. 2001. Beyer. Vol. 1999) VTTR1133706 (Knuttilla.D. Pacific Northwest National Laboratory.E. PNNL13576. Pacific Northwest National Laboratory. Geelhood.4 Material properties handbook NA Material properties handbook FRAPTRAN code description FRAPTRAN code description FRAPTRAN code description NA (input parameter) NA (input parameter) FRAPTRAN code description FRAPTRAN code description FRAPTRAN code description Material properties handbook NA Material properties handbook FRAPTRAN code description FRAPTRAN code description FRAPTRAN code description FRAPTRAN code description Optional models and properties not developed at PNNL NA VTTR1133706 (Knuttilla. Berna. Including MixedOxide Fuel Properties. G. FRAPTRAN: Integral Assessment.2 References Lanning. 2006) NA = not applicable VVER = watercooled. Medvedev. 4. Beyer. Panisko. 1. ultimate stress..E. and plastic deformation Cladding failure criteria Cladding waterside corrosion Cladding hydrogen pickup Cladding high temperature oxidation Cladding ballooning model Cladding mechanical deformation Oxide thermal conductivity Crud thermal conductivity Gas conductivity Gap conductance Plenum gas temperature Rod internal pressure Coolant temperature and heat transfer coefficients VVER fuel and cladding models Cladding FEA model FRAPCON3.J. M. Richland. NUREG/CR6739. Vol.4 Material properties handbook Material properties handbook Material properties handbook FRAPCON3 code description FRAPCON3 code description NA FRAPCON3 code description FRAPCON3 code description NA NA FRAPCON3 code description Material properties handbook FRAPCON3 code description Material properties handbook FRAPCON3 code description FRAPCON3 code description FRAPCON3 code description FRAPCON3 code description FRAPTRAN 1. NUREG/CR6534. and H. PNNL11513.Model/Property Cladding emissivity Cladding axial growth Cladding Meyer hardness Cladding annealing Cladding yield stress.G.E.A. F.E. WA. WA. Cunningham.H. Scott. 2006) NUREG/IA0164 (Shestopalov et al... 2005. FRAPCON3 Updates.3 .
Vol. PNNL19418. 2008. 1. Pacific Northwest National Laboratory.J. Beyer. Improvements on FRAPCON3/FRAPTRAN Mechanical Modeling.K. W. SCDAP/RELAP5/MOD 3. Nuclear Regulatory Commission.E. U. WA. ID.J. Richland. Finland. 1. Knuutila. Hohorst. NUREG/CR7023.3 Code Manual MATPROA Library of Materials Properties for LightWaterReactorAccident Analysis. NUREG/CR7022. 1999. G. FRAPTRAN 1. K. Geelhood. Luscher. ThermalMechanical Behavior of Oxide Fuel Rods for High Burnup.. Vol. Cuta. Abyshov. Richland. Coryell. PNNL17700. 2006. VTTR1133706. Richland.S.W. K. 2010a. Geelhood.4: A Computer Code for the Calculation of SteadyState.G. 4. Pacific Northwest National Laboratory. 2. Luscher. Harvego.4: A Computer Code for the Transient Analysis of Oxide Fuel Rods. Idaho National Engineering and Environmental Laboratory.4 . L. Lioutov. 1.Siefken. Vol. Shestopalov.. Beyer. INEL96/0422. and C. 1.. Washington. and C. VTT.A. Geelhood. 1. PNNL Stress/Strain Correlation for Zircaloy. K. WA. Beyer.. NUREG/CR6150. K. WA. 2001.G. 2010b. Modification of USNRC’s FRAPT6 Fuel Rod Transient Code for High Burnup VVER Fuel.C. E. FRAPCON3. E. Pacific Northwest National Laboratory. and J. C.G. Idaho Falls. and K. Mikitiouk. NUREG/IA0164.. W. A. A. Vol.M.J. Rev. PNNL19400.E. L. D.J. Vol. W. and J.E. Yegorova. Luscher.
which was determined experimentally by Brassfield (1968).15. thermal expansion.468390 x10 3 c 2 liqdus (c) 2840 3. These correlations include the melting temperature. These correlations are presented in Equations 2.1.41395 c 7. thermal conductivity.0 Oxide Fuel Properties Material property correlations for UO2 and MOX fuel pellets are described in the following subsections. FHYPRP.15K as the melting temperature of urania.15) where c sldus (c) liqdus (c) ftmelt FBu fdelta = = = = = = Pu content (wt%) solidus temperature as a function of Pu content (°C) liquidus temperature as a function of Pu content (°C) fuel melting temperature (K) burnup (MWd/MTU) temperature range between solidus and liquidus (K) 2. The subroutine containing the material correlation in the FRAPCON3/FRAPTRAN codes is given in parentheses in the heading of each subsection. some correlations account for gadolinia (Gd2O3) additions in UO2 fuel pellets.15 5 FBu / 10000 fdelta liqdus (c) sldus (c) 5 FBu / 10000 (2. sldus (c) 2840 5.1 . specific heat capacity and enthalpy.4 and exhibits smaller burnup dependence than the corresponding MATPRO correlation. emissivity.14) (2.21860 c 1. 2.15 5 FBu / 10000 ftmelt sldus (c) 273.13) For c > 0 (2. and swelling of the fuel pellets. The same subroutine is used in both FRAPCON3. 2. and a least squares fit to parabolic equations for the solidus and liquidus boundaries from the Lyon and Baily (1967) phase diagram for the stoichiometric (U. These temperatures are calculated as a function of burnup and plutonia content. When indicated.1 Model Development and Comparisons The equations used to calculate the UO2 and MOX melting points were derived by using 3113.11 through 2.1 Fuel Melting Temperature (PHYPRP) The subroutine PHYPRP is used to calculate the temperature of the first appearance of liquid phase (solidus) and the temperature of the melting point of the last solid phase (liquidus) of UO2 and MOX.12) ftmelt 3113. Pu)O2 mixed oxide.11) (2.4 and FRAPTRAN 1.2. densification.448518 x10 2 c 2 For c = 0 (2.
respectively. However.12. (2000) and is based on data from Adamson et al. However. (1985) and Komatsu et al. Also.2 . 2. Results for burnup varying from 0 to 62 GWd/MTU and Pu contents of 0. This change was suggested by Popov et al.13 through 2. (1988). these figures reveal a smaller decline in solidus and liquidus temperatures for the FRAPCON3 and FRAPTRAN subroutine (PHYPRP) than the MATPRO subroutine (FHYPRP). the burnup dependence of the PHYPRP subroutine differs from the FHYPRP subroutine. and 7 wt% are presented. However. This is a result of reducing the burnup dependence from 3. the term 3.5 K/GWd/MTU between the FHYPRP and PHYPRP subroutines.21860 in PHYPRP.21660 in FHYPRP to 3. A burnup dependence of 3. Solidus Temperatures 3125 3075 Temperature (K) 3025 2975 2925 2875 2825 0 MATPRO 0 wt% FRAP 0 wt% 10 20 MATPRO 3 wt% FRAP 3 wt% 30 40 MATPRO 7 wt% FRAP 7 wt% 50 60 Burnup (GWd/MTU) Figure 2.11 A comparison between solidus temperatures calculated by MATPRO using the FHYPRP subroutine and FRAPCON3/FRAPTRAN (FRAP) using the PHYPRP subroutine. 3.5 K/GWd/MTU is used in PHYPRP.12 is slightly different from the term used in FHYPRP. The coefficient was changed from 3. while a burnup dependence of 0.2 K/GWd/MTU is used in FHYPRP. Both subroutines predict solidus and liquidus temperatures that decrease with increasing burnup and plutonium concentration.The correlations defined above for the subroutine PHYPRP are almost identical to those used in the subroutine FHYPRP documented in MATPRO.21860 c in Equation 2. Comparisons between solidus and liquidus temperature calculated by PHYPRP and FHYPRP are presented in Figures 2.11 and 2.2 to 0. this change does not appreciably affect the liquidus temperature. The burnupdependent term in Equations 2.15 reflects this change in burnup dependence.
“Experimental and thermodynamic evaluation of the melting behavior of irradiated oxide fuels.3125 3075 Liquidus Temperatures Temperature (K) 3025 2975 2925 2875 2825 0 MATPRO 0 wt% FRAP 0 wt% 10 20 MATPRO 3 wt% FRAP 3 wt% 30 40 MATPRO 7 wt% FRAP 7 wt% 50 60 Burnup (GWd/MTU) Figure 2.E.1.. Oak Ridge National Laboratory. Ivanov.W. OH. Thermophysical Properties of MOX and UO2 Fuel Including the Effects for Irradiation. 1967.3 . M.3 References Brassfield.G.. E. 3. Due to the experimental data and data interpretations on which these temperature correlations are based.” Journal of Nuclear Materials. there are plans to develop a correlation in a future revision of the code. and G. 2. 22:332. V. Popov.F. 2. Results for burnup varying from 0 to 62 GWd/MTU and Pu contents of 0. and R.. Recommended Property and Reactor Kinetics Data for Use in Evaluating a LightWaterCoolant Reactor LossofCoolant Incident Involving Zircaloy4 or 304SSClad UO2.K. 130:349365. Aitken. H. TN.. Adamson. Yoder. Carbajo.J. 1985. Cincinnati. A conservative estimate of uncertainty based on engineering judgment is ± 50K. GEMP482. Oak Ridge. Caputi. Baily. S. J. W. Lyon.12 A comparison between liquidus temperatures calculated by MATPRO using the FHYPRP subroutine and FRAPCON3/FRAPTRAN (FRAP) using the PHYPRP subroutine. and 7 wt% are presented. However. General Electric Company. 2.C.G. “The Solid Liquid Phase Diagram for the UO2PuO2 System.A.L. et al. it is difficult to express a quantitative measure of uncertainty. there are no models for uraniagadolinia (UO2Gd2O3) fuel melting. Currently. 1968. 2000.1. ORNL/TM 2000/351.” Journal of Nuclear Materials.2 Applicability and Uncertainty The correlation used in PHYPRP is valid through the melting temperature of UO2 and MOX for burnups up to 62 GWd/MTU. and W.
respectively.2 Fuel Specific Heat Capacity (FCP) and Fuel Enthalpy (FENTHL) The subroutines FCP and FENTHL are used to calculate the specific heat capacity and enthalpy of the fuel pellet.21. composition.Komatsu. but with different constants. The same subroutine is used in FRAPCON3. FCP YK 3 E D K 1 2 exp ( / T ) K 2T exp( E D / RT ) T 2 exp( / T ) 1 2 2RT 2 K 1 K T2 Y 2 K 3 exp ( E D / RT ) exp ( / T ) 1 2 2 (2. molten fraction.21) FENTHL where FCP FENTHL T Y R ED = = = = = = = (2. J.22) specific heat capacity (J/kg*K) fuel enthalpy (J/kg) temperature (K) oxygentometal ratio universal gas constant = 8. The basic relationships are presented in Equations 2. The specific heat capacity and enthalpy of nuclear fuel are modeled empirically as functions of four parameters: temperature. T.” Journal of Nuclear Materials.4. respectively. 1988. and oxygentometal (O/M) ratio. “The Melting Temperature of Irradiated Oxide Fuel. FRAPTRAN 1. 2.. and MATPRO.1 Model Development and Comparisons Equations for the specific heat capacity and enthalpy of solid UO2 and plutonium dioxide (PuO2) are assumed to have the same form. Tachibana. 2. and K.22. Konashi.3143 (J/mol*K) the Einstein temperature (K) activation energy for Frenkel defects (J/mol) and the constants are given in Table 2.4.2.21 and 2. 154:3844. 2.4 .
Table 2.21
Constant K1 K2 K3 ED
Constants used in UO2, Gd2O3, and PuO2 heat capacity and enthalpy correlations.
UO2 296.7 2.43 x 10
2
PuO2 347.4 3.95 x 10
4
GdO2* 315.86 4.044 x 10 0.0 348.0
5 2
Units J/kg*K J/kg*K2 J/kg K J/mol
8.745 x 107 535.285 1.577 x 10
5
3.860 x 107 571.000 1.967 x 10
0.0
* Gadolinia additions are only accounted for in FRAPCON3.4.
The constants in Table 2.21 were determined by Kerrisk and Clifton (1972) for UO2 and Kruger and Savage (1968) for PuO2. The specific heat capacity of UO2 in the liquid state (see Equation 2.23) was determined by Leibotwitz (1971) and assumed to be valid for PuO2 in the liquid state. FCPLiquid = 503 J/kg*K (2.23)
Inspection of Equations 2.21 and 2.22 reveals that fuel enthalpy, or stored energy, is calculated by integrating the fuel specific heat capacity correlation from 0 to T in kelvin. Stored energy is important in reactor transient analysis because the severity of the transient is significantly affected by the initial stored energy of the fuel. Since UO2 and PuO2 are the principal oxides in lightwater reactor (LWR) fuels, they are the constituents considered. The correlations for specific heat and enthalpy in both UO2 and MOX fuel pellets are identical in FRAPCON3.4, FRAPTRAN 1.4, and MATPRO. However, FRAPCON3.4 can accommodate gadolinia additions whereas FRAPTRAN cannot. For a mixture of UO2, Gd2O3, and PuO2, the specific heat capacity of the solid is determined by combining the contribution from each constituent in proportion to its weight fraction. When the material is partially molten, the heat capacity is determined similarly with a weighted sum of the solid and molten fractions. Since the specific heat correlation is only valid above a fuel temperature of about 300K, the fuel enthalpy correlation is not valid below a temperature of about 300K. In addition, it is necessary to calculate fuel enthalpy with respect to a reference temperature 300K. Thus, the fuel enthalpy at any desired temperature, T, is calculated by evaluating Equation 2.22 at T and a reference temperature, Tref, of 300K and taking the difference [FENTHL(T) – FENTHL(Tref)]. For temperatures greater than 2K below melting, the molten fraction and heat of fusion are used to interpolate between the enthalpy of unmelted fuel and justmelted fuel at the melting temperature. Specific heat correlations are compared with independent experimental data in Figure 2.21 and Figure 2.22 for UO2 and PuO2, respectively. Comparison between the UO2 model and experimental data from Hein (1968), Leibowitz (1969), and Gronvold (1970) in Figure 2.21 reveals agreement between predicted and measured values up to high temperatures (a few hundred kelvin below melting). At this point the data begin to fall lower than the model. This is attributed to partial melting due to a nonuniform temperature distribution within the sample. With the exception of the Affortit (1970) data, relatively good agreement is observed in Figure 2.22, which compares the values measured by Gibby (1974), Leibowitz (1972), and Affortit with predicted heat capacities for a MOX fuel, (U0.8, Pu0.2)O2. Since the Affortit results are known to be generally low in comparison to results from other investigators, the correlation is considered to be in good agreement with the experimental data. 2.5
800
Hein
Specific Heat Capacity (J/kg*K)
700 600 500 400 300 200 300
Leibowitz Gronvold FRAP
800
1300
1800 Temperature (K)
2300
2800
3300
Figure 2.21
Specific heat capacity of UO2 from three experimenters compared with the FRAPCON3/FRAPTRAN/MATPRO (FRAP) correlation (solid line) for UO2.
2.6
700 650 Specific Heat Capacity (J/kg*K) 600 550 500 450 400 350 300 250 200 300 Gibby Leibowitz Affortit FRAP
800
1300
1800 Temperature (K)
2300
2800
3300
Figure 2.22
Specific heat capacity of (U0.8Pu0.2)O2 from three experimenters compared with the FRAPCON3/FRAPTRAN/MATPRO (FRAP) correlations (solid line) for MOX.
2.2.2
Applicability and Uncertainty
The correlations for fuel specific heat and enthalpy are valid for temperatures from 300K to more than 4000K. The standard error of the UO2 specific heat capacity correlation is 3 J/kg*K. However, this error increases at temperatures above 2300K, as seen in Figure 2.21, and may lead to underpredictions and overpredictions of 59 and 66 J/kg*K, respectively. As for the mixedoxidespecific heat capacity correlation, it is 6 to 10 J/kg*K, depending on the fraction of PuO2. For nonstoichiometric fuels, these uncertainties are approximately double.
2.2.3
References
Kerrisk, J.F., and D.G. Clifton. 1972. “Smoothed Values for the Enthalpy and Heat Capacity of UO2,” Nuclear Technology, 16:531535. Kruger, O.L., and H. Savage. 1968. “Heat Capacity and Thermodynamic Properties of Plutonium Dioxide,” Journal of Chemical Physics, 45:45404544. Leibowitz, L., et al. 1971. “Enthalpy of Liquid Uranium Dioxide to 3500 K,” Journal of Nuclear Materials, 39:115116. Hein, R.A., L.H. Sjodahl, and R. Swarc. 1968. “Heat Content of Uranium Dioxide from 1,200 to 3,100 K,” Journal of Nuclear Materials, 25:99102.
2.7
phonon) and electronhole pair (i. “Chaleur Specifique a Haute Temperature des Oxydes D'Uranium et de Plutonium. However. “Enthalpy of Uranium Plutonium Oxides (U0.3 Fuel Thermal Conductivity (FTHCON) The subroutine FTHCON is used to calculate the thermal conductivity of the fuel pellet. 2.. Unlike the models used in 2. “Enthalpy of Solid Uranium Dioxide from 2. 2:665679. and J.. 50:155161. 1970.Leibowitz. the calculation of fuel thermal conductivity has evolved from the model originally proposed in MATPRO to the modified Nuclear Fuels Industries (NFI) and Duriez/modified NFI models used in the FRAPCON3. Leibowitz. and M.Pu0. respectively. and plutonium content. that led to its replacement by a modified model developed at NFI. 2. L. 1970.017 and UO2..W. An additional fuel thermal conductivity model for MOX developed at Halden is included as an option in FRAPCON3. These updates were incorporated in FRAPCON3.4 codes for UO2. This historical summary of how thermal conductivity modeling has evolved provides insight to model development and previous motivations for change. L.3. it is only used to calculate the melt temperature. Each of the models is discussed in detail in the following sections along with the evolution of the MATPRO model and the applicability and uncertainty of the models. 1972. O/M ratio. and MOX. an overview of subsequent thermal conductivity models is given.. 7:236241. Chasanov. 42:113116. Fischer. and M. however. FRAPTRAN 1. and MATPRO.8. However. “Thermodynamics of the UO2+x Phase I. Gronvold.4. Further refinements of this model were implemented by Pacific Northwest National Laboratory (PNNL) to provide a better fit to UO2 data. 1974. Chasanov. 1969.07 from 2. For example..” Journal of Chemical Thermodynamics. 29:356358. et al. Since accurate predictions of fuel rod behavior are strongly dependent on temperature.4 and FRAPTRAN 1.4. The FTHCON subroutine is used in FRAPCON3. 1981). “Analytical Expressions for Enthalpy and Heat Capacity for UraniumPlutonium Oxide. Gibby.1 Model Development and Comparisons The original MATPRO model for fuel thermal conductivity is based on a mechanistic description of thermal conductivity. electronic) contributions (Hagrman et al.G.. UO2Gd2O3.000 K and Electronic Contributions.8 . D.F.254 from 300 to 1.4. there were several issues with the Lucuta model. The MATPRO thermal conductivity model is a function of temperature..500 K to its Melting Point.L. L.” Revue Internationale des Hautes Temperatures et des Refractaires.” Journal of Nuclear Materials.. including its nonstandard form and weak burnup degradation term.000 K.2) O1.e. the calculation of fuel thermal conductivity is critical. Burnup is also included. Following the initial description of the MATPRO fuel thermal conductivity model.” Journal of Nuclear Materials. C. Heat Capacities of UO2. In addition. PNNL added refinements to include the Duriezderived modifications to the NFI model. density. including lattice vibration (i. Mschler. Marcon.” Journal of Nuclear Materials. F. which permit MOX fuels to be modeled. Affortit. R. the MATPRO model was followed by a model by Lucuta that accounted for burnup degradation and was incorporated in FRAPCON3. et al.e.3 and continue to be used in FRAPCON3.350 to 3.4.G.
5 0. The MATPRO model used to express thermal conductivity of solid fuel is described with Equation 2. but this effect is significantly decreased above 1600K due to radiation effects and gas conductivity within the pores at these temperatures. a factor proportional to the point defect contribution to the phonon mean free path (m*s/kg*K). T = 6. for temperature in the range 1800 to 2300K. linear strain caused by thermal expansion when temperature is > 300K (unitless).FRAPCON3. 2.. fraction of theoretical density (unitless).4 and FRAPTRAN 1.69x103).31. The dominant contributor to thermal conductivity (i. estimates of liquid fuel thermal conductivity are made from physical considerations because data for molten fuel have not been found. 2 (2. CV D k 1 6. MATPRO does not account for the degradation of thermal conductivity with increasing burnup.50 – T*(4. T is found by interpolation. The specific heat capacity of the fuel (CV) and the phonon scattering factor (B) are dependent on the plutonium content. The temperature and density dependence of this term can be seen in Equation 2.00469 T '1 D A BT ' '1 3e th 5. fuel temperature if < 1800K. T is equal to 2050K.0 – O/M ratio).6 x absolute value (2. phonon contribution to the specific heat at constant volume (J/kg*K). At higher temperatures (> 3000K).06867 x (1+ 0.31 represents the phonon contribution to specific heat.9 . The first term of the correlation for fuel specific heat capacity is used for this factor. the dependence of the first term on O/M ratio and plutonium content is also included. thermal conductivity decreases with increasing porosity. whereas electronic contributions are dominant above 2000K. while the point defect contribution factor (A) is dependent on the O/M ratio. For temperatures > 2300K. the correlation used for this factor is 0. Temperature dependence can be observed in the first and second terms of Equation 2. In general.31. the MATPRO correlation for fuel thermal expansion is used for this factor fuel temperature (K).2997 x10 3 Texp 13358 / T 1 0. Although not as evident. for temperature < 1364. for temperature > 1834.31. phonon or electron) is dependent on temperature.339 + 12. porosity correction for temperature.e. which represents the electronic contribution.169 13358 / T 2 where k D CV eth T T T A = = = = = = thermal conductivity (W/m*K).6238 x plutonium content of the fuel).31) = = B = The first term of Equation 2. and for temperatures in the range from 1364 to 1834K. T is found by interpolation. T = 1. Phonon contributions to thermal conductivity are dominant below 1500K.4. a factor proportional to the phononphonon scattering contribution to the phonon mean free path (m*s/kg*K). the correlation used for this factor is 0.
This MATPRO model (Hagrman et al., 1981) was initially replaced in FRAPCON3 and FRAPTRAN with a model from Lucuta (Lucuta et al., 1996) which accounts for the degradation of thermal conductivity due to burnup. The Lucuta model for fuel thermal conductivity, K, is presented in Equation 2.32
where Ko FD FP FM FR = = = = =
K K o FD FP FM FR
thermal conductivity of unirradiated, fully dense urania factor for dissolved fission products factor for precipitated fission products factor to correct for the Maxwell porosity effect factor for the radiation effect
(2.32)
Thermal conductivity of unirradiated, fully dense urania and factors included in the Lucuta model are described by the Equations 2.33 through 2.37.
Ko
4.715 x10 9 1 16361 exp 4 2 0.0375 2.165 x10 T T T
(2.33)
1 1.09 0.0643 FD 3.265 T arctan B 1.09 0.0643 T B B 3.265 B
(2.34)
1 0.019B FP 1 3 0.019B 1 exp( T 1200 ) 100
FM 1 p 1 (s 1) p
(2.35)
(2.36)
FR 1
0.2 T 900 1 exp( ) 80
(2.37)
where Ko T B p s = = = = = conductivity of unirradiated, fully dense urania (W/mK) temperature (K) burnup in atom% (1 atom% = 9.383 GWd/MTU at 200 MeV/fission) porosity fraction (asfabricated plus swelling) shape factor ( = 1.5 for spherical pores)
Although the Lucuta model for thermal conductivity accounted for burnup degradation, this model tended to overpredict thermal conductivity values at high temperature (> 2200K).
2.10
Furthermore, the model had too little burnup degradation, making it nonconservative above 30 GWd/MTU. In addition, the form of the Lucuta model is nonstandard, which does not facilitate comparison between models. As a result, the Lucuta model was later replaced with a modified version of a thermal conductivity model proposed by NFI (Ohira and Itagaki, 1997) (Lanning et al., 2005). Both FRAPCON3.4 and FRAPTRAN 1.4 currently model urania fuel pellet thermal conductivity with the modified version of the pellet thermal conductivity model proposed by NFI (Ohira and Itagaki, 1997). The original NFI model was modified to alter the temperaturedependent portion of the burnup function in the phonon terms and change the electronic term (Lanning et al., 2005). The original (unmodified) and modified versions of the NFI thermal conductivity model are presented in Equations 2.38 and 2.39, respectively.
K 95
1 CT 2 DT 4 A BT f ( Bu ) g ( Bu ) h (T )
(2.38)
K 95
where K T Bu f(Bu) g(Bu) h(T)
1 E 2 exp( F / T ) A BT f ( Bu ) (1 0.9 exp( 0.04 Bu )) g ( Bu ) h (T ) T
(2.39)
= = = = = = = = =
thermal conductivity, W/mK temperature, K burnup, GWd/MTU effect of fission products in crystal matrix (solution) 0.00187 * Bu effect of irradiation defects 0.038 * Bu0.28 temperature dependence of annealing on irradiation defects
1 1 396 exp(Q / T)
temperaturedependent parameter (“Q/R”) = 6380K 0.0452 mK/W 2.46 x 104 mK/W/K 5.47 x 109 W/mK3 2.29 x 1014 W/mK5 3.5 x 109 WK/m 16,361K
Q A B C D E F
= = = = = = =
The NFI model was further modified in FRAPCON3.4 with a data correlation presented by Massih (Massih et al., 1992) to include gadolinia content. The modified NFI with gadolinia dependency used in FRAPCON3.4 is presented in Equation 2.310.
K
1 (2.310) A a gad BT f ( Bu ) (1 0.9 exp( 0.04 Bu )) g ( Bu ) h (T )
E exp( F / T ) T2
2.11
where a gad = = constant = 1.1599 weight fraction of gadolinia
As applied in FRACON3.4 and FRAPTRAN 1.4, the above models are adjusted for asfabricated fuel density (in fraction of theoretical density [TD]) using the Lucuta recommendation for sphericalshaped pores (Lucuta et al., 1996), as follows:
where Kd K95 d = = =
K d 1.0789 * K 95 * [d /{1.0 0.5 (1 d )}]
thermal conductivity adjusted for asfabricated fuel density, d thermal conductivity for 95 percent dense fuel asfabricated fuel density
(2.311)
The phononterm modification in the modified NFI model reduces defect annealing at low burnup from that by Ohira and Itagaki (NFI). However, the original temperaturedependent annealing is restored at higher burnups such that, for burnups greater than 40 GWd/MTU, the phonon term is equivalent to that in the original NFI model. The electronic terms (which in either case become significant above 1500K) are altered in the modified NFI model to a more theoretically based equation. The magnitude is slightly lower than the original model at high temperature. This adjustment was indicated by the Institute for Transuranium Elements data on unirradiated pressurizedwater reactor (PWR) pellet material at temperatures approaching fuel melting (Ronchi et al., 1999). At low burnups (< 20 GWd/MTU) and low temperatures (< 1000K), the modified model is higher than the unmodified NFI model and roughly equivalent to the Lucuta model without its radiation term. At higher burnup (> 30 GWd/MTU), the modified model is equivalent to the original NFI model with the exception of the small reduction at very high temperatures. Fuel thermal conductivity modeling was extended to include MOX in FRAPCON3 and FRAPTRAN by adopting a model that was a combination of the Duriez stoichiometrydependent correlation, which is derived from diffusivity measurements on unirradiated fuel pellets, and the modified version of the NFI model, which includes burnup degradation effects. The Duriez data places greater dependence of MOX thermal conductivity on stoichiometry, but only a minor dependence on plutonia content (Duriez et al., 2000). The Duriez/modified NFI model is presented in Equation 2.312.
K 95
1 (2.312) A ( x ) a gad B( x )T f ( Bu ) (1 0.9 exp( 0.4Bu )) g ( Bu ) h (T ) C D mod exp( ) 2 T T
where K95 x T A(x) = = = = conductivity at 95 percent TD, W/mK 2.00 – O/M (i.e., oxygentometal ratio) temperature, K 2.85x + 0.035, mK/W
2.12
The modified NFI model at a burnup level of 30 GWd/MTU is shown for comparison and reveals significant degradation of thermal conductivity.038*Bu0.86 – 7.1599 x 0. Finally. 500 to 3000K) for unirradiated UO2.28 temperature dependence of annealing on irradiation defects 1 1 396 exp(Q / T) temperature dependent parameter (“Q/R”) = 6380K 1. whereas the Duriez/NFI model is used for MOX.1599 weight fraction gadolinia (not expected in MOX) burnup in GWd/tHM effect of fission products in crystal matrix (solution) 0. but the modified NFI model in FRAPTRAN 1.4 does not. The Halden correlation tends to overpredict thermal conductivity data at high temperatures (> 2500K) (Lanning et al. The MATPRO model predicts lower thermal conductivities than the modified NFI model across the applicable temperature range (i.31. 2005).= = = = = = = = g(Bu) = = h(T) = = Q Cmod = = B(x) C D a gad Bu f(Bu) (2. although no gad additions are included in MOX at this time.92 (2.520K 1. the modified NFI model is used for UO2.689x109. FRAPCON3. WK/m 13. which does not account for burnup degradation of thermal conductivity..313) 0.00187*Bu effect of irradiation defects 0. K 95 0.5x109 WK/m FRAPCON 3 provides an option to use a fuel thermal conductivity model derived at Halden. W/mK temperature.4 has the option for Halden thermal conductivity correlation.13 .0040 B 2. This represents an improvement relative to the MATPRO model. The Halden correlation does not include dependency on stoichiometry. but does emphasize burnup effects.1148 a gad 1. 2.475 x10 4 (1 0.1599 weight fraction gadolinia (not expected in MOX) burnup in MWd/kg UO2 minimum of 1650°C or current temperature in °C To summarize. The modified NFI model used in FRAPCON3. unlike the MATPRO model. The Duriez/NFI models account for gad in both codes.. A comparison between the MATPRO model and the modified NFI model is presented in Figure 2. However.4 accounts for gad. °C 1.e.00188 T ) where K95 T a gad B = = = = = = conductivity at 95 percent TD. the modified NFI model can account for degradation of thermal conductivity with increasing burnup.00333 B) 0. but FRAPTRAN does not.15x)*1E4 m/W 1.3.0132 exp( 0.
In addition to Ronchi (1999).00 6.33). 2.32. with a standard deviation of about 11 percent.32 through 2.3.00 0. Based on Figure 2.3.31 MATPRO and modified NFI thermal conductivity models for UO2 as a function of temperature. Based on the smaller bias and standard deviation.00 8.00 Thermal Conductivity (W/m*K) 9. The modified NFI model also underpredicts the unirradiated data. but only by about 2.00 7.10.3.3.00 1.00 5.3.5 percent. it can be seen that the modified NFI model predicts the thermal conductivity of UO2 much better than the MATPRO model.4 percent and with a standard deviation of 8 percent (see Figure 2. the unirradiated UO2 data sets from that were used to derive the MATPRO model are also included.32 underpredicts thermal conductivity by about 9.00 3.33. the modified NFI model can account for burnup and is presented at burnup levels of 0 and 30 GWd/MTU for comparison. Unlike the MATPRO model.00 2. Modeltodata comparisons for the MATPRO and modified NFI models and data collected from unirradiated UO2 are made in Figures 2.14 .3.00 0 500 1000 1500 2000 MATPRO Modified NFI 0 GWd/MTU Modified NFI 30GWdMTU 2500 3000 Temperature (K) Figure 2. The average bias in Figure 2.00 4. it can be seen that the MATPRO model tends to underpredict the thermal conductivity of the unirradiated UO2.
32 MATPRO modeltodata comparison for thermal conductivity of unirradiated UO2. Modified NFI (UO2) 8 Predicted Thermal Conductivity (W/m*K) 7 6 5 4 3 2 1 1 2 3 4 5 6 7 8 Measured Thermal Conductivity (W/m*K) Predicted = Measured Bates Goldsmith and Douglas Christensen Gibby Hobson Godfrey Weilbacher Ronchi Figure 2.3.MATPRO (UO2) 8 Predicted Thermal Conductivity (W/m*K) 7 6 5 4 3 2 1 1 2 3 4 5 6 7 8 Measured Thermal Conductivity (W/m*K) Predicted = Measured Bates Goldsmith and Douglas Christensen Gibby Hobson Godfrey Weilbacher Ronchi Figure 2.3. 2.33 Modified NFI modeltodata comparison for thermal conductivity of unirradiated UO2.15 .
3.35. Based on Figure 2. The modified NFI model also tends to overpredict the small amount of irradiated data on average.34 MATPRO modeltodata comparison for thermal conductivity of irradiated UO2.16 .3.34 indicates that the thermal conductivity is overpredicted by about 40 percent. with a standard deviation of about 31 percent. MATPRO (UO2) 5 Predicted Thermal Conductivity (W/m*K) 4 3 2 1 1 2 3 4 5 Measured Thermal Conductivity (W/m*K) Predicted = Measured Ronchi 5055 GWd/MTU Ronchi 96 GWd/MTU Ronchi 3035 GWd/MTU Ronchi 82 GWd/MTU Carrol 40 GWd/MTU Ronchi 92 GWd/MTU Figure 2. it can be seen that the modified NFI model predicts the thermal conductivity of irradiated UO2 better than the MATPRO model.34.3. 2. it can be seen that the MATPRO model predicts significantly higher values for thermal conductivity than the measured values obtained from irradiated UO2. but only by about 5 percent and with a standard deviation of about 8 percent (see Figure 2.3. The average bias in Figure 2.34 through 2.3.3. Based on the significantly smaller bias and standard deviation. which is accounted for in the modified NFI model but not the MATPRO model.35).Modeltodata comparisons for the MATPRO and modified NFI models and data collected from irradiated UO2 are made in Figures 2. The disparity between models is attributed to the degradation of thermal conductivity with increasing burnup.
which does not account for burnup degradation of thermal conductivity. The MATPRO model predicts lower thermal conductivities than the Duriez/modified NFI model across the applicable temperature range (i. the Duriez/modified NFI model can account for degradation of thermal conductivity with increasing burnup.Modified NFI (UO2) 5 Predicted Thermal Conductivity (W/m*K) 4 3 2 1 1 2 3 4 5 Measured Thermal Conductivity (W/m*K) Predicted = Measured Ronchi 5055 GWd/MTU Ronchi 96 GWd/MTU Ronch 3035 GWd/MTU Ronchi 82 GWd/MTU Carrol 40 GWd/MTU Ronchi 92 GWd/MTU Figure 2.3.3. 500 to 3000K). unlike the MATPRO model. The Duriez/modified NFI model at a burnup level of 30 GWd/MTU is shown for comparison and reveals significant degradation of thermal conductivity.35 Modified NFI modeltodata comparison for thermal conductivity of irradiated UO2. This represents an improvement relative to the MATPRO model..e. A comparison between the MATPRO model and the Duriez/modified NFI model is presented in Figure 2. However.17 .36 for MOX fuel with a plutonium concentration of 7 wt%. 2.
18 .3. Unlike the MATPRO model.00 Thermal Conductivity (W/m*K) 9. the Duriez/modified NFI model can account for burnup and is presented at burnup levels of 0 and 30 GWd/MTU for comparison. The older data from which the MATPRO model was developed were not as close to being stoichiometric as the more recent data.00 8. the Duriez/modified NFI model exhibits better agreement with the limited experimental data from which it was derived.38.37. The current MOX fuel is fabricated closer to being stoichiometric than fuel typically fabricated more than 25 years ago.00 2.00 6. Based on Figure 2..3. Based on the smaller bias and standard deviation.00 5. Data collected from MOX with plutonium concentrations of 3 and 15 wt% were used (Duriez et al.10. with conductivity decreasing with a deviation from stoichiometry (2.00 7. it can be seen that the MATPRO model tends to underpredict thermal conductivity.38.0). 2000).36 MATPRO and Duriez/modified NFI thermal conductivity models as a function of temperature for MOX fuel with 7 wt% plutonium content. it can be seen that the Duriez/modified NFI model fits this MOX thermal conductivity data better than the MATPRO model.00 4.00 0 500 1000 1500 2000 MATPROMOX DuriezModified NFI DuriezModified NFI 30GWd/MTU 2500 3000 Temperature (K) Figure 2. As expected. Based on Figure 2.3.3.00 1. The average bias in Figure 2. The average bias indicates that the MOX model overpredicts by less than 2 percent and the standard deviation between the model and the data is less than 3 percent. The comparisons utilize the data used to derive the Duriez/modified NFI model. It should be noted that the MOX fuel thermal conductivity is strongly influenced by O/M ratio.3. Modeltodata comparisons for the MATPRO and Duriez/modified NFI models and unirradiated MOX fuel are made in Figures 2.00 3. The standard deviation of the MATPRO predictive bias at both plutonium concentrations is under 6 percent. it can be seen that plutonium content has a relatively minor effect at concentrations less than 15 wt%.37 and 2.00 0.37 underpredicts thermal conductivity by about 9 percent at both 3 and 15 wt% plutonium concentrations. 2.3.
3.3.37 MATPRO modeltodata comparison for thermal conductivity of MOX modeled with 7 wt% of plutonium.MATPRO (MOX) 5 Predicted Thermal Conductivity (W/m*K) 4 3 2 1 1 2 3 4 5 Measured Thermal Conductivity (W/m*K) Predicted = Measured 3 wt% Pu 15 wt% Pu Figure 2.19 . 2.38 Duriez/modified NFI modeltodata comparison for thermal conductivity of MOX with modeled with 7 wt% of plutonium. DuriezModified NFI (MOX) 5 Predicted Thermal Conductivity (W/m*K) 4 3 2 1 1 2 3 4 5 Measured Thermal Conductivity (W/m*K) Predicted = Measured 3 wt% Pu 15 wt% Pu Figure 2.
the average and standard deviations of the bias are 9 and 6 percent. whereas a positive bias indicates that the model overpredicts the data For the modified NFI and the Duriez/modified NFI models. The relative error for the limited number of irradiated UO2 data points is approximately 51 percent. Temperature must range between 500 (the Debye temperature) and 3000K. while the relative error is approximately 13 percent. The bias and relative error of the MATPRO fuel thermal conductivity model was assessed based on the modeltodata comparisons previously presented for unirradiated UO2. For irradiated UO2. UO2Gd2O3.2 Applicability and Uncertainty The following are limits that restrict the applicability of the MATPRO fuel thermal conductivity model (Equation 2. several assumptions have been made to provide an approximate model for effects of variation in the plutonium content and O/M ratio of ceramic fuels: The effect of variation in theoretical density for MOX fuels has been assumed to be described by the porosity correction derived with UO2 data. and nonstoichiometric fuels. In addition. irradiated UO2. respectively. Above 3000K. the average and standard deviations of the bias are 40 and 31 percent.2. while the relative error is approximately 9 percent. which has a limited number of data points for 2. UO2. For reference. The average and standard deviations of the bias for unirradiated UO2 are approximately 9. and MOX.20 . The hightemperature electronic contribution to thermal conductivity has been assumed to be the same for PuO2.4 and 8 percent.31): Theoretical density must range between 90 and 100 percent.5 and 11 percent. Variation in O/M ratio has been assumed to affect only the defect term. Variations in plutonium content have been assumed to affect only the phononphonon scattering factor. respectively. respectively. it is assumed that the fuel becomes molten and the uncertainties on predictions made above this temperature are large. irradiated UO2. For irradiated UO2. the recommended ranges for application for modeling the thermal conductivity of UO2 are as follows: Temperature: 300 to 3000K Rodaverage burnup: 0 to 62 GWd/MTU Asfabricated density: 92 to 97 percent TD Gadolinia content: 0 to 10 wt% For the Duriez/modified NFI model. The average and standard deviations of the bias for unirradiated UO2 are approximately 2. which has a limited number of data points for comparison. a negative bias indicates that the model underpredicts the data. The bias and relative error of the modified NFI and the Duriez/modified NFI fuel thermal conductivity models were assessed based on the modeltodata comparisons previously presented for unirradiated UO2. for MOX. while the relative error is approximately 19 percent. and MOX. the thermal conductivity model is applicable for plutonia particle sizes of less than 20 microns. respectively. Finally.3.
” Journal of Applied Physics. Idaho Falls. Including MixedOxide Fuel Properties.” in Proceedings of the ANS International Topical Meeting on LWR Fuel Performance. 541549. March 26. NUREG/CR0479 (TREE1280). Christensen. 1997. “Uranium Dioxide Thermal Conductivity. and I. Matzke. while the relative error is approximately 3 percent.D. 1973. respectively..comparison.A. 232:166180.L.. D. Vol. “Thermal Conductivity of Uranium Dioxide Up to 2900K from Simultaneous Measurement of the Heat Capacity and Thermal Diffusivity. “Measurements of the Thermal Conductivity of Uranium Dioxide at 6701270 K. M. C. Weiss. G. C. “Thermal Conductivity Measurements of High Burnup UO2 Pellet and a Benchmark Calculation of Fuel Center Temperature. For reference. Geelhood. Ronchi..” Journal of Nuclear Materials.L. 188:323330. A Handbook of Materials Properties for Use in the Analysis of Light Water Reactor Fuel Rod Behavior. NUREG/CR6534. 50:652656.. J.” in Proceedings of Symposium E on Nuclear Materials for Fission Reactors of the 1991 EMRS Fall Conference. Massih.P. Douglas. 1981. Reymann. ID. 1997. Bates.G. PNNL11513. 1967. respectively.Gd)O2 Fuel Behavior in Boiling Water Reactors.E. T. J. C. whereas a positive bias indicates that the model overpredicts the data 2. “Electrical Conductivity of Uranium Dioxide.E. Nuclear Regulatory Commission.A. Lucuta. and K. These are values are based on comparisons to the small amount of nearstoichiometric MOX data. “A Pragmatic Approach to Modeling Thermal Conductivity of Irradiated UO2 Fuel: Review and Recommendations. 4. Sheindlin. Hinman. Journal of Nuclear Materials.S. Washington. for the Duriez/modified NFI fuel thermal conductivity model for MOX. and N. C.A. Oregon. P. 2005. M. Musella.” Transactions of the American Nuclear Society.M.” Journal of Nuclear Materials. MATPRO Version 11 (Revision 2). a negative bias indicates that the model underpredicts the data. 1992. 1999.. and J. Ohira. L. 47:3142. the average and standard deviations of the bias are less than 2 and 3 percent. D. H. Allesandri. Inc.” Journal of the American Ceramic Society.A.C. A. Goldsmith. Philipponnea. and Z. Duriez.3. 7:391392. “Thermal Conductivity of Hypostoichiometric Low Pu Content (U. Finally.R.. 2000.A..” Journal of Nuclear Materials. Mason.J. Persson. Richland. “Modeling of (U. 277:143158. the average and standard deviations of the bias are approximately 5 and 8 percent. Gervais. for the U. Beyer.. while the relative error is approximately 10 percent. Hyland. 2. and T. prepared by EG&G Idaho. Hastings. and Y. S. D.. FRAPCON3 Updates. 1996.. Lanning.Pu)O2x Mixed Oxide.S. K.3 References Hagrman.G.21 . and G. Pacific Northwest National Laboratory. et al.J. 1964. pp. Kawada. WA. 85(2):776789. Itagaki. and G. J. Portland.
TN. C. D. 38:163177.4 Fuel Emissivity (FEMISS) The subroutine FEMISS is used to calculate the total hemispherical fuel emissivity (emissivity integrated over all wavelengths) as a function of temperature.42) The first term of Equation 2.Gibby. ORNL3556.1 Model Development and Comparisons According to the StefanBoltzmann law. 7:10031015. and J.G. and plenum gas. Weilbacher.672 x 108 W/m2K) temperature (K) (2. 1971. The FEMISS subroutine used by FRAPCON3 and FRAPTRAN is the same as the subroutine documented in MATPRO. Sheindlin.” Journal of Nuclear Materials.78557 1. 4:431438. 1964.C.5263 x 10 5 T (2. 327:5876. R.42 is slightly less than the value used in MATPRO. Godfrey. M. depending on the gap size. temperature gradient across the gap. Section D: Applied Physics. Hobson.000 MWdt1. Oak Ridge National Laboratory.. Kinoshita. Ainscough.7856. Oak Ridge.41) The expression used in the FEMISS subroutine to describe total emissivity is e 0. et al. “Diffusivite Thermique de l’Oxyde d’Uranium et de l’Oxyde de Thorium a Haute Temperature. R. 2004.B. 1974... Fuel emissivity is defined as the ratio of radiant energy emitted from a material to that emitted by a black body at the same temperature.L. J. 0. the total radiant power per unit area emitted by a body at temperature T is P e T 4 where P e T = = = = radiant power per unit area (W/m2) total hemispherical emissivity (unitless) the StefanBoltzmann constant (5.Pu)O2 Solid Solutions. which is in 2. “Effect of Porosity and Stoichiometry on the Thermal Conductivity of Uranium Dioxide. I. Taylor. Radiant energy transfer can be a significant heat transfer mechanism.” High TemperaturesHigh Pressure. Thermal Conductivity of Uranium Dioxide and Armco Iron by an Improved Radial Heat Flow Technique.42 with respect to its database is 6. Ronchi. T. “Effect of Burnup on the Thermal Conductivity of Uranium Dioxide up to 100.” Journal of Nuclear Materials. 1972. The subroutine FEMISS is used to calculate radiant energy transfer from fuel to cladding in conjunction with thermal conduction.C.796 percent.” Journal of Physics. and M. 2.4.22 . “The Effect of Plutonium Content on the Thermal Conductivity of (U. Staicu. The standard error of Equation 2. 2.
264. 2. so Equation 2. SC. there were no data to develop a MOX emissivity equation.agreement with the error reported in MATPRO. Jones and D.2 0. “Optical Properties of Uranium Oxides. “High Temperature Hemispherical Spectral Emittance of Uranium Oxides at 0. The emissivity data were measured at temperatures up to approximately 2400K.M.8 0. Cabannes.41 Data to model comparison for FEMISS. and D. P.3 References Held.M.G.G. Stora.F.23 .P. M.R. 1965. The hemispherical spectral data of Held and Wilder (1969) and the emissivity data of Cabannes (1967) and J.65 and 0.8 percent.” C.5 0. Acad.70 m. At the time of model development. J. 1967.3 0. Paris. 205(4972):663665. The emissivity data used to model comparison for FEMISS are presented in Figure 2.41. 1. Tsakiris.0 300 Held and Wilder Cabannes et al.” Journal of the American Ceramic Society. and use of FEMISS above this temperature is speculative because of possible hightemperature effects that are not modeled.2 Applicability and Uncertainty The correlation used in the FEMISS subroutine to calculate fuel emissivity is applicable at temperatures up to 2400K.7 0.M.C. 6.4. The standard error of this correlation with respect to the database from which it was derived is approximately 6.8 percent 2.9 Emissivity (unitless) 0. 1969. Jones and D. FEMISS Standard Error 800 1300 1800 2300 2800 Temperature (K) Figure 2. 52(4):182185.R. These small differences are attributed to rounding error and do not significantly alter emissivity predictions.4 0.. and J. J.1 0.0 0.42 is also recommended for MOX.4.6 0. “ORTIQUEMOLECULAIREFracteurs de re’flexion et d’e’mission de UO2 a’ haute Temperature. Murchison (1965) were used in developing the FEMISS model. Murchison. Wilder. 2.” Nature.
4. this effect is ignored in modeling thermal expansion. Gronvold (1955).7 x 103 7. FRAPTRAN 1.0) is greater than 0. or PuO2 in solid. and Kempeter and Elliott (1958) were used to determine the correlation constants for UO2 used in MATPRO.24 . When the departure from stoichiometry (O/M – 2.4/FRAPTRAN 1. In the solid phase. (1991) required the constants to be updated to improve the fit between the correlation and hightemperature data.0 x 105 3. and MATPRO. These updated constants are included in FRAPCON3. since typical reactor fuels only deviate less than a tenth this much from the stoichiometric composition. which is a major factor in determining gap heat transfer and thus the stored energy. Dimensional changes in the fuel affect the pellettocladding gap size. However.0 x 102 6. liquid. UO2Gd2O3. The correlation given in Equation 2. K2. there is clearly an effect. or solidliquid states and includes expansion due to the solidliquid phase change. MATPRO UO2 1.51 is used.4.2.32 x 1019 PuO2 9.51) linear strain caused by thermal expansion (equal to zero at 300K) (unitless) temperature (K) energy of formation of a defect (J) Boltzmann’s constant (1.80 x 106 2. However.5 Fuel Thermal Expansion (FTHEXP) The subroutine FTHEXP models dimensional changes in unirradiated fuel pellets caused by thermal expansion.51 was originally used in MATPRO and was incorporated into both the FRAPCON3 and FRAPTRAN codes. Table 2. an important quantity for safety analysis. and PuO2 have the same form. the FRAPCON3 and FRAPTRAN codes use updated fitting coefficients based on more recent data.16 x 101 1.61 x 103 3. Table 2.9 x 1020 FRAP UO2 9. newer data provided by Martin (1988) and Momin et al. However.4 (FRAP).0 x 106 2. Burdick and Paker (1956). and the fraction of fuel which is molten. fraction of PuO2.0 x 102 7.51 presents the original MATPRO constants and the updated constants for UO2. The subroutine FTHEXP is used in FRAPCON3. The equations for thermal expansion of UO2. L / L 0 K 1T K 2 K 3 exp( E D / kT ) where L/L T ED k = = = = (2. It can handle any combination of UO2. The O/M ratio is not included.51 Parameters used in UO2 and PuO2 solidphase thermal expansion correlations. Data collected by Baldock et al. and K3 are constants to be determined. 2. UO2Gd2O3. Christensen (1963). (1966).0 x 103 4. as well as the constants for PuO2.2.5.1 Model Development and Comparisons The FTHEXP subcode models fuel thermal expansion as a function of temperature.38 x 1023 J/K) and K1.0 x 1020 Units K1 Unitless Unitless J Constant K1 K2 K3 ED 2. Equation 2.
0. For MOX.6 x10 5 T Tm Tm (2.010 0. denoted in Equation 2.045 Fractional Length Change ( L/L) 0.25 .51 Modeltodata comparison for UO2 correlation used in MATPRO and FRAPCON3/FRAPTRAN with updated constants.53 by Tm.0.005 0. an expansion equal to a linear strain of 0.4/FRAPTRAN 1.035 0.52.51.015 0.043 FACMOT thermal expansion strain of solid fuel from equations with T = Tm melting temperature of the fuel (K) fraction of the fuel which is molten (unitless) If FACMOT = 0.4 (FRAP) for UO2 provide a better fit to the hightemperature data than the previous constants used in MATPRO. the thermal expansion of the solid is found by combining the contribution from each constituent in proportion to its weight fraction.53.050 0. the strain due to thermal expansion is given by Equation 2.53) The solidtoliquid phase transition is isothermal only for pure UO2 or pure PuO2.51 and 2. the fuel is all molten (2.For mixed UO2 and PuO2.000 0 500 1000 1500 Temperature (K) Baldock Other MATPRO References FRAP UO2 Burdick Martin References Gronvold MATPRO UO2 2000 2500 3000 Figure 2. During melting. the transition occurs over a finite temperature range.030 0. The comparison in Figure 2. If the fuel is partially molten.020 0.025 0. it can be seen that the updated constants used in FRAPCON3. where L/L0(Tm) Tm FACMOT = = = L / L 0 L / L 0 (T m ) 0. Comparisons between the databases for UO2 and PuO2 with their respective correlations are presented in Figures 2.52) The correlation used to describe the expansion of entirely molten fuel is given by Equation 2. respectively. 0.043 3.52. 2. If FACMOT = 1. the fuel is all solid. L / L 0 L / L 0 (T m ) 0.040 0.52 reveals good agreement between the correlation and the database. In Figure 2.043 occurs.
P. Therefore. C. 2.002 0.D. and H. 1963. Christensen.012 0. 1:357370. 46:607608.5.010 0.. J..5.” Journal of the American Ceramic Society. 30:15241526. Burdick.P. 18:305313.2 Applicability and Uncertainty The fuel thermal expansion correlation used in the FTHEXP subroutine is applicable through the melting temperature of the fuel (3111. The uncertainty of the pooled data was found to be temperature dependent. 2.016 0. “The Xray Thermal Expansion of NearStoichiometric UO2.004 0. and ThO2. increasing approximately linearly with temperature. a percentage error is given rather than a fixed number. and R.S.” Journal of Chemical Physics.014 Fractional Length Change ( L/L) 0.. UO2ThO2. 39:181187.J. F. Elliott.000 0 500 1000 1500 2000 Temperature (K) FRAP and MATPRO PuO2 Tokar and Brett References Figure 2.006 0. 1955. et al. 1966.15K for UO2). Parker.A.E. Kempter.” Journal of Nuclear Materials.26 . “Thermal Expansion of UN.0. “High Temperature Xray Study of Uranium Oxides in the UO2U3O8 Region. “Thermal Expansion and Change in Volume of Uranium Dioxide on Melting” Journal of the American Ceramic Society. 1958.52 Modeltodata comparison for PuO2 correlation used in MATPRO and FRAPCON3/FRAPTRAN. UO2. 2. Gronvold. The limits were found to be within 10 percent of the calculated value. 1956.008 0. M.3 References Baldock.” Journal of Inorganic and Nuclear Chemistry. “Effect of Particle Size on Bulk Density and Strength Properties of Uranium Dioxide Specimens.
62 and 2. Brett. This was done during the development of the FUDENS model. both MATPRO and FRAPCON3. and (U0.D. D. pp.66 while FRAPCON3.61 or Equation 2. 1960. The RSNTR method uses the density change observed during resintering tests (1973K for 24 hours based on Regulatory Guide 1.1 Model Development and Comparisons The subroutine FUDENS uses one of two methods to calculate the maximum density change during irradiation.27 . Grenoble. it is not included in FRAPTRAN. “The Thermal Expansion of PuO2 and Some Other Actinide Oxides Between Room Temperature an 1000°C. ThO2. The data used to develop FUDENS were taken from irradiated fuel which also included swelling (Rolstad. the subcode FUDENS calculates densification as a function of fuel burnup. If a resintering density change is not input (i.66. the TSINT method uses the initial unirradiated density of the fuel and the fuel fabrication sintering temperature and burnup for density calculations. depending on the fuel temperature. 1991. 1960. In the MATPRO version of FUDENS..8)O2 Doped with Fission Products.” Transactions of the American Nuclear Society. data from resintering tests should be input by the user to determine densification (Regulatory Guide 1.. FTEMP (Siefken et al. If available. 2. RSNTR = 0). The FUDENS subroutine is used in FRAPCON3. Nutt.62 is used.126 [NRC 1978]) in a laboratory furnace and is the preferred input for the calculation. 10:210211. 2. and A. swelling can be neglected during this period.. “Thermal Expansion of PuO2 from 25 to 1420°C. 185:308310. Momin. “High Temperature XRay Diffacractometric Studies on the Lattice Thermal Expansion Behavior of UO2.C. Both the RSNTR and TSNTR methods for MATPRO and FRAPCON3. 2. M. France. N. Mirza.66.Martin.H..6 Fuel Densification (FUDENS) The subroutine FUDENS calculates fuel dimensional changes due to densification of UO2 and MOX during the first few thousand hours of water reactor operation. Since densification occurs over longer time scales.4 and is similar to the correlation described by MATPRO. temperature.” in Proceedings of the Second International Conference on Plutonium Metallurgy. Russel.” Journal of Nuclear Materials. E.e. if a nonzero value for the resintering density change is input.6..2Th0. 2001).” Journal of Nuclear Materials.E.W.63 and 2. 1972. 152:94101. For TSNTR calculations. 1974).4 use either Equation 2. When RSNTR > 0.65 and 2.4 uses Equations 2. “The Thermal Expansion of Solid UO2 and (U.Pu) Mixed Oxides – A Review and Recommendations. then. and M.4 are described by the following equations to calculate the maximum densification length change during irradiation.126 [NRC 1978]). Otherwise. to a first approximation. Tokar.G. If fuel densification is much greater than swelling during the first 1000 hours of irradiation.B. However. and L.61 or 2. 397410. then either Equation 2. and initial density.64 or 2. MATPRO uses either Equation 2. A. April 1922. 1988. Mathews. FUDENS is not included in FRAPTRAN because FRAPTRAN is intended for transient events occurring over a short time scale.
64 or Equation 2. FTEMP (Rolstad et al.62) In FRAPCON3. depending on the fuel temperature.00285RSNTR L m (2.65) L L m RSNTR FTEMP DENS TSINT FDENS = = = = = = maximum possible dimension change of fuel due to irradiation (percent) resintered fuel density change (kg/m3) fuel temperature (K) initial density (percent theoretical) sintering temperature (K) starting density (kg/m3) Densification as a function of burnup is calculated using L L 3 FBU B 2. L For FTEMP < 1000K L m 22.61) For FTEMP 1000K (2. 1974).For FTEMP < 1000K L 0.64) L For FTEMP 1000K where L m (2. if the resintering density is a nonzero input.28 .65.2 100 DENS TSINT 1453 66.63) If zero is input for the resintering density change. then the following equation is used to calculate the maximum dimensional change.4. then the MATPRO and FRAPCON3.0 * FDENS) L m (2.6 100 DENS TSINT 1453 (2..4 models use either Equation 2.66) where L L FBU = = dimension change (percent) fuel burnup (MWd/kgU) 2.0015RSNTR L m L 0. L 100 * RSNTR /(3.0 e 35FBU B e L L m (2.
The pellets were fabricated with sintering temperatures from 1748 to 2008K and had starting density values ranging from 91 to 95 percent TD. Two pins were irradiated with fuel centerline temperatures up to 2048K and burnup to 4. which is used in both FRAPCON3. Since these initial studies. 2.6 MWd/kgM. and then the densification from the previous time step is subtracted to obtain the incremental densification.66). 1976) that was collected shortly thereafter.. The mechanism of densification was observed to be sintering of pores with initial diameters of 0.61 compares measured and predicted values of densification using the TSNTR correlation.B = a constant determined by the subcode to fit the boundary condition: L/L = 0 when FBU = 0 The FUDENS subcode presented in Equation 2.64. Figure 2. Densification was determined by measuring porosity before and after irradiation. fuel currently fabricated has very few pores less than 2 m in diameter. and 2. Three of the four types of pellets were fabricated with pores smaller than 2 m in diameter to more carefully examine the mechanisms behind densification. The solid pellets experienced centerline temperatures of 733 to 2443K over the burnup range 0. 2.37 to 23.0025 to 2. Small (1987) irradiated four types of UO2 pellets in the Harwell Materials Test Reactor.65.64. Freshley et al. The incremental densification for the time step being considered is the output of the subcode FUDENS and is determined by Newton’s method (Hamming. (1974). (1978) reported densification data on MOX pellets irradiated in the radially adjustable facility tubes of the General Electric Test Reactor.66 is used to calculate total densification. Banks (1974) reported densification data on solid and annular UO2 pellets irradiated in the Winfrith Steam Generating Heavy Water Reactor and the Halden Boiling Water Reactor. For reference. 1971).. 1974) and EPRI data (Freshley et al. Pellet centerline temperatures ranged from 623 to 1823K and burnup ranged from 0. Therefore.7 MWd/kgU. and 2.29 . 2.6 MWd/kgU.65. At centerline temperatures above 1273K. only data for temperatures less than 1423K were included in comparison with the FUDENS model (Equations 2. This relationship was adopted for use in the FUDENS subroutine because it successfully describes the burnup dependence of both the original data (Rolstad et al.4 and MATPRO and incorporates Equations 2. They concluded that dilute additions of PuO2 in UO2 (up to 6 wt% PuO2) do not change the densification behavior of the fuel.66.66 was originally suggested by Rolstad et al. there has been additional inreactor densification data collected for UO2 and MOX fuels irradiated in thermal reactors.5 m or smaller. The mechanism for densification was determined to be the loss of pores with initial diameters less than 2 m. The relationship between densification and burnup in Equation 2. but at higher temperatures dimensional change was affected by intergranular fission gas swelling. densification was caused by sintering of pores less than 1 m in diameter. They concluded that pellets with centerline temperatures less than 1423K densified during irradiation.
1978).00 5.07 MWd/kgU).00 3. et al.00 6.00 8. which is atypical of current fuel fabrication.6.” Enlarged HPG Meeting on Computer Control and Fuel Research. Exreactor resintering tests conducted by Freshly et al. 1974. However. (1976) study. possessed a nonprototypic pore size distribution that was heavily weighted toward smaller diameters.00 1.3 References Rolstad. which exhibited very high densification at very low burnup (less than 0.5 to 0.8. 2.00 1.00 2. E. it is difficult to establish a meaningful measure of uncertainty.00 Banks Freshley Small Freshley 1976 0. 2. “Measurements of the Length Changes of UO2 Fuel Pellets During Irradiation..6. 1974. Due to the scatter in the experimental data.00 6.00 1.00 3. exreactor resintering tests are the preferred method for predicting inreactor fuel densification (NRC.00 1.00 Experimental Densification (% TD) Figure 2. it should be noted that the Small (1987) data.00 Predicted Densification (% TD) 5. Three of the four fuel types had 100 percent of their porosity in pores smaller than 2 m diameter. Since inreactor densification is strongly influenced by fuel fabrication.61 Comparison of densification predictions generated by the FUDENS model and experimental densification measurements determined in reactor. 2.00 7.00 4.30 . (1976 and 1978) at 1600 and 1700°C provided reasonable estimate of inreactor densification tests with standard deviations of 0. This data comparison demonstrates that basing densification on the sintering temperature provides a large degree of uncertainty. June 47.00 0.00 7.6 percent of the theoretical density.00 4.00 2.2 Applicability and Uncertainty The densification correlation used in the subroutine FUDENS is applicable to fuels with pore size distributions that are similar to those included in the Freshley et al.
J.S.126. 1978. “Densification of Uranium Dioxide at Low BurnUp. INEL96/0422. and irradiation (FUDENS) to calculate the overall dimensional changes in fuel.4 is intended for shorter time scales than those that result in significant solid swelling. Rev. SCDAP/RELAP5/MOD 3. D. 1971. E. Hohorst. 2001. The FSWELL subroutine is not used in FRAPTRAN 1.4 because FRAPTRAN 1.K.C. SS 2..W. and P. Washington. An Acceptable Model and Related Statistical Methods for the Analysis of Fuel Densification.” Journal of Nuclear Materials. 62:138166. Banks.D. Hart. U.L. the FSWELL solid swelling correlation used in FRAPCON3. Nuclear Regulatory Commission.. Freshley. 2. which is caused by the buildup of solid and gaseous fission products during irradiation.5 x10 29 B S where SS = fractional volume change due to solid fission products (m3 volume change/m3 fuel) (2.L. Freshley. The gaseous swelling correlation in FSWELL is not used in FRAPCON3 because it significantly overpredicts swelling and its effects on cladding deformation. 1978. J.1. New York: McGrawHill Book Company. Hamming. M. D. Vol. Rev. The FSWELL correlation for solid swelling is used in both MATRPO and FRAPCON3. 1974.S. Harvego.A. densification due to pressure sintering (FHOTPS). ID.31 . Nuclear Regulatory Commission. NUREG/CR6150..3 Code Manual MATPROA Library of Materials Properties for LightWaterReactorAccident Analysis.” Journal of Nuclear Materials. 54:97107.A. J. 1987. Inc. Daniel. “IrradiationInduced Densification and PuO2 Particle Behavior in MixedOxide Pellet Fuel. However. Introduction to Applied Numerical Analysis. 2. D. Fuel swelling (FSWELL) is combined with creepinduced elongation (FCREEP). Hart.J. Regulatory Guide 1.1. L.7 Fuel Swelling (FSWELL) The subroutine FSWELL calculates fuel swelling.1 Model Development and Comparisons In MATPRO. “Some Observations of Density and Porosity Changes in UO2 Fuel Irradiated in WaterCooled Reactors. Siefken. J. “IrradiationInduced Densification of UO2 Pellet Fuel. Idaho National Engineering and Environmental Laboratory. Idaho Falls.NRC – U.71) 2. R.7. 4. 1976.D. D.” Journal of Nuclear Materials. Brite. M. Brite.E. and P.W. 148:302315.4 has evolved from the original model described in MATPRO. 81:6392. G.” Journal of Nuclear Materials. Small.W.E. 2. the correlation used by FSWELL to calculate swelling due to the buildup of solid fission products is give by Equation 2. Coryell.W.7. E. Daniel.
the swelling due to solid fission products was decreased to 0.62 percent per 10 GWd/MTU for burnup less than 80 GWd/MTU and increased to 0.BS = burnup during a time step (fissions/m3) At temperatures below 2800K.699 percent (MATPRO and FRAPCON2) to 0. Sg is zero because the gas that causes swelling is assumed to have released.71 and 2. Results of an NRCsponsored test rod were compared with predictions made by FRAPCON3 and revealed that swelling models that included gaseous fission product buildup tend to overpredict test results at burnup levels up to 10 GWd/MTU (Lanning and Bradley.8 x10 56 2800 T where Sg T B = = = 11 . respectively. The correlations used to calculate fuel swelling under these conditions are given in Equations 2.32 . Lanning et al. FRAPCON3.0162 2800 T e 8. in the FSWELL subcode described by MATPRO.74.72.211x10 24 ) where soldsw sigswell bus fdens bu bul and = = = = = = (2.315 x10 24 ) For burnup > 80 GWd/MTU (2.72) S fractional volume change due to gas fission products (m3 volume change/m3 fuel) temperature (K) total burnup of fuel (fissions/m3) For temperatures greater than 2800K. swelling due to the buildup of gaseous fission products is given by Equation 2. The correlations presented in Equations 2.0 x10 27 B B (2.72 are used to determine swelling due to solid and gaseous fission product buildup. 1997).73 e 0. Unlike the MATPRO version of FSWELL.74) fractional volume change due to solid fission products (m3 volume change/m3 fuel) userdefined parameter that incrementally changes error fuel burnup during time step (analogous to BS in MATPRO model) initial pellet density (kg/m3) burnup at end of time step (MWs/kgU) burnup at end of previous time step (MWs/kgU) 2.73) soldsw bus * (3.315 x10 23 sigswell * 2.4. In FRAPCON3.73 and 2. It was determined that swelling could be modeled better by eliminating the contribution from gaseous fission products and increasing the contribution of solid fission products from 0.86 percent per 10 GWd/MTU for burnup greater than 80 GWd/MTU. 1984.4 does not include a correlation for swelling due to the buildup of gaseous fission products.211x10 23 sigswell * 3.. Sg 8.77 percent (FRAPCON3) per 10 GWd/MTU. respectively. For burnup < 80 GWd/MTU soldsw bus * ( 2.
00 2.71 Data comparison between MATPRO and FRAPCON3. it can be seen that the FRAPCON3. which tends to overpredict the experimentally determined data points.974 x1010 * bu bul The variable “bus” in Equations 2. However.4 predictions and data from instrumented fuel assemblies tested at Halden are presented in Figure 2.4 model predicts less swelling than the MATPRO model. “bus” permits the fractional volume change due to solid fission products to be determined as a function of burnup in MWs/kgU as opposed to fissions/m3.50 4.50 0.00 1.4 version of FSWELL.4 Garde (1986) Newman (1986) Smith (1994) Dideon and Bain (1983) Figure 2.71 along with experimental data from immersion density measurements.4 versions of FSWELL is presented in Figure 2. A comparison between the MATPRO and the FRAPCON3. The diametral cladding change data versus burnup is particularly valuable because it implicitly includes the effects of cladding restraint on fuel swelling. 4.74 is analogous to the variable “BS” in Equation 2. For burnup > 80 GWd/MTU.50 2. Data comparisons between FRAPCON3.bus fdens * 2.73 and 2.71.50 V/V (%) 3.00 0 10 20 30 40 50 60 70 80 Burnup (GWd/MTU) MATPRO FRAPCON3.72 for varying levels of burnup. These Halden swelling data are based on measurements of fuel stack length and cladding diameter change with burnup and converted to volume swelling assuming isotropic swelling. Based on this figure.00 3.00 0.4 versions of FSWELL. 2.50 1. a higher swelling rate is used in the FRAPCON3.33 .
34 IFA655 Rods 2 &4 IFA534.R.4 IFA629.2 Applicability and Uncertainty Based on the data from which it is derived. 1986. A. NUREG/CR3071..2 1 0.4. and E.72 Comparison between swelling rates determined for various instrumented fuel assemblies tests at Halden and swelling rates predicted by FRAPCON3. 2. Lynchburg. C.9 IFA504 99 GWd/MTU . Painter. Combustion Engineering. Bradley. NUREG/CR6534.10 IFA515. Above 80 GWd/MTU.3 IFA597.S.14 IFA515. VA. a 1sigma uncertainty of 0. Newman.6 17 GWd/MTU62 GWd/MTU62 GWd/MTU65 GWd/MTU68 GWd/MTU70 GWd/MTU75 GWd/MTU78 GWd/MTU90 GWd/MTU90 GWd/MTU99 GWd/MTU 0. Babcock and Wilcox.W. D. Pacific Northwest National Laboratory for the U. C. Garde.D. 2.4 Swelling Rate.16 percent V/V per 10 GWd/MTU has been calculated.3 References Lanning.C. 1986.L. Nuclear Regulatory Commission. Vol.8 0.M. %DV/V per 10 GWd/MTM 1. D.D.E. 1984. Windsor.C. Washington. a 1 sigma uncertainty of 0.S. the correlation used in FSWELL is applicable to fuels with burnups up to 100 GWd/MTU. CT.08 percent V/V per 10 GWd/MTU has been calculated. D. DOE/ET3403011 (CEND427). Lanning. D. FRAPCON3: Modifications to Fuel Rod Material Properties and Performance Models for HighBurnup Application. Irradiation History and Interim Postirradiation Data for IFA432.10 IFA633 IFA610 IFA629. 2. L. Hot Cell Examination of Extended Burnup Fuel from Fort Calhoun. Below 80 GWd/MTU.1.7.2 0 IFA655 Rods 1 &3 Figure 2. Beyer.7. The Hot Cell Examination of Oconee 1 Fuel Rods after Five Cycles of Irradiation.3 IFA 519. DOE/ET3421236. Nuclear Regulatory Commission. Pacific Northwest Laboratory for the U. Washington.4 0. 1997. 1.
R. Babcock and Wilcox..Smith. DOE/ET3421238.R. Fuel Performance under Extended Burnup for the B&W 15x15 Design. C. Bain. Schrire.M.35 . Lynchburg. Windsor.. and G. CT. 2. DOE/ET3401315. Freeburn. VA. 1994. H.P. and D. 1983. Pirek. Dideon. Combustion Engineering. G.C. The Evaluation and Demonstration of Methods for Improved Nuclear Fuel Utilization.G.
determines the true specific heat at constant pressures for cladding. CCP returns linear interpolations for the points listed in Table 3. For the 57 data points between 300 and 800K. Specific heat calculations are based on interpolation of measured data.4 codes assume a standard error of 10 J/kg*K below 1090K.4/FRAPTRAN 1. The correlation in FRAPCON3. As a result. For the alpha phase of the Zircaloy alloys (temperature less than 1090K). the standard error of the CCP interpolation was based on 90 points in the Brooks and Stansbury (1966) database and was found to be temperature dependent. values of specific heat proposed by Deem and Eldridge (1967) are adopted.4 contains correlations to calculate the specific heat of ZrNb1. Between 800 and 1090K. and M5 alloys. In addition. the standard error is 1.0 Cladding Material Properties Material property correlations for fuel claddings are described in the following subsections. The Deem and Eldridge (1967) values are based on measurements of enthalpy and temperature which provide considerably less precise specific heat data than the results of Brooks and Stansbury (1966).4 codes.7 J/kg*K. and M5. it is 2. FRAPTRAN 1. 3. However.4/FRAPTRAN is assumed to be 25 J/kg*K between 1090 and 1300K.1 . 3.1. 3.1 Model Development and Comparisons The CCP subcode requires temperature as an input to calculate specific heat. a constant value of 355.11.3. The standard error in FRAPCON3. ZIRLO. These data points are based on precise data taken by Brooks and Stansbury (1966) with a Zircaloy2 sample that had been vacuumannealed at 1075K to remove hydrogen.3 reported by Coughlin and King (1950) for pure beta zirconium. the MATPRO standard error estimated from the Deem and Eldridge (1967) data in the region of 1090 through 1310K is 10.8 J/kg*K.4/FRAPTRAN 1. ZIRLO. The standard errors associated with this interpolation technique differ between MATPRO and FRAPCON3. This value agrees well with a value of 365.4/FRAPTRAN 1. FRAPTRAN 1.4.4 subroutines include additional correlations to describe a Russian zirconium alloy containing 1 wt% niobium (ZrNb1).1 Cladding Specific Heat (CCP) The specific heat subcode. In MATPRO.4. which would have otherwise affected the measurement. since only a single data source is used. This standard error is a measure only of the precision of the fit. The FRAPCON3. The standard error of specific heat calculations made above 1300K is assumed to be 100 J/kg*K in the FRAPCON3.7 J/kg*K was reported by Deem and Eldridge (1967). Zircaloy4. 4. CCP.4. and MATPRO is applicable to Zircaloy2. Unless otherwise specified. the correlations below are applicable to Zircaloy2. Above the alpha + beta to beta transformation temperature (about 1250K) and up to about 1320K.1 J/kg*K. For temperatures from 1090 to 1300K (where Brooks and Stansbury do not report results). The subroutine used to determine the material correlation in either the FRAPCON3 or FRAPTRAN codes is given in parentheses. several FRAPTRAN 1. This calculation can be based on data collected at either fast or slow heating rates and the user can designate the appropriate data set.
Temperature (K) Specific Heat Capacity (J/kg*K) Source .Low Heating Rate ( < 1000 K/s) 280 473 573 673 773 873 883 973 1025 1073 345 360 370 380 383 385 448 680 816 770 Volkov B. 3. FRAPTRAN 1. Yu et al. There is no standard error described for specific heat calculations based on these data sets.2 .7 10. Yu et al. The data used to interpolate specific heat values for the ZrNb1 alloy are presented in Table 3. Zircaloy4.In addition to Zircaloy alloys. Yu et al. Volkov B.1 2.4 has been modified to include specific heat calculations for ZrNb1.1 1. Volkov B.12. Volkov B. Yu et al. Yu et al.8 10 10 10 10 Beta Phase1093 1113 1133 1153 1173 1193 1213 1233 1248 2098 2099 502 590 615 719 816 770 619 469 356 356 356 Deem and Eldridge Deem and Eldridge Deem and Eldridge Deem and Eldridge Deem and Eldridge Deem and Eldridge Deem and Eldridge Deem and Eldridge Deem and Eldridge Coughlin and King Coughlin and King 10.12 Specific heat capacity database for ZrNb1 alloys. Yu et al.7 10. Standard Error Specific Heat Capacity (J/kg*K) Source MATPRO (J/kg*K) FRAPCON3 / FRAPTRAN (J/kg*K) Alpha Phase300 400 640 1090 281 302 331 375 Brooks and Stansbury Brooks and Stansbury Brooks and Stansbury Brooks and Stansbury 1.7 10.7 10.7 100 100 25 25 25 25 25 25 25 25 25 100 100 Table 3. Yu et al.7 10.7 10.1 1. and M5. specific heat calculations can be based on either the fast or the slow heating rate data. Volkov B. Volkov B. Table 3.7 10. These calculations are based on data collected at two different heating rates. ZIRLO.7 10. Yu et al. Volkov B. Depending on the user input. Volkov B. Volkov B.11 Temperature (K) Specific heat capacity database for Zircaloy2. Yu et al. Yu et al. Volkov B.
3 References Brooks. This correlation is applied to Zircaloy2. In addition. “Experimental research of zirconium reactor materials thermal properties: 110alloy. C.A.4. et al. Higher Temperature Thermal Physics.E. et al.R. et al. 1989. J. Moscow. and minor composition differences. and E.11. and E.. et al. King. 3. may have secondary effects on thermal conductivity. “The Specific Heat of Zircaloy2 from 50 to 700°C. H. crystal orientation. V.2 Applicability and Uncertainty The CCP subroutine is valid over the temperature range from which the data sets have been collected on zirconium alloys (300 to 2099K).E. However the standard error increases with increasing temperature as indicated in Table 3.. ZIRLO. Specific Heats of Transformation of Zirclaoy2 and Low Nickel Zircaloy2. et al. 1967.” M. FRAPTRAN 1. which is required for accurate predictions of fuel temperature. Yu. 72:2262. and FRAPTRAN 1. 18:223. Yu et al.W. 1600 Ljusternik V.. 3. 600 Ljusternik V.. et al. B. et al.” Journal for the American Chemical Society. The thermal conductivity of the cladding is primarily a function of temperature. et al. 1950. . Library of Subprograms on Physical and Mechanical Properties of the N1Alloy Fuel Rod Cladding Material. Stansbury. 420 Ljusternik V. 3.3 . ZIRLO. 1000 Ljusternik V.1. Yu et al.J. Eldridge.E.4. Yu et al.G. 3. 31(4).4 includes a correlation to calculate the cladding thermal conductivity of ZrNb1. Deem. Battelle Memorial Institute.2 Cladding Thermal Conductivity (CTHCON) The subroutine CTHCON is used to calculate cladding thermal conductivity.P. and E.E. 4.” Journal of Nuclear Materials. Coughlin. This correlation is applied to Zircaloy2.E. such as residual stress levels. Institute of Automation and Electrometry. The correlation used in CTCHCON to calculate cladding thermal conductivity is the same in MATPRO.E. 1400 Ljusternik V. 1400 Ljusternik V.E. 480 Ljusternik V.E. 1966.1153 1173 1248 1100 1110 1120 1134 1142 1150 1155 1161 400 392 356 Volkov B.1. Preprint IAE4941/11.E. Ljusternik. “HighTemperature Heat Contents of Some ZirconiumContaining Substances. Volkov. Volkov B.. 1993. Columbus. et al. and M5. Other characteristics.E. 4. FRAPCON3.High Heating Rate ( > 1000 K/s) 412 Ljusternik V. Russia. and M5. USAEC BM11803. Volkov B. et al. OH.
4. extrapolated to the melting point of Zircaloy (2098K).51 2. the thermal conductivity of the ZrNb1 cladding is assumed to be 36 W/m*K.4 with the data on which it is based. k 15.3.24) Equation 3.21 is 20 to 30 percent of the value of the constant. For temperatures less than 2133K. Similar findings were reported by Korber and Unger (1974).0636 e 0.22.21) (3.25) For temperatures above 2133K.21 predicts k very well from room temperature to the data limit of about 1800K (Figure 3.23) k 5 where k T k = = = thermal conductivity of Zircaloy (W/m* K) temperature (K) standard deviation (W/m*K) (3. The standard deviation (k) of the data with respect to this correlation appears to be temperature independent over the data range. the thermal conductivity of Zircaloy for temperatures less than 2098K is described by CTHCON in Equation 3.1 Model Development and Comparisons Considering only temperature as the defining parameter.09 x10 2 T 1. the thermal conductivity of the ZrNb1 alloy is calculated by FRAPTRAN 1.000461843 * T (3.23 and 3. The uncertainty of the conductivity calculations is given in Equation 3.4 . and FRAPTRAN1.45 x10 5 T 2 7. Figure 3.21.21 compares the CTHCON model as incorporated in MATPRO. k 7. k 36 (3. The standard deviation of the constants in Equation 3.2. the thermal conductivity and uncertainty are given in Equations 3. Jensen (1969) performed a parametric analysis of several variables involved in the estimation of fuel and cladding temperatures.24.4 with the correlation presented in Equation 3. FRAPCON3.01 (3.8K. Figure 3. respectively.21) and may be extrapolated with some confidence to the melting point.22 compares the CTHCON model with the Zircaloy thermal conductivity and 3.25. Fuel centerline temperatures are more sensitive to cladding thermal conductivity and showed variations of 28K.67 x10 9 T 3 k 1.22) For temperatures greater than or equal to 2098K. Both steadystate and transient analysis showed that variations of 20 percent resulted in calculated cladding temperature variations of about 2.
(1962) Chirigos et al. The scatter in the data is distributed relatively uniformly around the curve representing the CTHCON model. (1961) Thermal Conductivity (W/mK) 50 Feith (1966) Lucks and Deem (1958) 40 Powers (1961) Scott (1965) 30 20 10 0 0 500 1000 Temperature (K) 1500 2000 2500 Figure 3.diffusivity data that were not included in development of the model.5 .21 CTHCON model and data used for development. The scatter in the data due to differences in alloy. form. there were no discernable differences between the alloys. Although different alloys were examined. 3. or measurement direction is comparable to the variation between individual data sets. (1983) data above 1300K. Thermal diffusivity data were converted to thermal conductivity using the recommended models in Fink (2000) for specific heat and density. except for anomalously high Peggs data below 1000K and the anomalously low Bunnell et al. 70 MATPRO 60 Anderson et al. respectively.
Bates.S.insc. and J.J.3 References Jensen.php.. 3. and H. S.. “Zircaloy Thermal Conductivity.22 CTHCON model and data collected since development. Argonne. “Zirconium Alloys. Philips Petroleum Company. 3. 18:234235.L.” Reactor Structural Materials: Engineering Properties as Affected by Nuclear Reactor Service. 1962. Fink. C.gov/matprop/zircaloy/index. 1969. (1975) Peggs et al.E. Beck.2 Applicability and Uncertainty The correlation used in CTHCON to calculate the cladding thermal conductivity may be applied to temperatures up to the melting temperature of Zircaloy (2098K). H. 1974.70 MATPRO Krett and Cleveland (1997) . Above the melting temperature.” Journal of Nuclear Materials. Bunnell. Theilacker. “Sensitivity Study on Core Heatup and Meltdown by Variation of Heat Conductivity and Thermal Emissivity.24. Anderson.China 60 Krett and Cleveland (1997) .22. ID. 1983.6 . W. IL.anl. Idaho Falls. (1994) Thermal Conductivity (W/mK) 50 30 20 10 0 0 500 1000 Temperature (K) 1500 2000 2500 Figure 3.K. Argonne National Laboratory. Parametric Studies of Fuel Pin Temperature Response. Korber..E. the uncertainty increases to the value given by Equation 3. Unger.K.AECL Gilchrist (1976) Bunnell et al.2. J. Mellinger. J. http://www. Atomic Energy Division.” Transactions for the American Nuclear Society. “Some HighTemperature Properties of ZircaloyOxygen Alloys.R. (1983) Murabayashi et al. A. and G.B.” International Nuclear Safety Center Material Properties Database. 2000.R. 3. The uncertainty through this range of temperature is given by Equation 3. L. Kephar. (1976) 40 Maglic et al. 116(23):219232.2. IDO17295.
The correlation used in FRAPCON3. GEMP669. October 1966. eds. IAEATECDOC949. NY. pp. Murabayashi. 8(4):441450. Knolls Atomic Power Laboratory. Deem. West Conshohocken. 1976.4 is based on measurements obtained by Kingery (1954) from fully dense and porous (87 percent TD) zirconium oxides. C.E. and Y..W. and A. 1958. PA. 1994. and D.7 . “Calorimetric and Transport Properties of Zircalloy 2. Tanaka. Westinghouse Electric Corporation. Pittsburgh. Missile and Space Division.W. Peggs. pp. Progress Relating to Civilian Applications during June.D. The correction was applied to better fit data that was more representative of the oxide layer that forms on zirconium alloys. Vienna. pp. Physical and Mechanical Properties of Zircaloy 2 and 4.4 and FRAPTRAN 1. 1975. Powers. “Development of Zircaloy4. M. 4 and Unalloyed Zirconium.4 codes. 1955.P. Zircalloy 4. Takahashi. Tipton. Feith.D.j. Battelle Memorial Institute. Dayton and C. Thermal Conductivity and Electrical Resistivity of Zircaloy4..M. 1997. This database contained measurements from a variety of zirconium oxide materials and test methods.F. General Electric Company. D. International Atomic Energy Agency. 1966. S. 1961. 15(4):741755. The MATPRO correlation is based on an arbitrary correction to an earlier correlation that was fit to a large database of oxide thermal conductivity measurements. A. I.3 Cladding Oxide Thermal Conductivity (ZOTCON) The ZOTCON subroutine calculates the thermal conductivity of the zirconium oxide layer that forms on zirconium alloys. Perovic. and H. Cladding temperature is the only parameter used to calculate zirconium oxide thermal conductivity.L. 3. “Thermophysical Properties of ZirconiumAlloy Fuel Channel Components. KAPL2146. 6293.ASTMSTP314.N. Austria. 9. Chirigos. Stanimirovic.” Fuel Element Fabrication. pp. OH... WCAP326941.” Journal of Nuclear Science and Technology. Thermophysical Properties of Materials for Water Cooled Reactors. Gilchrist... Scott.E. K. Maglic. Atomic Power Division. “Thermal Conductivity and Heat Capacity of Zircaloy2. BMI1273. and Inconel 625. Schenectady. pp. N.” International Journal of Thermophysics. 3. New York: Academic Press.” High TemperaturesHigh Pressures. K. 5. Godin. General Electric Company.B. Columbus. V. Application of the Ewing Equation for Calculating Thermal Conductivity from Electrical Conductivity from Electrical Conductivity. A. 62:257264. 1976.R. Cleveland. 1961.M. R. American Society for Testing and Materials. OH. J. Stadnyk. 1965. et al. PA. Lucks. “Thermal Property Measurements on Zircaloy2 and Associated Oxide Layers up to 1200°C. Cincinnati. and J.. 12(10):661662.” Journal of Nuclear Materials. Krett.4/FRAPTRAN 1..D. 79. A. Jr. 1958. 6784. The correlation used in the ZOTCON subroutine differs between MATPRO and the FRAPCON3.
41x10 4 T * (6. The data are from two different samples. Two sets of data attributed to Waldman by Maki are also shown in the table.. K (W/m*K). In addition. After the cladding temperature is established.4 databases. However. and both kinds of oxide have conductivities that are significantly lower than the stabilized Zircaloy dioxide conductivities reported by Adams (1954).4 are only a function of temperature. The equations used to describe these models are presented in Equations 3.4/FRAPTRAN 1.31 and 3.946 x10 10 )) (3.e.8 . Figure 3.835 1. The Gilchrist (1976) data includes two types of oxide films. these films do not have the same O/M ratio as cladding oxide and may not adhere well to the substrate. and Gilchrist (1976).81x10 4 T K FRAP 1. the earlier Gilchrist (1976) data is associated with considerable uncertainty and the data from 3.4 model is based on data from Kingery (1954). In addition. the thermal conductivity of the oxide layer can be calculated using the correlation determined from either the MATPRO or the FRAPCON3.31 reveals that the FRAP model fits the Kingery data from which it was derived and agrees well with the measurements performed by Adams (1954). The data of Lapshov and Bashkatov (1973) are from films formed by plasma sputtering of zirconium dioxide on tungsten substrates. respectively. The database for the MATPRO (2001) model includes data from a variety of zirconium oxide materials and test conditions. Lapshov and Bashkatov (1973). The FRAP model tends to overestimate both the Gilchrist data sets and the data from Lapshov and Bashkatov (1973).3.4/FRAPTRAN 1.31) (3. to cladding temperature.31 along with both cladding oxide thermal conductivity models.. The MATPRO (2001) correlation is derived from a database that includes data collected by Adams (1954). One is a nodular oxide and the other a black oxide characteristic of the kinds of layers usually reported in hightemperature tests with cladding. one was a bulk sample of 100 percent dense ZrO2 and the other with approximately 13 percent porosity.4/FRAPTRAN 1.43x10 7 T * 1. these data may not be representative of Zircaloy cladding oxide thermal conductivity because sputtered coatings are quite porous.32 for MATPRO and FRAPCON3. These models relate the thermal conductivity of the cladding oxide layer. The Maki (1973) data include two samples oxidized in steam over a small temperature range and show a sharp increase in conductivity between 400 and 500K. Each database was fit with a different model to relate the thermal conductivity of the cladding oxide layer to temperature.32) Both databases are presented in Figure 3. The correlation used in FRAPCON3. The database for the FRAPCON3.4 is derived from data collected by Kingery (1954). K MATPRO 0.3.9599 T * ( 2. Maki (1973).4/FRAPTRAN correlation and database was adopted from an earlier version of MATPRO (Hagrman et al. The FRAPCON3. as stated earlier. The principal recommendation for the data is that they were taken with black oxide from Zircaloy tubes.4 (i. However. FRAP). a more recent data set from Gilchrist (1979) is included.4 and FRAPTRAN 1. 1981). T (K). Considerable uncertainty is reported by Gilchrist (1976) because of difficulty in measuring oxide film thickness.4/FRAPTRAN 1. The nodular oxide thermal conductivities are much lower than the black oxide thermal conductivities.1 Model Development and Comparisons Oxide thermal conductivity models in MATPRO and FRAPCON3.
The MATPRO model was initially fit to the data collected by Adams (1954).Lapshov and Bashkatov (1973) is likely to be low due to different O/M ratios and high porosity in the sputtered films.9 .31 to better fit the Gilchrist (1976) data for black oxide. 3. The rationale for this correction was that the black oxide was more prototypic of the cladding oxide layer and was achieved by dividing the original equation by two. but it was adjusted to the form presented in Equation 3.
4 (FRAP) cladding oxide thermal conductivity models.31 Oxide thermal conductivity measurements as a function of temperature compared with the MATPRO and FRAPCON3. 3.9 Adams 8 Maki Gilchrist (black oxide) Kingery (100% Dense) Maki (Waldman data) Gilchrist (nodular oxide) Gilchrist (1979) Lapshov and Bashkatov Kingery (13% Porosity) Thermal Conductivity (W/m*K) 7 6 5 4 3 2 1 0 300 F MATPRO 500 700 900 1100 1300 1500 1700 1900 Temperature (K) Figure 3.10 .4/FRAPTRAN 1.
A Handbook of Materials Properties for Use in the Analysis of Light Water Reactor Fuel Rod Behavior. 3.” Journal of the American Ceramic Society. “Thermal Conductivity: III. is assumed to be described by Equation 3. and A. Kingery.. Idaho Falls. 5:1922. but is assumed to be within 10 percent for materials that are known to be dense (>87%TD) ZrO2.2 Applicability and Uncertainty The subroutine ZOTCON is applicable over the temperature range for which the data was collected. 37:7479. MATPROVersion 11 (Revision 2). KLiquid (W/m*K).” Journal of Nuclear Materials. ID.4 Cladding Surface Emissivity (ZOEMIS) The subcode ZOEMIS returns the cladding surface emissivity. Prolate Spheroidal Envelope Method. 37:107100. Inc. V. 2). it is difficult to establish a meaningful measurement of uncertainty.4.3. Gilchrist.” Journal of Nuclear Materials. Gilchrist.” Journal of the American Ceramic Society. 1954. 1976. “Thermal Conductivity of Oxide Deposited on Zircaloy Fuel Tube Material – A Continuation of Previous Work. “Thermal Conductivity of Coatings of Zirconium Dioxide Applied by the Plasma Sputtering Method. 3.33) The standard error for the MATPRO model is approximately 0.11 . Since the relatively high density of the bulk ZrO2 materials bias the thermal conductivity measurements high. NUREG/CR0479 (TREE1280. M. “Thermal Conductivity: X. EG&G Idaho.D. The ZOEMIS model described in MATPRO is the same as the model used in FRAPCON3. Maki.4 and FRAPTRAN 1.33 (Hagrman et al..3. K Liquid 1. W. Data for Several Pure Oxide Materials Corrected to Zero Porosity. 1954.” Journal of Nuclear Science and Technology.L. 1973. Hagrman.3 References Adams. 1973. et al. 10:107175.” Heat TransferSoviet Research. H. 82:193194. 375 to 1673K. Upon melting the thermal conductivity. 62:257264.. Bashkatov. 1979. 1981. K.75 W/m*K. which is directly proportional to the radiant heat transfer from the cladding surface during an abnormal transient. K. “Thermal Property Measurements on Zircaloy2 and Associated Oxide Layers. Rev. E..4 (3. D. The MATPRO model is applicable from 300K to the melting temperature (2973K). et al. “Heat Transfer Characteristics of Zircaloy2 Oxide Film. Lapshov..E. V.3. 3. N. 1981).
43 relates the emissivity above 1500K. to 1 and the maximum cladding temperature. Equation 3.88x106 m thick.12 . 1 0. emissivity is taken to be the larger of 0.808642 50. The model for emissivity was constructed by considering measured emissivities reported by several investigators.325 0. Both equations relate the hemispherical emissivity. T (K).43) Comparisons between data and the high and low temperature models are presented in Figure 3. When the cladding surface temperature has not exceeded 1500K.42) When the maximum cladding temperature has exceeded 1500K.1246 x10 6 d 1 0.41. 3.42.41 is used for oxide layers less than 3. d (m). to the oxide layer thickness. emissivities are modeled by Equations 3.3.41) (3. the oxide surface thickness is only a few wavelengths of near infrared radiation and is partly transparent.43.325 and the result of Equation 3. 1 (unitless). Equation 3.88x106 m thick and Equation 3. Oxide thickness is an important parameter for these thin coatings.4. which is affected by temperature and by chemical environment. but variations in crud on the external cladding surface and chemical reaction products on the inside surface are not modeled explicitly. 2 (unitless).42 is used for oxide layers equal to or greater than 3. 1500 T 2 1 exp 300 (3.0 d (3. The effect of temperature has been modeled. Expressions used to predict the emissivity of Zircaloy cladding surfaces are summarized below. Thicker oxide layers are opaque. so the oxide thickness is not as important as the nature of the outer oxide surface.1 Model Development and Comparisons Surface emissivities are significantly affected by surface layers on the cladding. For cladding with thin oxide coatings.41 and 3.
4 0.0 0.6 0.3 0. Murphy and Havelock Burgoyne and Garlick Uniform Oxide Burgoyne and Garlick with Crud Model Prediction > 1500K Juenke and Sjodahl Burgoyne and Garlick Nodular Oxide Model Prediction < 1500 K 3.41 Modeltodata comparison for cladding oxide emissivity.8 Emissivity (unitless) 0.7 0.9 0.2 0.13 .1 0.1.5 0.0 0 20 40 60 80 100 120 140 Surface Layer Thickness (mm) Figure 3.
expected from the use for Equation 3.4. 1 0 . the value 2 is returned as the expected standard error.1 Model Development and Comparisons In the MATPRO model. Values of thermal expansion between 1073 and 1273K are determined in the FRAPCON3.4/FRAPTRAN 1. General Electric Co. GEMP1008. If the prediction 2 + 2 is greater than 1 or if 2 . The MATPRO model calculates these strain components for single crystal Zircaloy and applies these to polycrystalline cladding materials through the use of pole figures collected over a cladding section. the FRAPTAN 1. 239242. In addition. basal plane symmetry (11 = 22) is assumed.14 .” Advanced Pressure Vessel Materials. and F. 3.2 at 0. “The effect of oxidation and crud deposition on the emissivity of Zircaloy2 cladding. Garlick. 3.44) T 1500 2 0. 3.F. The FRAPCON3. which is described by Equation 3. 3.5 Cladding Thermal Expansion (CTHEXP) The subroutine CTHEXP returns the axial and diametral components of thermal expansion in the cladding as a function of temperature.B.45 is modified to limit 2 + 2 at 1 and/or 2 . 1. Cincinnati.” Journal of Nuclear Materials. the standard error of Equation 3. T. Murphy.0).1 When the maximum cladding temperature exceeds 1500K. and L. Norway. 1976. the expected standard error is estimated by 2.” Paper presented to the OECDCSNI Meeting on the Behavior of Water Reactor Fuel Elements under Accident Conditions.44.45 predict values of 2 and 2 that fall inside the range of physically possible values of emissivity (0.5. 1968. In addition.2 is less than 0.4 model includes a calculation of the thermal expansion for ZrNb1.4 codes by linear interpolation. a total of six correlations that are functions of temperature only are used to find single crystal thermal strains. OH. E. 60:167176. 1976. Spåtnid. Havelock.41 to predict emissivity in a reactor when cladding surface temperature has never exceeded 1500K is given by Equation 3.2 Applicability and Uncertainty The absolute standard error.0 – 1.45. pp. Sept.4/FRAPTRAN 1.3 References Juenke.V.H.1 exp 300 (3.4 models use the correlation from Mehan and Wiesinger (1961) for temperatures below 1073K and a constant value of thermal expansion above 1273K.4.43 and 3.. Burgoyne. “Physical and Mechanical Properties: Emittance Measurements. Sjodahl. Missile and Space Division.. (3. 1316. and A. “Emissivity of Zirconium Alloys in Air in the Temperature Range 100400°C.3.45) If Equations 3.. 1976. E.
a typical 2 percent volume increase at melt was assumed (Siefken et al.51) (3.77763 1.76 x10 6 T 4.09822 cos x10 3 161 T 1083 33 2.95 x10 6 T 1.57) thermal expansion strain in liquid Zircaloy (m/m) circumferential thermal expansion strain of a single crystal of Zircaloy at 2098K (m/m) axial thermal expansion strain of a single crystal of Zircaloy at 2098K (m/m) where the 11 and 33 are calculated by Equations 3.7 x10 6 T 1. The expressions used for the thermal strain in liquid zirconium (temperatures 2098K) are thus: p p 11 33 = = = 2 1 11 33 0.0067 3 3 (3. For 1244 T < 2098K (3.77763 1. The correlations for single crystal thermal strains are: For 300 < T < 1083K 11 4. but comparisons with Zircaloy2 and zirconium data also show good agreement for these materials..56) For temperatures 2098K.53 and 3. consideration of the volume change associated with melting is required.78 x10 3 where 11 33 = = circumferential thermal expansion (m/m) axial thermal expansion (m/m) (3.55) (3.09822 cos x10 3 161 where the arguments of the cosines are in radians.26 x10 5 T 3.52) For 1083 T < 1244K T 1083 11 2.04 x10 2 33 9. respectively. 2001). Since no data were found for the MATPRO model.53) (3.485 x10 3 33 1.54) 11 9.The model was developed for as fabricated Zircaloy4.4 x10 3 (3. 3.15 .54.
axial 2. These correlations will be examined further and possibly revised in a future revision of FRAPCON3. respectively.5060 10 5 4. the coefficient of thermal expansion is the constant value of 9.59) The correlations used to calculate the axial and diametral components of thermal expansion in the cladding above 1273K are presented in Equations 3.511) Modeltodata comparisons for the MATPRO and FRAPCON3. respectively. axial 8.7210 10 6 T (3.59.70 10 6 T (3.51 and 3.4.7x106. and Kearns (1965).16 . Strain is given a function of temperature. T (°C).To obtain cladding strains from these single crystal strains. The model for cladding thermal expansion in the FRAPCON3. the thermal expansion components are determined by linear interpolation.70 10 6 T diametral 6.300 10 3 9. Scott (1965). Strain is given a function of temperature.58) (3.4/FRAPTRAN 1.4/FRAPTRAN (FRAP) codes is different from the MATPRO model but provides similar predictions of expansion (< 6 percent).4 models are presented in Figures 3. respectively. Above 1273K. 3.58 and 3. Such an averaging requires the collection of pole figures and is described in greater detail in the MATPRO (2001) handbook.800 10 3 9.511.4410 10 6 T diametral 2.52 for axial and diametral thermal expansion. The data for the FRAP correlation used from room temperature to 1273K was taken from Mehan and Wiesinger (1961).4/FRAPTRAN 1. The input parameters for typical cladding are not readily available. as recommended by Lustman and Kerze (1955). The correlations used to calculate the axial and diametral components of thermal expansion in the cladding between room temperature and 1073K are presented in Equations 3.3730 10 5 6.510) (3. These comparisons illustrate that the axial and diametral components of cladding thermal expansions are being underpredicted. it is necessary to do a volume weighted averaging of crystalline orientation over the entire cladding section. T (°C).510 and 3. Between 1073 and 1273K (approximately the alphabeta transition range for Zircaloy).
0E03 7.4 (FRAP) and MATPRO axial thermal expansion.0E03 6.0E03 1.0E03 5. which included temperatures up to 1100K.0E03 Thermal Expansion (m/m) 8.4/FRAPTRAN 1. and Mehan Axial Temperature (K) Figure 3.0E03 Thermal Expansion (m/m) 8.0E03 5.0E03 1. Scott.0E03 4.0E03 3. 1.0E03 2. Data from cladding tubes. 3.4/FRAPTRAN 1.0E+00 200 400 600 800 1000 1200 1400 1600 1800 FRAP Axial MATPRO Axial Bunnel Axial Kearns.0E03 2.0E02 9.0E03 0.0E03 4.0E03 7.1.52 Modeltodata comparison for FRAPCON3.0E03 3. and Mehan Diametral Temperature (K) Figure 3. Data from cladding tubes. 3.0E+00 200 400 600 800 1000 1200 1400 1600 1800 FRAP Diametral MATPRO Diametral Bunnel Diametral Kearns. Scott.5.0E03 6. The uncertainty of the axial and diametral thermal expansion correlations has not been quantified.0E03 0.0E02 9.51 Modeltodata comparison for FRAPCON3.4 (FRAP) and MATPRO diametral thermal expansion.17 .2 Applicability and Uncertainty The subroutine CTHEXP is applicable over the temperature range for which the experimental data was collected.
B. Idaho National Engineering and Environmental Laboratory. Elastic moduli are affected primarily by temperature and oxygen content. Harvego. Bunnell. p. Whitmarsh. The subcode CELAST would only be required if the Young’s modulus and shear modulus for an anisotropic cladding is desired and. P. 3. WAPDTM472. 1955. 1962). Wiesinger. Rev. 1961. KAPL2110. but they are not as important as temperature and oxygen content for typical LWR fuel rod cladding. Siefken. cold work. Fisher and Renken (1964). since these data include the best description of texture for the temperature range in which they were used.5. et al. In practice. WCAP326941. Lustman and F. Pittsburgh. Schenectady. However. pp. The models are based primarily on data published by Bunnell et al. High Temperature Properties of Zircaloy Oxygen Alloys. ID. Fast neutron fluence. J. General Electric Company. Battelle Pacific Northwest Laboratory. Thermal Expansion and Preferred Orientation in Zircaloy. Mechanical Properties of Zircaloy2. cladding is frequently assumed to be an isotropic material.. and Padel and Groff (1976).0x1010 is used to avoid dividing by zero). R.4 (FRAP) codes to determine the Young’s modulus and the shear modulus. the cladding is assumed to be isotropic. and J. L. The Metallurgy of Zirconium.J. 4. L. In such a case. 1718.18 . Coryell. New York. E. Pittsburgh.A. SCDAP/RELAP5/MOD 3. as stated previously. Vol. and F.3 References Mehan.R. Bettis Atomic Power Laboratory. 2. Atomic Power Division. which is included in MATPRO and determines the compliance matrix for isotropic cladding. 1965. 1. 1977.L. 1965.K.J. Westinghouse Electric Corporation. NUREG/CR6150.. CELAST is not used in the FRAP codes. NY. Idaho Falls. To calculate Zircaloy elastic moduli at temperatures greater than the melting temperature of Zircaloy (2098K). 355. Scott. INEL96/0422. Kerze. 2001.3 Code Manual MATPROA Library of Materials Properties for LightWaterReactorAccident Analysis. and texture effects are also included in the models described herein. PA. McGrawHill Book Company.W. PA. E. The subcodes CELMOD and CSHEAR are used in the MATPRO and the FRAPCON3. (1977). 1957. Armstrong and Brown (1964). Data from several other sources are used to evaluate the expected standard error of the CELMOD and CSHEAR codes and to estimate the effect of fast neutron fluence (Shober. Kearns.. The tensor from which these moduli are derived is calculated by CELAST.3.6 Cladding Elastic Modulus (CELMOD) and Shear Modulus (CSHEAR) Elastic moduli are required to relate stresses to strains.W. the moduli are set to zero (actually. Richland. EPRI NP524. Hohorst. Physical and Mechanical Properties of Zircaloy2 and 4. 3. only two independent elastic moduli are needed to describe the relation between elastic stress and strain: the Young’s modulus and the shear modulus.4/FRAPTRAN 1. WA. Knolls Atomic Power Laboratory. The elastic moduli are defined by the generalized form of Hooke’s law as elements of the fourth rank tensor that relates the second rank stress and strain tensors below the yield point.
Y (1.0012 kg oxygen/kg Zircaloy) that the exact magnitude of the asreceived concentration will not affect the correlation predictions. fast neutron fluence (n/m2).19 . The differences in form between the codes will be discussed in this subsection.4.4x109 Pa. The MATPRO expressions used to model the effects of oxidation.912 x108 T ) K 2 2.12 exp 25 10 where = (3.64) (3.61 and 3. 3.6. MATPRO determines Y by linearly interpolating between values calculated at the alpha to alpha plus beta and the alpha plus beta to beta phase boundaries.65) C = = average oxygen concentration minus oxygen concentration of asreceived cladding (kg oxygen/kg Zircaloy).62 to describe the Young’s modulus in the alpha phase and the beta phase.088x1011 5. MATPRO uses Equations 3.63) (3.1 CELMOD Model Development and Comparisons The CELMOD subcode used in MATPRO is the same equations written in a different formas the one used in FRAPCON3.63 through 3.475x10 7 T K 1 K 2 ) / K 3 Y 9. cold work (unitless ratio of areas).05x10 7 T where Y T K1 K2 K3 = = = = = Young’s modulus for Zircaloy2 and 4 with random texture (Pa) cladding temperature (K) modification to account for the effect of oxidation (Pa) modification to account for the effect of cold work (Pa) modification to account for the effect of fast neutron fluence (unitless) (3.62) In the alpha plus beta phase. cold work. The standard error of CELMOD is 6. This standard error is not included in FRAPTRAN 1. and fast neutron fluence are presented in Equations 3. respectively.61) (3.65. Asreceived oxygen concentrations are so small (0.4 is coded to include this error.61x1011 5. K1 (6.6 x1010 C K 3 0. Although FRAPCON3.88 0. this value is not used.4 (FRAP).4/FRAPTRAN 1.3.21x1010 4.
cold work.610) Both FRAPCON3. 3. c3.4 uses these errors.67 through 3.66) where celmod ctemp deloxy cwkf = = = = Young’s modulus (Pa) cladding temperature (K) input average oxygen concentration excluding oxide layer (kg oxygen/kg Zircaloy)(hardwired to zero in FRAPCON3. c1 1. FRAPTRAN 1.4 and FRAPTRAN 1.88 * 1 exp fnck / 1x10 25 exp fnck / 1x10 25 input effective fast fluence (n/m2) (3.611 and 3. and fast neutron fluence respectively.49 x1010 1. c2 is given by Equation 3.612) In the alpha plus beta phase.68) (3. and c3.66 x10 7 T (3.66. c2.69) c 3 2.67) (3.088 x1011 5.037 x108 * 5. Equations 3. cel mod (1. and c2 are expressions that account for oxygen content. neither FRAPCON3.475 x10 7 * ctemp c1 * deloxy c 3 * cwkf ) / c 2 (3.66.4 nor FRAPTRAN 1. G 4.69 describe the constants c1.4 includes the coding to return the standard error of the model.6 x1010 For neutron fluences greater than 1x1022. CSHEAR In MATPRO.611) G 3.168x10 7 T K1 K 2 K3 (3.0 (3. the expressions used to determine shear modulus in the alpha and beta phases are given in Equations 3.610.612. respectively.04x10 10 2.4 includes additional coding to determine Young’s modulus for ZrNb1. the expression for Young’s modulus in the alpha phase is presented in Equation 3. c1.20 . MATPRO determines G by linearly interpolating between values calculated at the alpha to alpha plus beta and the alpha plus beta to beta phase boundaries.4 use the same correlations to determine Young’s modulus in the subcode CELMOD.The CELMOD subcode included in the FRAP codes differs only in formy from the MATPRO version.4) input effective cold work (unitless ratio of areas) In Equation 3. where fnck = c 2 0. For instance.4/FRAPTRAN 1. respectively.7015 c 2 1 .16 x1011 ctemp * 1. Although FRAPCON3.
K2.4 includes the coding to return the standard error of the model. For instance.07 x1011 2.615 through 3.4 nor FRAPTRAN 1.613) The CSHEAR subcode included in the FRAP codes differs slightly from the MATPRO version. c2.615) (3. c2 is given by Equation 3.4 returns the error. However. Although FRAPCON3. and K3 have the same definitions as stated before in CELMOD. neither FRAPCON3.618.168x10 7 * ctemp c1 * deloxy c 3 * cwkf ) / c 2 (3. (3. Figure 3.4 use the same correlations to determine Young’s modulus in the subcode CSHEAR.The constants K1.88 * 1 exp fnck / 1x10 25 exp fnck / 1x10 25 input effective fast fluence (n/m2) (3. and fast neutron fluence.21 . the expression for shear modulus in the alpha phase is presented in Equation 3.614. and Poisson’s ratio as a function of temperature for ZrNb1.614. and c2 are expressions that account for oxygen content.616) c 3 0. respectively. shear modulus. cold work. measured Young’s modulus values for data sets used in the development of the CELMOD models as well as more recent data. the expression used to model the effect of oxidation for shear modulus is given in Equation 3. a modeltodata comparison for Young’s modulus is presented. cshear (4.315 x108 c 2 1 .07 x1011 ctemp * 2.4 and FRAPTRAN 1.04 x1010 2. FRAPTRAN 1.4 includes additional coding to determine Young’s modulus.617 describe the constants c1. and c3. c3.618) Both FRAPCON3. Equations 3.61 presents predicted vs. c1 7.614) where cshear ctemp deloxy cwkf = = = = shear modulus (Pa) cladding temperature (K) input average oxygen concentration excluding oxide layer input effective cold work (unitless ratio of areas) In Equation 3. K 1 (7. respectively.867 x1010 For neutron fluences greater than 1x1022.0 (3. Since there is limited data available from shear modulus measurements.315 x10 8 T ) The standard error of the CSHEAR code is 9x109 Pa.613. where fnck = (3. c1. 3.617) c 2 0.
22 .The upper and lower bounds of Figure 3. 3. Nevertheless the CELMOD model. This is attributed to the relatively low temperatures that were used to collect these data (e.4 GPa.61 represent an interval of ± 6. provides a satisfactory fit to the data within the temperature range it will be applied for predicting inreactor performance (> 250°F). a quarter of Bolmaro’s data is collected at < 250°F). It may be noted that the data fit is slightly better below 80 GPa ( ~ < 4 percent) than above 80 GPa (~ > 5 percent)..g. in general.
61 Upper Bound Busby Northwood Armstrong and Brown Spasic Bolmaro and Povolo Predicted vs. 3.120 Predicted Young's Modulus.23 . GPa Lower Bound Padel and Groff Mehan Measured = Predicted Figure 3. GPa 80 40 0 0 40 80 120 Measured Young's Modulus. measured values for Young’s modulus for various data sets.
ORNL3281. and C. Armstrong. Northwood. Padel.” J. 1976. L. Materials Science.R.” Journal of Nuclear Materials. 1961.L. 55:299310. 1988. p. 3. TN.E. USAEC CONF681086.6. and F. et al. High Temperature Properties of Zircaloy Oxygen Alloys.L.4 GPa (the upper and lower bounds). Whitmarsh. pp.W. 277284.. et al. Above this temperature. The CELMOD and CSHEAR correlations are similar between MATPRO and the FRAPCON3. “Anelastic behaviour of materials under multiaxial strains. P. 1975. R. Shober.. KAPLMRLM16. Schenectady. Battelle Pacific Northwest Laboratory. Schenectady.. D. Bolmaro. Mehan.” Physical Review. and Hf. Battelle Memorial Institute. Spasic. WAPDTM585.3 References Bunnell. Bettis Atomic Power Laboratory. et al.L.L. Groff. General Electric Company. 1112.3. the moduli are assumed to be zero (but actually set at 1x1010 to avoid dividing by zero).C.E. Czechoslovakia. EPRI NP524. Modulus of Elasticity of Zircaloy2 Between Room Temperature and 1.. C. A.4. “Variation du Module de Young du Zirconium b en Function de la Temperature. Fisher.6. BMI1168. General Electric Company. Z. and A.24 . 962966.” Journal of Nuclear Materials 59:325326.. “Dynamic Young’s Modulus Measurement above 1000°C on Some Pure Polycrystalline Metals and Commercial Graphites. R. NY. Savannah Reactor Design. Review of Zircaloy2 and Zircaloy4 Properties Relevant to N. Marlanske Lanze.. WA. pp. E. 1964. Wiesinger. 23:371.4/FRAPTRAN 1. Columbus. 1962. 1964.. pp. KAPL2110. OH. Properties of Zircaloy4 Tubing. Brown. The uncertainty of the predictions for CELMOD are within 6. Renken. Mehan. Busby.O.B. “Elastic Constants of Zirconium Alloys.. 1968.J. Oak Ridge. NY. Knolls Atomic Power Laboratory. 3. 1977. and F.000°F. 135(2A):A482494. Povolo.. Oak Ridge National Laboratory. and typically < 5 percent below 80 GWd/MTU. Richland. Conference on the Use of Zirconium Alloys in Nuclear Reactors. August 1964. The uncertainty of the CSHEAR predictions is estimated to be within 9 GPa. “Single – Crystal Elastic Moduli and the CHPBCC Transformation in Ti. and H. PA. Mechanical Properties of Zricaloy2. Pittsburgh. 1958. 65. 1968.” Transactions of the Metallurgical Society of AIME 230. Knolls Atomic Power Laboratory. et al. 1966. 1957. C.S.S. F.2 Applicability and Uncertainty The CELMOD and CSHEAR correlations are applicable below the melting temperature of the cladding (2098K). Zr. The Mechanical Properties of Zirconium and Zircaloy2. R.
time (s). Consequently. as determined by Xray diffraction analysis. the CAGROW model was updated in FRAPCON3. The original CAGROW model used in MATPRO was modified in FRAPCON3.4 version includes correlations for Zircaloy2 and 4.71) = fractional change in length due to growth.4 to better describe the cladding axial growth at higher fluence.1 Model Development and Comparisons The MATPRO correlation used to describe cladding axial growth is presented in Equation 3. A value of fz = 0.4 to better model axial growth under greater fluences ( > 1x1025 n/m2).72 and relates the axial growth. The MATPRO subroutine CAGROW was based on lowburnup.407x1016 (n/m2)1/2. cladding temperature (K). M5. In addition.02CW (3. and Daniel (1971. texture factor for the tubing. This model is presented in Equation 3.0 MeV).4. These correlations may be applied to either PWR or boiling water reactor (BWR) conditions.71 and growth above 360°C is approximated by using T = 360°C. lowfluence data found in Harbottle (1970). ax 2. Axial growth for temperature below 40°C is approximated by using T = 40°C in Equation 3.7. A model proposed by Franklin (1982) that was based on high fluence PWR data was adopted for FRAPCON3.71. 1. 1972). cold work (fraction of crosssectional area reduction).25 . rod growth increases the internal void volume that impacts the rod internal pressure calculation. This equation has been developed to model the irradiation growth of Zircaloy tubes at temperature between 40 and 360°C (the normal range of cladding temperatures in LWRs). and ZIRLO.72) 3.3. The CAGROW subroutine is not used in FRAPTRAN 1.4 because significant irradiation axial growth is not expected under the short time scales considered by FRAPTRAN 1. The change in length of commercial fuel rods due to irradiation growth is small. of stress relief annealed (SRA) Zircaloy4 in a PWR to fluence (n/cm2). However.8 / T t where L/L A T t fz CW = = = = = = 1/ 2 1 f z 1 0. growth can be a significant fraction of the clearance between the rod and the top and bottom assembly nozzles. fast neutron flux (n/m2s) (E > 1.845 (3. ax (m/m).18 x10 21 * 0.7 Cladding Axial Growth (CAGROW) The subcode CAGROW calculates the factional change in length of Zircaloy tubes due to irradiationinduced growth.05 is typical. Contact with the nozzles can cause rods to bow and possibly fail at points where the rods contact each other. The MATPRO version of CAGROW was applied to both Zircaloy2 and 4. Kreyns (1960). The variable fz is the effective fraction of cells aligned with their <0001> axis parallel to the tubing axis. The FRAPCON3. 3.4. L / L Aexp 240 .
respectively.where ax = = axial growth increment (m/m) fast neutron fluence (n/cm2) (E > 1. M5. Based on these figures.26 .75 through 3.73 and 3.78 for SRA Zircaloy4.81787 For ZIRLO cladding: (3.72 is used to determine the axial growth at the beginning and the end of a time increment.013 10 21 0.5. The variables have the same definitions as those described in Equation 3. In addition to updating the cladding axial growth model for SRA Zircaloy4 and RXA Zircaloy2. For M5 cladding: ax 7. it appears that the cladding axial growth of these alloys can be predicted well with the CAGROW subroutine. which is the value returned by the CAGROW subroutine. ax (m/m). The difference is used to determine the axial growth for the time increment. and ZIRLO. to fluence (n/cm2).73) ax 9. which is the value returned by the CAGROW subroutine.7893 10 25 0. RXA Zircaloy2.0 MeV) The expression in Equation 3.72 and the time difference is used to determine the axial growth during the time increment. new correlations have been added to describe the cladding axial growth of M5 and ZIRLO.72 can be used to describe the axial growth of fully recrystallized annealed (RXA) Zircaloy2 cladding in a BWR by multiplying by a factor of 0. 3.74) Modeltodata comparisons are presented in Figures 3. respectively. The expression in Equation 3.98239 (3.74. These equations relate the cladding axial growth. The cladding axial growth correlations for M5 and ZIRLO are presented in Equations 3.
The standard error of L/L (%) is 0.5 0.8 0.1 0. 3.6 0.0 0 2E+21 4E+21 6E+21 8E+21 1E+22 1.4 0.3 0.7 L/L.2E+22 Fast Neutron Fluence.0 0 2E+21 4E+21 6E+21 8E+21 1E+22 1.76 Modeltodata comparison for RXA Zircaloy2 (BWR).2 0.8 0. % 0.0 0.1. % 0.9 0.4 0.2E+22 Fast Neutron Fluence.2 0.3 0.07.1 0.11. n/cm²s FRAPCON Data Gilbon Zr2 BWR Model Figure 3. n/cm²s Newman Franklin Gilbon Zr4 PWR Model Figure 3.75 Modeltodata comparison for SRA Zircaloy4 (PWR). 1.7 L/L.9 0. The standard error of L/L (%) is 0.5 0.0 0.27 .6 0.
78 Modeltodata comparison for ZIRLO.5 0.0 0 2E+21 4E+21 6E+21 8E+21 1E+22 1.6 0.6 0. % Figure 3.77 Modeltodata comparison for M5. E>1MeV M5 Data PWR M5 Model Figure 3.4 0.1 0.0 0 2E+21 4E+21 6E+21 8E+21 1E+22 1.1 0. (2000). 1.2E+22 Fast neutron fluence.9 0.7 0.8 0.05. % 0.2E+22 Fast neutron fluence.3 0.2 0. n/cm². n/cm². Vandellos data was collected by Irisa (2000) while North Anna and BR3 data was collected by Sabol (1994). The standard error of L/L (%) is 0. The standard error of L/L (%) is 0.2 0.7 0.05.3 0.8 Fuel rod growth.5 0.28 .1. E>1MeV Vandellos ZIRLO Data BR3 ZIRLO Data North Anna ZIRLO Data PWR ZIRLO Model Fuel rod growth.0 0. M5 data was collected by Gilbon et al.9 0.0 0.4 0. 3.
K. S. R.7.J. InPile Dimensional Changes of Zircaloy4 Tubing Having Low Hoop Stresses (LWBR Development Program). Lynchburg. PA. PA.G. J. 235267. ASTMSTP1354.11. Garde and E.W. “InReactor Corrosion Performance of ZIRLO and Zircaloy4. Doriot.3.. 2000.. quoted by E.” Zirconium in the Nuclear Industry: Twelfth International Symposium.C. Larouere.7. Y. These three alloys are typically used in PWRs. “InPile Dimensional Changes of Zircaloy4 Tubing Having Stresses (Light Water Breeder Reactor Development Program).M. WAPDTM973. S. E > 1MeV). Soniak. P. R.N. R... 3. ASTMSTP1425. “The Temperature and Neutron Dose Dependence of Irradiaton Growth in Zircaloy2. “Irradiation Creep and Growth Behavior. 1982. Irisa. Daniel. M5. Alonso. R. Stanutz. D.3 References Harbottle. UT. 0. Babcock and Wilcox Company. The absolute uncertainties for the growth strains predicted for stressrelief annealed Zr4. and 0.M. Comparison with Experiment of Calculated Changes and Failure Analysis of Irradiated Bulk Oxide Fuel Test Rods Using the CYGRO01 Computer Program. West Conshohocken. L. VA.. DOE/ET/3421250. Pittsburgh. whereas recrystallized Zr2 is used in BWRs... ASTMSTP485. 1972. ASTMSTP754. and ZIRLO are 0.” Nuclear Technology.P. G. 14:171186. 3. 2000. American Society for Testing and Materials. Suzuki. GP Sabol and GD Moan eds. eds.4 is applicable over the temperature range from 700 to 900K for local burnups up to 65 GWd/MTU (or 12x1025 n/m2 fluence. R. and R. pp. Newman. respectively. Mardon. The Hot Cell Examination of Oconee 1 Fuel Rods After Five Cycles of Irradiation.” Irradiation Effect on Structural Alloys for Nuclear Reactor Applications. Pittsburgh. West Conshohocken. Bradley.P. J.H. pp. Murai. 1986. Bettis Atomic Power Laboratory. pp. 1996. Daniel. pp. WAPDTM583.29 . PA.. A.” In Proceedings of the Fifth International Symposium on Zirconium in the Nuclear Industry.05. D. Duncombe et al.07. 5173. American Society of Testing and Materials. Goto. “Segmented Fuel Rod Irradiation Programme on Advanced Materials for High Burnup.C. and Microstructural Evolution of Advanced ZrBase Alloys. Gilbon. Kreyns. 1970. 1971. “Zircaloy Cladding Deformation during Power Reactor Irradiation.” Zirconium in the Nuclear Industry: Tenth International Symposium. PA. PA.05. The absolute uncertainty in growth strains for Zr2 is 0.E. Sabol.A. P. Park City. and J. Philadelphia. K. Weiner. American Society of Testing and Materials. 2000.2 Applicability and Uncertainty The CAGROW model in FRAPCON3.” in Proceedings of the 2000 Light Water Fuel Performance Meeting.R. April 1013. Comstock. 724744. Bettis Atomic Power Laboratory. American Society of Testing and Materials. West Conshohocken. Franklin.. A. 287299. Sabate. 1994. PA.
The steadystate thermal and irradiation creep rates are given by Equations 3.008314 kJ/molK) fast neutron flux (n/m2s) The variables A. This model uses a thermal creep model described by Matsuo (1987) and an empirical irradiation creep rate with tuned model parameters that were fit to data by Franklin et al./in.4 is based on the model given by Limbäck and Andersson (1996).30 .8.82) th irr T eff R = = = = = = thermal strain rate (in. This model has different parameters for SRA and RXA cladding types. Several of the fitting coefficients were consequently changed to accommodate this modification. C1. Also. In addition. The FRAPTRAN calculation of cladding ballooning as a result of a LOCA is handled with a separate subroutine named BALON2 3. because creep occurs over much longer time intervals than those considered in FRAPTRAN 1. CREPR is not used in FRAPTRAN 1. Co. Q. respectively./hr) temperature (K) effective stress (MPa) universal gas constant (0.4 of a minute or less.81.4. E.4 to determine the cladding creep rate.8 Creep Rate (CREPR) The subcode CREPR is used by FRAPCON3. The Limbäck model was further modified by PNNL to use effective stress rather than hoop stress as an input so that the difference in creep behavior during tensile and compressive creep would be modeled correctly. accounting for both thermal and irradiation creep.1 Model Development and Comparisons The creep model used by FRAPCON3. (1983). n. C2. a temperaturedependent term was added to the formula for irradiation creep strain rate.81 and 3. ai.82./in.3. and cladding type (SRA or RXA). and the function f(T) have different values depending on temperature.81) irr C o C1 C2 f (T) eff where (3. 3.3. flux./hr) irradiation strain rate (in. E a Q th A sinh i eff exp T E RT n (3. The CREPR subcode was not used in MATPRO. since creep is a timedependent deformation process. and provides much more reasonable creep strains in the LWR range of temperatures and cladding hoop stresses than the previous model used in FRAPCON3. The values for these variables under different conditions are presented in Table 3.
81 Parameter A* E ai n Q Co C1 C2 f(T)* Parameters for FRAPCON3.08x10 * 5 9 RXA Cladding 5.0216 0.3)]} =fast neutron fluence (n/cm2) 2.86) The effective stress in the cladding is found using the principal stresses at the midwall radius using the thick wall formula.87) t Pi ri2 Po ro2 ri2 ro2 Po Pi r2 ro2 ri2 (3. which is obtained by taking the derivative of Equation 3.4x1027*1.00699132T 1. r Pi ri2 Po ro2 ri2 ro2 Po Pi r2 ro2 ri2 (3. and time.85. in FRAPCON3.89. the combined thermal and irradiation creep rates.85 1.84) The total thermal strain can then be calculated as a function of time. The principle stresses can be determined with Equations 3.0985x1024 0.5 201 4.18562+0.0 0. H s 1 exp 52 th irr t th irr t p (3.109 2 tanh 35500 th irr p (3.47x108* unitless kJ/mole (n/m2s)Cl MPaC2 unitless unitless unitless 1.4763 *Parameters changed from original Limbäck equation.83) (3.0 3.4 the strain rate is used.85.0136T T > 625 1.88) 3. th irr th irr 2.9*T 650{10.149x10 59.0237+0.7994 3.7283 570< T< 625 7.84. Units K/MPa/hr MPa MPa 1 SRA Cladding 1. in hours.83 and used to calculate the saturated primary hoop strain.56[1exp(1. which is presented in Equation 3.87473x1024 T < 570 0.05 th irr s 0. t. which relates the total thermal strain rate to the saturated primary hoop strain. t (hours). This derivative is presented in Equation 3.4 creep equation for SRA and RXA cladding. as shown in Equation 3.86.Table 3.1840 1.87 through 3. However.85) H th 2 52 s 1 / irr p 2 t1/ 2 exp 52 th irr t th irr (3.31 . The thermal and irradiation creep rates may be added together as shown in Equation 3.
. and Gilbon et al.8. 3.5 percent for SRA and 21.32 . It has been found that the Zircaloy RXA model adequately describes the creep behavior of M5 (Gilbon et al.81 and 3. The reduction factor is the result of studies that have shown that ZIRLO exhibits about 80 percent of the Zircaloy4 creepdown (Sabol et al. (2000).6 percent for RXA cladding. 2002). eff 0. The relative standard error for the predictions is 14. The Zircaloy SRA model is used for ZIRLO with a reduction factor of 0. 1994). Soniak et al.. Modeltodata comparisons are presented in Figures 3. 2000. Soniak et al.. Creep data was obtained from irradiated SRA and RXA tube from Franklin (1983).l where Pi Po ri ro r r t l = = = = = = = = inner pressure outer pressure inner radius outer radius radius within tube radial stress tangential stress longitudinal stress Pi ri2 Po ro2 ro2 ri2 (3.82 for SRA and RXA Zircaloy cladding. (2002).89) The effective stress can then be calculated by Equation 3.810) The subcode CREPR can be used for newer alloys like M5 and ZIRLO.5 l t t r r l 2 2 2 (3.810.
33 . measured hoop strain for RXA Zircaloy model. in/in Franklin Soniak 350°C Soniak 320°C Gilbon Predicted=Measured Figure 3.02 0.004 Measured hoop strain.002 0. 3.015 0.01 0.025 Measured hoop strain.004 0. This covers the nominal 3. measured hoop strain for SRA Zircaloy model. in/in 0.01 0.008 Predicted hoop strain. an effective stress range of 40 to 130 MPa.81 Predicted vs.025 Predicted hoop strain. 0. The database spans a temperature range of 570 to 625K.0.015 0.8.005 0.002 0 0 0. in/in 0. in/in Franklin Soniak 350°C Soniak 320°C Gilbon Predicted=Measured 0.005 0 0 0.006 0.006 0.2 Applicability and Uncertainty The subroutine CREPR is applicable over the range of conditions for which the data was collected. and a fast neutron flux range of 1x1017 to 2x1018 n/m2s.008 Figure 3.82 Predicted vs.02 0.
448468..3 References Limbäck. Bernaudat. “Creep of Zirconium Alloys in Nuclear Reactors. the points of contact enlarge due to localized plastic deformation and the solidtosolid thermal conductance is improved. Soniak. The same CMHARD subroutine is used in the MATPRO.operating conditions for LWRs.R. Mardon. Doriot. A.E.” Journal of Nuclear Science and Technology. E. Sabol. 724744. A. Garde and E. Boffioux. R. pp. Bradley and G.N.” ASTM STP 815.P. eds. 5173. 2002. 1983.9 Cladding Meyer Hardness (CMHARD) The subroutine CMHARD calculates Meyer hardness as a function of cladding temperature. A. ASTM STP 1245.L. “Thermal Creep of Zircaloy4 Cladding under Internal Pressure.” Zirconium in the Nuclear Industry: Tenth International Symposium. C. ASTM STP 1295. P. Moan and P. 3. R. Franklin. V. Bement.6 percent for RXA cladding.P. West Conshohocken. Comstock. The Meyer hardness is used by Ross and Stoute (1962) in their heat transfer correlation as an indication of the hardness of resistance to deformation of the softer (Zircaloy) material. L’Hullier.94x108 N/m2 (the highest temperature data point) and includes provisions for ZrNb1. PA. G. Lucas. American Society for Testing and Materials.4 includes additional coding that ensures that the minimum hardness returned is 1.” Zirconium in the Nuclear Industry: Thirteenth International Symposium. However. ASTM STP 1423.8. 3. G. D. and T. The relative standard error for the predictions is 14. As the contact pressure between the two surfaces increases. G. ASTM International. American Society for Testing and Materials. PA. R. 1987. American Society for Testing and Materials.. “Irradiation Creep and Growth Behavior. “InReactor Corrosion Performance of ZIRLO™ and Zircaloy4. PA.. 3.M. 2000. JP.” Zirconium in the Nuclear Industry: Twelfth International Symposium. Rudling. American Society of Testing and Materials. eds..4 codes. Bradley.R.4.D. N.A.P. West Conshohocken. 1994.J. Larouere. eds.G.P. Matsuo. pp. eds. Philadelphia.34 . “A Model for Analysis of the Effect of Final Annealing on the In. D. and J. Sabol and G. ASTMSTP1354..” Zirconium in the Nuclear Industry: Eleventh International Symposium. Weiner.. 1996. Gilbon.and OutofReactor Creep Behavior of Zircaloy Cladding. pp.. Stanutz. P.. Mardon. PA. West Conshohocken. S.. PA. and Microstructural Evolution of Advanced ZrBase Alloys. The uncertainty for predicting cladding creep in a fuel rod will most likely be greater than these values because there may be greater uncertainty in the estimate of fast neutron flux and cladding temperature than those from the creep data used to determine the creep model uncertainties.. and FRAPTRAN 1.837862. M. 24(2):111119. A. Andersson.D. Y. pp. Soniak. FRAPTRAN 1. Moan. Sabol. FRAPCON3. Hardness is one of the parameters required for calculating fueltocladding contact conductance.. West Conshohocken.5 percent for SRA and 21. Rebeyrolle. “Irradiation Creep Behavior of ZrBase Alloys. G.
5621x10 8 (3. Meyer hardness numbers for temperatures from 298 to 877K were taken from Peggs and Godin (1975).92) where MH T = = Meyer hardness (N/m2) temperature (K) The Meyer hardness decreased rapidly with increasing temperature.6034 x101 T 2. Rockwell. However. the routine CMHARD was created to provide information required by the Ross and Stoute gap conductance model that includes a dependence on Meyer hardness.92.0x105 N/m2. etc. Other hardness numbers are available (Brinell.35 . and conversion from one to another is possible.1 Model Development and Comparisons In MATPRO.3502 x10 5 T 2.91.6394 x10 2 T 4. A regression analysis of the reciprocal of the Meyer hardness values versus the log of temperature was used to obtain the analytical expression used in CMHARD. A comparison between the CMHARD correlation and the data from which it was derived is presented in Figure 3.91. The correlation used is given by Equation 3.).9. MH exp2. beginning at 2x109 MPa at room temperature and decreasing to 2x108 MPa at 875K. The minimum Meyer hardness number of Zircaloy cladding is 1. the Meyer hardness number is a measure of indentation hardness and is defined in conjunction with Meyer’s law. where L d n a = = = = L ad n load the diameter of impression at the surface of a specimen in a static ball test the Meyer work hardening coefficient a material constant (3.3. The hardness is presumed to continue its rapid rate of decrease at temperatures above 875K. which is presented in Equation 3.91) The Meyer hardness number (MH) is defined as 4L/d2. 3.
” Journal of Nuclear Materials. and R.D.2000 1800 1600 Meyer Hardness (MPa) 1400 1200 1000 800 600 400 200 0 200 400 600 800 1000 1200 1400 1600 1800 2000 2200 Temperature (K) Figure 3.. I. 57:246248. AECL1552.2 Applicability and Uncertainty The subroutine CMHARD is applicable for temperatures up to 875K.L... 3.3 References Ross. Ltd. 1975.P.36 . A. Atomic Energy of Canada. An estimate of the uncertainty has not been established for this correlation due to the limited data. Heat Transfer Coefficient Between UO2 and Zircaloy2. 3.M. Ontario.91 CMHARD correlation as a function of temperature with data from Peggs and Godin. Chalk River. Stoute.9. Godin.9. 3. 1962. and D. “The Yield StrengthHot Hardness Relationship of Zircaloy4. Peggs.
nitrogen (N).4 codes use similar correlations to determine the gas thermal conductivity.11) The constants A and B are fitting parameters. K AT B where K T = = thermal conductivity (W/mK) gas temperature (K) (4. hydrogen (H). The subroutine used to determine the material correlation in either the FRAPCON3/FRAPTRAN codes is given in parentheses in the heading of each subsection.4 and FRAPTRAN 1.12.4 use updated fitting parameters to better estimate gas thermal conductivity at higher temperatures.314x104 3 B 0. However. 4.097x103 5.7224 0. FRAPCON3.1 Gas Conductivity (GTHCON) The subroutine GTHCON calculates the gas thermal conductivity as a function of temperature and gas fraction for seven gases: helium (He).4 (FRAP) are given in Table 4. FRAPCON3.986x104 8.4. The correlations used for the gases of interest are all of the form presented in Equation 4.4 and FRAPTRAN 1.8616 0.6898 4.11 Gas He Ar Kr Xe H2 N2 Constants used in gas thermal conductivity correlations in MATPRO. 4. xenon (Xe). The values for A and B used in MATPRO for each gas are given in Table 4.11 and the values used in FRAPCON3.639x10 2. and water vapor (steam).351x105 1.1 Model Development and Comparisons The heat conductance of gasfilled gaps or pores is dependent on the thermal conductivity of the gas mixture when the dimensions of the gasfilled regions are large compared to the mean distance between gas molecule collisions (mean free path of the gas molecules). The MATPRO. and FRAPTRAN 1.7085 0.11. argon (Ar). A 2.1 .8363 0.1.4.247x105 4. This section presents data and correlations for the thermal conductivities of the gases of interest in fuel rod analysis.8785 0. Table 4.0 Gas Material Properties This section describes material property correlations for gap gases. krypton (Kr).
08 x10 7 * T 273 . is in pascals.009 * P 2 / T 2 / T 273 . the correlation presented in Equation 4.15 M M j M i 0.45 x10 5 * ( 2.15K.16) Kronecker delta = 1 for i = j.8408 0.4 codes.3 k P / T * 2. k 4.13) The thermal conductivity of gas mixtures is calculated with Equation 4.Table 4.6 x10 4 5.12 Gas He Ar Kr Xe H2 N2 Constants used in gas thermal conductivity correlations in FRAP. T.142 M j ij ij 1 2.092x104 1.16 (4.005 x10 14 * T 2 4.2 .45 9.7799 The MATPRO steam correlation is also used in FRAPCON3.44 x10 6 * T1. For temperatures less than or equal to 973. P.531x10 4.15) and ij n Mi = = = k 1 / 2 M 1 / 4 1 i i kj Mj ij 1/ 2 M 2 3 / 2 1 i M j 2 (4. A 2.2 2 3 (4.4/FRAPTRAN 1. For temperatures greater than 973.15 4.15 1.1668 x10 9 * P / T )1.424 x10 10 * T 6. the correlation presented in Equation 4. is in kelvin and pressure.14) where ij is given in Equation 4.12 is used.7006 0.984x104 3 B 0.15 (4.7146 0. k mix kixi n i x i 1 ij ij x j j1 n (4.41 i 2 M i M j and ij is given in Equation 4.14.8516 x10 8 9. Temperature.825x105 1.966x104 9.51x10 11 * T 273 .15K.87 x10 5 * T 273 .13 is used to determine the thermal conductivity of steam.6748 0.12) 1. 0 otherwise (unitless) number of components in mixture (unitless) molecular weight of component i (kg) 4.15 ] [17.7334 0.349x103 2.
10 0.30 0.3 .40 0.70 0.xi ki = = mole fraction of component i (unitless) thermal conductivity of the component i (W/mK) Modeltodata comparisons for the gas conductivity models are presented in Figures 4.60 0. (1977) FRAP Jain and Saxena (1975) 2000 2500 3000 Figure 4. K MATPRO Faubert and Springer (1973) Jody et al.11 through 4. Thermal Conductivity.11 Modeltodata comparison for He thermal conductivity.80 0.17.00 0 500 1000 1500 Temperature. 4.50 0.20 0. W/mK 0.
10 0.00 0 500 1000 1500 Temperature. 4.02 0.0. K MATPRO Timrot and Umanski (1966) Faubert and Springer (1972) Stefanov et al.03 0.08 0. W/mK 0.01 0.04 0.4 .09 Thermal Conductivity.06 0.05 0. (1976) FRAP Saxena and Saxena (1968) Springer and Wingeier (1973) 2000 2500 3000 Figure 4.12 Modeltodata comparison for Ar thermal conductivity.07 0.
02 0. W/mK Figure 4. K MATPRO Saxena and Saxena (1969) Vargaftik and Yakush (1971) FRAP Stefanov et al.06 0.03 0.0.5 .01 0.04 0.13 Modeltodata comparison for Kr thermal conductivity.05 0.00 0 500 1000 1500 Temperature.07 0. (1976) 2000 2500 3000 Thermal Conductivity. 4.
20 Thermal Conductivity. 4.80 0.03 0.04 Thermal Conductivity.01 0.40 0.05 0. K MATPRO Stefanov (1976) Saxena and Saxena (1969) FRAP Springer and Wingeier (1973) Vargaftik and Yakush (1971) 2000 2500 3000 Figure 4.00 0 500 1000 1500 Temperature. K 2000 2500 3000 MATPRO FRAP Saxena and Saxena (1970) Figure 4.00 0.20 0.15 Modeltodata comparison for H2 thermal conductivity.02 0.04 0.00 0 Modeltodata comparison for Xe thermal conductivity.60 0.0. W/mK 1. 500 1000 1500 Temperature.01 0. W/mK 0.03 0.14 1.02 0.6 .
04 0.16 0. K MATPRO Vargaftik and Zimina (1964) Chen and Saxena (1973) FRAP Faubert and Springer (1972) 2000 2500 3000 Figure 4.02 0.1.10 0.04 0.17 Modeltodata comparison for steam thermal conductivity.08 0. MATPRO/FRAP Data 700 800 Temperature (K) 900 1000 Figure 4.08 0.06 0.00 0 500 1000 1500 Temperature. The standard error for each correlation is given in Table 4.12 0.16 Thermal Conductivity (W/mK) 0. 4.14 0.13.12 0.0. 4.7 .2 Applicability and Uncertainty The gas thermal conductivity correlations for used in GTHCON are applicable over the temperature range for which data was collected.1 0.14 0. W/mK 0.02 0 600 Modeltodata comparison for N2 thermal conductivity.16 Thermal Conductivity.06 0.
.” Journal of Chemical Physics.” Journal of Physics A. S. Argon. L.” Chemical Physics Letters. Jody. Saxena. Vargaftik. Aziz.97x104 1. Stefanov. 2(1):4446. and L.. F...86x104 5. “Thermal Conductivity of Krypton and Xenon in the Temperature Range 3501500K.. and G. Wingeier.” High Temperature.S.. “Thermal Conductivity of Neon.C.J. 22:5358.S.8 . “Thermal Conductivity of Nitrogen at High Temperatures. V. 14(1):4857.Table 4. 21(3):11561161.C. 36(4):489491. 59(5):27472750.” Chemical Physics Letters.C. “Measurement of the Thermal Conductivity of Argon.S.66x104 8. and R. and G. “Thermal Conductivity of Helium: A Problem for the Repulsive Wall of the Interatomic Potential.B. Saxena. 1968b. Saxena. “Transport Properties of Helium in the Temperature Range 4002300K.” Journal of Engineering Physics. Krypton.” Journal of Chemical Physics.S. and V.” Journal of Chemical Physics.B. 1975.K.13 Gas He Ar Kr Xe H2 N2 Steam Standard deviation of gas thermal conductivity correlations.1. 1973.” High Temperature. Springer. and D. 1977.” Journal of Chemical Physics. 3:285287. D. 3(3):309320.. 51(8):33613368.P. Springer. Zimina.99x103 9.L. 4.C. N. Oliver. and Nitrogen in the Range 8002000°K.34x104 1.67x102 1. 58(10):40804083. Saxena. G. and S. 1964. “Measurement of the Thermal Conductivity of Argon Using HotWire Type Thermal Diffusion Columns.C. Nain. Faubert. Springer. Standard Deviation 8. and N. and S. Jain. 57(6):23332340. S. F. “Measurement of the Thermal Conductivities of Neon. “Thermal Conductivity Data for Hydrogen and Deuterium in the Range 1001100°C.W.K. 1976. “Measurement of the ThermalConductivity Coefficient of Gases and Vapors up to 2500°K. Umanskii. V. B.69x103 4. and E.3 References Faubert. Krypton. and Xenon at High Temperatures. 1972.. and S.S.Kh. Timrot. Saxena.. P. 1971. Zarkova. 1966. 1970..” High Temperature. B. “Thermal Conductivity of Hydrogen and Argon. Saxena. Yakush. and A. V.M. N.. 1969.A.K.M.V. 2(6):782790. “Measurement of the Thermal Conductivity of Helium up to 2100°K by the Column Method. Saxena.” Chemical Physics.C.. and Xenon over a Wide Range of Temperatures. 1973. Vargaftik. Saxena.
4.” High Temperature Science.. 1973.Chen. and S. Saxena.P.9 .C. 5(3):206233.H. “Experimental Determination of Thermal Conductivity of Nitrogen in the Temperature Range 1002200°C. S.
Appendix A Subroutine Source Codes .
G. 22 (1967) p 332 c phypro was coded by v.leibowitz et al.4 if (comp. and zirconium and zircaloy alpha to beta c transition temperatures.w.nuc.comp.brassfield et al gemp482. Geelhood in June 2004 based on c S. new york." ORNL/TM2000/351.05.0) go to 100 ftmelt = 3113.7 kcal/mole from l. olsen in feb. Popov. + ctrane.ctranz. c beginning and end of alphabeta transus are from data in c figure iii.ctranb.l.f.Appendix A Subroutine Source Codes A.chefus.J. cladding melting point of 2098 k from m.oxygen concentration c of as received cladding (kg oxygen/kg zircaloy) c the uo2 and zr cladding melting points and heats of fusion are c taken from the following references.ctmelt. (u.gt. j.l.pu)o2.deloxy c ftmelt = output uo2 or mixed oxide fuel melting points (k) c fhefus = output uo2 or mixed oxide fuel heat of fusion (j/kg) c ctmelt = output zr clad melting point (k) c chefus = output zr clad heat of fusion (j/kg). 1955 c mixed oxide melting point was obtained from lyon et al. picklesimer c private communication. gemp482. + locidx / 2 / sldus(c) = 2840. c mat.03.0*fbu/10000.h' data on / 1 /.s.c.bu.1 Uranium Dioxide/Mixed Oxide Properties A. c ctranb = output start of zr4 alphabeta transus temperature (k) c ctrane = output end of zr4 alphabeta transus temperature (k) c ctranz = output zr isothermal alphabeta transus temperature (k) c fdelta = output liquidsolid coexistence temperature range (k) c bu = input burnup (mws/kgu) c comp = input puo2 content (wt%) c deloxy = input oxygen concentration .0 A. nuc.4: PHYPRP *deck phyprp subroutine phyprp c c phyprp returns uo2. "Thermophysical Properties of MOX and UO2 Fuels c including the Effects of Irradiation. + off / 2 /.41395*c+7.468390e3*c*c liqdus(c) = 2840. uo2 heat of fusion c of 17.mat. and zircaloy melting points c and heats of fusion. include 'lacmdl. common / phypro / ftmelt.fhefus. cladding heat of fusion of 4.kerze "the metallurgy of zirconium" c mcgrawhill book co..1 Fuel Melting Temperature (PHYPRP) FRAPCON3. 1977 c modified by d. 39 p 115 c (1971).155.21860*c1..0. baston in may 1974 c modified by c.fdelta. uo2 fuel melting point of c 3113k from h. j.1 .1. et al.9 kcal/mol c from brassfield et al.2K/GWd/tHM c to 0. burnham in nov.. 1975 c modified by b.33 of anl7649 c isothermal zirconium alphabeta transus temperature is 1135 k c taken from b.448518e2*c*c fbu = bu/86. hagrman in june 1979 c burnup dependance for fuel melting temperature changed from 3.5 K/GWd/tHM by K.lustman & f.
*deloxy+0. c c beginning and end of alphabeta transus are from data in c figure iii. cladding heat of fusion of 4.914e+05) if (wfox.1242807)**2+3. 2 routine modified by PNNL for use in FRAPT c c phyprp returns uo2. Rev.c.8281e+04*(wfox0. j.oxygen concentration c of as received cladding (kg oxygen/kg zircaloy) c c the uo2 and zr cladding melting points and heats of fusion are c taken from the following references.100 110 120 130 fdelta = 1.4.15 return end FRAPTRAN 1.0e10 go to 110 c1 = comp ftmelt = sldus(c1)+273.289e+03+wfox*7.*deloxy+0.lt.15 chefus = 22.l.9 kcal/mol c from brassfield et al.lt.4: PHYPRP *deck phyprp c subroutine phyprp (icm) c implicit real (ah.0.0012 ctranb = 1094.33 of anl7649 c isothermal zirconium alphabeta transus temperature is 1135 k A.7413 continue ctranz = 1135.5e+04 wfox = deloxy+0.0 fdelta = liqdus(c1)sldus(c1)5. cladding melting point of 2098 k from m.4+3.12)*491.0*fbu/10000.2 .1417) if (deloxy.h' c c ftmelt = output uo2 or mixed oxide fuel melting points (k) c fhefus = output uo2 or mixed oxide fuel heat of fusion (j/kg) c ctmelt = output zr clad melting point (k) c chefus = output zr clad heat of fusion (j/kg). and zircaloy melting points c and heats of fusion.+wfox*(1.mat.025) ctrane = 392.oz) c real liqdus c c This is a MATPRO11. 39 p 115 c (1971).155. picklesimer c private communication.4e+4 ctmelt = 2098.leibowitz et al. c include 'phypro. uo2 heat of fusion c of 17.157+1081. (u. c ctranb = output start of zr4 alphabeta transus temperature (k) c ctrane = output end of zr4 alphabeta transus temperature (k) c ctranz = output zr isothermal alphabeta transus temperature (k) c fdelta = output liquidsolid coexistence temperature range (k) c c bumtp = input burnup (mws/kgu) c compmt = input puo2 content (wt%) c deloxy = input oxygen concentration .nuc.46*((100.7308937e03) go to 130 ctrane = (100.pu)o2.brassfield et al gemp482.7 kcal/mole from l.0 fhefus = 27.0*fbu/10000. uo2 fuel melting point of c 3113k from h. gemp482. and zirconium and zircaloy alpha to beta c transition temperatures.025) go to 120 ctranb = 1556.
phypro was coded by v.gt.025d0) 30 continue ctrane = 392.12d0)*491.15d0 .448518d2*c*c fbu = bumtp/86.914d+05) if(wfox . January 1997.400)icm c 400 continue generic Zry properties ctmelt = 2098. 4.4d0 + 3.0d10 go to 20 c 10 continue c1 = compmt ftmelt = sldus(c1) + 273. 1975 modified by b.0d0 .w.5.0d+4 50 continue c c 5/7/03: ctranb.0d0 c 20 continue fhefus = 27.0d0) go to 10 c ftmelt = 3113.289d+03 + wfox*7.0d0 fdelta = 1. et al.400.0d0*fbu/10000.G.0d0*fbu/10000.4d0 if (compmt . olsen in feb.5d+04 go to 50 600 continue c Zr1%Nb properties from RRCKI ctmelt = 2133.1242807d0)**2 + 3.400.5.d0*deloxy + 0. 0.3.600. nuc.. 22 (1967) p 332 was obtained from lyon et al. ctrane.1. burnham in nov.f.025d0) go to 30 ctranb = 1556. to clean up coding and delete sensitivity uncertainty analysis coding burnup dependance for fuel melting temperature changed from 3.5. ctranz phase transition temperatures c not currently used in FRAPTRAN.46d0*((100. j.0d0*fbu/10000.8281d+04*(wfox .7308937d03) go to 40 ctrane = (100.5. baston in may 1974 modified by c.15d0 c return c A.d0 chefus = 21.lustman & f.Geelhood in June 2004 based on S.0." ORNL/TM200/351 sldus(c) = 2840. RRCKI does have different c transition temperatures for Zr1%Nb wfox = deloxy + 0.0d0 ..2 K/GWd/tHM to 0.5 K/GWd/tHM by K.1417d0) if(deloxy .kerze "the metallurgy of zirconium" mcgrawhill book co. 1955 mixed oxide melting point mat.157d0 + 1081.15d0 .lt. new york.468390d3*c*c liqdus(c) = 2840.0d0 fdelta = liqdus(c1) .0012d0 ctranb = 1094..sldus(c1).l.21860d0*c .d0 + wfox*(1.J.400.c c c c c c c c c c c c c c c c c c taken from b.lt.600.400.400.41395d0*c + 7. hagrman in june 1979 Modified by PNNL.15d0 chefus = 22. Popov.3 .d0*deloxy + 0.7413d0 40 continue ctranz = 1135.4d+4 c go to (400.400.400. "Thermophysical Properties of MOX and UO2 Fuels including the Effects of Irradiation.s. 0. 1977 modified by d.
puo2.860e07.044e02 .16 (1972).comp.95e04.a.t. + 348.fotmtl.4 .2 Fuel Specific Heat Capacity (FCP) and Fuel Enthalpy (FENTHL) FRAPCON3.0 .marcon. 3.c3.1.bu.pu)o2 fuels as a function of temperature.gadoln) c c the function fcp is used to calculate the specific heat capacity c of uo2. 0. common / phypro / ftmelt.th.edu / 296.ctranb. c quantities contained in it used in this subcode are.fuel is all molten c facmot = 0. and oxygentoc metal ratio.c3pu. data c1u .7 (1970).savage.577e05 / data c1pu. puo2 content. nuclear technology. and comp must be c input.0 .0 / cp(c1. c the effect of oxygentometal ratio was estimated from c equations published by.fotmtl) = c1*(th**2)*exp(th/t)/((t**2)*((exp(t +h/t)1. not as part of the c inel code.reymann in may 1978.4 .fdelta. c last modified by g. c the specific heat capacity of molten uo2 was taken from c (4) l.baston in march 1974. 8. and (u. + 571. c fraction of fuel which is molten.0 . 1.ed.c2gd.thu .c2.)*(c3*ed/(8. c enter 2.c A. c Modification for gadolinia additions were made by DD Lanning c in 1996.thpu. + ctrane.clifton.c2u .kruger and h.0.7 .chefus.fuel is all solid c fotmtl = input oxygen to metal ratio (unitless) c fotmtl = (atoms oxygen)/(atoms metal) if not known.deloxy c the phypro common block is set by call to the subroutine c phypro which is part of the materials properties package c used at the inel.%) c when the function fcp is used alone. 1.39 (1970) c fcp was originally coded by v. c ftmelt = fuel melting temperature (K) c fdelta = liquidsolid coexistence temperature (K) c comp = puo2 content (wt.745e07. c (1) j.kerrisk and d. journal of nuclear materials. revue internationale des hautes c temperatures et des refractaires. c (3) c.c3gd.0 .ctranz. 2.967e05 / data c1gd.leibowitz et al.facmot. 4.)**2))+c2*t+(fotmtl/2.86 .0 c gadoln = weight fraction of gadolinia in the fuel c ufcp = estimated standard error for uo2 (j/(kg*K)) c (not currently returned) c pufcp = estimated standard error for puo2 (j/(kg*K)) c (not currently returned) c the equations used in this function are based on data from. values for ftmelt. fdelta.f.43e02.45 c (1968).314*t)) t = ftemp A. 3.285. journal of chemical physics. c (2) o. + 535.ctmelt.314*(t**2)))*exp(ed/ +(8.affortit and j.thgd. 0.fhefus.edpu/ 347.edgd/ 315.4: FCP *deck fcp function fcp (ftemp.c2pu.c3u . c fcp = output fuel specific heat capacity (j/(kg*K)) c ftemp = input fuel meshpoint temperature (K) c facmot = input fuel fraction molten (unitless) c facmot = 1.
c2pu.c2gd.f.(tm0.savage.fuel is all solid c fotmtl = input oxygen to metal ratio (unitless) c fotmtl = (atoms oxygen)/(atoms metal) if not known.thgd.edu.t.5 .r = facmot tm = ftmelt fcomp = comp/100. c fraction of fuel which is molten.0 .affortit and j.clifton.lt.leibowitz et al.0gadoln) + gadoln*cp(c1gd.pu)o2 fuels as a function of temperature. 2 routine modified by PNNL for use in FRAPT c c the function fcp is used to calculate the specific heat capacity c of uo2.c3u.fuel is all molten c facmot = 0. puo2 content.frden) implicit real (ah.gt.6 110 continue return end FRAPTRAN 1. c (1) j.a. to clean up coding and delete lic A. puo2.fotmtl)*(1.oz) c c This is a MATPRO11. revue internationale des hautes c temperatures et des refractaires.kruger and h. c the specific heat capacity of molten uo2 was taken from c (4) l.fotmtl.thu. c last modified by g.0 fcpmol = 503. nuclear technology. c (2) o.t.kerrisk and d.c2u.16 (1972).fotmtl)*fcomp fcp = fcp*(1.4: FCP *deck fcp function fcp (ftemp.(tm+fdelta)) go to 100 fcp = cp(c1u.reymann in may 1978.0 pufcp = 5. c (3) c. c enter 2.0 c ufcp = estimated standard error for uo2 (j/(kg*k)) c (not currently returned) c pufcp = estimated standard error for puo2 (j/(kg*k)) c (not currently returned) c c the equations used in this function are based on data from. January 1997.edpu.t. and (u. journal of chemical physics.0r)*fcp+r*fcpmol go to 110 100 fcp = fcpmol ufcp = 2.0 c fcpmol = specific heat capacity of molten fuel (j/(kg*K)) if (t.0 .facmot. journal of nuclear materials.7 (1970).marcon.45 c (1968).thpu.39 (1970) c c fcp was originally coded by v. c c fcp = output fuel specific heat capacity (j/(kg*k)) c c ftemp = input fuel meshpoint temperature (k) c facmot = input fuel fraction molten (unitless) c facmot = 1. + fotmtl) if (t.edgd. and oxygentoc metal ratio. c Modified by PNNL.0fcomp)+cp(c1pu. c c the effect of oxygentometal ratio was estimated from c equations published by.c3gd.1)) go to 110 fcp = (1.baston in march 1974. Rev.c3 +pu.
0d0 . fdelta.c3pu.6 .4d0 .fotmtl)*(1.gadoln) c implicit real * 8 (ah. values for ftmelt.gt.fotmtl) = c1*(th**2)*exp(th/t)/((t**2)*((exp(th/t)1. ftmelt = fuel melting temperature (k) fdelta = liquidsolid coexistence temperature (k) compmt = puo2 content (wt. (tm+fdelta)) go to 50 fcp & c if (t .577d05 / 3.edu .thpu.0d0 specific heat capacity of molten fuel (j/(kg*k)) cp(c1.c2pu.0d0 503.c2.thpu. data c1pu.edu / 296.0d0 pufcp = 5. 3. 1.t.1d0)) go to 100 fcp = (1.fcomp.314d0*(t**2)))*exp(ed/(8.c2u . & 571. 535.0d0 .c3pu.c3.7d0 .0d0 .th. not as part of the inel code.c2u .fotmtl.thu .r)*fcp + r*fcpmol go to 100 c 50 continue fcp = fcpmol ufcp = 2. data c1u .fotmtl)*fcomp c 100 c FRAPCON3.facmot.t. c & & c t r tm fcomp fcpmol fcpmol = = = = = = ftemp facmot ftmelt compmt/100.d0)*(c3*ed/(8.860d07.967d05 / c c c if (t .745d07.c3u .edpu. and compmt must be input.c2pu.oz) c ******************************************************************** c this function is called by subroutine energy and computes the c enthalpy of fuel at a point relative to zero degrees absolute c temperature A.c3u .h' the phypro common block is set by call to the subroutine phypro which is part of the materials properties package used at the inel.285d0.lt.ed.ftmelt.t. 8. 1.fcomp) + cp(c1pu.95d04.thu .6d0 continue return c end = cp(c1u .4: FENTHL *deck fenthl c function fenthl (temp. quantities contained in it used in this subcode are.%) when the function fcp is used alone. (tm0.fhefus.314d0*t)) 2.c c c c c c c c c c c c c c c & analysis and sensitivity uncertainty coding include 'phypro.43d02.edpu/ 347.d0)**2)) + c2*t + (fotmtl/2.
thpu.c2.(ftmelt+2.fotmtl.0 . facmot. c3u.fraction of molten fuel fcomp .tx.285. 0./(exp(th/t)1.otm.c2gd.oz) c c c ********************************************************************* c c this function is called by subroutine energy and computes the c enthalpy of fuel at a point relative to zero degrees absolute c temperature c c ********************************************************************* c input arguments c ********************************************************************* c c facmot .c2pu. thu.860e07.c3pu)*fcomp fenthl = fenthl*(1gadoln)+ &gadoln*cpdt(c1gd.95e04. fcomp.044e2. 8.7 .)) go to 100 fenthl = fenthl+(tempftmelt)*fcpmol 100 continue return end FRAPTRAN 1. 3. fotmtl.edgd.7 . 1.th. 3.e0 tx = temp if (tx.c2pu.thgd. + 571. 2.967e05 / data c1gd.thu.e0 / c the following data statements contain constants from matpro11 fcp data c1u.4: FENTHL *deck fenthl c function fenthl (temp.e0) go to 100 fenthl = fenthl+fhefus*facmot if (temp.c2gd.86 .0 . 4.c3gd) if (temp.weight fraction of gadolinia in the fuel temp .le.edpu.4 . + 348.edgd / 315.fuel melting temperature (K) gadoln . 0.edu.le.fraction of molten fuel A.local fuel enthalpy relative to zero degreesK (j/kg) ******************************************************************** fcpmol = specific heat capacity of molten fuel ( j/(kg*K) ) data fcpmol / 503.43e02.t.ftmelt2. edu / 296.fotmtl. 1. ftmelt.))+c2*t*t/2.c3u)*(1.gt.745e07.tx.c3pu.local temperature (K) ******************************************************************** output ******************************************************************** fenthl .c3) = c1*th*(1.thpu.c3gd.fuel oxygen to metal ratio ftmelt . + 535.ed. c2u.577e05 / data c1pu. fhefus) c implicit real (ah.fotmtl.0 .thgd.0 / c the following equation is the integral of the fuel specific heat c with respect to temperature cpdt(c1.e0fcomp) fenthl = fenthl+cpdt(c1pu.c c c c c c c c c c c c c c c c ******************************************************************** input arguments ******************************************************************** facmot .314e0))/2.puo2 fraction of the fuel fhefus .tx.c2u.edpu / 347.ftmelt) tx = ftmelt fenthl = cpdt(c1u.e0 ++c3*otm*exp(ed/(t*8.heat of fussion of the fuel (j/kg) fotmtl .
d0)) & + c2 * t * t / 2.c2u .t. ftmelt) tx = ftmelt c fenthl = cpdt(c1u & . 571.d0))go to 100 c c fenthl = fenthl + (temp .c2.c c c c c c c c c c c c c c c c c c c fcomp fhefus fotmtl ftmelt temp  puo2 fraction of the fuel heat of fussion of the fuel (j/kg) fuel oxygen to metal ratio fuel melting temperature (K) local temperature (K) ********************************************************************* output ********************************************************************* fenthl .2.3 Fuel Thermal Conductivity (FTHCON) FRAPCON3.c3pu) * fcomp c if( temp .c2pu & .le.d0 / (exp(th/t)1.4d0 . c2u.fotmtl .285d0.thu .c3u .gt.314d0) )/2. thu. 1.(ftmelt+2.d0 & + c3 * otm * exp(ed /(t * 8. ftmelt . 535.745d07.edpu / 347.fcomp ) A.d0 / the following data statements contain constants from matpro11 fcp data & data & c1u.0d0 . 8.967d05 / c c c c the following equation is the integral of the fuel specific heat with respect to temperature cpdt(c1.1.8 .th.4: FTHCON *deck fthcon A.43d02.95d04.d0 ) go to 100 c c c fenthl = fenthl + fhefus * facmot c if(temp.d0 c tx = temp if(tx .c2pu.ftmelt) * fcpmol 100 continue c return end .le.fotmtl .d0 ) fenthl = fenthl + cpdt(c1pu .860d07.thpu . c3u. 1.ed.thpu.7d0 . c1pu.tx c if( fcomp .c3) = & c1*th * ( 1.c3pu. 0. edu / 296.tx .gt.otm.d0 .edpu . 2.577d05 / 3. 3.edu ) * ( 1.local fuel enthalpy relative to zero degreesK (j/kg) ********************************************************************* fcpmol = specific heat capacity of molten fuel ( j/(kg*K) ) data fcpmol / 503.
burnup + . proceedings pp.fotmtl.sigh2 common / phypro / ftmelt. Japan.fraden. density.ctmelt. Proceedings of the 2000 International Topical Meeting on LWR Fuel Performance burnup con ftemp fraden current local burnup (MWd/MTU) output fuel thermal conductivity (W/(m*K)) current fuel ring temperature (K) input fuel density (ratio of actual density to theoretical density) fotmtl = input oxygen to metal ratio of fuel (atoms oxygen/ atoms metal) gadoln = input weight fraction of gadolinia in the fuel the following inputs are by common block comp = input puo2 content of fuel (percent puo2 in total fuel weight) bu = input burnup (mws/kgu) emflag(12) = input switch for evaluation model. the matpro model for fuel thermal conductivity is replaced by the subcode emfton common /uncert/sigftc.deloxy = = = = c include 'lacmdl. composition and burnup. + locidx / 12 / A.. et al. at the May 1997 ANS Topical Meeting on Light Water Reactor Fuel performance in Portland. Applies to UO2.sigswell . OR: (Ohira. 541549. "Thermal Conductivity Measurements of High Burnup UO2 Pellet and a Benchmark Calculation of Fuel Center Temperature".sigcor . and N.9 . UO2 Fuel (IMOX = 0) The equation used in this subroutine is that proposed by staff at NFI.bu.Wiesnack. sigftex. 143158 2000) and the burnup degradation conatined in a modified version of the NFI fuel thermal conductivity model Option number 2 (IMOX = 2) The MOX fuel thermal conductivity formulation is based on the OECD Halden Reactor Project report "Thermal Performance of of High Burnup Fuel Inpile Temperature Data and Analysis" W.subroutine fthcon (ftemp. + ctrane.siggro .Duriez.fdelta. if this variable is equal to 1.imox) c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c fthcon calculates the fuel thermal conductivity and its derivative with respect to temperature as a function of temperature.fhefus. J.ctranz. Tverberg. sigfgr & . derived from diffusivity measurements on unirradiated fuel pellets (C.ctranb.Itagaki. T.gadoln.comp. K.con.Nuclear Materials 277. MOX: Option number 1 (IMOX = 1) The 100% dense solid MOX fuel thermal conductivity formulation is based on a combination of the Duriez stoichiometrydependent correlation.sigcreep . 1997.0. + off / 2 /.chefus.h' data on / 1 /.
5*(1.10 .00187*bug+1.0789*fraden/(1.0 if(imox.0+0.tc) buguo2=bug*0.00187*bug+1.0fraden)) c fm is multiplied by 1.0789*fraden/(1.867.0/t)) rphonon= 1.0132*exp(0.0/(0.92/(0.15*x)*1.0 + 0.5*(1. not MaxwellEuken c as proposed by Duriez et al.eq.0452+0.04*bug))*0.0+396.50e9/t**2)*exp(16361/t) base = rphonon + elect & c c c c c c c c fm is the Lucuta porosity correction factor(applied to 100% TD fuel) fm = fraden/(1.0*exp(6380.0/(ax + cx*t + 0. so multiply by 1.00.000246*t + 1.eq.038*bug**0.1599*gadoln+ & 2.5*(1.038*bug**0.00.0fraden)) base=0.035 cx=(2.0.1148+0.28*h) elect = (3. Topical mtg.1599*gadoln + (1.00333*buguo2)*tco)+ & 0. ANS LWR Fuel perf.0789 to account for 95% TD c Porosity correction is Lucuta correction.28*h) elect = (1.0+0.15 tco=min(1650. x = 2.475e4*(1.0*0.0fraden)) NFI base equation is for 95% TD fuel. then multiply by fm con = base*fm*1.50e9/t**2)*exp(13520/t) base = rphonon + elect con = base*fm c else if(imox.0fotmtl ax=2.1) then c c c c c Using the Duriez/NFI Mod correlation combination base term for MOX where X = deviation from stoichiometry (2O/M) fm = 1.0*exp(6380. t = ftemp c c c c Burnup in GWd/MTU bug = burnup/1000.0) then c c c c c NFI formula (Ohira & Itagaki.00188*tc) A.eq.079 to raise to 100% TD fuel conductivity.0/t)) rphonon = 1.079 else if(imox.0e4 c h = 1/(1.9*exp(0.004*buguo2+1.85*x+0.8815 fm = 1.00.1599*gadoln &+ (1. h = 1/(1.04*bug))*0.2) then c c Using the Halden correlation c tc=t273.9*exp(0. 1997) MODIFIED in January 2002 to raise lowburnup thermal conductivity at low temperature and lower thermal conductivity at very high temp.0+396.c c c find constants frpu = comp/100.
01800.088) if(imox." proceedings pg. c reference: Ohira and Itakaki.2 then stop the calculations else stop 'fthcon .eq. fraden. c c This is the same formulation used in FRAPCON3. compmt) c implicit real (ah.2*(1.ne. 1997.IMOX not within bounds' end if c c find uncertainty if(sigftc.2 (February 2003).lt. c c burnup = current local burnup (MWd/MTU) c con = output fue thermal conductivity (W/(m*K)) A.0." in Proceedings of c ANS Annual Meeting.on) call emfton (ftemp.07 ucon=frac*con endif endif if (emflag(locidx). "Thermal Conductivity c Measurements of High Burnup UO2 Pellet and a Benchmark c Calculation of Fuel Center Temperature.0.con) return end FRAPTRAN 1. Hollywood.0)/(3100. dkdt.0+sigftc*0. composition and burnup.ftmelt.0) then ucon = 0.0. "Revised UO2 Thermal c Conductivity for FRAPCON3 NRC Fuel Performance Codes.0sigftc*0.lt.4: FTHCON *deck fthcon c subroutine fthcon (ftemp. & gadoln. burnup. con. Japan.088) if(sigftc. c The PNNL reference is: Lanning and Beyer. density. c c The NFI formula was modified in January 2002 by PNNL to raise c lowburnup thermal conductivity at low temperature and lower thermal c conductivity at very high temperature.0 endif else if(t.eq.200.07)+0.fraden. fotmtl.0) con = con/(1.0)*(0.gt. Oregon. 2002. oz) c c FTHCON calculates the UO2 fuel thermal conductivity c as a function of temperature. June 913. Florida.ftmelt) then ucon = 0. 541549.1.1800.con=base*fm c c c If IMOX. at the May 1997 ANS Topical Meeting on c Light Water Reactor Fuel Performance in Portland.le.0fotmtl)*10.07*con else frac=(t1800. c minus the formulations for MOX fuel.) else ucon = con/2.11 . c c The equation used in this subroutine is that proposed by c staff at NFI.0) then if(t.0) con = con*(1.0+abs(2. 2002.
0*exp(6380.50e9/ftemp**2)*exp(13520/ftemp) base = rphonon + elect con = base*fm c 200 continue return end A.9*exp(0. ANS LWR Fuel perf.0/ftemp)) phonon term rphonon = 1.28)*h) electronic term elect = (3.c c c c c c c c ftemp = current fuel ring temperature (K) fraden = input fuel density (ratio of actual to theoretical) fotmtl = input oxygen to metal ratio of fuel (atoms oxygen/atoms metal) gadoln = input weight fraction of gadolinia in the fuel compmt = input weight % of Pu in Pu and U convert burnup to GWd/MTU bug = burnup/1.9*exp(0.0*exp(6380.0) go to 100 UO2 Model NFI formula (Ohira & Itagaki. so multiply by 1. 1.4 Fuel Emissivity (FEMISS) A.00187*bug & + (1.d10) bug = 0.d3 if(bug .0789*fraden/(1.000246*ftemp + 0.038*bug**0.0 + 0.lt.28*h) elect = (1.00187*bug+1.038*bug**0.04*bug))*0. Topical mtg.0789 to account for 95% TD c Porosity correction is Lucuta correction. calculate terms temperature dependence of annealing of irradiation effects h = 1.0/(1.001 use modified NFI model for UO2 and Duriez/NFI Mod model for MOX if(compmt.1. x = 2.0fraden)) NFI base equation is for 95% TD fuel.079 go to 200 c c c c c c c c c c c c c c c c c MOX Model c 100 continue c Using the Duriez/NFI Mod correlation combination c c base term for MOX c where X = deviation from stoichiometry (2O/M) fm = 1. then multiply by fm con = base*fm*1.0 .0 + 396.00.1599*gadoln &+ (1.0+0.04520 + 0.0+396.50d9/ftemp**2)*exp(16361. not MaxwellEuken c as proposed by Duriez et al.0/(0.04*bug))*(0.5*(1.035 cx=(2.gt.12 .0e4 c h = 1/(1.867. 1997) MODIFIED in January 2002 to raise lowburnup thermal conductivity at low temperature and lower thermal conductivity at very high temp.0.0.5*(1./ftemp) base is conductivity before apply porosity correction base = rphonon + elect apply lucuta porosity correction factor (applied to 100% TD fuel) fm = fraden/(1.0/(ax + cx*ftemp + 0.0fotmtl ax=2.0fraden)) c fm is multiplied by 1.15*x)*1.0/ftemp)) rphonon = 1.079 to raise to 100% TD fuel conductivity.85*x+0.
oz) c c This is a MATPRO11. (1969) c cabannes.5263e05*ftemp c the following calculations provide the upper and lower bounds. journ. Rev. femiss = 0. (1969) c cabannes. 52.5263d05 * ftemp c c the following calculations provide the upper and lower bounds. ceram. et at. paris. soc.d0 + 0. c femisu is the upper bound femisu = femiss*(1. ser.06796) 100 continue return end FRAPTRAN 1. c Modified by PNNL.d0 .. c femisu is the upper bound femisu = femiss * (1. 52. sci. mason in october 1978. c.13 .06796d0) c femisl is the lower bound femisl = femiss * (1. c vol. amer.4: FEMISS *deck femiss function femiss (ftemp) c c femiss calculates fuel emissivity as a function of temperature. c the upper and lower bounds are not an output unless the user c desires to modify the subcode appropriately.FRAPCON3. e. c ftemp = input fuel temperature (k) c femiss = output fuel emissivity (unitless) c data used to develop the model are c held and wilder. c vol. acad. c. e. paris. 2 routine modified by PNNL for use in FRAPT c c femiss calculates fuel emissivity as a function of temperature. mason in october 1978. r. journ. acad. r. ceram.4: FEMISS *deck femiss function femiss (ftemp) implicit real (ah. ser.. c the upper and lower bounds are not an output unless the user c desires to modify the subcode appropriately. amer.78557+1. et at. to clean up coding and delete lic c analysis and sensitivity uncertainty coding c femiss = 0.06796d0) c return c A.0. b c (1967) c femiss was coded by r.78557d0 + 1. sci.06796) c femisl is the lower bound femisl = femiss*(1. c c ftemp = input fuel temperature (k) c femiss = output fuel emissivity (unitless) c c data used to develop the model are c held and wilder. soc. January 1997. b c (1967) c c femiss was coded by r.0.+0.
sigfgr & .burdock and h.fuel is all solid c the correlations used in this subroutine are based on data from.1 c (1955) pp 357370.4: FTHEXP *deck fthexp function fthexp (ftemp.a.russel. c (5) j. journal of the american ceramic society.momin.comp.parker.30 (1958) pp 15241526.a. Luscher 101707 c A. anl5053 (1956).pu)02 c as a function of temperature.siggro .sigswell .h. c (4) j. c (10)m.ctranz.christensen.6 (1963).facmot) c c the subprogram fthexp is used to calculate the strain of the c fuel caused by thermal expansion.hoch and a. c ftemp = input fuel temperature (k) c facmot = input fuel fraction which is molten (unitless) c facmot = 1.e. c (7) m.fhefus. and fraction of c the fuel which is molten. sigftex.ctranb.lambertson and j. c fthexp was originally coded by v.h. or (u. journal of the american c ceramic society. c 46 (1963) pp 607608. and fdelta must c be input.j.5 Fuel Thermal Expansion (FTHEXP) FRAPCON3.end A.sigh2 common / phypro / ftmelt. c (3) m.reymann in july 1978 common /uncert/sigftc.fuel is all molten c facmot = 0. c thermal expansion strain is calculated for uo2. 17 (1973) pp 147152.handwerk. journal of inorganic and nuclear chemistry.p. c (9) w. journal of nuclear materials.deloxy c the phypro common block is set by call to the subroutine c phypro which is part of the materials properties package c used at the inel.bu.a.0 . not as part of c the inel code. grenoble.elliott. c (8) c.G.sigcor . puo2 content. 1975 c last modified by g. c france (1960) pp 397410.fdelta. ftmelt. c fthexp = output fuel strain due to thermal expansion (unitless) c ufthex = estimated standard error of fthexp (per cent) c (not currently returned).d. high temperatureshigh pressures.o.brett and l. c3u.14 .baston in march 1974 c modified by c.1 c (1969) pp 401407.39 (1956) pp 181187. nuclear technology.sigcreep .1.0 .gronvold. c comp = puo2 content (wt.conway et al.ctmelt. the journal of chemical c physics.kempter and r.baldock et al. c (1) p. edu parameters updated by W.c. c quantities contained in it used in this subcode are.b.olsen in feb. values for comp.pu02. :) c ftmelt = fuel melting temperature (k) c fdelta = liquidsolid coexistence temperature range (k) c when the function fthexp is used alone. 18 (1966) c (2) n.tokar et al. c c c1u. c2u. c the correlations were derived with the assumption that the c thermal expansion strain is zero at 300k. + ctrane.s. c (6) f.f.chefus. proceedings of the 2nd c international conference on plutonium metallurgy.s. transactions of the american nuclear c society.
39 (1956) pp 181187.043*facmot go to 120 110 fthexp = fthexm+. 18 (1966) c (2) n.0.a.c3u. 2.16e01.bk. 3.bk. journal of inorganic and nuclear chemistry.christensen.russel. c (7) m.0) fthexp = fthexp/(1. journal of the american ceramic society.ftmelt)*(fcomp) if (t. +edpu.bk.0 return end FRAPTRAN 1. puo2 content. c the correlations were derived with the assumption that the c thermal expansion strain is zero at 300k. high temperatureshigh pressures.edu. or (u.15 . 7.0sigftex*0.edpu/ 9.t) = c1*tc2+c3*exp(ed/(bk*t)) t = ftemp bk = 1.ftmelt) go to 100 fthexp = ftx(c1u.brett and l.1 c (1955) pp 357370.t)*fcomp go to 120 100 fthexm = ftx(c1u.c.parker. 2.38e23 c bk = boltzmann's constant (j/k) fcomp = comp/100.c2u. Rev.baldock et al.32e19 / data c1pu. c 46 (1963) pp 607608.4: FTHEXP *deck fthexp c function fthexp (ftemp. and fraction of c the fuel which is molten.fuel is all molten c facmot = 0. c france (1960) pp 397410.s.6 (1963).0) fthexp = fthexp*(1.103) ufthex = 10.bk.j.ed.pu)02 c as a function of temperature. 7. journal of nuclear materials.edu.ftmelt)*(1.0.0e02.e.d.gt.edu / 9. c (4) j.61e03.b.oz) c c This is a MATPRO11. c (1) p.0e06.data c1u .(ftmelt+fdelta)) go to 110 fthexp = fthexm+.c3pu.c3u .c2u .h.0fcomp)+ftx(c1pu.c3pu.conway et al.0e20 / ftx(c1.c2pu.103) if(sigftex.gt. 1.ge.lt. c (6) f. c (3) m. grenoble.7e03. 2 routine modified by PNNL for use in FRAPT c c the subprogram fthexp is used to calculate the strain of the c fuel caused by thermal expansion.0 .c2u. transactions of the american nuclear c society.momin. c thermal expansion strain is calculated for uo2.edpu.burdock and h.0fcomp)+ftx(c1pu.gronvold.t)*(1.c2pu. +c3pu. journal of the american c ceramic society. proceedings of the 2nd c international conference on plutonium metallurgy.1 A.0 .bk.hoch and a.0 if (t. c c fthexp = output fuel strain due to thermal expansion (unitless) c ftemp = input fuel temperature (k) c facmot = input fuel fraction which is molten (unitless) c facmot = 1. c (5) j.6e05*(t(ftmelt+fdelta)) 120 continue c add on uncertainty if(sigftex.0+sigftex*0.facmot) c implicit real (ah.c2.pu02.fuel is all solid c c the correlations used in this subroutine are based on data from.c2pu.c3.c3u.8e06.043+3.
ftmelt) * fcomp if(t .t) * (1. and fdelta must be input.c3pu. compmt = puo2 content (wt.c2pu.tokar et al. 2. not as part of the inel code. 3.reymann in july 1978 Modified by PNNL.olsen in feb.6 Fuel Densification (FUDENS) A.p.s.c3u . and edu parameters updated by W. to clean up coding and delete lic analysis and sensitivity uncertainty coding include 'phypro. values for compmt.edu .0d02.baston in march 1974 modified by c.elliott.043d0 * facmot go to 100 c 75 c 100 c return end continue continue fthexp = fthexm + 0. the journal of chemical physics.0d20 / bk is Boltzmann's constant (J/k) data bk / 1.fcomp) + ftx(c1pu.gt.handwerk.c2u .0d0 .bk.38d23 / ftx (c1.a.bk.G.bk.c2pu. Luscher 101707 data c1u . quantities contained in it used in this subcode are.0d0 if(t .6d05 * (t.o. (10)m. 1975 last modified by g.0d0 .c3pu.bk. (ftmelt+fdelta)) go to 75 c fthexp = fthexm + 0.t) * fcomp go to 100 c A.0d06.c2pu.8d06. January 1997.(ftmelt + fdelta)) fthexp = ftx(c1u . 2.t) = c1 * t .c2u . (9) w.edpu/ 9.16d01. c3u.fcomp) & + ftx(c1pu.bk. c1u.ge.c3. fthexp was originally coded by v. anl5053 (1956).edu / 9.edpu.ftmelt) * (1. ftmelt) go to 25 c & c 25 continue fthexm = ftx(c1u .h' the phypro common block is set by call to the subroutine phypro which is part of the materials properties package used at the inel.30 (1958) pp 15241526.c2.lambertson and j. 7.edu .c2u . :) ftmelt = fuel melting temperature (k) fdelta = liquidsolid coexistence temperature range (k) when the function fthexp is used alone.32d19 / data c1pu. nuclear technology. ftmelt.043d0 + 3.1.kempter and r.c3pu. 17 (1973) pp 147152.c3u .f.edpu. (8) c.7d03.ed.c3u .c c c c c c c c c c c c c c c c c c c c c c c c c c (1969) pp 401407.61d03.16 . c2u.c2 + c3 * exp(ed/(bk * t)) c c c t fcomp c = ftemp = compmt/100. 7.a. 1.h.
mason (november 1978).96.prvden) c c fudens calculates irradiationinduced densification.0/x) if (err.bu.0al1 al4 = 0. dimension c(2).1000.17 .al4) y1 = dlen2(al3. c updated and corrected by b.and.0 al1 = dlen1 al3 = 3.00.(rsntr.6) go to 160 100 continue 110 continue x1 = x3 do 120 j=1.0 if (al3. 2.0de)/(t +s1180.rsntr.2*(100.x4.2.0) dlen1 = 100.1000. c fudens = output fuel dimensional change (%) c ftemp = input fuel temperature (k) c bu = input burnup (mws/kgu) c fdens = input fuel density (kg/m**3) c rsntr = input maximum density change determined by a resintering c test of 1973 k for 24 hours (kg/m**3) c tsint = input fuel sintering temperature (k) c comp = input plutonia content (weight percent) c prvden = input total densification from previous time step (%) c fudens was developed and programmed by c.0.al4) if (y1*y2. 0. e.0) x3 = 0.27e03) go to 130 do 100 i=1.4.le.0de)/(t +s1180.0) if ((ftemp.0.le.01*comp*c(1)+(1.bu.157e05 ts = tsint2.(rsntr.0.b(5) data c / 10.fdens.and. 35.0 * fdens) if ((ftemp.0.)) dlen1 = 66.le.x3.lt.gt. 3.4: FUDENS *deck fudens function fudens (ftemp.comp.50 x = x1dlen2(al3.x1.0) if (rsntr . 1.00.abu) = b(1)+alen+b(2)*exp(b(3)*(bu+abu))+b(4)*exp( +b(5)*(bu+abu)) dlen3(bu) = b(2)*b(3)*exp(b(3)*bu)b(4)*b(5)*exp(b(5)*bu) fbu = bu*1.).0) ts = 1600.le. if (tsint.0 * rsntr/(3.0e04) go to 140 x1 = x 120 continue 130 al3 = 2.01*comp)*c(2)) de = fdens/(roth*10.).6*(100.)) dlen1 = 22. burnham (october 1975).le.00. c fudens was modified by r.tsint.FRAPCON3.0.7315e02 c if rsntr or tsint is not defined by user. s.eq.0 if (i.0. the default value is c tsint = 1873 k.) go to 110 x3 = x4 x4 = x4+1.46/ data b / 3.0 x4 = 1. olsen (january 1975). 11.ge.0 roth = c(1)*c(2)/(0.le.00 / dlen2(alen.996 al2 = 5.384 go to 150 140 al2 = x 150 continue A.6 y2 = dlen2(al3. w.al4)/dlen3(x1) err = abs((xx1)*100.
974e+10*(bubul) soldsw = bus*(2. e.2 Cladding Material Properties A. sigftex.abs(prvden)) fudens = fudenprvden return format (1x.abs(prvden)) fudens = 0.1. w. A. mason ./' no roots found between 0 and 6000 mws/mt uo2') end A. D.J.sigh2 bus = fdens*2.lt.7 Fuel Swelling (FSWELL) FRAPCON3.sigcor . journal of nuclear c materials 18 (1966) p 233 c (2) e.fbu.bul.160 170 180 fuden = dlen2(al3. stansbury.1 Cladding Specific Heat (CCP) FRAPCON3. Geelhood c .comp. May 4.sigswell .le.sigcreep .siggro .0 if (abs(fuden). brooks and e.18 . deem. Lanning c recommendations in ltr to L. eldridge and h. Berna per D. Siefken.4: CCP *deck ccp function ccp (ctemp) c c ccp calculates the specific heat at constant pressure c for zircaloys c ccp = output cladding specific heat at constant c pressure (j/kgk) c ctemp = input cladding temperature (k) c this code is based on data from c (1) c.315e23+sigswell*2. e.al2) if (bu.0 continue if (abs(fuden). 1995 c fswell was revised October 2008 by K.june 1978. r.soldsw) c c comp = input plutonia content (weight percent) c fdens = input initial density of the fuel (kg/m**3) c ftemp = input temperature of the fuel ring (k) c bu = input burnup to end of time step (mws/kgu) c bul = input burnup to end of last time step (mws/kgu) c soldsw = output fuel swelling due to solid fission products c (fraction) c fswell was developed and programmed by r. a.211e23+sigswell*3.4: FSWELL *deck fswell subroutine fswell (fdens.315e24) if(bu. specific heats and heats c of transformation of zircaloy2 and low nickel zircaloy2 A.ftemp.6912000.ge. sigfgr & .bu. c fswell was revised July 1995 by G.program common /uncert/sigftc.211e24) return end A.1728) fuden = 0.2. "the specific heat 0f c of zircaloy2 from 50 to 700 c".0) soldsw = bus*(3.gt.0 go to 170 print 180 fuden = 0.
icm) implicit real (ah.is (1) for temperature less than 1090 k.eq. deem. + 469.5 . eldridge and h. miller october 1974 modified by d. 356.1248..1213.(sum of squared residuals/(number of residualsdegrees of freedom))**0.1090. 331. 302.. 25j/kgk) (3) for temperature above 1300 k .... 770. 10j/kgk) (2) for temperature 1090 k to 1300 k.1193.1153. e. r. + locidx / 4 / if (emflag(locidx). 590.1133. l.npcp. 1967) conversion from j/(kg*k) to cal/(gm*c) is 2..1/ data on / 1 /. 619.ctemp. Rev. 300..... + 375... journal of nuclear c materials 18 (1966) p 233 c (2) e. 100j/kgk) ccp coded by r. 816. 1967) c c conversion from j/(kg*k) to cal/(gm*c) is c 2. a. + off / 2 /.. brooks and e.1233. 502..39006d4 (cal/(gm*c))/(j/(kg*k)) c c estimated standard error of the prediction of c ccp for the specific heat of zircaloy cladding c samples . 2 routine modified by PNNL for use in FRAP c c ccp calculates the specific heat at constant pressure c for zircaloys c c ccp = output cladding specific heat at constant c pressure (j/kgk) c ctemp = input cladding temperature (k) c c this code is based on data from c (1) c.640. l.on) go to 110 if (ctemp..(sum of squared residuals/(number of A.iu) go to 120 100 ccp = 356..oz) c c This is a MATPRO11.19 .4: CCP *deck ccp function ccp (ctemp..c c c c c c c c c c c c bmi1803 (may 31.... go to 120 110 ccp = emccp(ctemp) 120 continue return end FRAPTRAN 1. iu / 13..ge. + 719. / data npcp.1113. 615.1173.39006d4 (cal/(gm*c))/(j/(kg*k)) estimated standard error of the prediction of ccp for the specific heat of zircaloy cladding samples .. w. "the specific heat 0f c of zircaloy2 from 50 to 700 c"... stansbury. 400.1093.h' dimension cpdata(26) data cpdata/ 281..0) go to 100 ccp = polate(cpdata. hagrman may 1976 include 'lacmdl.1248. specific heats and heats c of transformation of zircaloy2 and low nickel zircaloy2 c bmi1803 (may 31.
0d0. 25j/kgk) (3) for temperature above 1300 k .600.d0.400)icm c c c c c c c 400 continue generic Zry ccp = 279.d0.le.1153.d0. & 719.d0. iu) if (ctemp .d0.d0.d0 .d0.d0.d0.d0.d0.d0. & 816.d0/ data npww / 12 / high heat rate (>1000K/s) dimension cpwwr2(26) data cpwwr2/412. 302. to clean up coding. 400. & 348.d0. 1989.d0.400.1173. 448. 331. 392. January 1997."Material property library for H1alloy cladding".1190.1233.d0. 883. 385.d0. & 1000.1155.d0 / data npcp.d0.d0. 873. This property are induced by low heating rate.c c c c c c c c c c c c residualsdegrees of freedom))**0.d0) ccp = 356.d0. 600.1177.d0.1180.d0. 400.1193. 816.d0.E.d0.d0 return 600 continue RRCKI Zr1%Nb property This block calculates the specific heat at constant pressure for H1alloy cladding. al. 356. 5/5/03: default to iwwer=1 until can set this flag up iwwer = 1 low heat rate c c c c c c c c c c c c c c c A.1200.d0. 1993. If iwwer = 2 then data is derived from Ljusternik V.d0. et.6d0. 770.290.d0.d0.d0. npcp. 640. & 469.J." v. 100j/kgk) ccp coded by r.1110.1400. et. ctemp.d0. 590.d0.5 . M.d0.d0.d0.d0. 600.400.and.400.d0. & 360.1173.Yu.d0. hagrman may 1976 Modified by PNNL.d0.1153.d0. 973.d0.1120."Experimental research of zirconium reactor materials thermal properties:110alloy".d0.d0. 619.d0.d0.d0.1213. 480.d0.d0) & ccp = polate (cpdata. 673. & 281.1025.1090. 380. miller october 1974 modified by d. 400.d0.ge.1150. iu / 14 .d0.d0..d0.1600.d0. If iwwer = 1 then data is derived from Volkov B.1168.d0. 773.400.d0.31 n.280. 770.d0 if (ctemp.290.1133.d0. & 420. 360.d0.1113.d0. 10j/kgk) (2) for temperature 1090 k to 1300 k.d0/ data npw2 / 13 / go to (400.d0. 473. 502. 615.is (1) for temperature less than 1090 k.d0. including removed licensing assistance coding and sensitivity uncertainty coding Zry properties dimension cpdata(28) data cpdata/279. 1248. ctemp. l.400.d0.4.d0. & 375.1248.1142. 300.1073.20 .d0. Preprint IAE4941/11.1161.d0.d0.400.d0.d0.1248. al.d0. & 1000.d0.400.d0. 370.d0.1134.ge. l.1100.d0. 1 / Zr1%Nb properties low heat rate (<1000K/s) dimension cpwwer(24) data cpwwer/345.600.d0. 573. High Temperature Thermal Physics. 1400.d0. 680.d0. & 383.d0. This property are induced by highrate heating>=1000 K/s.1093.
d0) & ccp = polate(cpwwr2.1100.w. scott.h' data on / 1 /. miller.91d2*ctemp if(ctemp.oz) c A. dec 1975 c last modified by e.d. ccon. gemp669 (oct 1966) c c. cdkdt. kass. 1961.ccon) c c cthcon calculates cladding thermal conductivity as a function c of temperature. pp 19 .r.r. flux. flux or coldwork c as parameters in calculation of zircaloy thermal conductivity c one standard deviation of this function = 1.2) go to 601 if(ctemp .1173. anderson.lt.d0) ccp = 392.j. 1962.eq. and cold work c ccon = output thermal conductivity of zircaloy4 (w/mk) c ctemp = input cladding meshpoint temperature (k) c time = input time at temperature and flux (s) c flux = input fast neutron flux ((neutrons/m**2)/s) c coldw = input cold work (unitless ratio of areas) c the equation used in this subcode is based on data from c w. include 'lacmdl.d0) ccp = polate(cpwwer. beck.d0) ccp = 199. chirigos.ge. feith.4: (CTHCON) *deck cthcon subroutine cthcon (ctemp. w.ctemp. + off / 2 /. bmi1273 (1958) pp 79 c a.ccon) 110 continue return end FRAPTRAN 1.7d0+12.n. kapl2146 (1961) c d.flux.2.55 c a. coldw. carlson in june 1978. flux. pp 62 .d0 .coldw.k.93 c j. wcap326941 (1965) pp 59 c this version of cthcon does not use time. lucks and h. icm) implicit real (ah.ctemp.1200.le.e. time.iu) if(ctemp.l.s.f.668e09)) go to 110 100 call emcton (ctemp. + locidx / 7 / if (emflag(locidx). kirk and g.450e5+ctemp*7.iu) if(ctemp.time.088e2+ctemp*(1.if(iwwer.511+ctemp*(2.4: (CTHCON) *deck cthcon subroutine cthcon (ctemp.gt. 1173. powers.5d0+15. academic press.and.on) go to 100 ccon = 7.d0 return c c 601 continue high heat rate ccp = 237. time.ge.npw2. deem.2 Cladding Thermal Conductivity (CTHCON) FRAPCON3.j. ctemp.flux. kephart and j.21 .s.1200.b.eq.01 w/mk c this version of cthcon was developed by r.npww.coldw.time.364d2*ctemp return end c A. a. c. theilacker c astmstp314. c. salvaggio c fuel element fabrication.
e.000461843d0*ctemp)*15.511d0 + ctemp*(2. c. 1961. c if(ctemp.22 . wcap326941 (1965) pp 59 c c this version of cthcon does not use time. c Modified by PNNL. carlson in june 1978. lucks and h. academic press. time.400) icm 400 continue c Generic Zircaloy property from MATPRO c c the equation used in this subcode is based on data from c w. scott.le. to clean up coding and delete c licening analysis and sensitivity uncertainty coding c ccon = 7. January 1997.400. 2 routine modified by PNNL for use in FRAPT cthcon calculates cladding thermal conductivity as a function of temperature.000461843d0*ctemp) else ccon = 36.d0 cdkdt = 0. c. kephart and j.Yu.w. Data is derived from: c Volkov B. 1962. pp 62 .400.088d2 + ctemp * (1.400. feith.k. kapl2146 (1961) c d.400. flux or coldwork c as parameters in calculation of zircaloy thermal conductivity c c one standard deviation of this function = 1.b."Material property library for H1alloy c cladding". gemp669 (oct 1966) c c. et.s. Preprint IAE4941/11.r. theilacker c astmstp314.l. a. chirigos. pp 19 . kirk and g.668d09)) cdkdt = 2.d. beck.400. Rev.006957018215d0*dexp(0. powers. bmi1273 (1958) pp 79 c a. deem.93 c j. anderson.088d2 + ctemp*(2.600.r.j.0636d0 cdkdt = 0.450d5 + & ctemp*7.2133.0d0) then ccon = dexp(0.01 w/mk c c this version of cthcon was developed by r.d0 endif return c A. salvaggio c fuel element fabrication. flux.f. dec 1975 c last modified by e. w. 1989.j. al.c c c c c c c c c c c c c c c c This is a MATPRO11. miller.400.9d5 + ctemp*2.3d8) return c 600 continue c E110 property as coded by RRCKI and provided to PNNL c c This code block calculates thermal conductivity (w/(m*k)) and c derivative of thermal conductivity with respect to temperature c (w/(m*k*k)) of H1alloy.600.400.n.s. kass. and cold work ccon cdkdt = output thermal conductivity of zircaloy4 (w/mk) = output derivative of thermal conductivity with respect to temperature = = = = input input input input cladding meshpoint temperature (k) time at temperature and flux (s) fast neutron flux ((neutrons/m**2)/s) cold work (unitless ratio of areas) ctemp time flux coldw go to appropriate model for cladding type go to (400.55 c a.
4 Cladding Oxide Emissivity (ZOEMIS) FRAPCON3. kingery. r.t*1.946d10)) c return end A.43d7 . r. miller may 1974 c use caution above 1700k t = ctemp zotcon = 1. amer.oz) c c zotcon calculates zirconium dioxide thermal conductivity as a c function of temperature. coble and t. c c zotcon = output thermal conductivity of zro2 (W/mK) c ctemp = input cladding meshpoint temperature (K) c c the data used to generate this correlation were taken from: c w.2. kingery. j.4: ZOTCON *deck zotcon function zotcon (ctemp) implicit real (ah.41e4t*(6.i (1954) pp 107110 c zotcon was coded by r.l.2. c modified by r.i (1954) pp 107110 c c zotcon was coded by r. francl. ceram. miller may 1974 c c use caution above 1700K c t = ctemp zotcon = 1.4: ZOTCON *deck zotcon function zotcon (ctemp) c c zotcon calculates zirconium dioxide thermal conductivity as a c function of temperature.. coble and t.. soc.miller in march 1974.43e7t*1.end A.emissv) c c zoemis calculates the emissivity of the cladding surface A. vasilos c j. c zotcon = output thermal conductivity of zro2 (w/mk) c ctemp = input cladding meshpoint temperature (k) c the data used to generate this correlation were taken from: c w.41d4 .l.9599t*(2.4: ZOEMIS *deck zoemis subroutine zoemis (ctmax. ceram.946e10)) return end FRAPTRAN 1.l.23 . amer.3 Cladding Oxide Thermal Conductivity (ZOTCON) FRAPCON3. j.t*(2.9599d0 . vasilos c j. soc.t*(6.l. 37 no. francl.l.miller in march 1974. c modified by r.l.d.d. 37 no.zroxid.
60 (1976) pp 167176 zoemis coded by r.25e01+1.. emissivity of zirconium alloys in air in the temperature range 100 .24 . spatind norway (september 1976) (3) e. 76.gt. juenke and s.1 if (ctmax.lt.88e06) go to 100 emissv = 3. expected standard error of the calculated emissivity compared to inreactor data is also returned.1008 (1968). usaec report gemp . paper presented at specialists meeting on the behaviour of water reactor fuel elements under accident conditions. v.4: ZOEMIS *deck zoemis subroutine zoemis (ctmax.00emissv if (uuemis. l.08642e015.ge. zroxid.50e03ctmax)/3.oz) c c This is a MATPRO11.246e05*zroxid go to 110 100 emissv = 8. f. c c emissv = output cladding surface emissivity (unitless) c puemis = output positive standard error expected in emissv c when compared to inreactor data (not currently returned) c uuemis = output negative standard error expected in emissv A. murphy and f.00e02) if (emissv. hagrman october 1976 model for temperatures below 1500k follows if (zroxid. burgoyne and a. nuc.gt.1500. Rev. emissv) implicit real (ah. garlick. havelock.c c c c c c c c c c c c c c c c c c c c c c c c c as a function of maximum cladding temperature and oxide thickness. emissv = output cladding surface emissivity (unitless) puemis = output positive standard error expected in emissv when compared to inreactor data (not currently returned) uuemis = output negative standard error expected in emissv when compared to inreactor data (not currently returned) ctmax = input maximum cladding temperature (k) zroxid = input oxide layer thickness (m) the equations used in this subroutine are based on data from (1) aec fuels and materials development program progress report no. l.325 puemis = puemis/exp((1. miller sept 1974 modified by d.00e02) uuemis = puemis c standard error cut off at impossible values follows if (puemis. j. b. expected standard error of the c calculated emissivity compared to inreactor c data is also returned. section by e.00e01*zroxid 110 puemis = 0. 2 routine modified by PNNL for use in FRAPT c c zoemis calculates the emissivity of the cladding surface c as a function of maximum cladding temperature and c oxide thickness. p 239 (2) t.00emissv)) puemis = 1.325) emissv = 0.le. sjodahl.) go to 120 c modification for maximum temperatures above 1500 k follows emissv = emissv*exp((1.50e03ctmax)/3.emissv) uuemis = emissv 120 continue return end FRAPTRAN 1.400 c j. mat.3.(1.1 uuemis = 0.0.
January 1997.b.j.w. section by e. nuc. (1. "thermal expansion A.4: CTHEXP *deck cthexp subroutine cthexp (ctemp. mat. delete sensiti uncertainty coding. and correct error in PUEMIS for T > 1500K model for temperatures below 1500K: emissv = 3. 76.gt.ctmax)/3. v.50d03 .25 .emissv if(uuemis .lt. 0. paper presented at specialists meeting on the behaviour of water reactor fuel elements under accident conditions. f. garlick. 60 (1976) pp 167176 zoemis coded by r. emissv) uuemis = emissv 50 continue return c c c c c c c end A..00d0 . b.1008 (1968).25d01 + 1.00d02) if(emissv .00d02) uuemis = puemis standard error cut off at impossible values follows if(puemis .ge.cdthex) c c cthexp calculates axial and diametral thermal expansion of c zircaloy lwr cladding. kearns.5. l.246d05*zroxid if (zroxid . spatind norway (september 1976) (3) e. hagrman october 1976 Modified by PNNL.5 Cladding Thermal Expansion (CTHEXP) FRAPCON3. scott. 1500.le. burgoyne and a.0d+01*zroxid puemis = 0. to clean up coding. p 239 (2) t. j.400 c j.c c c c c c c c c c c c c c c c c c c c c c c when compared to inreactor data (not currently returned) ctmax = input maximum cladding temperature (k) zroxid = input oxide layer thickness (m) the equations used in this subroutine are based on data from (1) aec fuels and materials development program progress report no.88d06) emissv = 8.08642d01 . usaec report gemp . and j. c wcap326941 (1965).cathex.1d0 uuemis = 0.1d0 if(ctmax . "mechanical properties c of zircaloy2".2. c "physical and mechanical properties of zircaloy2 and 4". juenke and s.l. c cathex = output axial thermal expansion of zircaloy (m/m) c cdthex = output diametral thermal expansion of zircaloy (m/m) c ctemp = input cladding meshpoint temperature (k) c data for the correlation from room temperature to 1273k are c taken from r.gt. mehan and f. sjodahl. l.d0) go to 50 modification for maximum temperatures above 1500K: emissv = emissv * exp((1.325d0 puemis = puemis*exp((1. wiesinger.ctmax)/3.00d0emissv)) puemis = 1. emissivity of zirconium alloys in air in the temperature range 100 . havelock. miller sept 1974 modified by d. usaec report kapl2110 (1961). murphy and f. 3. d.50d03 .325d0) emissv = 0.
npcatx. 5. 1233. + 3.2200e03.15/ data npdexp / 22 / if (ctemp.8800e03. 1173.15 . 1193. 293.2200e03.1273. + 5. reymann in june 1978. 1253.8600e03.1 / data dthexp/ 4.0800e03. 1193.15.15) go to 100 if (ctemp.iu) go to 120 100 cathex = 2. 1.15)*6. 2. 3. 1083.15. icm) implicit real (ah.300e03+(ctemp273.15 .2400e03. 3.50000e03. 1133.15. 3. cthexp was originally coded by r.30000e03.15.13000e03. 1203. 1213.41000e03.15. 1183. 1. + 3.07000e03.9000e03.l.16000e03. miller in nov.53000e03.52774e03. c c cathex = output axial thermal expansion of zircaloy (m/m) c cdthex = output diametral thermal expansion of zircaloy (m/m) c c ctemp = input cladding meshpoint temperature (k) c A. lustman and f.ge.30000e03.00000e03. 1143. 1.15.15 . 1273. 1183.15) go to 110 cathex = polate(cathxd. 1203. 1103. 373. 1243.50000e03.800e03+(ctemp273. 1.15 .15. 1974.le.15.7210e06 go to 120 110 cathex = 8.15 . iu / 22. wapdtm472 (1965) above 1273k the coefficient of thermal expansion used is the constant value 9.15)*9.1000e03.7e06/k recommended by b.15 . cdthex. 1093.2800e03. 1113. kerze. 2.50000e03. + 1.15. 1223.15 / data npcatx. + 3. + 2. 2. 1073. 2.1500e03.15 .70e06 120 continue return end FRAPTRAN 1. 3.2500e03. 3.15 . 1.iu) cdthex = polate(dthexp. + 3. last updated by g.13000e03. 3. + 4.1073.1395e03.21000e03.9000e03. 5.11000e03. 1263.15.46000e03.3480e04.15 . + 1. 5. new york (1955) page 355.7000e03.ctemp.15 . + 1.15 .15 . + 4.3730e04+(ctemp273.33000e03.15.15.15 . dimension cathxd (44) dimension dthexp (44) data cathxd/ 0.9700e03. 1213. 1123. 1273.15)*9.15 . 5. + 2.10000e03.4500e03.15. + 1.15. 1113. 1143.15 .15. cathex.26 .4: CTHEXP *deck cthexp subroutine cthexp (ctemp. 1263. 1133. 1233. 5.15. + 1.1300e03.0 .8700e03.oz) c c cthexp calculates axial and diametral thermal expansion of c zircaloy lwr cladding. 1163.15 .15 . + 2.15 . 1073. 1153.40000e03.80000e03.ctemp.15 . 1123.2800e03. 1173. 1243.15.c c c c c c c c c c and preferred orientation in zircaloy". 1103. mc grawhill book company. 1153. + 5.4410e06 cdthex = 2.15.15 .9200e03.70e06 cdthex = 6.15.a.5060e05+(ctemp273.15. 1093. 5. 1223. 1083. 1253.npdexp. 1163.15)*4.20000e03.5000e03. 4.15 . "the metallurgy of zirconium". 2.15. 4. between 1073 and 1273k (approximately the alphabeta transition range for zircaloy) cthexp uses the linear interpolation routine polate to find the thermal expansion. + 5. + 2.
15d0.1223.2200d03.460d03.7210d06 return c 402 continue cathex = 8.ge.1243. wapdtm472 (1965) above 1273k the coefficient of thermal expansion used is the constant value 9.1203.15d0.1223.400.9200d03.15d0.a. 5. & 3. 3.200d03. kerze.1000d03. iu /22.15d0.1103. wiesinger.1263.w. kearns.15d0.8700d03.1300d03.1253.8600d03.1133. wcap326941 (1965).15d0.300d03. scott.b. 1273.1103.1193.15d0.5300d03.15d0.7e06/k recommended by b.300d03 + (ctemp273. 1.15d0.3000d03. 1974.j. & 2.1143.15d0.400.15d0.1173. 1.iu) cdthex = polate(dthexp.1300d03.l.1093. 1.5000d03. mc grawhill book company.1600d03. "thermal expansion and preferred orientation in zircaloy".1300d03.1233. 1. & 1.15d0)*6. lustman and f. 5.15d0.2200d03. 5.1093. 1. miller in nov.800d03 + (ctemp273.293.15d0.1143.15d0.2400d03. & 2.15d0.15d0/ data npdexp.1163.5000d03.1163.npdexp.3480d04.15d0.1073.0000d03.15d0.15d0. & 4. 4.1083.15d0. & 2. "physical and mechanical properties of zircaloy2 and 4".0d0.15d0.15d0.15d0) go to 402 cathex = polate(cathxd. 2.15d0.15d0)*9.1253. 3.1243.2500d03.4410d06 cdthex = 2.1113.le.1193. 5.15d0. 3.400. 1073.0700d03.15d0/ data npcatx /22/ data dthexp/ 4.1123.15d0.70d06 cdthex = 6.7000d03. & 1.1183.ctemp. & 5. & 5. 1/ go to (400. usaec report kapl2110 (1961). 2.15d0. mehan and f.iu) return c 401 continue cathex = 2.15d0. reymann in june 1978. Licensing analysis coding deleted 9/23/97 by ME Cunningham.400.52774d03.1173. 3. 2.15d0.4100d03. "mechanical properties of zircaloy2". & 1.70d06 return c c A.15d0.15d0.1203.1395d03. 5.15d0.400)icm c c c 400 continue generic Zircaloy thermal expansion if(ctemp .3730d04 + (ctemp273.15d0.0800d03.400.15d0.1263.1213.1100d03.3300d03. 2. 2.500d03.600.1133.15d0)*9. 5. 3.1273. 1.15d0.2800d03.15d0.210d03.npcatx.15d0. cthexp was originally coded by r.15d0. & 3.27 . PNNL generic Zry dimension cathxd (44) dimension dthexp (44) data cathxd/ 0.400.600.c c c c c c c c c c c c c c c c c c c c c data for the correlation from room temperature to 1273k are taken from r.9000d03. d.15d0.8800d03.1153.15d0.1213.4500d03.5060d05 + (ctemp273.1273.100d03. & 4.1500d03.9700d03.400. between 1073 and 1273k (approximately the alphabeta transition range for zircaloy) cthexp uses the linear interpolation routine polate to find the thermal expansion.15d0)*4.9000d03. "the metallurgy of zirconium".5000d03.1153. new york (1955) page 355.l. 3. & 5. and j.15d0.15d0.15d0) go to 401 if(ctemp . last updated by g. 1.4000d03.15d0.8000d03.1183.15d0.1123.2800d03.1073. & 3. 2.373.1113.ctemp. 4.1083. 3.1233.
28 .312d8*(ctemp883.7211d10*(ctemp883.gt. l.1338985d8*ctemp**2+3.7d6*(ctemp1153. and average c oxygen concentration.65390d6*ctemp0.199649865d2 c else cathex=0.gt.fnck. journal of nuclear materials 59 A. c the expected standard error is computed but not returned c (ucelmd).d0)**2+1.cwkf.7d6*(ctemp1153.13725577d2+5. cold work. s.5977000d3+2.deloxy) c c celmod calculates cladding young's modulus as a function c of temperature.199649865d2 endif c return end A. armstrong and h. grain orientation is assumed random. 883.312d8*(ctemp883. groff.358d8*(ctemp883.127813365d2 cdthex=0.d0) then cathex=1. fast neutron fluence.d0) then cathex=0.076459d3+9. brown. bates. mellinger and j.d0) cdthex=0.d0) & 7.average oxygen concentration of c asreceived cladding (kg oxygen/kg zircaloy) c the young's modulus calculated by this function is based on c data from the following references c (1) l. padel and a.c c 600 continue RRCKI thermal expansion if(ctemp . l. high c temperature properties of zircaloy .2. r. j. epri c report np .65390d6*ctemp0.d0) & 7. c (2) e.85875d6*ctemp0.d0)**3 c else if(ctemp .524 (1977).7211d10*(ctemp883.3336985d8*ctemp**2+5.0465577d3+2. fisher and c.6 Cladding Elastic Modulus (CELMOD) and Shear Modulus (CSHEAR) FRAPCON3.358d8*(ctemp883.627600d3+9. 573.3133600d2 c else if(ctemp . e.4d6*(ctemp573.494.ge. transactions of the c metallurgical society of aime 230 (august 1964) c pp 962 .d0)then cathex=3.d0)**3 cdthex=5. 2133.oxygen alloys.966 c (4) a. physical review 135 2a c (20 july 1964) pp a482 .gt. c celmod = output young's modulus for zircaloy 2 and 4 c with random texture (pa) c ucelmd = output standard error expected in celmod (pa) c (not currently returned) c ctemp = input cladding temperature (k) c fnck = input effective fast fluence (neutrons/(m**2)) c cwkf = input effective cold work (unitless ratio of areas) c deloxy = input average oxygen concentration excluding c oxide layer . renken.d0) cdthex=3.0582459d2 cdthex=1.d0)**2+1.d0) then cathex=1. g. bunnell.3336985d8*ctemp**2+5. b.d0) c else if(ctemp .1153.4: CELMOD *deck celmod function celmod (ctemp. c (3) p.
lt. simis.7015 c2 = 1.7308937e03) go to 110 tabb = (100*deloxy+0.gt. a.gt. conf681086 (1968) pp 277 . van echo.1. l. mehan.h' data on / 1 /.4: CELMOD *deck celmod A.05e+07 go to 140 em model young's modulus celmod = emclem(ctemp) continue if (celmod.) go to 140 calculate a to a + b and a + b to b boundaries wfox = deloxy+0. j.157+1081. l. bmi1168 (1957) celmod was coded by r. o. busby and c.6e+10 celmod = (1.0e+25) c3 = 2.12)*491. shober.1.21e+10tabb*4.88*(1.1242807)**2+3.8281e+04*(wfox0. m. part i: materials. l.lt. northwood.lt. e.eq.0 if (fnck.46*((100*deloxy+0.025) tabb = 392.1090. woods (eds. r.475e+07*ctemp+c1*deloxy+c3*cwkf)/c2 if (ctemp. the mechanical properties of zirconium and zircaloy2. and j. r. kaplmrlm16 (july 1958) (4) d.1417) if (deloxy. hagrman december 1977 include 'lacmdl.4.16e+11+ctemp*1. saxton core ii fuel performance evaluation.088e+115.914e+05) if (wfox. + locidx / 5 / if (emflag(locidx). pavlovic and g. burnham october 1977 modified by d. burnham in october 1975 modified by b.0. miller in march 1974.025) go to 100 taab = 1556. smalley. journal of nuclear materials 55 (1975) pp 299310 (5) f.037e+08)*5.tabb) go to 120 amodl = (1.0exp(fnck/1. modulus of elasticity of zircaloy2 between room temperature and 1000 f. keeler.on) go to 130 best estimate model young's modulus c1 = (1.lt. l.) "properties of zircaloy4 tubing". c.4+3.0e+22) c2 = 0. m london.0012 taab = 1094.289e+03+wfox*7. usaec report wapdtm585 (december 1966) p 65 (2) z.0) celmod = 1. i.29 . l.284 (3) r.088e+115. bahen. wcap338556 (september 1971) the expected standard error was derived using additional data from the following references (1) c. r.c c c c c c c c c c c c c c c c c c c c c c c 100 110 120 c 130 140 (1976) pp 325326 (5) w. marianske lanze. marsh jr.0e+25))+exp(fnck/ +1.taab) go to 140 if (ctemp.r. conference on the use of zirconium alloys in nuclear reactors.21e+10ctemp*4. czech. spasic. w. and l.7413 continue if (ctemp.0 ucelmd = 6. updated and corrected by b.lt. w.+wfox*(1.05e+07 celmod = amodl+(ctemptaab)*(amodramodl)/(tabbtaab) go to 140 celmod = 9.4e+09 return end FRAPTRAN1.475e+07*taab+c1*deloxy+c3*cwkf)/c2 amodr = 9. + off / 2 /.
l. cwkf. cold work. shober. deloxy. marianske lanze.400.400)icm A. woods (eds. and l. fnck.400. smalley. bunnell. kaplmrlm16 (july 1958) (4) d.600. w. w.966 (4) a. a. l. simis. j. celmod = output young's modulus for zircaloy 2 and 4 with random texture (pa) ctemp = input cladding temperature (k) fnck = input effective fast fluence (neutrons/(m**2)) cwkf = input effective cold work (unitless ratio of areas) deloxy = input average oxygen concentration excluding oxide layer . icm) c implicit real (ah. part i: materials. usaec report wapdtm585 (december 1966) p 65 (2) z. c. renken. journal of nuclear materials 59 (1976) pp 325326 (5) w. g. saxton core ii fuel performance evaluation. delete licensi analysis and sensitivity analysis coding go to (400.600.16d+11 + ctemp * 1. (3) p. bmi1168 (1957) celmod was coded by r. i. grain orientation is assumed random.0d0 This is a MATPRO11. bates.037d+08) * 5. transactions of the metallurgical society of aime 230 (august 1964) pp 962 . physical review 135 2a (20 july 1964) pp a482 . l. r. mellinger and j. modulus of elasticity of zircaloy2 between room temperature and 1000 f. busby and c. czech. conf681086 (1968) pp 277 .284 (3) r. m. spasic.7015d0 c2 = 1. j. m london. armstrong and h.30 . miller in march 1974. b. hagrman december 1977 Modified by PNNL. epri report np .494.400. l. conference on the use of zirconium alloys in nuclear reactors. wcap338556 (september 1971) the expected standard error was derived using additional data from the following references (1) c. bahen.400.c function celmod (ctemp. and j. e. (2) e. mehan.) "properties of zircaloy4 tubing". e. padel and a.400. pavlovic and g. groff. marsh jr. journal of nuclear materials 55 (1975) pp 299310 (5) f.oxygen alloys. Rev. s.524 (1977). van echo. o. burnham in october 1975 modified by b. 2 routine modified by PNNL for use in FRAPT celmod calculates cladding young's modulus as a function of temperature.oz) c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c 400 continue generic Zry model best estimate model young's modulus c1 = (1. l. fisher and c. brown. l. updated and corrected by b. fast neutron fluence.r. burnham october 1977 modified by d. r. high temperature properties of zircaloy .400. l. r. and average oxygen concentration. northwood. r. January 1997. to clean up coding.400. keeler.average oxygen concentration of asreceived cladding (kg oxygen/kg zircaloy) the young's modulus calculated by this function is based on data from the following references (1) l. the mechanical properties of zirconium and zircaloy2.
lt.5.4d0 + 3. 1.gt.475d+07 * ctemp + c1 * deloxy + c3 * cwkf) & /c2 if(ctemp .5d07*ctemp 602 continue if(celmod .al.if(fnck .05d+07 c 401 continue if(celmod .lt.lt. tabb) go to 435 amodl = (1.31 .6d+10 celmod = (1.88d0 * (1. 0.4: CSHEAR *deck cshear function cshear (ctemp.129d10 .lt.al.gt.5.475d+07 * taab + c1 * deloxy + c3 * cwkf) & /c2 amodr = 9.tabb * 4.46d0 * ((100 * deloxy + 0. 4.21d+10 . taab) go to 401 if (ctemp .lt.d0) go to 401 c c calculate a to a + b and a + b to b boundaries wfox = deloxy + 0.0d+25) c3 = 2.0d0) go to 601 celmod = 1.121d11 ."Material property library for H1alloy cladding". 1.088d+11 .deloxy) c c cshear calculates the shear modulus of zircaloy2 and 4 A. et. 1989.0d0) celmod = 1.12d0) * 491.0d+25)) & + exp(fnck/1."Influence of oxygen content and gamma irradiation on the elastic module and internal friction pure zirconium" Metal Physics vol.fnck.05d+07 celmod = amodl + (ctemp .exp(fnck/1.7413d0 425 continue if (ctemp .7308937d03) go to 425 tabb = (100 * deloxy + 0.1417d0) if (deloxy .taab) go to 401 c 435 continue celmod = 9.438d07*ctemp celmod = celmod + 3.gt.025d0) go to 415 taab = 1556.914d+05) if (wfox .21d+10 .Yu.0d0 return c end c FRAPCON3.021d12*deloxy go to 602 601 continue celmod = 9.0012d0 taab = 1094.0d0) celmod = 1.cwkf.0d0 return c c c c c c c c c c c c 600 continue Zr1%Nb model from RRCKI This code block calculates cladding Young's modulus as a function of temperature .taab) * (amodr .amodl)/(tabb .025d0) 415 continue tabb = 392.lt. 1971 if(ctemp . data is derived from Volkov B.ctemp * 4.6. Preprint IAE4941/11.157d0 + 1081.088d+11 .0d+22) c2 = 0. Fedorov et. 1073.0.1242807d0)**2 + 3.8281d+04 * (wfox .d0 + wfox * (1.0d0 .289d+03 + wfox * 7.4. 1090.621. 1.32(3) p.
renken. l.46*((100*deloxy+0. padel and a. grain orientation is assumed random. (3) p.04e+102.lt.1.lt.4.49e+10ctemp*1.0012 taab = 1094.32 . smalley.07e+11ctemp*2.168e+07*ctemp+c1*deloxy+c3*cwkf)/c2 if (ctemp. l.4: CSHEAR A. brown.66e+07 cshear = amodl+(ctemptaab)*(amodramodl)/(tabbtaab) go to 130 cshear = 3. l.524 (1977) (2) e.gt.0e+25) c3 = .1242807)**2+3.0e+09 return end FRAPTRAN 1.66e+07 continue if (cshear.lt.oxygen alloys.7308937e03) go to 110 tabb = (100*deloxy+0.315e+08 c2 = 1. transactions of the metallurgical society of aime 230 (august 1964) pp 962 . part i: materials. bates. journal of nuclear materials 59 (1976) pp 325 .taab) go to 130 if (ctemp.) go to 130 calculate a to a + b and a + b to b boundaries wfox = deloxy+0.0e+25))+exp(fnck/ +1. epri report np .289e+03+wfox*7. fisher and c.04e+102. b.0 if (fnck. wcap338556 (september 1971) cshear was coded by r. physical review 135 2a (20 july 1964) pp a482 .0exp(fnck/1. june 1974 modified by d.49e+10tabb*1. r.tabb) go to 120 amodl = (4. r.8281e+04*(wfox0. armstrong and h. j.494.168e+07*taab+c1*deloxy+c3*cwkf)/c2 amodr = 3. groff. hagrman december 1977 c1 = 7.1090. (5) w.0.326.average oxygen concentration of asreceived cladding (kg oxygen/kg zircaloy) the shear modulus calculated by this function is based on data from the following references (1) l.0) cshear = 1. s. mellinger and j.12)*491.gt.4+3. and average oxygen concentration.867e+10 cshear = (4.0 ucsher = 9. fast neutron fluence.1.c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c 100 110 120 130 as a function of temperature. cold work. saxton core ii fuel performance evaluation.lt.966 (4) a.0e+22) c2 = 0. l.025) go to 100 taab = 1556. bunnell.88*(1.+wfox*(1. e.914e+05) if (wfox.025) tabb = 392. high temperature properties of zircaloy .157+1081. g. miller.1417) if (deloxy.7413 continue if (ctemp.lt. the expected standard error is computed but not returned (ucsher) cshear = output shear modulus for zircaloy2 and 4 with random texture (pa) ucsher = output standard error expected in cshear (pa) (not currently returned) ctemp = input cladding temperature (k) fnck = input effective fast fluence (neutrons/(m**2)) cwkf = input effective cold work (unitless ratio of areas) deloxy = input average oxygen concentration excluding oxide layer .
025d0) 415 continue tabb = 392.0d0 .7308937d03) go to 425 tabb = (100 * deloxy + 0. smalley. c (5) w.1242807d0)**2 + 3. epri c report np .d0) go to 401 c c calculate a to a + b and a + b to b boundaries wfox = deloxy + 0. January 1997.400. fisher and c. l. armstrong and h.exp(fnck/1.46d0 * ((100 * deloxy + 0.400. l. c cshear = output shear modulus for zircaloy2 and 4 c with random texture (pa) c ctemp = input cladding temperature (k) c fnck = input effective fast fluence (neutrons/(m**2)) c cwkf = input effective cold work (unitless ratio of areas) c deloxy = input average oxygen concentration excluding c oxide layer . padel and a.d0 + wfox * (1.12d0) * 491. bunnell.867d+10 cshear = (4.400)icm c 400 continue c generic Zry properties c1 = 7. brown. grain c orientation is assumed random. icm) c implicit real (ah. 1090. r.914d+05) if(wfox . Rev. b. 1.88d0 * (1.0d+22) c2 = 0.600. saxton core ii fuel performance evaluation.04d+10 .524 (1977) c (2) e.2.oz) c c This is a MATPRO11.7413d0 A.315d+08 c2 = 1. c cold work.8281d+04 * (wfox .07d+11 .400. j. e. 0. r.157d0 + 1081.1417d0) if(deloxy .0012d0 taab = 1094.33 .lt.*deck cshear c function cshear (ctemp. cwkf. june 1974 c modified by d.400. transactions of the c metallurgical society of aime 230 (august 1964) c pp 962 .0d0 if(fnck . journal of nuclear materials 59 c (1976) pp 325 . l. to clean up coding and delete c sensitivity uncertainty analysis coding c go to (400. high c temperature properties of zircaloy .400. and average oxygen concentration. s. g.966 c (4) a. c (3) p. mellinger and j.400. bates.494.0d+25)) & + exp(fnck/1.600. l. 1 routine modified by PNNL for use in FRAPT c c cshear calculates the shear modulus of zircaloy2 and 4 c as a function of temperature.ctemp * 2.0d+25) c3 = 0. c part i: materials.average oxygen concentration of c asreceived cladding (kg oxygen/kg zircaloy) c c the shear modulus calculated by this function is based on c data from the following references c c (1) l. deloxy.289d+03 + wfox * 7.oxygen alloys.0.025d0) go to 415 taab = 1556.lt. hagrman december 1977 c Modified by PNNL.400. physical review 135 2a c (20 july 1964) pp a482 .326.gt. 4. wcap338556 (september 1971) c c cshear was coded by r. fnck. renken. miller.4d0 + 3. groff.lt.168d+07 * ctemp + c1 * deloxy + c3*cwkf)/c2 if(ctemp . fast neutron fluence.
fnck.sigcreep . Hagrman. August 1975 c revised April. G.845 if(icm.42628d0 .sigcor .235267.amodl )/(tabb .49d+10 .5d0*ey/(1.425 continue c if(ctemp .168d+07 * taab + c1 * deloxy + c3*cwkf)/c2 amodr = 3.556d5*ctemp c c shear modulus cshear = 0.013e21*f1**0.4: CAGROW *deck cagrow function cagrow (flux. D.fluenc.gt.49d+10 .Yu. 1982.81787 A.5. Preprint IAE4941/11. f1=(fluencflux*dtime)/10000.lt.dtime) common /uncert/sigftc.tabb * 1.0d0) cshear = 1. ax1=2. 1.18e21*f1**0.0d0 return c end A. sigfgr & .sigswell . Franklin.eq. pp.lt.sigh2 c c cagrow calculates axial cladding growth strain increment as a functi c flux = input fast neutron flux ((neutrons/m**2)/s) c fluenc = input fast neutron fluence (neutrons/m**2) c dtime = input time increment at temperature (s) c icm = input cladding material index:4 . L. f2=fluenc/10000. cwkf.taab) go to 401 435 continue cshear = 3.7 Cladding Axial Growth (CAGROW) FRAPCON3. J.04d+10 . Geelhood c to conform to EPRI model by D. CladType) c c poisson's ratio: xnu = 0.Zircaloy 4 c 2 .0d0 return c 600 continue c Zr1%Nb from RRCKI c c This code block calculates cladding young's modulus and c poisson's ratio as a function of temperature .icm.2."Material property library c for H1alloy cladding". c c elastic modulus ey = celmod (ctemp. tabb) go to 435 amodl = (4.al. deloxy.5) then ax1=7.2. "Zircaloy4 c Cladding Deformation During Power Reactor Irradiation.taab ) * (amodr .Zircaloy 2 c cagrow = output axial cladding growth straini increment (m/m) c cagrow coded by D.d0+xnu) if(cshear .18e21*f2**0. 1995 by D. et.34 . 1989. data is c derived from Volkov B. 1.845 ax2=2. Lanning and K.0d0) cshear = 1. sigftex.66d+07 401 continue if(cshear . taab) go to 401 if(ctemp .66d+07 cshear = amodl + (ctemp ." c ASTM STP 754.ctemp * 1.siggro .lt.
temperature.gt.98239 endif cagrow = ax2ax1 if( icm . 4.GapPress.and.lt.creep strain rate (1/hr) c ******************************************************************** c c NEW MODEL c New model uses a modified version of the Limback and Andersson equation c ASTM STP 1295 pp.sigcreep . c crepr computes (transverse) cladding creep strain rate c as a function of effective stress.0siggro*0.edot.lt.7893e25*f1**0.0siggro*0.lt.sigcor .0.0.cladding hoop stress per node (psi) c ******************************************************************** c output arguments c ******************************************************************** c edot .0)**2) &/(rco**2rci**2) st=(GapPress*rci**2CoolantPress*rco**2 &rci**2*rco**2*(CoolantPressGapPress)/((rco+rci)/2.CladAveTemp &.and.cldwks.35 .203) if(siggro.lt.sigswell .lt.c c c ax2=7.rci.223) if(siggro.4) &cagrow = cagrow*(1.0.0 ) cagrow = cagrow * 0.sigh2 c implicit real*8 (ah.0.total accumulated creep strain (dimensionless) c phi .8 Cladding Creep Rate (CREPR) FRAPCON3.0.and.7893e25*f2**0. fast c neutron flux and accumulated total creep strain c ******************************************************************** c input arguments c ******************************************************************** c CreepStrain .4) &cagrow = cagrow*(1.phi.eq.0+siggro*0.gt.siggro .5 add on uncertainty PWR cladding if(siggro.oz) c ******************************************************************** c crepr is called from cladf.98239 ax2=9.ge.0. 448468 c calculate stress in each direction sr=(GapPress*rci**2CoolantPress*rco**2 &+rci**2*rco**2*(CoolantPressGapPress)/((rco+rci)/2.4) &cagrow = cagrow/(1.223) BWR cladding if(siggro.icm.0+siggro*0.ge.013e21*f2**0.cladding average temperature (F) c sig . ProblemTime) c common /cladtype/ icm common /uncert/sigftc. sigfgr & .time to end of power step (hr) c CladAveTemp .4: CREPR *deck crepr c subroutine crepr(sig.4) &cagrow = cagrow/(1.flux (neutrons/m**2/sec) c ProblemTime .0)**2) A.icm.0.fluence.rco.2.6) then ax1=9.203) return end A.and.0.81787 elseif(icm.CoolantPress. sigftex.icm.icm.
eq.c c c &/(rco**2rci**2) sl=(GapPress*rci**2CoolantPress*rco**2)/(rco**2rci**2) sag is effective stress in MPa sag = sig*6894.1) then esat=0.36 .0*(1.0.0.0.145) if(sigcreep.5)* &exp(52.0.00.cldwks.cldwks.gt.18562+0.0.145) RXA cladding if(sigcreep.ne.08E9 if(cldwks.625.8 continue add on uncertainty SRA cladding if(sigcreep.67)/1.0E6 sag=0.0+sigcreep*0.8 fluenceLimback = fluence/10000.5/((ProblemTime)**0.eq.1547 if(icm.and.3))) edottherm = A*(E/tcak)*(sinh(ai*abs(sag)/E))**an*exp(Q/(R*tcak)) c c c c c c Irradiation Creep Rate c0SR = 4.eq.85 c2 = 1.0.or.0.eq.lt.9*tcak Q = 201000.0.0.0) tmod=570.0.lt.0.109* &(2.5) edottherm=edottherm+edotprimary endif c edot = edottherm + edotirr edot = edot*1.0.87473E24 c0 = C0SR if(cldwks.eq.0*(edottherm+edotirr)**0.0.5 E = 1.0) &edot = edot/(1.ne.0136*tmod if(cldwks.0E6 tcak = (CladAveTemp+459.6) edot = edot*0.0237+0.0 if(cldwks.or.gt.148E5 .5*esat*52.0) ctemp=3.216) return 100 c c c A.0sigcreep*0.0tanh(3.eq.05 edotprimary= &0.570.0sigcreep*0.707*sqrt((slst)**2+(stsr)**2+(srsl)**2)*6894.0.0 R = 8.00699132*tmod edotirr = c0*phi**c1*(abs(sag))**c2*ctemp c c c Primary Creep Rate if(ProblemTime.47E8 an = 2.0*(edottherm+edotirr)**0.and.0 ctemp=7.0exp(1.cldwks.0.0) &edot = edot*(1.and.0216*(edottherm+edotirr)**0.7573/1.4e27*fluenceLimback**1.0 if(tcak.0.0) &edot = edot*(1.5) A = 5.and.7573/1.ge.0) c0 = c0RXA c1 = 0.0985E24 c0RXA = 1.icm.0+sigcreep*0.5) an = 3.lt.59.eq.0.icm.0) &edot = edot/(1.55e4*(edottherm+edotirr)))**2.cldwks.eq.0.0 tmod=tcak if(tcak.eq.5*(ProblemTime)**0.314 ai = 650.0.216) if(sigcreep.0) tmod=625.gt.0 Thermal Creep Rate A = 1.56*(1.
c c cmhard = output meyer hardness of zircaloy cladding (n/m**2) c c ctemp = input cladding temperature (k) c c the equations used in this function are based on data from c (1) a. m.end A.2.94d+08) cmhard = 1.400. peggs and d.6034e012. morgan june 1978 t = ctemp cmhard = exp(2. p.5621e08*t** +3) return end FRAPTRAN 1. godin. the yield strength . peggs and d.2. d.2.baston in may 1974. 2 routine modified by PNNL for use in FRAPT c c the routine cmhard calculates meyer hardness as a function of c cladding temperature. godin.baston in may 1974.3502d05*ctemp**2 & .9 Cladding Meyer Hardness (CMHARD) FRAPCON3.5621d08*ctemp**3) if (cmhard . aecl .lt. icm) implicit real (ah.400.2.400.400. 1.2. stoute.3502e05*t**22. heat transfer coefficient c between uo2 and zircaloy . ross and r. morgan june 1978 c Modified by PNNL.400)icm c 400 continue c generic Zry property cmhard = exp(2.4.400. January 1997. a. d. m.248 (1975) c cmhard was coded by v.1552 (june 1962) c (2) i. c cmhard = output meyer hardness of zircaloy cladding (n/m**2) c ctemp = input cladding temperature (k) c the equations used in this function are based on data from c (1) a. aecl .hot c hardness relationship of zircaloy .6394e02*t+4.600. to clean up coding.4: CMHARD *deck cmhard function cmhard (ctemp. a.600. delete sensiti c uncertainty coding c go to (400. journal of nuclear c materials 57 pp 246 .6394d02*ctemp + 4. heat transfer coefficient c between uo2 and zircaloy . ross and r. p.400.6034d+01 . c modified by m.f.f.oz) c c This is a MATPRO11.1552 (june 1962) c (2) i. l.37 . stoute.4: CMHARD *deck cmhard function cmhard (ctemp) c c the routine cmhard calculates meyer hardness as a function of c cladding temperature.248 (1975) c c cmhard was coded by v. journal of nuclear c materials 57 pp 246 . Rev.94d+08 A.hot c hardness relationship of zircaloy .4. l. c modified by m. the yield strength .400.
ge.young march 1975 c adapted from routine cmix by p.3201 .G.lt. 6. 1981 c "aa" and "bb" parameters updated by W. A. the american society of c mechanical engineers (1967) c gthcon coded by r.3 (1968) c also: wisconsin electric power co. cvna127 (1962) c steam equation is from meyer. vol. c gmix(i) = input mole fractions of the gas mixture c the seven elements of gmix must sum to 1.10.1d0 .c.4586 .3666 . c gpthk = input effective gap thickness for knudsen domain (m) c (maximum of gap dimension or surface roughness) c not used if 0.034d0 .d0) cmhard=exp(26.brokaw. 1.5621d8*ctemp**3) if (cmhard .278d5*ctemp**3 + 1.016 / c r = sqrt(a) data r /2.1 Gas Thermal Conductivity (GTHCON) FRAPCON3. 9.80 .gandhi and c s.0d+05) cmhard = 1. + 5.d6*(2172. data. report nasa tr r81 c (1960). no.3. 1. aa(6).h' dimension gmix(7)..944 . 1.a. "thermodynamic and c transport properties of steam".131.3 Gas Material Properties A.2.1542 .. a(7).38 . jour.83. 50301 (jan 1973) c the accommodation factor is from r.gpthk) c c gthcon calculates gas thermal conductivity as a function of c temperature and gas fraction for seven gases: c gthcon = output gas thermal conductivity (w/mk). and eng. r(7). Luscher 101707 .3.6394d2*ctemp & + 4.s. conductivity of rare gases is based on j. docket no. bb(6) data a /4.2445 / c c The "aa" and "bb" parameters are from MATPRO11 Rev.macdonald include 'lacmdl.gtemp.13. 800.0d+05 end c A.2.dean.saxena.al.d0) cmhard=1. 800.lt.m. 2.0 c constituent gas number key c 1 helium c 2 argon c 3 krypton c 4 xenon c 5 hydrogen c 6 nitrogen c 7 water vapor c gtemp = input gas temperature (k) c gpres = input gas pressure (pa) c used for knudsen domain correction and for steam c not used if 0.41986 . et. c formula for gas mixtures is from r.02765d0*ctemp**2 . c(7).30 .8 + 18. 4.00075 .016 .7055d0*ctemp & + 0.28.return c c 600 continue Zr1%Nb from RRCKI if(ctemp . chem.gpres. 39.003 .e.c.3502d5*ctemp**2 .423d8*ctemp**4) if(ctemp .4 *deck gthcon function gthcon (gmix.11.
e9 / data on / 1 /./'0fractions'.973.15d0 if (gtemp.le. + off / 2 /.825e5.' k'.1.1668d09*gpres & /gtemp)**1.349e3.44d6*gtemp**1.le. 0.on) gthcon = emgton(gmix. 0.on) go to 170 if (gtemp.eq. do 160 i=1.6748. 0. gtemp='.8516d08 + 9.1pe12. Luscher 101707 tc = gtemp .15d0) & c(7) = gpres/gtemp*(2.gtemp.45d0 + 9.15d0) & c(7) = 4.le.4.eps) go to 150 rc = c(i)/c(j) ra = a(i)/a(j) fij = 1.973.gpres.7334.1.+2.7006. + locidx / 16 / if (emflag(locidx).5555.)*(ra.0.gt.l if (j.51d11*(tc)**3) if (gtemp.*(1.39 .le.4. do 150 j=1.2d0 + (17.G.+ra)) sum = sum+fij*gij*gmix(j) continue gthcon = gthcon+c(i)*gmix(i)/(gmix(i)+sum) continue continue return format ('1gthcon input bad.eq.( +10e12.009d0*gpres**2/gtemp**2 & /(tc)**4.273.45d005*(2.7146.eps) go to 120 l = i c(i) = aa(i)*gtemp**bb(i) continue if (gmix(7).eps) go to 140 l = 7 steam c(7) defination updated by W.l if (gmix(i).142)/(1.4 *deck gthcon function gthcon (gmix. oz) A.0000) go to 110 write (6.le.gmix stop l = 0 do 120 i=1. 0.gpres.424d10*gtemp & 6.gpthk) if (emflag(locidx).966e4.4)) end 100 110 c 120 c c 140 150 160 170 180 FRAPTRAN 1.eps) go to 160 sum = 0.gt.6d4 + 5.eq.0) go to 100 gthcon = 0.6 if (gmix(i).) go to 100 if (gtemp.092e4.c data aa/2.180) gtemp.005d14*gtemp**2) + 1.08d7*(tc)**2 .i) go to 150 if (gmix(j).gt.87d5*(tc) & + 1.+ra)**2 gij = (1.984e4/ data bb/0.9.3d0 if (l.41*(ra1. 0.le.gtemp.2.gpthk) implicit real (ah.531e3.7799/ c c data eps/1.+sqrt(rc*r(i)/r(j)))**2/sqrt(8.4.8408.
3666d0. the american society of mechanical engineers (1967) gthcon coded by r.c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c gthcon calculates gas thermal conductivity as a function of temperature and gas fraction for seven gases: gthcon = output gas thermal conductivity (w/mK).d9 / c c c c c c if (gtemp.003d0.0 constituent gas number key 1 helium 2 argon 3 krypton 4 xenon 5 hydrogen 6 nitrogen 7 water vapor gtemp = input gas temperature (K) gpres = input gas pressure (pa) used for knudsen domain correction and for steam not used if 0.c.a.2.13.1. vol.macdonald dimension gmix(7).2445d0 / The "aa" and "bb" parameters are from MATPRO11 Rev.825d5.6748d0.30d0. data.eps) go to 120 l = i c(i) = aa(i)*gtemp**bb(i) A.531d3.brokaw.180) gtemp.02d0.4.944d0.1542d0. 4.02d0/ r = sqrt(a) data r /2.41986d0.7006d0. 9.3201d0.gmix stop c 110 continue l = 0 c do 120 i=1. 2.5555.092d4.7146d0. conductivity of rare gases is based on j.d0) go to 100 if (gtemp.0. no.7334d0.6 if (gmix(i).0..al.0. 18. 8/1981 "aa" and "bb" parameters updated by W. jour. & 28. report nasa tr r81 (1960). 83. 131.7799d0/ data eps/1. bb(6) data a /4..3 (1968) also: wisconsin electric power co. r(7). aa(6).young march 1975 adapted from routine cmix by p.966d4. a(7).saxena.gt.80d0.0. 2.G.s.0.m. formula for gas mixtures is from r. 11.0000d0) go to 110 100 continue write (6.le.dean.349d3.016d0.0.c.gandhi and s.e.00075d0. 39.9.4586d0. c(7). 50301 (jan 1973) the accommodation factor is from r.40 .984d4/ data bb/0.1. docket no. gpthk = input effective gap thickness for knudsen domain (m) (maximum of gap dimension or surface roughness) not used if 0. cvna127 (1962) steam equation is from meyer. 1. et. "thermodynamic and transport properties of steam". gmix(i) = input mole fractions of the gas mixture the seven elements of gmix must sum to 1. Luscher 101707 data aa/2. & 5. 6. chem. and eng.8408d0.gt.
51d11*(tc)**3) c c if (gtemp .142d0)/(1.le.41d0*(ra1.1pe12.87d5*(tc) & + 1.l if (gmix(i). Rev.G./' fractions'.4)) c end A.45d05*(2.15d0) & c(7) = gpres/gtemp*(2.08d7*(tc)**2 .424d10*gtemp & .41 .le.6.8516d08 + 9. Luscher 101707 tc = gtemp 273.4.d0 do 150 j=1. 2) c(7) definition updated by W.l if (j.04e74.15d0)**4.i) go to 150 if (gmix(j).6d4 + 5.009d0*gpres**2/gtemp**2 & /(gtemp273.44d06*gtemp**1.eq.le.51e11*tc)) 140 continue if (l.' K'.d0 c do 160 i=1.120 continue c if (gmix(7). 973.87e5+tc*(1.d0+sqrt(rc*r(i)/r(j)))**2/sqrt(8. & (10e12.d0*(1.d0+ra)**2 gij = (1.1668d09*gpres & /gtemp)**1. gtemp= '.0) go to 100 gthcon = 0.eps) go to 160 sum = 0.15d0) & c(7) = 4.eps) go to 150 rc = c(i)/c(j) ra = a(i)/a(j) fij = 1.gt.d0)*(ra.le.45d0 + 9.d0+ra)) sum = sum+fij*gij*gmix(j) 150 continue gthcon = gthcon+c(i)*gmix(i)/(gmix(i)+sum) 160 continue c return c 180 format ('gthcon input bad.15d0 if (gtemp .005d14*gtemp**2) + 1.d0+2. 973.le.eps) go to 140 l = 7 steam (from MATPRO11.6e3+tc*(5.4.3d0 c c(7) = 17.2d0 + (17.
This action might not be possible to undo. Are you sure you want to continue?
We've moved you to where you read on your other device.
Get the full title to continue reading from where you left off, or restart the preview.