ML20009D421
| ML20009D421 | |
| Person / Time | |
|---|---|
| Site: | Calvert Cliffs  |
| Issue date: | 07/31/1981 |
| From: | ABB COMBUSTION ENGINEERING NUCLEAR FUEL (FORMERLY |
| To: | |
| Shared Package | |
| ML20009B075 | List: |
| References | |
| CEN-161(B)-NP, NUDOCS 8107240046 | |
| Download: ML20009D421 (69) | |
Text
_ _ _ _ - -..-
CEN-161 (B)-NP IMPROVEMENTS TO FUEL EVALUATION MODEL JULY,1981 g, i c=l SYSTEMS POWER i =i 8107240046 810707 DR ADOCK 05000 COMBUSTION ENGINEERING, INC.
CEN-161(8)-NP IMPROVEMENTS TO FUEL EVALUATION MODEL REACTOR DESIGN FUELS DEVELOPMENT JULY, 1981 2 't -
Combus.'. ion Engineering -Inc.
i l
l l
l l
ABSTRACT A new version of the FATES fuel performance code has been developed whien gives improved predictions of fuel rod temperature distributions and internal gas pressures as a function of mechanical design and operating history.
The improvements to FATES are of particular significance at high fuel burnups.
This report describes the new models which have been incorporated in the improved version, denoted as FATES 3, the most significant of which is a new fistion gas release model. A data base is provided which demonstrates the precictive capability of the FATES 3 code.
e e
1
TABLE OF CONTENTS Page
1.0 INTRODUCTION
1-1 1.1 Purpose 1-1 1.2 Scope and Description 1-1 1.3 References for Section 1.0 1-2 2.0 FISSION GAS RELEASE 2-1 2.1 Introduction 2-1 2.2 Sumary 2-1 2.3 Discussion 2-4 2.4 Application to a Variable Fuel Temperature History 2-6 2.5 Correlation Data Base and Results 2-7 2.6 References for Section 2.0 2-8 3.0 FUEL PELLET SWELLING 3-1 3.1 Introduction 3-1 3.2 Discussion 3-1 3.3 References for Section 3.0 3-4 4.0 FUEL-CLAD INTERFACE 4-1 4.1 Introduction 4-1 4.2 Pellet-Clad Contact Loading 4-1 4.3 Gap Conductance 4-2 4.4 References for Section 4.0 4-2 5.0 ANNULAR FUEL PELLETS 5-1 5.1 Introduction 5-1
[,
5.2 Temperature Distribution 5-1 5.3 Thermal Expansion 5-3 l
5.4 Void Volume 5-4 5.5 References for Section 5.0 5-4 l
6.0, FUEL PELLET RELOCATION 6-1 i
l 6.1 Introduction 6-1 l
6.2 Discussion 6-1 63 References for Section 6.0 6-1 7.0 CLAD AXIAL IRRADIATION GROWTH 7-1 7.1 References for Section 7.0 7-1 8.0 PLENUM GAS TEMPERATURE 8-1 l
l t
11 l
As TABLE OF CONTENTS Page 9.0 PREDICTIONS OF EXPERIMENTAL DATA 9-1 9.1 Purpose for Data Selections 9-1 9.2 Data and Results 9-2 9.3 Conclusions 9-5 9.4 References for Section 9.0 9-6 I
e l
l l
I l
iii
LIST OF TABLES Table Pace 2-1 The Correlation Data Base - FATES 3 Predictions of 2-10 Gas Release from Calvert Cliffs - I Rods 2-2 The Correlation Data Base - FATES 3 Predictions 2-11 of Gas Release from Over-Ramp Program Rods 2-3 Summary of Design Parameters for Rods in the 2-12 Correlation Data Base 2-4 Summary of Thermal-Hydraulic Parameter for Rods 2-13 in the Correlation Data Base 9-1 Summary of Irradiation Parameters for Rods 9-7 in the Verification Data Base 9-2 Summary of Design Parameters for Rods in the 9-8 Verification Data Base 9-3 Summary of Thermal-Hydraulic Parameters for Rods in 9-9 the Verification Data Base 9-4 Irradiation and Design Parameters for Petten Fuel Rods 9-10 with Measured and Predicted Fission Gas Release Comparison of Gas Release Model Predictions wjth Data 9-11 9-5 of Bellamy and Rich (Gap Resistivity = 2
'C/W) cm 9-6 Comparison of Gas Release Model Predictions with Data 9-12 of Bellamy and Rich (Gap Resistivity = 1.5 cm2
- C/W)
I I
iv
LIST OF FIGURES l
Figure Page 2-1 Instantaneous Portion of the Total Fission 2-14 Gas Release 2-2 Grain Size Dependence of the Instantaneous 2-15 Fission Gas Release 2-3 Time to Achieve an Equilibrium Grain Size 2-16 as a Function of Initial Grain Size and Local Fuel Temperature 2-4 Instantaneous and Grain Growth Related 2-17 Portions of the Total Fission Gas Release as Functions of Local Fuel Temperature and Time at Temperature for a 5 um Initial Grain Size 2-5 Instantaneous and Grain Growth Related 2-18 Portions of the Total Fission Gas Release as Functions of Local Fuel Temperature and Time at Temperature for a 10 um Initial Grain Size 2-6 Instantaneous and Grain Growth Related 2-19 Portions of the Total Fission Gas Release as Functions of Local Fuel Temperature and Time at Temperature for a 20 nm Initial Grain Size 2-7 Power and Time Dependence of Fission Gas 2-20 Release at 10 MWD /KgU F-8 Power and Time Dependence of Fission Gas 2-21 Release at 25 MWD /Kgu 2-9 Example Application of the Gas Release 2-22 Model to a Variable Temperature History 2-10 A Comparison of FATES 3 Predictions with 2-23 Measured Fission Gas Release from Calvert Cliffs I Rods and Studsvik OVER-RAMP Rods (The Model Correlation Data Base)
~
3-1 Fuel Density versus Burnup for an Assumed 3-6 Fuel Densification of 1.6% TD 9-1 A Comparison of Model Predictions vs. Measured 9-13 Fission Gas Release for the Bellamy and Rich Data V
t LIST OF FIGURES Figure Page 9-2 Comparison of Measured and Predicted Fission 9-14 Gas Release 9-3 Predicted-Measured Gas Release versus Burnup 9-15 9-4 Comparison of Measured and Predicted Fuel 9-16 Centerline Temperatures 9-5 Comparison of Measured and Predicted Void 9-17 Volume for Calvert Cliffs I Test Rods G
9 l
l l
vi
1.0 INTRODUCTION
1.1 PURPOSE The purpose of this report is to describe the improvements made to the Combustion Engineering (C-E) fuel performance analysis code, FATES, which is a component of the C-E Fuel Evaluation Model described in Reference 1-1.
The improved code version is referred to as FATES 3.
FATES 3 is used to predict fuel rod temperature distributions and internal gas pressures as a function of mechanical design and operating history.
Results from FATES 3 are used in fuel rod design (e.g.,for setting init:a1 fill gas pressure) and safety analyses.
This report will:
a.
Describe the additions and modifications to the FATES fuel performance code, which have resulted in the new code version referred to as FATES 3.
b.
Describe the experimental data base used to develop the new models incorporated in FATES 3.
c.
Provide comparisons of FATES 3 predictions with experimental data from commercial and test reactors.
1.2 SCOPE AND DESCRIPTION C-E compiles and reviews fuel behavioral data from the open literature and its own programs on a continuing basis in order to better understand fuel behavior and to develop new models for incorporation in its fuel performance codes.
Data developed since 1974, the date of inception of the FATES model, extend to higher burnups and are applicable to a wider variety of operating conditions. The improvements made to FATES which more accurately predict fuel behavior, especially at high burnup, are in the following areas:
1-1
- fission gas releasa
- fuel pellet swelling
- pellet-clad interface treatment
- fuel relocation In addition, the capability of modeling annular fuel pellets has been included
~
in FATES 3.
This capability is particularly useful for the prediction of fuel temperatures in experimental rods where centerline thermocouples have been incorporated to produce temperature measurements, and in the evaluation of the performance of advanced fuel designs containing annular fuel pellets.
The plenum gas temperature and clad axial irradiation growth were input to the previous FATES code version. These calculations are performed internally in FATES 3.
Axial growth is calculated with a previously approved model.
The most significant improvement to FATES is in the area of fission gas release modeling.
The fission gas release rate depends on the amount of retained fission gas, burnup, temperature, and fuel grain size.
Fission
~
gas release concomit. ant with fuel grain growth is also taken into account.
Fuel pellet swelling consists of solid and gaseous components and comences after maximum fuel densification is reached at 4000 MWD /MTU.
The gaseous component decreases fuel density and decreases fuel thermal conductivity.
Fuel density was held constant after 4000 MWD /MTU in the previous version of FATES. A new value for the overall fuel volumetric swelling rate also has been incorporated.
The previous FATES model used a preassigned limit on gap conductance in lieu of an explicit treatment of the pellet-clad interface. A model has been incorporated l
in FATES 3 which explicitly treats pellet-clad mechanical interaction. The gap conductance is no longer limited to the values given in Reference 1-1.
Fuel pellet relocation is a term used to describe the outward displacement of pellets into the gap region between fuel and clad resulting from fuel cracking.
The FATES 3 pellet relocation model is the model originally proposed in Reference 1-1.
l
1.3 REFERENCES
FOR SECTION 1.0 l
l-1 Combustion Engineering, Inc., "C-E Fuel Evaluation Model Topical Report", CENpD-139, July,1974. (Froprietary) l 1-2 l
l
2.0 FISSION GAS RELEASE
2.1 INTRODUCTION
The prediction of fission gas release is important in the evaluation of fuel perfonnance because gas conductivities, fuel temperatures and fuel rod internal pressures are dependent on the amount of fission gas released from the fuel.
An empirical model to calculate fission gas release was developed by C-E for use in FATES and was described in CENPD-139 (Ref. 2-1).
The model was calibrated against the data from U0 fuel available at that time 2
which were limited to burnups below 10,000 MWD /MTU.
Although from theoretical considerations, fission gas release is known to be dependent on temperature, burnup and fuel microstructure, the lack of a sufficiently well-characterized data base precluded a separation of the effects of these variables on gas release.
As a result, temperature was used as the only explicit variable in most LWR models for fission gas release (Ref. 2-2).
In the CENPD-139 model, a burnup dependence was indirectly accounted for, in the temperature region below columnar grain growth, by a term which varied with the square root of irradiation time.
Recently, with the availability of some gas release data at higher burnups, the burnup effects on gas release have received increased attention (Ref. 2-3).
A review of some of these data led the NRC to conclude that the burnup sensitivity of gas release is stronger than recognized earlier.
C-E initiated an analytical and experimental investigation of fission gas release.
Emphasis was placed on well characterized fuel rods representing mcdern PWR's and irradiated over a wide range of monitored parameters.
The new model presented herein accounts for the effects of temperature, burnup, and grain size.
The model is described and the comparisons with experimental data are also presented.
2.2
SUMMARY
In the C-E fission gas release model, gas release is calculated by following the local inventory of retained fission gas in the fuel. At each axial region of the fuel column, the fuel is divided into ten rings of equal thick-ness and the local inventory of fission gas is followed in each of these rings.
Local fuel temperature, burnup, grain size and irradiation history are 2-1
variables affecting the inventory of retained fission gas in the following manner:
(2-1)
(2-2)
(2-3) i The percent of generated fission gas that is released, F, is calculated from:
(2-4)
The functional relationships assumed in Equations 2-1 and 2-2 are based on l
an inspection of the shapes of the experimentally determined curves of the retained inventory of fission gas in small UO fuel samples at high burne:c 2
(Refs. 2-4 and 2-5).
The specific values of the constants in the expression for K, given by Equation 2-2, have been arrived at by correlating the gas release predictions of the overall gas release model, when employed in the FATES 3 code, to the experimenta:
2-2
data obtained from the steady state irradiation of commercial fuel rods in Calvert Cliffs-1 (Refs. 2-6 and 2-7) and from ramp tests performed at Studsvik as part of the OVER-RAMP program which included C-E segmented commercial fuel rods irradiated in Obrigheim (Ref. 2-8). The maximam inventory 'btained by applying Equation 2-3 is equivalent to the release predicced by the low-temperature gas release model developed by the ANS 5.4 Committee (Ref. 2-9).
Fission gas release that is accompanied by grain growth (via grain boundary sweeping) is accounted for in the model by an additional term which depletes the fission gas previously retained in the volume of fuel which is swept by moving grain boundaries. The local inventory of fission gas that remains in each ring of fuel after a local grain growth from G to G is given by:
7 (2-6)
The kinetics of grain growth are followed in each fuel ring M
ppm GMm (2-6) l f
2-3 l
l
(2-7)
O L_.
2.3 DISCUSSION The basic approach that 5as been adopted in calculating fission gas release in this model is to follow the retained inventory of fission gas in each of ten radial rings of the fuel in up to 20 axial nodes.
The release is then calculated by subtracting the amount of gas retained from the total amount of gas generated.
This approach was based on 3 review of data on retained fission gas in high burnup UO fuels 2
published by Zimmermann of Karlsruhe (Refs. 2-4 and 2-5).
These data were 9enerated by utilizing specially designed test rigs for irradiating small and thin discs of UO at nearly isothermal conditions.
Fuel temperatures 2
were monitored with thermocouples, and burnups of up to 89 MWD /KgU were achieved by utilizing ' specimens of 15-20% enrichment.
Fuel temperatures ranged from 1100 to 1900*K.
These results show that at any temperature, the fission gas retained in the UO fuel increases with burnup and tends to 2
reach a saturation level. The burnup at wnich the saturation inventory is reached varies with temperature, and the saturated inventory goes down exponentially with increasing temperature.
These experimentally observed trends were used as inputs in developing the functional instantaneous dependency of the retained inventory, and the released inventory, of fission gas on fuel temperature and burnup.
Figure 2-1 illustrates this relationship for a fuel with an initial grain size of 5 um based on Equations 2-1 through 2-4.
2-4
In addition to temperature and burnup, the dynamic grain size is also used as a variable affecting fission gas retained in the fuel.
From theoretical grounds, a grain size effect is believed to originate primarily from the following two factors:
- 1) variation in the diffusion distance of fission gas to the grain boundaries as a function of grain size and 2) variation in the rate of grain growth with grain size.
Experimental support for both of these effects up to 4 K4D/Kgu is available from the work of Turnbull (Ref. 2-11). A continued influence of grain size on fission gas release at higher burrups has been documented in a number of C-E publications (Refs. 2-12 and 2-13).
Both aspects of the grain size effect discussed above are treated in the model.
This effect is illustrated in Figure 2-2 for fuels of various grain sizes at a burnup of 24 MWD /KgU, based on Equations 2-1 through 2-4.
The grain growth effect is modeled by incorporating the additional inventory reduction term of Equation 2-5, which accounts for fission. gas release caused by grain boundary sweeping.
E 1
Figure 2-3 shows the number of hours necessary
~
to achieve an equilibrium grain size for v6rious starting grain sizes and
[
local fuel temperature when these equations are employed.
Figures 2-4 through 2-6 show the additional gas release tht occurs through the grain growth mechanism depending on time at temperature and the initial grain size (the zero-hour curves represent the instantaneous release contribution previously illustrated alone in Figure 2-d.
Figures 2-7 and 2-8 show the instantaneous and grain growth portions of gas release when integrated across a fuel pellet cross section at 10 and 25 K4D/KgU for instantaneous transients from a power level of 10 kw/ft.
2-5 l
2.4 APPLICATION TO A VARIABLE FUEL TEMPERATURE HISTORY m
m.
l ll
- 1 end
\\
l I
b e-e em W
i a
l l
~
l l-l m
N (2-8) p e
W 4
e em M
uns 2-6
D 2.5 CORRELATION DATA BASE AND RESULTS The dependancies of the fission gas release model on burnup, temperature and grain size were established by correlating the model predictions with measured gas release data from t'wo different sources ( a steady state irradiation source and a ramp test data source).
Both sources include prepressurized PWR fuel rods that were fabricated by C-E. ~ The final correlation which resulted in the model presented herein was done through the explicit use of this model in the FATES 3 code and consisted of establishing the final form of the burnup dependence in Equation 2-1 and the values of the constants in Equation 2-2.
The results of the correla, tion exercise are discussed in this section.
Additional data l
comparisons have been made to verify the model agairist an independent data base in Section 9.0.
The correlation data base included:
l measured gas release values from full-length prepressurized C-E
~~fuel rods irradiated at Calvert Cliffs-I through 1 to 3 cycles
~~
measured gas release values from C-E and KWU fuel rod segments
~~
irradiated at Obrigheim through 1 to 3 cycles, and subsequently ramp tested without failure at the R-2 reactor as part of the Studsvik OVER-RAMP Program
~~
measured gas release values from fuel rods irradiated
'a~t BR-3 during Cycle 4A Snd subsequently ramp tested without failure at the'R-2 reactor, also as part of the OVER-RAMP Program.
The important design and operating variables as well as comparisons of measured and FATES 3 predicted gas release values are summarized for these data sources in Tables 2-1 through 2-4.
The rods included represent a range in peak burnup a range in peak LHGR and a range 2-7
inir.itialgrainsize}
A sumary of the comparisons of measured and predicted fission gas ralease is given in Figure 2-10 for
~~
all rods in the correlation data case. This figure demonstrates the excellent
~~
prediction capability of the FATES 3 code with the use of C-E's new gas release model.
It is noted that fission gas release. from the Calvert Cliffs I rods having an old-type densifying fuel is more conservatively predicted as a group.
2.6 REFERENCES
FOR SECTION 2.0 2-1 Combustion Engineering, Inc., "C-E Fuel Evaluation Model Topical Report", CENPD-139, July,1974. (Proprietary) 2-2 Core Perfonnance Branch, USNRC, "The Role of Fission Gas Release in Reactor Licensing", NUREG-75/077, November 1975.
9 2-3 R. O. Meyer, C. E. Beyer and J. C. Voglewede, " Fission Gas Release from Fuel at High Burnup", Office of Nuclear Reactor Regulation, USNRC, NUREG-0418, March 1978.
2-4 H. Zimmermann, " Investigation of Swelling and Fission Gas Retention on Oxide Nuclear Fuel Under Neutron Irradiation", Kernforschung. entrum, Karlsruhe, KFX-2467, June 1977.
2-5 H. Zimermann, " Investigation of Swelling and Fission Gas Behavior in Uranium Dioxide", J. Nucl. Mat., Vol. 75 (1978) pp.154-161.
i 2-6 S. R. Pati, " Gas Release and Microstructural Evaluation of One-and Two-Cycle Fuel Rods from Calvert Cliffs-I", Combustion Engineering, Inc.,
l NFSD-75, March 1979.
l 2-7 S. R. Pati, " Gas Release and Microstructural Evaluation of Three-Cycle Fuel Rods from Calvert Cliffs -
I", Combustion Engineering, Inc.,
C-E NPSD-ll9, December 1980.
2-8 T. Hollowell et. al., "The International Over-Ramp Project at Studsvik, Proposed Paper for ANS Topical Meeting on LWR Extended Burnup Fuel Performance and Utilization," April 4-8, 1982, Williamsburg, Virginia.
l 2-8 1
J
2-9 ANSI /ANS-5.4, " Proposed America.. idtional Standard Method for Calculating the Release of Fission Products from 0xide
~
Fem.s", ANS 5.4, Draft, November, 1979.
2-10 J. B. Ainscough, B. W. Oldfield and J. O. Ware, " Isothermal Grain Growth Kinetics in Sintered U0 Pellets", J. Nucl. Mat.
2 Vol. 49 (1973/74) pp.117-128.
0 2-11 J. A. Turnbull, "The Effect of Grain Size on the Swelling and Gas Release Properties of UO During Irradiation", J. Nucl. Mat.
2 Vol. 50 (1974) pp. 62-68.
2-12 S. R. Pati, " Structure Sensitivity of Fission Gas Release in LWR Fuels", American Ceramic Society Bulletin, Vol. 56, No. 8 (1977) p. 735.
2-13 S. R. Pati, " Gas Release and Fuel Microstructure", American Ceramic Society Bulletin, Vol. 57, No. 3 (1978), p. '357.
2-14 F. Sontheimer, P. Dewes, and H. Knaab, " Correlating Fission Gas Release Due to Power Ramping with Observed Microstructural Changes in the Fuel by Use of a Simple Diffusion Model", Paper Presented at the Enlarged Halden Prograane Group Meeting on Water Reactor Fuel Performance", June 14-19,1981.(Proprietary)
I 2-9 i
Table 2-1 The Correlation Data Base-FATES 3 Predictions of Gas Release From Calvert Cliffs-I Rods Rod rae! Stability Initial Peak Averaged Rod to In-Reactor Grain LHGR Burnup
% Gas Release Numoer Densification Size, um kw/ft(BOL)
Mwd /kgU Measured Predicted 01 Densifying 2.5 9.1 18.7 0.27 8.25 46 Nondensifying 15 11.1 21.6 0.71 0.84 50 Nondensifying 4
9.1 18.7 0.33 1.41 05 Densifying 2.5 9.1 25.8 0.34 6.71 47 Nondensifying 15 11.1 29.1 0.64 0.72 51 Nondensi fying 4
9.1 25.8 0.35 1.15 11 Densifying 2.5 9.1 33.0 0.36 5.53 12 Densifying 2.5 9.1 33.0 0.35 5.46 39 Nondensifying 15 11.1 37.0 0.71 0.61 42 Nondensifying 15 11.1 37.0 0.72 0.69 53 Nondensifying 4
9.1 33.0 0.33 1.00 60 Nondensifying 7
9.1 33.0 0.59 0.84 i-(
{
2-10 i
l
Table 2-2 The Correlation Data Base FATES 3 Predictions of Gas Release From Over-Ramo Program Rods Ramp Rod Initial Peak Averaged Rod Grain LHGR, a,b
- Burnup,
% Gas Release Nurter Size, um kw/ft Nd/kgU Measured Predicted o
9 e
M e
e e
a~
}
4 b
~
eau m
2-11 i
Table 2-3 Sumary of Design Parameters For Rods in the Correlation Data Base Parameter Over-Ramp Calvert Cliffs 1 Clad 00, IN
.440 s
Clad ID, IN
.388 Initial Pellet-Clad Diametral Gap, Mfis 8.5 Initial Grain size, um 2.5 - 15 Fuel Column Length, IN 136.7 m
Initial Fuel Density, % TD 93 - 95 0
Fill Gas Pressure at 70 F, psia
- 315. 465 Enrichment, % U235 2.5, 2.8 M
W e
em A
Table 2-4 Summry of Thennal-Hvdraulic Parameters Fa - Rods in the Correlation Data Base i
Parameter Over-Ramp
- Calvert Cliffs 1 1
j Coolant Pressure, psia 2250 i
Coolant Inlet Temperature. F 548 ro
.L w
- The coolant pressure and inlet temperature during power ramping are
]psiaand respectively.
i
E i
Ui 5
w NL w
a 0
z2 m
u_*
~
LOCAL FUEL TEMPERATURE, OC Figure 2-1 Instantaneous Portion of the Total Fission Gas Release
d l
1 i
b 1
(
1 i
w i
m l
w
~,
m
=
m I
<o zo e--e m
m i
C i
1 l
LOCAL FUEL TEMPERATURE, DC Figure 2-2 Grain Size Dependence of the Instantaneous fission Gas Release
O t
I HOURS TO ACHIEVE EQUILIBRIUM GRAIN SIZE l
l 4
1 2
io c
5
'?
I
? * :!
- o. R i
- to E E m
ao O
E, r+ n, F
c o
w m cu o -
O
.' E " E 3>
C m a o,, a r cn s$ M
- 3., s:
c5 C.$.. m~D o,
C
~d
-* w cn b a '. Ef
-z ra m rn LA m.. cn
- i m
1 n
M i=
n r-N 3
8 i
I
'i:,!
- i :,
1 i
"LN
.JoO<a Cwwooz o<w Ew$wu e e
I F
i gure 2
4 aGI t anss T
t e
Ra
-men p l t eea ran aseeo tu u
ras es L
a O
ffn oud C
rn A
cG atr L
ia Soi F
nn U
e s E
m G
L or ifo T
n w
E iLtoh M
t
- i. c P
i aR l l e ER l
Cfa A
- r. ut iee T
i l d U
n TP R
seo
,E imr nptei O
ro C
an ts u
ro ef at nh de TT io rrtaa l
F i
ss ion
i I
i allt uI m
t.u cr to m
.L m
o Z
.O.
i m
g E
a<
O O
._J LOCAL FUEL TEMPERATURE, OC figure 2-5 Instantaneous and Grain Growth Related Portions of the Total fission Gas Release as functions of local fuel Teniperature l
.itul 1 it.:#5.ii Teioner.it aire 1 o.i 10 i:in liii i i.il r,r.i ivi s i r..
l
61-2 2
e C
a
~
e en g a? g LOCAL FISSION GAS RELEASE, %
, -m G W W W =
-- *ii E!
o nc
-C M
= _.. =.
t c. a wa d 3 O M
- 1 O
3 o-
,%3 W
srO Lo. o=c o N -e O
p c x t--
-. r1 W c
- " m
$$I is > ;;:l.
2 Si ig: -
E1 u 9 o m o3 y c, " 3>
soH
- c. % c
--4 e x g o" -
_-d* = (T)
~ o
$0 0 rt (D m
. = *
- E8 2 !!?
m I
Figure 2-7
~
POWER AND TIME DEPENDENCE OF FISSION GAS RELEASE AT 10 MWD /KGU l
ui E
"i us E
I e
5 i
E l
l POWER, KW/FT 2-20
Figure 2-8 POWER AND TIME DEPENDENCE OF FISSION GAS RELEASE e
AT 25 MWD /KGU l
e 5
d=
4e 29 8
ir 4
e M
POWER, KW/FT 2-21
w tE E
3 d
2 te o
3 l
ACTUAL BURNUP FOR FISSION GAS GENERATION to O
2O cnd u
O 8
1 -
5e i
Eg i
ct:
l l*
PSEUD 0 BURNUP FOR RETAINED INVENTORY OF FISSION GAS Figure 2-9 Example Application of the Gas Release Model to a Variable Temperature History 2-22
ui I*
.swz Eo S
U5 m
E MEASURED GAS RELEASE, %
Figure 2-10 A Comparison of FATES 3 Predictions with Measured Fission Gas Release from Calvert Cliffs I Rods
(
and Studsvik OVER-RAMP, Rods (The Model Correlation Data Base) l l
t l
2-23
0.0 FUEL PELLET SWELLING
3.1 INTRODUCTION
The fuel swelling modal in FATES has been revised by taking into consideration some of the recent data in the open literature as well as the data en post-irradiation density changes of C-E fuels irradiated in Calvert Cliffs-1 through three cycles.
These data indicate that in the range of comercial operation, the unrestrained swelling rate of LWR fuels is lower than the rate used in Reference 3-1.
The technical basis for the modification of the swelling rate is discussed below.
3.2 DISCUSSION The previcus model for calculating diametral swelling of fuel pellets in FATES (Ref. 3-1) assumed isotropic swelling in the burnup range from 4,000 MWD /MTU until hard contact occurs between fuel and cladding.
The diametral swelling rate was set equal to one-third the pellet average volumetric rate of 0.6% AV/V per 4,000 MWD /MTU burnup. The basis for this volumetric swelling rate was the data on plate fuel elements obtained by Bettis in the early 1960s (Ref. 3-2).
These data were reanalyzed by Rowland et al (Ref. 3-3),
and swelling rates of 0.3 to 0.5% aV/V per 4,000 MWD /MTU were obtained by a least-squares fit of the data. Additional data and analyses were also presented by these authors which suggested that the typn.41 swelling rate for LWR fuels would be significantly lower than that originally published fuel rods irradiated by Bettis.
For example, precise measurements of UO2 in GETR up to 91,000 MWD /MTU yielded a swelling rate of 0.4% AV/V per 4,000 MWD /MTU. A number of other investigators also reported swelling rates in UO which are lower than the value used before.
For example, 2
Brucklacher and Dienst (Ref. 3-4) derived a rate of 0.32%AV/V per 4~,000 MWD /MTU from in-pile creep measurements.
Collins and Hargreaves (Ref. 3-5) reported
~
that measurements on high-burnup fuel yield a swelling rate of 0.2% AV/V per 4,000 MWD /MTU, whereas interpretation of length changes of fuel pins irradiated in Windscale AGR yielded a rate of 0.4% AV/V per 4,000 MWD /MTU.
More recently, the swelling behavior of a numbe' d PWR fuels have been characterized by analyzing the changes in d%1rca Of fuel pellets as a 3-1 w.
._,,..,.w.
,__._..__._,_-_..__-_,7-
function of burnup.
Data presented by Assmann and Manzel of KWU (Ref. 3-6) for standard fuel of 95% TD showed a pore-free matrix swelling rate of 0.4% per 4,000 MWD /MTU in the burnup range of 12,000 to 45,000 MWD /MTU.
High burnup swelling characteristics of three fuel types (one densifying and two nondensifying fuels) were recently characterized by C-E (Ref. 3-7) as part of an EpRI program.
The swelling rates were deduced by analyzing the rates of density dec eases of these fuels as a function of burnup in the range of 20,000 to 41,000 MWD /MTU.
Specimens for density measurements were taken from fuel rods with two and three cycles in Calvert Cliffs-1. To musmize the perturbation from in-reactor densification, only the data above 20,000 MWD /MTU burnup were used.
Each of the three fuel types showed lower bulk swelling rates than the pore-free matrix swelling rate of 0.4% AV/V per 4,000 MWD /MTU. The observed variation in the bulk swelling rate was related to the variations in the fuel microstructure.
For example, the 93%
TD nondensifying fuel showed the lowest bulk rate of 0.24% per 4,000 MWD /MTU compared to the highest observed rate of 0.37% per 4,000 MWD /MTU in the 95%
TD nondensifying fuel.
It is believed that the lower swelling rate of the lower density fuel resulted' from greater swelling accommodation in this fuel type. Higher swelling accomodation in this fuel was facilitated by the presence of a large fraction of initial porosity as open pores distributed in the inter-agglomerate region of the pellet microstructure. These results indicate that although the bulk swelling rate deduced from pest-irradiation imersion densities may vary appreciably due to different degrees of swelling accommodations, the rate of 0.4% AV/V per 4,000 MWD /MTU provides a good estimate of the unrestrained swelling rate of U0 fuel at high burnups.
2 This unrestrained rate is equivalent to the swelling rate of a pore-free l
matrix published by Assmann and Manzel (Ref. 3-6).
Further justification for the swelling rate selected is available from the good l
agreement that is observed between the internal fuel rod void volume calculated by using this swelling rate and the void volume raeasured in twelve Calvert Cliffs-1 fuel rods.
The internal void volumes of these rods were measured l
in a hot cell at the end of each of the first three cycles of irradiation (Refs.
l
[
3-8 and 3-g).
Fuels from the two-and three-cycle rods from this group were subsequently used for post-irradiation density measurements discussed above.
These fuel rods were well characterized in terms of the as-fabricated dimensions, l
fuel pellet attributes, and detailed irradiation histories.
The beginning-of-life 3-2
(BOL) void volumes were estimated from puncturing archive rods, from the initial as-fabricated dimensions and from the open porosity measured in archive fuel pellets (Refs. 3-8 and 3-9).
The dimensional changes of the fuel rods, during irradiation were measured during poolside inspections (Refs. 3-10 and 3-11).
The void volume changes due to fuel densification were estimated by an upward
~
ad,justment of the resintering test data so that a higher estimate of the fuel swelling is obtained.
The void volumes at the end of each cycle were calculated by FATES 3 following the detailed power histories. As shown in Section 9.1 good agreement exists between the predicted and measured void volumes.
i l
The above considerations provide a basis for the use of an unrestrained swelling rate of 0.4% AV/V per 4,000 MWD /MTU.
No change in the swelling model (Ref. 3-1
) has been made in the range of zero to 4,000 MWD /MTU (i.e.,
co swelling is assumed below 4,000 MWD /MTU) and after fuel-cladding contact is predicted to occur by FATES 3.
Since the two-and three-cycle Calvert Cliffs-1 fuel rods operated well beyond the onset of fuel-cladding contact, and the predicted void volumes agree well with the measured void volumes, the integrated swelling mcdel (including accommodation of swelling volume in closed and open pores, into dishes, and the fuel-clad gap) is, therefore, inferred to be valid to a significant range of burnup (38 MWD /KgU rod average).
l During periods of peDet-clad contact, the difference between restrained and unrestrained swelling fills the fuel column internal voids in proportion to BOL amounts of internal volume, l
l The total swelling rate of 0.4% AV/V per 4,000 MWD /MTU results from the combination of swelling due to solid and gaseous fission products.
Theoretical estimates of solid fission product swelling in U0 vary fr m 0.13 to 0.35%
2 AV/V per 4,000 MWD /MTU (Refs. 3-12, 3-13, and 3-14).
The minimum rate of these estimates 'is generally obtained by assuming a complete utilization of vacancies created by fissioning of uranium atoms and the maximum rate by assuming no utilization.
The average of the above estimate 0.24% aV/V per 4,000 MWD /MTU, is taken as the volumetric swelling rate due to solid fission products. The remaining swelling rate of 0.16% AV/V per 4,000 MWD /MTU is assumed to be due to gaseous fission products and is used to reduce fuel l
density and, therefore, fuel conductivity, beyond 4,000 MWD /MTU.
Gaseous fission product swelling was not treated in the previous FATES code version.
3-3
Figure 3-1 shows fuel density as a function of burnup for the combined fuel densification (Ref. 3-1) and swelling models. The assumed densification change is 1.6% of theoretical density (TD) for this example.
3.3 REFERENCES
FOR SECTION 3.0 3-1 Combustion Engineering, Inc., "C-E Fuel Evaluation Model Topical Report", CENPD-139, July,1974. (Proprietary) 3-2 R. C. Daniel, M. L. Bleiberg, H. B. Meieran and W. Veniscavitch, Bettis P :mic Laboratories, WAPD-263,1962.
3-3 T.C.Rowland,M.b.MarloweandR.B.Elkins,"FissionProduct Swelling in BWR Fuels", Tran. Am. Nucl. Society, Vol. 18 (1974) 124.
3-4 D. Brucklacher and D. W. Dienst, J. Nucl. Mat. Vol. 42 (1972) 285 3-5 D. A. Collins and R. Hargreaves, " Performance Limiting Phenomena in Irradiated U0 ", BNES Conf. London (1973) 50.1 2
3-6 H. Assmann and R. Manzel, Kraftwerk Union AG, "The Matrix Swelling Rate of UO ", J. Nucl. Mat. Vol. 68 (1977) 360-364.
2 3-7 N. Fuhrman and P. A. Van Saun, Combustion Engineering, Inc., "Densi-fication and Swelling Behavior of Two-and Three-Cycle UO Fuels 2
from Calvert Cliffs-1", CE NPSD-135, March 1961.
3-8 S. R. Pati, Combustion Engineering, Inc., " Gas Release and Micro-structural Evaluation of One-and Two-Cycle Fuel Rods from Calvert Cliffs-1", NPSD-75, March,1979.
3-9 S. R. Pati, Combustion Engineering, Inc., " Gas Release and Micro-structural Evaluation of Three-Cycle Fuel Rods from Calvert Cliffs-1",
C-E NPSD-119, December,1980.
3-10 D. E. Bessette, et al, Combustion Engineering, Inc., " Examination i
of Calvert Cliffs-1 Test Fuel Assemblies at End of Cycles 1 and 2,"
NPSD-72, September, 1978.
l 3-4 l
l
3-11 E. J. Ruzauskas, J. G. Schneider and P. A. Van Saun, Combustion Engineering, Inc., " Examination of Calvert Cliffs-1 Test Assemblies after Cycle 3", NPSD-87, September,1979.
3-12 Ansilin, "The Role of Fission Products in the Swelling of Irradiated 00 and U, Pu) 0 Fuel", GEAP-5583,1969.
2 2
3-13 Findlay, J. R., " Calculations of Fission Products Swelling in Oxide and Carbide Fuels", AERE M-1943 (1967).
3-14 Ewart, F. T., Private Comunication as Referred to in Reference 3-5.
1 l
S 3-5
Figure 3-1 Fuel Density versus Burnup for an Assumed Fuel Densification of 1.6% TD
~
~
o at be2 wc
.Jwm W
l t
l BURNUP, MWD /KGU l
t 3-6
4.0 FUEL-CLAD INTERFACE
4.1 INTRODUCTION
The previous FATES model used preassigned maximum values for gap conductance and pellet-clad mechanical interfacial pressure in lieu of an explicit treat-ment of the movement of the pellet-clad interface after pellet-clad contact.
A pellet-clad mechanical interaction model, which explicitly treats the interface, has been included in FATES 3.
This model is described in the following sections.
4.2 PELLET-CLAD CONTACT LOADING After the pellet and clad make contact, a mechanical interfacial pressure exists which is calculated from the elastic strain necessary to force the clad to ccnform to the fuel pellet.
The interfacial pressure is used in the previous FATES model in the calculation of the contact conductance ccmponent of the gap conductance.
In FATES 3, the interfacial pressure is also used in. calculating the hoop stress for'the clad creep calculation.
The resulting expression for hoop stress is a modified form of Equation 30 in Reference 4-1:
'e = [P -P,+P,] [R)+R ]/[2(R -R )3 I4-l) g 2
j 2
where:
a, = clad hoop stress, psi P' = fuel rod internal gas pressure, psi l
P,= reactor coolant pressure, psi P, = mechanical interfacial pressure, psi Rj = input clad outside radius, in R2 = input clad inside radius, in l
The mechanical interfacial pressure relaxes through clad creep.
All components are calculated such that pellet and clad are in equilibrium.
4-1 i
I._
_ _ _ _ _ _. -,,.... _ _.,. _.. = _,. _ _ _. _ _ _ _, _ _... _ _ _.. _ _ _ _ _ _. _.. _ _ _ _ _. _
4.3 GAP CONDUCTANCE The interfacial pressure calculated by the previous FATES model did not account for creep and necessitated an upper limit on the value of computed gap conductance.
An upper limit on gap conductance is not used in FATES 3 because of the inclusion of an explicit treatment of the pellet-clad interface.
The predictions of experimental data in Section 9.') and fission gas release model calibration in Section 2.0 were performed without a maximum value.
4.4 REFERENCES
FOR SECTION 4.0
~4-1 Combustion Engineering, Inc., "C-E Fuel Evaluation Model Topical Report",
CENPD-139, July,1974. (Proprietary) 1 l
l l
l 4-2 I
5.0 ANNULAR FUEL PELLETS
5.1 INTRODUCTION
l The capability of modeling annular fuel pellets has been included in FATES 3.
The models used to calculate the fuel pellet temperature distribution, fuel pellet thermal expansion, and rod internal void volume have been modified for annular fuel pellets.
These changes are discussed in the following sections.
5.2 TEMPERATURE DISTRIBUTION The derivation of the equations used to calculate the radial power and temperature distributions for annular fuel pellets is given in this section.
As is the case for solid fuel pellets in Reference 5-1, the radial power distribution in an annular fuel pellet is assumed to be described by:
2 4
q" ' ( r/ r,) = q" ', [A+B ( " ) + C (h ) ]
(5-1) o o
where:
3 q'" = local volumetric heat generation rate, BTU /hr-ft q"', = volumetric average heat generation rate in the fuel, clad, and moderator, 3
BTU /hr-ft A B.C = flux depression constants r/r, = dimensionless pellet radius r, = pellet outer radius, inches Conservation of energy requires that:
r
[A+B(h)2 + C (k ) ] r dr l
rh o
o 1,0 (5-2 )
'o l
r dr rh 5-1
~
wh;re:
h = Pellet inner radius, inches r
Solving Equation 5-2, the normalized flux depression factors must obey the relation:
h B
"h 4
P 6
2 r
2 (A[1-(7)]+y[1-(7)]+y[1-(7)
/ [1-([r ) 3 = 1.0 (5-3)
C h
o o
o o
The steady state radial temperature distribution through an annular cylindrical pellet with internal heat generation and temperature dependent thermal conductivity is given by the one-dimensional heat conduction equation:
h[hr(krh)]=-q"',F[A+B(f)2 +C(h)
]
(5-4) 4 f
o o
where:
k = temperature dependent fuel thermal conductivity, BTU /hr-ft *F (Ref. 5-1)
Ff = fraction of q"', generated in the pellet T = temperature at radius r, 'F Integrating between r and r with the pellet inside surface assumed to be an h
adiabatic boundary gives:
kr h = q"' Ff[f(r - r h) +
2 (r -g)+
(r-rh )]
(5-5) 2 4
4 6
6 Integrating between T, and T and rearranging gives:
T T0
/
k dt = /
k dT + q',Ff 95 95 o
o h
4nFp [1-( 7 ) ]
o 2
2
{A[1-(h) +2([r ) in (f- )]
4 (5-6)
+B[h[1-({o)4]+([r
) In(k)]
o o
+C[h[1-(h)]+h(o) in(f)]}
o o
5-2
where:
k95 = thermal conductivity of 95% dense UO, BTU /hr-ft *F, (Ref. 5-1) 2 F,
= Maxwell-Euken fuel thermal conductivity porosity correction factor, (Ref. 5-1) q', = linear heat generation rate in the fuel, clad, and moderator, kw/ft T,
= pellet surface temperature, 'F Equation S-6 is used in, FATES 3 to calculate the fuel pellet temperature
. distributions in annular fuel pellets.
5.3 THERMAL EXPANSION The derivation of the equations used to calculate thermal expansion in annular fuel pellets is given in this section.
The resulting expressions reduce to those in Reference 5-1 for the case of solid fuel pellets.
Thermal expansion of the uncracked portion of the pellet is computed by finding the radially averaged displacement.
At each radial ring, the displace-ment is computed with the coefficient of thermal expansion defined in Reference 5-1. The displacement in the uncracked r etion of the pellet is given by:
~~
~
l r
Trdr AR * "c #
"f u
rh I
c (5-7) f I
r dr i
ry where:
AR = change in pellet radius due to thermal expansion in the uncracked u
l portion of the pellet, inches
= pellet crack radius, taken at the 1400*C isotherm, inches (Ref. 5-1) rc = thermal expansion coefficient for UO I"/I" -
'(*'
'I}
a 2
f 5-3 l
Integrating Equaticn 5-7 givas:
2r r
C
" Tr dr (5-8)
=
2 f
(r
-"h )
h r
c The integration required by Equation 5-8 is replaced by the numerical sumation in Reference 5-1.
Thermal expansion in the cracked portion of the pellet is identical to that in Reference 5-1.
5.4 VOIO VOLUME The fuel rod internal void volume is adjusted for the presence of a central hole. The central hole volume is referenced to the plenum
'emperature, as are the other components of void volume in Reference 5-1, as follows:
2 N
i 4
"'f (r
( pg + 60)/(T
+460)
(5-9) h*
h c
.=1 where:
V = total effective volume in the fuel column 3
3 central. hole, in i
L
= length of fuel column axial segment i, in f
j N
= total number of fuel column axial segments r
= fuel pellet inner radius in axial segment i, in h
T
= plenum gas temperature, *F pg T I = temperature at the fuel pellet inner radius in g
axial segment i, 'F The central hole and planum volumes are also corrected for a fuel centerline thermoccuple, if one is present.
5.5 REFERENCES
FOR SECTION 5.0 5-1 Combustion Engineering, Inc., "C-E Fuel Evaluation Model Topical Report", CENPD-139. July, 1974. (Proprietary) 5-4 v--p---
--,--,,,-_.,---w g,
y--n-_,m.-y-,.-__-y ny
-w.
.-,.,,_.,,,,,n_-----_,,,,,,,,,,,_~a
6.,0 FUEL PELLET RELOCATTON
6.1 INTRODUCTION
Fuel pellet relocation accounts for the outward displacement of pellets into the gap region between fuel and clad that results from the process of cracking and crack healing. A previously derived relocation model has been included in FATES 3 and is described in the following section.
6.2 DISCUSSION Equation 42 of Reference 6-1 is used to calculate fuel relocation in FATES 3.
Fuel relocatica continued to occur after pellet-clad contact in the previous FATES model.
Depending on the specifics of a given duty cycle, all, or part, of the relocation calculated with the referenced equation may be calculated to v eur during a given time step in FATES 3.
All of the relocation is calcul
, to occur if the hot pellet-clad gap is open.
If fuel swelling, thermal expansion, and relocat.!on calculated to occur in previous time steps are sufficient to close the pellet-clad gap, no additional relocation is calculated.
As show, in Section 9.0, use of this relocation model along with th Mher gap closure models in FATES 3 results in good predictions of fuel temperature, fission gas release, and rod internal void volume.
Use of this model is, theref ore, considered to be appropriate.
~
6.3 REFERENCES
FOR SECTION 6.0 6-1 Combustion Engineering, Inc., "C-: 7uel Evaluation Model Topical Report", CENPD-139, July,1974. (Proprietary) l l
6-1 l
7.0 CLAD AXIAL IRRADIATION GROWTH Irradiation induced axial growth of the clad was numerically input to the previous version of FATES. The model for fuel rods given in Reference 7-1 has been incorporated in FATES 3.
7.1 REFERENCES
FOR SECTION 7.0 7-1 Combustion Engineering, Inc., "In.leactor Dimensional Changes in Zircaloy -4 Fuel Assemblies", CENPD-198, December,1975. (Proprietary)
O I
e 7-1 l
i 8.0 PLENUM GAS TEMPERATURE The fuel rod plenum gas temperature was numerically input to the previous version of FATES.
In FATES 3, the fuel rod plenum gas temperature is approximated by a weighted average of the fuel pellet volumetric average temperature adjacent to the plenum and the coolant temperature at the channel outlet:
T
= (A T +Apj gg)/(A +Apj)
(8-1)
T pg ff f
where:
l T
plenum gas temperature
- F
=
pg 2
cross sectional area of pellet acjacent to end plenum, in A
=
f T
= volunetric average temperature of pellet adjacent to end plenum, 'F f
2 A)
= cylindrical area of end plenum, in p
T,
= channel coolant outlet tempeerature, *F c
O O
8-1
9.0 PREDICTIONS OF EXPERIMENTAL DATA 9.1 PURPOSE FOR DATA SELECTIONS The purpose of this section is to demonstrate the accuracy of the predictive capability of the FATES 3 fuel performance code by comparing predictions with experimentally measured data.
This data was not used in the development of the FATES 3 code.
Three experimentally measured parameters, which give an excellent measure of a feel performance code's predictive capability, were selected for comparison.
These are:
fission gas release Tuel temperature fuel rod internal uid volume
. Fission gas release is important because of its effect on conductivity of the gas in the fuel-clad gap and because of its partial pressure which increases with burnup.
Fuel temperature is obviously important since it represents stored energy and is the ultimate parameter being determined by FATES 3.
Internal void volume has been selected for comparison because it, along with fission gas release, is very important in the ci.lculation of rod internal pressure.
In assembling the data base, only rods were selected which were well characterized with respect to mechanical design and irradiation history.
Fuel rods typical of C-E fuel rod design were used to the maximum extent possible.
Of the total of rods used for verification, were prepressurized with helium.
A large number of fuel rods of Kraftwerk Union (KWU) design (which are also similar to C-E fuel design) irradiated in the Obrigheim (KWO) and Stade (Kp ) commercial PWR's, both pressurized and unpressurized with helium, l
were used for fission gas release comparisons.
Data for these rods, obtained through a technical information exchange agreement between C-E and KWU, were believed to be not as well defined as the remainder of the data base.
They are, however, an excellent source of data for rods irradiated under PWR conditions.
9-1 i
A wide variety of irradiation histories are represented in the data base.
Many rods experienced duty cycles typical of those encountered in comercial reactors.
Rods which were irradiated in a comercial pWR prior to ramp testing in a test reactor were used to verify fission gas release predictive capability for transient power conditions.
Only rods equipped with fuel centerline thermocouples were selected for fuel temperature comparisons. Although temperatures may be inferred from fuel grain growth or other microstructural temperature markers, they are thought to be unsatisfactory due to the inherent uncertainties associated
~~
with microstructural changes. Of the rods selected for temperature comparisons,
" ~
are of C-E design.
The " remainder are of an older, unpre-
~
pressurized design, but have been historically used for fuel temperature comparisons.
Rods irradiated in the Calvert Cliffs 1 commercial PWR were used for rod internal void volume comparisons.
These rods were selected because they are C-E designed and are well characterized with respect to both mechanical design and irradiation history.
The following sections describe the verification data base and the results of the coniiparisons.
l ~
9.2 DATA AND RESULTS l
~~
l -
Calculations were made for a total of fuel rods to verify the predic-
~
~~
tive capability of FATES 3.
rods were modeled with detailed
~
design data and irradiation histories.
Published fuel temperatures were modeled l
for calculating fission gas release for the remaining rods.
Table 9-1 gives the number of rods in each data set and their uses.
The important design and operating variables are sumarized in Tables 9-1 through 9-3.
Comparisons of measured and predicted fission gas release, fuel tempe"ature, and void volume data are given in Tables 9-4 through 9-6 and Figures 9-1 through 9-5.
Figures 9-2 and 9-3 also show data from the fission gas release model correlation data base as darkened symbols. As can be seen from these results, FATES 3 does an excellent job of predicting tnese parameters.
The individual data sets are further discussed below.
9-2
The fission gas release data of Bellamy and Rich (Refs. 9-1 and 9-2) were used as an independent chack of the predictions of the C-E fission gas release model which is shown in Figure 9-1.
These predictions were performed external to FATES 3 with data as shown in Tables 9-5 and 9-6.
The rods were irradiated for up to r.ine reactor cycles at DID0 in a sodium coolant.
The clad operated at about 500*C, which is about 200*C higher than the typical operating temperature of Zircaloy cladding in commercial LWRs.
Peak rod linear heat ratings ranged from an estimated 3 to 9 kw/ft, which were estimated to match the reported end-of-life fuel centerline temperatures.
Gas release predictions have been made for 17 rods. These are the rods for which detailed fuel central temperature histories have been published by AERE, the test sponsor.
These rods experienced a variety of duty cycles.
Some had peak temperatures early in life while others had peak temperatures later or relatively constant thermal histories.
It is believed that for all rods in this experiment, even those with first cycle peaks in fuel temperature, the time-integrated. release of fission gas can be predicted based on the reported fuel temperatures for the last irradiation cycle.
The reported temperatures are calculated values based on nonpublished rod linear heat ratings and the range of reported gap resistivity.
The gap resistivity is 2
reported to be between 1.5 and 2.0 cm
- C/w (i.e., gap conductance between 2
- C).
Tables 9-5 and 9-6 show the resulting range of,
0.5 and 0.667 w/cm predicted versus measured gas release values.
The results are plotted in Figuro 9-1.
The Petten data set, plotted in Figures 9-2 and 9-3, consists of short length rodlets of C-E and KWU design which were pre-irradiated in KWO and subsequently ramp tested in a pressurized capsule in the poolside facility of the Petten reactor in the Netherlands. The purpose of these tests was to measure pellet-clad interaction failure propensity.
Extensive P.I.E. (Post Irradiation Examinations) was done on both failed and non-failed rods. Data from non-failed rods are an excellen-source of fission gas release data for transient power conditions.
Irradiation data and measured and predicted fission gas release values are given in Figure 9-4 for individual rodlets. All of the rodlets reported here did not fail during the ramp test.
9-3
The KWU rods were irradiated in the KWO and KKS reactors.
There are standard full-length rods, experimental rods of high power with a high enrichment, and short length rods from a power cycling experiment in this data set.
Due to this data being less well characterized as discussed in. Section 9.1 the predicted results contain more scatter,as would be expected.
The IFA 418 irradiation experiment is jointly sponsored by C-E and KWU at the Halden Boiling Water Reactor (HBWR) in Norway.
The IFA 418 assembly consisted of six instrumented rods positioned in a circular array.
Rods which contained C-E fabricated fuel, were selected for comparison.
Fission gas release was measured each contained a fuel centerline thermocouple (pellet ID =.07 in) in a sho5t section of annular pellets at the bottom of the fuel stack. L yielded fuel centerline temperature data out to burnups of Results of the fission gas release comparison is shown in Figures 9-2 and 9-3.
Comparison of FATES 3 temperatures with the thermocouple data is shown in Figures 9-4.
The IFA 428 irradiation' experiment provided thermocouple data as shown on Figure 9-4.
It is jointly sponsored by C-E and KWU at the HBWR.
The IFA 428 assembly consisted of two six-rod clusters, one upper and one lower.
C-E rods were selected for comparison.
Each rod
~
had two centerline thermocouples (pellet ID =.07 in), each extending about 4 inches into each end of the fuel stack.
Fuel temperatures were recorded at regular intervals (which are shown on the figure) and were obtained out to about 3,200 MID/MTU.
i l,
The IFA 11 and IFA 21 irradiation experiments (Ref. 3-3) were sponsored by AB j
Atomenergi at the HBWR.
Each rod was equipped with a fuel centerline thermocouple l
(pellt.t ID =.05 in) extenuing about midway into the fuel stack.
Four of six
(
thermocouples yielded data only at beginning of life.
The remaining two yielded data to 4,300 MWD /MTU.
These rods are an older design, but are included because they have been historically used for temperature comparisons.
Results of FATES 3 predictions versus measured temperatures are also shown on Figure 9-4.
The Calvert Cliffs 1 rods were described in Section 2.5.
They are used in this j
section to provide verification of predicted fuel rod internal void volume i
(or volume available for gases), which is important for calculating fuel rod internal pressure. These predicted and measured void volumes are at cold conditions and, therefore, give an excellent indication of expected changes 9-4 L
in internal pressure due to th irradiation induced changas in fuel rod dimensions (e.g., clad creepdown, fuel swelling, etc.).
Results of the comparison between FATES 3 and the measured data is shown in Figure 9-5.
Agreement is excellent.
9.3 CONCLUSION
S A comparison of FATES 3 predictions with measured data have been made for fission gas release, fuel rod centerline temperature, and fuel rod internal void volume.
Several conclusions may be drawn from these comparisons.
1.
Predictions of fission gas release from 00 fuel using FATES 3 agree 2
q0ite well with measured data (which contain a wide variation in power levels and burnups) as shown in Figure 9-2.
The more well-charac-terized the data the better the agreement, indicating an appropriate modeling of fission gas release mechanisms.
2.
The burnup dependency of fission gas release is well modeled as shown by Figure 9-3.
The data when normalized to burnup show no burnup trends for enhanced gas release beyond that inherent in the model.
3.
FATES 3 temperature predictions are shown in Figure 9-4 to be in fairly good agreement with measured data (but slightly high) through %22000 MWD /MTU burnup. Fuel-clad gap closure has usually occurred at this burnup so uncertainties in temperature due to the effect of gap size on gap conductance are not present at higher burnups.
This gives reasonable assurance that fuel temperatures at higher burnups will follow a similar I'
trend.
l 4.
The fuel rod internal void volume (available to accommodate fill gas and released fission gases) credic.ted by FATES 3 is in excellent agreement with measured data to fairly high burnups.
Therefore, prediction of fuel rod internal pressure would be expected to be excellent.
9-5
- e6
1 9.4 REFERENCES FOR SECTION 9.0 i
9-1 R. G. Bellamy and J. B. Rich, " Grain Boundary Gas Release and Swelling in High Burnup Uranium Dioxide", J. Nucl. Ma':., 33 (1969) i pp. 64-76.
9-2 Letter from R. G. Bellamy (AERE Harwell) to E. Roberts (W) dated March 13, 1975.
i l
l 9-3 G. Kjaerheim and E. Rolstad, "In-Pile Determination of UO Thermal 2
l Conductivity, Density Effects and Gap Conductance", HPR-80, December, 1967.
l t
I t
O 9-6 i
Table 9-1 Summary of Irradiation Parameters for Rods in the Verification Data Base 1
Number Used Peak Local Rod Average 4
Data of Rods For*
Power (kw/ft)
Burnup (MWD /KgU)
Petten
]
KWU Bellamy & Rich 17 F
3.1 - 8.7 8.0 - 47.8 IFA 418 IFA 428 4,
IFA 11 + 21 6
T 15.5 0, 4.3 Calvert Cliffs 1 12 V
9.1, 11.1 18.7 - 37.0 i
i l
F - Fission gas release T - Fuel temperature j
V - Void volume
Table 9-2 Sunmary of Design Parameters 4
For Rods in the Verification Data Base Bellamy &
IFA IFA IFA 11 Calvert l
Parameter
.Petten KWU Rich 418 428
& 21 Cliffs 1 Clad 00, in
.118.236
.532.541
.440 1
Clad ID, in
.330.594
.496.503
.388 l
Initial Pellet-
.001 1.9-6.6 8.5 j
CladDia. Gap, Mil]
Initial Grain 12, 15 14, 25 2.5 - 15
. Size, um i
?
Fuel Column
.8-1.2 67.5 136.7 te ngth, in Initial fuel 95, 98 ~
96, 98 93 - 95 j
Density, % TD FiliGasPressure3 15 315,"466 at 70 F, psia 5
2.5, 2.8 i
Enrichment, %U235 l
b I
i
~'the Bellamy & Rich rods which were fabricated in argon and helium atmospheres.
i, I
i
I Table 9-3 Sumary of Thermal-Hydraulic Parameters For Rods in the Verification Data Base Bellamy &
IFA IFA IFA 11 Calvert Parameter Petten*
KW Rich 418 428
& 21 Cliffs 1 420 2250 Coolant Pressure.
l psia i
Coolant Inlet s900 450 548
- Temperature, F e
i l
~
- The coolant pressure and inlet temperature during power ramping are psia and respectively.
i i
i 1
l
Table 9-4 Irradiation and Design Parameters for Petten Fuel Rods with Measured and Predicted Fission Gas Release Initial Fuel Ramp Peak Rod Average Fission Gas Test Grain Size,
- UlGR, Burnup, Release, %
Number um kw/ft*
MWD /Kgu Measured Predicted
?
5 h
m M
e m
9
t TABLE 9-5 I
Comparison of Gas Release Model Predictions 2
with Data of Bellamy and Rich (Gap Resistivity = 2 cm _o /w)
C Fuel Fuel Fuel Fuel Pred.
Exp.
Surf.
Central Pin Burnup Fuel Grain Diam.
Gas Gas No.
Mwd /Kgu
% TD Size,un cm kw/ft Temp., C" T_emp., UC Release, 1 Release, 1 l
5020 8.0 95 12 0.594 6.22 719 1279 2.47 1.2 5036 17.5 95 12 0.594 8.66 805 1669 17.95 13.8 5026 19.3 95 12 0.482 6.42 778 1381 6.42 2.4 5042 30.0 95 12 0.482 7.87 841 1629 19.39 11.1 5033 13.6 98 15 0.470 6.40 784 1357 4.08 0.9
[ 5030 13.9 98 15 0.374 4.16 732 1072 0.94 0.12
~
5029 14.1 98 15 0.330 3.15 699 943 0.39 0.09 5031 15.3 98 15 0.364 3.87 722 1033 0.74 0.16 5032 15.5 98 15 0.330 3.05 693 927 0.37 0.10 5037 18.2 98 15 0.330 3.15 699 943 0.45 0.2 5019 18.8 98 15 0.582 8.47 804 1602 11.13 1.5 5038 19.1 98 15 0.374 4.09 728 1061 1.02 0.2 5023 20.8 98 15 0.364 4.08 7 34 1068 1.13 0.22 5022 34.9 98 15 0.374 5.16
~
788 1238 3.90 2.52 5039 39.6 98 15 0.364 5.04 789 1228 3.97 3.1 5050 45.1 98 15 0.374 5.23 m 792 N 1250 4.70 4.1 5049 47.8 98 15 0.374 5.23 ev 792 ev 1250 4.81 7.1 8 Ts (fuel surface, UC) = Tc (clad ID, C)
+ 10.44 x kw/ft x Gap Resistivity (cm2 _ o /w) + Pellet Diameter (cm)
C U
where T = 500 C c
b Reported values of Rafs 9-1 and 9-2.
TABLE 9-6 Comparison of Gas Release Model Predictions 1
2 with Data of Bellamy and Rich (Gap Resistivity = 1.5 cm o /w) c Fuel Fuel Fuel Fuel Pred.
Exp.
Surf.
Central Pin Burnup Fuel-Grain Diam.
Gas Gas i
No.
Nd/KgU
% TD Size,e cm kw/ft Temp., C Temp..
C Release, %
Release, 1 5020 8.0 95 12 0.594 6.03 659 1176 1.32 1.2
)
5036 17.5 95 12 0.594 8.87 734 1590 11.91 13.8 5026 19.3 95 12 0.482 6.19 701 1251 3.21 2.4 5042 30.0 95 12 0.482 7.71 751 14 84 10.57 11.1 I
5033 13.6 98 15 0.470 6.19 706 1229 1.99 0.9 i
?
5030 13.9 98 15 0.374 3.96 665 971 0.43 0.12 i
N i
5029 14.1 93 15 0.330 2.98 641 860 0.20 0.09 l
5031 15.3 98 15 0.364 3.58 654 925 0.31 0.16 5032 15.5 98 15 0.330 2.79 632 835 0.18 0.10 5037 18.2 98 15 0.330 3.04 644 869 0.25 0.2 l
5019 18.8 98 15 0.582 8.28 723 1466 6.39 1.5 l
5038 19.1 98 15 0.374 c
NA 0.20 5023 20.8 98 15 0.364 3.87 666 965 0.51 0.22 5022 34.9 98 15 0.374 c
NA 2.52 s
5039 39.6 98 15 0.364 5.12 720 1146 2.28 3.1
)
5050 45.1 98 15 0.374 c
NA 4,1 5049 47.8 98 15 0.374 c
NA 7,g a Ts (fuel surface. C)=Tc (clad ID, C) + 10.44 x kw/ft x Gap Resistivity (cm2 C/w) Pellet' Diameter (cm) where T = 500 C c
b Reported values of Refs. 9-1 and 9-2 Data Not Available i
c
~
T Figure 9-1 A Comparison of Model Predictions vs. Measured Fission Gas Release for the Bellamy and Rich Data 20 a
j n
30.0 A, 7.5 3
10 18.8 O
A 4
)y.3[I A 47.8 A
34.9 A
BURNUP dt 39*6 p
- 13. 6 MWb/kgU>
g A
w O
8.0 a:
m 1 g a
o fI A
20.8 13.
l E
^
15.3 g
A O
l' O
O G
18.2 l
15.5 0 a 2.0 cm UCM (98%TD,15p m GR AIN SIZE) 2 14.1 O
2 O 1 'i cm _oCM (98%TD,15p m GRAIN SIZE)
O A 2.0 cm UCM (95%TD,12p m GRAIN SIZE) 2
~
i 2
e 1.5 cm _oCM (95%TD,12p m GRAIN SIZE)
I 0.1 0.1 1
10 20 EXPERIMENTAL GAS RELEASE, %
l 9-13
Figure 9-2 COMPARISON OF MEASURED AND PREDICTED FISSION GAS RELEASE
~
at w'
E
$we Eo Qw baw E
~
MEASURED GAS RELEASE,%
G 9-14
SI-6 PREDICTED - MEASURED GAS RELEASE, %
i l
3mo n
-4 m
O E
m m
E c=
E O
oa
>2 a,
N If a
pw C
m>
m<
m 2
B E=
k I
i
l l
Figure 9-4 COMPARISON OF MEASURED AND PREDICTED FUEL CENTERLINE TEMPERATURES e
s wc3
~.
E w&w3 s
mw
$wc-Jw 3m Ow l
E5w m
l
~
~
MEASURED FUEL CENTERLINE TEMPERATURE,0F l
l O
9-16 I
I
Figure 9-5 COMPARISON OF MEASURED AND PREDICTED VOID VOLUME FOR CALVERT CLIFFS I TEST RODS 3
i g
z m' 2 O
2 O
3 0
9 O>
8 ti51 w
E O
I I
O 1
2 3
3 MEASURED VOID VOLUME, IN l
l 9-17 I
., _, _ _, -. -,., _.