THE FATIGUE STRENGTH OF GRAPHITE AND CARBON ...Room temperature fatigue tests were carried out on graphite and carbon meterials, which are used for the components in the core region

JAERI-Research98-024

JAERI-Research--98-024

JP9808012

THE FATIGUE STRENGTH OF GRAPHITE AND CARBONMATERIALS FOR HTTR CORE COMPONENTS

Motokuni ETO, Taketoshi ARAI and Takashi KONISHI'

2 9 - 4 3

Japan Atomic Energy Research Institute

Hi,( T 319-H95

- (T319-1195

This report is issued irregularly.Inquiries about availability of the reports should be addressed to Research

Information Division, Department of Intellectual Resources, Japan Atomic EnergyResearch Institute, Tokai-mura, Naka-gun, Ibaraki-ken T319—1195, Japan.

©Japan Atomic Energy Research Institute, 1998

EP

JAERI-Research 98-024

The Fatigue Strength of Graphite and Carbon Materials

for HTTR Core Components

Motokuni ETO, Taketoshi ARAT and Takashi KONISHI *

Department of Materials Science and Engineering

Tokai Research Establishment

Japan Atomic Energy Research Institute

Tokai-mura,Naka-gun,Ibaraki-ken

(Received March 19,1998)

Room temperature fatigue tests were carried out on graphite and carbon

meterials, which are used for the components in the core region of the HTTR, in the

applied stress condition that R (= a min / a max) =— 3, — 1, 0 (PGX graphite) , =— 1,

0 (ASR-ORB carbon) and=— 1 (IG-11 graphite). The data were analyzed by Price's

method, homologous stress method and P-T-S diagram method to investigate which is the

most appropriate to derive desigh S-N curves. Fatigue tests were also carried out at 980

°C in vacuo on IG-llgraphite to clarify the effect of temperature on its fatigue strength.

The results indicated : (1) Price's method was the most appropriate to analyze the data

for a design S-N curve. (2) Fatigue strength decreased with decreasing R-value,with the

less pronounced tendency for ASR-ORB. (3) Design S-N curves were obtained on PGX

and ASR-ORB on the basis of the data analyzed by Price's method. (4) Fatigue strength

of IG-11 at 980 °C appeared to be almost the same as that for the room temperature

fatigue strength, if the applied stress was normalized to the mean tensile strength at

room temperature in vacuo.

Keywords : HTTR, Graphite,Carbon, Fatigue, Strength, Design Curve

+ Department of Advanced Nuclear Heat Technology, Oarai Research Establishment

* Toyo Tanso Co.,


HTTR &>bm

0

mm mn

(1998 ¥ 3 n 19 B

HTTR

T-S ^

( = a m i n / a max) = - 3 , - 1 , 0 ( P G X H f S ) , = - 1 , 0 (ASR-ORB

(IG-11 Hip) t L-fZo S i L f e r - ^ 5 : Price (Djjfe, MJfclfctlfelk.ZJ1- P-

fco IG-11 HfpM-o

(2)

(3)

Price

ASR-ORB

Price PGX £ ASR-ORB

t

: f 319 - 1195 2 - 4


Contents

1. Introduction • • 1

2. Experimental Procedure •• • • • • • • • • 2

2.1 Meterials and Specimens • • •• • • 2

2.2 Tensile and Compressive Tests •••••• • • • 3

2.3 Fatigue Tests • •-• • • • • • 3

3. Results and Discussion • ••• • • 4

3.1 Volume Dependence of Tensile and Compressive Strengths • 4

3.2 Static Tensile Strength , . . . . . . 4

3.3 Fatigue Life at Room Temperature • • • 5

3.3.1 Analysis of Data by Price's Method • • • 5

3.3.2 Analysis of Data by Homologous Stress Method 5

3.3.3 Analysis by P-S-N Diagram Method • 6

3.3.4 Dependence of S-N Curve on R-value • •••••• 7

3.3.5 Equi-fatigue Life Diagram • g

3.4 Fatigue Life at High Temperature • g

3.4.1 Tensile Strength of IG-11 Graphite • •• • > 8

3 . 4 . 2 Fatigue Strength a t 980 °C • • . . . . . . . . . • • • •,..-. • • • • • • • • • • • 9

4. Conclusions .••••• 10

Acknowledgments • 10

References • • • • 11

Appendix : Digital Date on Static Strength and Fatigue Strength 38

111


^ . i 4^r'ri*J&.v-/ff^i§5?/i * * * ^

*? *•? HI HFs 7& 71?j~c&fe =Y£LBf&i • Q

2 q i i t & d ^ ' ^ " ^ Q

3.2 #Stl5l5S^ifi 4

3.3.1 Price (DJffeK J; -S^ftff •• • 5

3.3.2 >frJ&/&;ftj£K:<fc<5$F#f ••••'• • • 5

3.3.3 P-S-NH fetd ££#?$? • •• •••••'• 6

3.3.4 S-N EfeS^R-fiS^^tt • 7

3.4.1 IG-11 H$&<^)3l5B§tS - 8

.4.2 yyu c t ^ - 3 b t j O R ^ K J X • • yii: S . , -in

• w Bra -i-U

10

/^?/V • y— $ .• 38

IV

JAERJ-Research 9 8 - 0 2 4

1. Introduction

The components made of graphite and carbon materials in the HTTR,High Temperature Engineering Test Reactor, are subject to cyclic stressescaused by earthquakes, the variation in the pressure of coolant helium andthermal cycles during operation of the reactor [1-3]. Therefore, it isessential to obtain data on fatigue properties of the graphite and carbonmaterials for the design and sefety evaluation of the components. From theaspect of design of these components, it is a requisite to establish a fatiguestrength diagram since the graphite structure design code for the HTTRrequires the fatigue analysis [4].

There have been a number of investigations into the fatigue behaviorof nuclear graphite. However, the fatigue data, in most cases, have shownlarge scatters so that it should be investigated what is the mostappropriate statistical method to analyze the data.

Price [5] employed an assumption that logarithm of fatigue life beexpressed by a linear equation of the logarithm of the ratio of themaximum applied tensile stress to the mean tensile strength, and that thedeviation from the linear equation is to obey the normal distribution. Hismethod gave rise to good agreement between the experiment and analysisfor various loading modes. Wilkins et al. [6] , Ishiyama et al. [7,8] , and Etoet al. [9] used a concept of homologous stress to analyze the data points,and indicated that one can estimate the fatigue life even when the numberof data points is small. There are another method to analyze the fatiguedata, what is called P-T-S diagram method, in which a statisticaldistribution is assumed to be applied to the data on fatigue life at eachstress level and the fatigue life as a function of stress level is plotted at acertain fatigue fracture probability [10].

The purpose of this report is to (1) summarize the fatigue data whichhave been obtained at room temperature on HTTR graphite and carbonmaterials, especially those used for the permanent reflector, plenum blockand thermal insulator, for establishing the database, i, e., the data obtainedin the test condition that the stress ratio R (= minimum applied stress amin/ maximum applied stress omax) is 0.0, -1.0, and -3.0, and (2) to derivedesign fatigue curves from the results of analysis of the data by means ofthe methods mentioned above. In addition, the results of the fatigue teston a graphite for fuel block and sleeve at around 1000°C were alsosummarized to make clear the effect of temperature on its fatigue strength,since the maximum temperature of coolant helium at the outlet issupposed to be 950°C in the HTTR.

- l -


2. Experimental Procedure

2.1 Materials and specimensMaterials used were a fine-grained isotropic graphite IG-11 (Toyo

Tanso Co.), semi-isotropic graphite PGX (UCAR) and a carbon material ASR-ORB (SIGRI), which are used as fuel sleeves, fuel blocks and replaceablereflectors; permanent reflectors and plenum blocks; and thermal insulatingblocks at the core bottom; respectively. Some properties of these materialsare summarized in Table 1. High temperature fatigue tests were done onlyfor IG-11. The size of the original billets from which the smaller blockswere cut for the specimen preparation was 300x540x850 mm for IG-11graphite, 1100 mm in diameter X1220 mm in length for PGX graphite, and1150 mm in diameter X55O mm in-length for ASR-0RB carbon. The size ofthe blocks from which specimens were machined was 155x155X500 mmfor IG-11, 300x300x200 mm for PGX. A fan-shaped block with an angle of45 degrees was used for specimens of ASR-0RB. Blocks in 115X115x400mm were machined from the original billets of IG-11 for specimens to beused in the high temperature fatigue test.

Fig. 1 shows the size of an original block of IG-11 and the cuttingplan for the specimen preparation. Table 1 summarizes mechanicalproperties of these materials. Dimensions of the specimen used in thepresent experiment are shown in Figs. 2 and 3 for room temperature andhigh temperature tests, respectively. For the room temperature test,specimens were machined either parallel to (L) or perpendicular to (T) thelongitudinal axis of the blocks. Specimens for the test were selectedrandomly from all the specimens prepared except PGX graphite specimensfor which the selection was carried out in the following way: First, theultrasonic wave velocity was measured for all the specimens. Then, theYoung's modulus, E, was estimated from the velocity, V and the mean valueof apparent density (p = 1.74 g/cm3), i, e., E = pV2. The results of theestimation is:

T-direction L-directionNumber of specimens 228 228Young's modulus (mean) 8.30 GPa 6.48 GPaStandard deviation 0.54 GPa 0.42 GPaUltrasonic wave velocity (mean) 2182 m/s 1928 m/s

Specimens which displayed the ultrasonic wave velocity smaller than themean value by 200 m/s or more, i. e., 0.70 GPa for the Young's modulus,were deleted from the specimens for the test. The number of specimensdeleted was 19 (8.3%) and 23 (10.1%) for T- and L-directions, respectively.

- 2 -


Specimens for the high temperature test were cut from 6 blocksshown in Fig. 1 in the way that their longitudinal axis was parallel to thelongitudinal axis of the blocks. No selection was made on specimens for thehigh temperature test.

2.2 Tensile and compressive testsSince there had been no data available on the effect of specimen size

on the tensile and compressive strengths of PGX graphite, tensile andcompressive tests were carried out on specimens with several sizes. For thetensile test three kinds of dumbbell type specimens were used: 5 mm indiameter X20 mm in gage length, 10 mm in diameter x20 mm in gagelength, and 20 mm in diameter x20 mm in gage length. For thecompressive test, five kinds of cylindrical specimens were used: 5 mm indiameter X10 mm in length, 10 mm in dia. X20 mm in length, 15 mm india. X 30 mm in length, 30 mm in dia. X 60 mm in length, and 50 mm in dia.x 100 mm in length . A screw-driven test machine was used for these testsat a cross-head speed of 0.1 mm/min. For the compressive test, 2 mm thicksheets of teflon were placed between a specimen and the upper and lowercompression plates to reduce the friction.

2.3 Fatigue testsTo determine the applied stress for the fatigue test tensile tests were

carried out for fatigue specimens of all the three materials, using aservohydraulic test machine. The number of specimens tested ranged from20 to 40. Before the high temperature fatigue test tensile tests were alsodone on specimens shown in Fig. 3 at room temperature and 980t invacuo using a servohydraulic test machine with the maximum allowableload of about 100 kN. The number of specimens tested was 5 for both roomtemperature and 980X:.

Room temperature fatigue tests were carried out in the loadingcondition that R (= c w / amax) values were 0, -1, and -3 (L-direction of PGXgraphite only). Several levels of the tensile applied stress were chosen,ranging from 0.7 to 1.0 of the mean static tensile strength. The loadingspeed was 8.75X103 N/s, 2.62x103 N/s, and 2.13x103 N/s for IG-11, PGXand ASR-0RB, respectively. The clyclic load in the triangular wave wasapplied to the specimen up to failure or 10s cycles.

High temperature fatigue tests were carried out in the loadingcondition that the maximum applied stress level was 0.75 to 0.9 of themean tensile strength with R~ 0.0. The frequency of the cyclic stress was


about 1 Hz. hi some cases the fatigue test was carried out to more than 105cycles. Those specimens which survived after 105 or more cycles ofrepeated stresses were subject to tensile tests at room temperature.

3. Results and Discussion

3.1 Volume dependence of tensile and compressive strengthsData on the effect of specimen volume on the tensile and compressive

strengths of IG-11 graphite were already summarized in the previousstudy, indicating that the size of the fatigue specimen employed here isappropriate [9].

The data on strength of PGX graphite measured in the present studyare shown hi Fig. 4, where one can see that there is no significant effect ofspecimen volume both for longitudinal and transverse directions. In thecase of compressive tests on the largest specimens, i. e., those 50 mm indiameter, the initiation and growth of a crack was often observed at theend or side surface of a specimen, being followed by its growth along thelongitudinal direction up to fracture. This is probably because some stressconcentration took place in the end surface of a specimen in contact withthe compression plate. On the other hand the smaller specimens showedclear shear fracture at an angle of about 45 degrees. It is to be noted thatdespite of the fact above there is no significant dependence of compressivestrength on the specimen volume. On the basis of the facts above andgeneral requisite that the specimen size should be at least 10 times largerthan the grain size the specimen dimensions shown in Fig. 2 weredetermined. The digital data on PGX graphite are shown in the Appendix.The diameter and gage length employed here for the fatigue test are twicelarger than those of the previous investigation on IG-11 graphite [7].

3.2 Static tensile strengthTo evaluate the variance of the tensile strength , the results of tensile

tests on fatigue specimens are shown on the normal probability paper forIG-11, PGX and ASR-0RB in Figs. 5, 6 and 7, respectively. The digital dataare shown in the Appendix. It is seen hi these figures that the distributionof the tensile strength of these materials is well expressed by the normaldistribution. The trend was more clearly found for the data set where themore data points taken from several lots of IG-11 and PGX werestatistically analyzed [10]. The data on tensile strength also obey theWeibull distribution fairly well, especially at low strength levels.


3.3 Fatigue life at room temperatureResults of the fatigue tests performed in the condition that R=O.O,

-1.0 and -3.0 are summarized in the Appendix, Tables A8 through A12 forIG-11, PGX (L)(T) and ASR-ORB (L, T), respectively. In case that a specimenfractured during the first tensile loading in each loading condition, thefatigue life was counted as 0.25 or 0.5 for R=-l or R=0.

3.3.1 Analysis of data by Price's methodAccording to Price model the fatigue life Nf is assumed to be

expressed as

8. (1)

Here, o*a, Smean are the maximum applied stress and the mean tensilestrength, respectively. Constants, A and B are determined by the leastsquare method, s is a normal random variable with the mean value = 0 andthe standard deviation = s. In the present analysis specimens with fatiguelives larger than 105 were omitted so that the derived equation isconsidered to be on the conservative side.

The data analyzed on the basis of Eq. (1) are shown in Fig. 8 through12 for the three materials with several R-values. Digital data for thesefigures are summarized in the Appendix. In the figures one can see thelines for 99% survival probability with 95% confidence as well as the bestfit lines to the data points. Table 2 summarizes values for the parametersA and B with standard deviations in Eq. (1). It is to be noted that theintercept with the ordinate, A and the slope B decrease with decreasing R.All the best fit lines obtained here pass very closely the point (normalizedapplied stress, fatigue life) = (1.0, 0.25) or (1.0, 0.5), which indicates thatthe fitness of the equations to the data points is good.

3.3.2 Analysis of data by homologous stress methodHomologous stress is defined as the ratio of the maximum applied

stress oa to the estimated tensile strength of i-th specimen, Si, OH = cya/Si.Si is estimated on the assumption that a specimen which displayed the i-thshortest fatigue life in the fatigue specimen set should be the i-th weakestif the static tensile tests are performed on the set of specimens, and thatthe variations of fatigue life and static strength can be expressed by thenormal or Weibull distribution. Here, the homologous stress at each appliedstress level was estimated on the basis of these distributions. When the

- 5 -


estimation was carried out for the lower applied stress levels, the run-outspecimens, i. e., those which had survived up to 105 cycles, were regardedto have a fatigue life of 105 to minimize the possible error that would bebrought about when those specimens would be omitted. There is no largedifference in the homologous stress values between the normal andWeibull distributions, although Weibull distribution gave a little largervalues at very low fracture probabilities. This is reasonably expected thatthe fracture probability in the low stress region is the higher for theWeibull than for the normal. However, the difference in the fatiguestrength between the two distributions are very small, as is shown in Fig.13.

Tables 3 and 4 summarizes the values for the parameters in thefollowing equation on the basis of the normal and Weibull distributions,respectively.

log aH = A + B log Nf (2)

In comparison with the previous calculation , B is a little smaller than thatobtained by the Price's method, i. e., the slope is the larger. The values of Aare larger than those obtained by Price's method. From these facts it isfound that the fatigue strength is much smaller than that in the case ofPrice's method for all the three materials tested. There is one crucial pointabout the homologous stress method from the aspect of its application tothe design and safety analysis of graphite components : Some homologousstresses become larger than 1.0 when the applied stress level is high. Thiswas also incurred in the previous study on IG-11 graphite [7]. The factabove could be a demerit of the homologous stress method if the methodwould be used for determining the fatigue life diagram for the design ofgraphite components. It is also to be noted that the static strengthestimated for the analysis depends on the log and heat of the material, thevariation within a log, the number of data points, etc.

3.3.3 Analysis by P-S-N diagram methodIn the P-S-N diagram method the best fit S-N curve is obtained in the

following way : (1) a probability distribution is assumed to the fatigue lifeat each stress level. (2) the fatigue life at a given fracture probability isestimated for each stress level. (3) plots of the applied stress vs. fatiguelife at a given fracture probability are drawn as the best fit S-N curve [10].

Here, the Weibull distribution was assumed to be applicable to thefatigue life at each stress level, i. e.,

- 6 -


P f =l -exp{ - (Nf /N 0 ) m } . (3)

In the above equation, Pf is the cumulative fracture probability at fatiguelife Nf, and m, No are constants. From Eq. (3),

In ln{ l / ( l -Pf)}=mlnNf-mlnN0 . (4)

Data on IG-11 and PGX graphites were analyzed on the basis of Eq. (4) ,which gives Figs. 14 through 16. Data on ASR-ORB was not analyzed by P-T-S method, because the number of specimens tested was not large enoughto be analyzed by the method.

P-S-N diagram method is, in principle, to be applied to the datawhere the fracture probability distribution is the same for all the stresslevels, i.e., the m-value in Eq. (4) is the same. However, in reality, as is seenin the figures, the slope decreased with decreasing stress level. Therefore,it is judged that the method is not appropriate for the data analysis on thegraphites examined in the present study.

3.3.4 Dependence of S-N curve on R-valueHere, the dependence of the fatigue life on R-value is discussed with

regard to S-N curves obtained by Price's method which is believed to bemost appropriate to analyze the fatigue data on nuclear graphite. It is to benoted tiiat there is no large difference in A-value between the grades ofmaterials as long as R is the same, i. e., R= 0 or = -1 . A-values seem todecrease with decreasing R-value, which is clearly seen in the case of PGX(L) in Table 2. Previous studies on IG-11 graphite indicated that the stresslevel at Nf= 1 decreased from 0.97 to 0.73 of the mean tensile strengthwith decreasing R from 0 to -3.5 [7,8]. In the present experiment on PGXgraphite (L) the stress level at Nf = 1 is 0.96 of the mean tensile strength atR= -3, much larger than that for IG-11. The stress levels in the compressionside in these cases, i. e., R= -3.5 for IG-11 and R= -3.0 for PGX (L), were0.85 and 0.83 of the mean compressive strength of each material,respectively. The fact indicates that the effect of compressive stress on thefatigue life is more pronounced for IG-11 than for PGX. This is probablybecause the defects introduced by high compressive stresses propagateless rapidly during the fatigue process for PGX because of its larger grainsize and higher deformability. In fact, a previous study on the effect ofcompressive prestress on the tensile strength of graphite indicates that thefine-grained, higher strength material showed the larger loss in the tensilestrength if the normalized prestress level was the same [12].

- 7 -


The absolute value of B increased with increasing R-value. Thisindicated that the stress in the compression side would affect the growth ofthe fatigue crack. It is also to be noted that the slope of the S-N curves isless steep for ASR-ORB, comparing with the other materials. This meansthat practically the fatigue behavior would not be observed in the carbonmaterial when the applied stress level is smaller than 0.8 of the meantensile strength, probably because of its pronouncedly brittle nature.

3.3.5 Equi-fatigue life diagramThe above discussion indicates that Price's method is the most

appropriate to analyze the fatigue data on graphite and carbon materials sothat the equi-fatigue life diagram for design, i. e., the modified Goodmandiagram, is proposed for PGX and 'ASR-ORB on the basis of the method.According to the HTTR Graphite Structural Design Code, the design fatiguelife curve is to correspond to 99% survival probability with 95% confidence[11]. The design fatigue life curve is given by

log (aa / Smean) = [ log (aa / Smean) ] mean " ( 2.326 + 1.645 ) S (5)

Here, [ log (aa / Smean) ] mean represents the stress value of the best fitcurve, and n is the number of data points, s, the standard deviation definedin Eq. (1). The design curves for Nf =1, 103 and 105 are shown in Figs. 17and 18 for PGX and ASR-ORB, respectively. In these figures, the meanapplied stress is assumed to be equal to the mean tensile strength for thecaseR=l.

3.4 Fatigue life at high temperature3.4.1 Tensile strength of IG-11 graphite

Results of the tensile test on high temperature fatigue specimen withdifferent treatments are summarized in Table 5. It is to be noted that thetensile strength at room temperature in vacuo is about 24% larger thanthat in air. This trend was also observed in the previous experiment [13]where the tensile strength was found to be about 15% larger in vacuo thanin air at room temperature. Though, in the present study, it is ratherdifficult to evaluate quantitatively the effect of atmosphere on thestrength, it is believed that the service condition of the HTTR wouldincrease the strength of the graphite in comparison with that in air. It isalso interesting to see that the heat treatment at 200*C in vacuo for 2hincreased the strength at room temperature by a few %, resulting in the

- 8 -


strength value which coincides with that at 980*0. The strength ofspecimens which survived after more than 105 cycles of repeated appliedstress seem to have kept the original strength, comparable with or evenlarger than that obtained at room temperature in vacuo. From these resultsit was decided that the mean tensile strength to be used in thenormalization of the applied stress in the high temperature fatigue testwas either that at room temperature in vacuo or that at 980*0 in vacuo, i.e., 40.3 MPa or 42.8 MPa, and the analysis of the fatigue data was done forboth cases.

3.4.2 Fatigue strength at 980t;The results of the fatigue test at 980*0 are shown in Fig. 19 (a), (b)

where one can see the plots of the normalized applied stress versus fatiguelife in comparison with the best fit equation for the room temperature dataobtained in the previous study [8]. In Fig. 19 (a) the applied stress isnormalized to the mean tensile strength at 980"C whereas the meanstrength at room temperature in vacuo is chosen as the normalizingstrength in Fig. 19 (b). The best fit line in the latter case is expressed as

log (c% / Smean) = - 0.0166 - 0.00770 log Nf (6)

which is very close to that obtained for the room temperature fatiguestrength in the condition of R= 0 [8], i. e.,

log (aa / Smean) = - 0.0136 - 0.00877 log Nf (7)

On the other hand, the fatigue strength appears to be rather low if thenormalizing strength is set at the mean strength at 980*0 in vacuo, i. e.,

log (aa / Smean) = - 0.0427 - 0.00776 log Nf (8)

The value of aa / Smean at "105 cycles is 0.881 and 0.829 for Eqs. (6) and (8),respectively. For the room temperature fatigue strength, a a / Smean is 0.876.From the above calculations it is concluded that the fatigue strength at980*0 is well described by an equation analogous to that for the roomtemperature fatigue strength when the mean strength at roomtemperature is chosen as the normalizing strength. If the normalizingstrength is set at the mean strength at 980*0, the best fit line gives thelower fatigue strength than is estimated from the data on the roomtemperature fatigue strength. This implies that the fatigue strength at980*0 is evaluated on the conservative side if the applied stress is

- 9 -


normalized to the mean tensile strength at 980°C. One thing to be noted isthat 17 out of 38 specimens which were subject to repeated stressessurvived even at 105 or more, and that the fatigue life of these specimenswas treated as 105 in the present analysis. The fact is believed to make theanalysis even more conservative.

4. Conclusions

Room temperature fatigue tests were carried out on PGX graphite,ASR-ORB carbon and IG-11 graphite which are used in the core region ofthe HTTR. The ratio of applied stress R= amin / omax was -3,-1,0 (PGX), -1,0 (ASR-ORB) and -1 (IG-11). The room temperature data were analyzed bythree different methods, i. e., Price's, homologous stress and P-T-S diagramto investigate which method is the most appropriate. Fatigue tests werealso carried out at 980°C in vacuo for IG-11 graphite to evaluate the effectof temperature on the fatigue strength.

Main conclusions are:1) Among the three methods above the Price's method was the most

appropriate to derive a design S-N curve from the data set. P-T-Sdiagram method could not be applied to the present data.

2) The fatigue strength of PGX and ASR-ORB decreased with decreasingR-value, which is consistent with the result which was obtainedpreviously on IG-11. Although the variation of data was the largestamong three materials, the fatigue strength normalized to the meantensile strength was the largest for ASR-ORB.

3) On the basis of the room temperature data analyzed by Price's method,design S-N curves were obtained for PGX and ASR-ORB.

4) Fatigue tests at 980°C on IG-11 graphite indicated that the S-N curvewas almost the same as that for the room temperature curve when theapplied stress was normalized to the mean tensile strength at roomtemperature in vacuo.

AcknowledgmentsThe authors are indebted to Messrs, M. Matsumoto and S. Sawahata

for the high temperature fatigue test. They are also grateful to Dr. T. Iyokufor valuable comments.

- 10 -


References1. T. Aral, et al., JAERI-M 86-192 (1986), p. 582. R. Vollman, et al., ibid., p.43.3. T. Iyoku et al., ibid., pl2O.4. HTTR Design Laboratory, JAERI-M 89-006 (1989).5. R. J. Price, Carbon 16, 367 (1978).6. J. B. S. Wilkins and A. R. Rice, AECL-4216 (1972).7. S. Ishiyama and T. Oku, JAERI-M 86-145 (1986).8. S. Ishiyama et al., J. At. Soc. of Japan 29, 651 (1987).9. M. Eto et al., JAERI-M 84-148 (1984).

10. J. Soc. for Mechanical Engineers, JSME S002 (1981).11. HTTR Design Laboratory, JAERI-M 89-006 (1989).12. T. Oku and M. Eto, Carbon 11,639 (1973).13. M. Eto etal., Carbon 29,11 (1990).

n -

Table 1 Some mechanical properties of specimens used in the present experiment

I

Grade

IG-11PGX

ASR-ORB

Direction

LLTLT

Apparentdensity(g/cm3)

1.781.741.741.651.65

Young'smodulus

(GPa)10.126.618.499.45

10.72

Tensilestrength

(MPa)27.83

8.299.186.516.78

Bendingstrength

(MPa)40.5110.9111.6111.3711.91

Compressivestrength

(MPa)80.9130.5930.2457.9550.39

L: Parallel to the longitudinal axis of the blockT: Perpendicular to the longitudinal axis of the block

00I

O

Table 2 Results of statistical fatigue data analysis by Price method

Grade

IG-11

PGX

ASR-ORB

Direction

LL

T

L

T

Stress ratio, R

omina max

-10

-1-30

-10

-10

-1

Intercept ofleast squaresline, A

-0.0072-0.0058-0.0092-0.0195-0.0094-0.0100-0.0046-0.0096-0.0057-0.0097

Slope ofleast squaresline,B

-0.0235-0.0098-0.0215-0.0360-0.0133-0.0230-0.0073-0.0078-0.0054-0.0138

Standarddeviation

s

0.01990.02930.03300.03230.02280.02570.02910.03040.03700.0385

Normalized strfor survival to50%

probability0.7500.8810.7650.6320.8400.7500.9100.8940.9270.835

~ r r o max105 cycle Smean99/95 lowertolerance limit

0.6680.7440.6320.5240.7360.6460.7660.7470.7450.664

toCO

oto

Table 3 Results of statistical fatigue data analysis by homologous stress method (Normal distribution)

Grade

IG-11

PGX

Direction

L

L

T

Stress ratio, Romina max

- 10

-1-30

-1


0.02060.02990.04480.00600.00770.0228

Slope ofleast squaresline,B

-0.0319-0.0253-0.0428-0.0432-0.0227-0.0354

Homologous stress limit, o maxfor survival to 105 cycles s mean

50% probability0.6610.6980.5510.6000.7560.632

Table 4 Results of statistical fatigue data analysis by homologous stress method (Weibull distribution)o

00

OCO4

Grade

IG-11

PGX

Direction

L

L

T

Stress ratio, Ramina max

-10

_ i

-30

-1


0.02140.03120.04730.00510.00980.0248

Slope ofleast squaresline, B

-0.0322-0.0262-0.0438-0.0433-0.0233-0.0360

Homologous stress limit, ° maxfor survival to 105 cycles s mean

50% probability^0.6570.6880.5420.6000.7480.624


Table 5 Tensile strength of IG-11 graphite in different test conditions

Condition Tensile strength(MPa)

98O°C in vacuo

RTinvacuo

42.8 ± 2.0 (8)

40.3 ± 3.7 (3)

RTinvacuoafter > 105 cycles

RTinvacuoafter 200*0 X2h

RTinair(including after > 105)

40.9 ± 1.4(17)

42.8 ± 0.7 (4)

32.4 ± 2.3 (3)

( ) : Number of specimen

Table 6 Fatigue life of IG-11 (L) at 980*0 in comparison withthat at room temperature (R= 0)

Temperature Stress normalization A B

RT Applied stress normalized to themean tensile strength at roomtemperature in air.

980°C Applied stress normalized to themean tensile strength at 980*0.

-0.0136 -0.00877

-0.0427 -0.00776

980*0 Applied stress normalized to themean tensile strength at roomtemperature.

-0.0166 -0.00770

15


[mm]

Size of the blocks: Large (L) : 155° x 500 mm Small (S): 115DX 430 mm

Fig. 1 Size of the original block and cutting plan for specimenpreparation for IG-11 graphite.

- 16 -


[ m m ]

Fig. 2 Specimen for the room temperature fatigue test.

- 17 -


[ mm ]

Fig. 3 Specimen for the high temperature fatigue tests.

- 18 -


1•n

gth

0)

essi

veC

ompr

35

30

25

40

35

30

Fig.4 (a)

25

1 | • 1 1 1 U l l

!

: .^—Li—•-: L-diredrtion

:T

'TT

'I ']

:• T-directlon

1 T

i i I I I I I I I i i i i 1 1 1 f

1

T

" • • • • • • • • • • • • • "

JI

• 1 1 1 M i l

1 1 1 1 1 I I I

l "I ^

. . • , , . M

10d 105

Specimen Volume (mm3)

1 0 f c

Isnoaa>{0

»—i

0)

<DH

15

10

5

20

15

10

1 1 1 f l l l l

: L-direction

" I; I

: T-direction

B—'— :

. . . . . .

i i i i i i i i

I

x r -i I :

«§• swr <|

Specimen Volume (mm3)

Fig.4 (b)

Fig. 4 Tensile and compressive strength of PGX graphite specimenswith different volumes, (a) tensile strength, (b) compressivestrength.

- 19 -


99.9999.9

. 99e- 955* 90

Iw

Bl

3 5018

1.1

.0124 25 26 27 28 29 30

Tensile Strength (MPa)31 32

Fig. 5 Plots of fracture probability versus tensile strength of IG-11graphite specimens (L-direction).

- 20


£GXGriaDMteiLdir.e£tionlx=8.29!MPa

7 8 9 10Tensile Strength (MPa)

Fig. 6(a)

99.9999.9

7 8 9 10Tensile Strength (MPa)

Fig. 6(b)

Fig. 6 Plots of fracture probability versus tensile strength of VGXgraphite specimens, (a) L-direction and (b) T-direction.

21


99.9999.9

>. 95

tJ 90

1 98I 5°S x°fe 1

.1.01 . , . , . . , , . . . ,

4.5

Hg. 7(a)

5.5 6.5Tensile Strength (MPa)

99.9999.9

t 955* 90

B 50

£ 103 5S !{X.

.1.014.5 5.5 6.5

Tensile Strength (MPa)

Fig. 7(b)

7.5

7.5 8

^

8

Fig. 7 Plots of fracture probability versus tensile strength of ASR-ORBcarbon specimens, (a) L-direction and (b) T-direction.

- 22 -

COoj

a.a,•<a

cds

Si

CO

<bH

aS

1

0.9

0.8

0.7

0.6

n « ;

IGR=

JAERI-Research

|••••GO--GH>-G ••

-11 (L)-1.0

^ ^

1 1 1 .

98-024

!

oo

. .......

! • ' " " " 1

|

]

4500 9TTTYy:?*l ^fc

•^ ^ ^ ^ <p OO -

110rl iou 101 ior

Number of Cycles to Failure N10fc

Fig. 8 Fatigue test data with R=-1.0 on IG-11 graphite (L-direction), theS-N curves obtained by Price method. A 99/95% lower tolerancelimit represents the limit above which at least 99% of all datawould fall with 95% confidence.

PGX Graphite (L)

10° 101 10z 103 10" 10D

Number of Cycles to Failure N

Fig. 9(a) Fatigue test data with R=0.0 on PGX graphite (L-direction), theS-N curves obtained by Price method. A 99/95% lower tolerancelimit represents the limit above which at least 99% of all datawould fall with 95% confidence.

- 23 -


tit<b

I

0.8

0.7

0.6

0.5

PGX Graphite (L)

IO"1 io° IO1 io2 io3 io4 io5 io6


Fig. 9(b) Fatigue test data with R=-1.0 on PGX graphite (L-direction), theS-N curves obtained by Price method. A 99/95% lower tolerancelimit represents the limit above which at least 99% of all datawould fall with 95% confidence.

S-53 0.8<

PGX traphife (L)' R=-3.0

I I I I 11I il I I I 111 III I I I I 11 111 I I I I II III 1 1 .1 I 11 I

S 2 0.6

0.510"1 10° 101 102 103 104 105


Fig. 9(c) Fatigue test data with R=-3.0 on PGX graphite (L-direction), theS-N curves obtained by Price method. A 99/95% lower tolerancelimit represents the limit above which at least 99% of all datawould fall with 95% confidence.

- 24 -


tit f3

fit

0.8

0.6

0.5

1 s i |

gI

: -~- -§-

" PG:R=

I !1 !1

ob ceo

—.— j____©

1CGrapliite(T)0.0 !

o oCO

— • —

CO OC

— - — - ^

i l l ;f t i i iii|l } i \ itt nl t i t i mi l i I I mi l l i i 11 m i

"o-©—.

- — -

ooQOO •

i | i i mil t i i inn

10"1 10° 101 102 103 104 10510*Number of Cycles to Failure N

Fig.lO(a) Fatigue test data with R=0.0 on PGX graphite (T-direction), theS-N curves obtained by Price method. A 99/95% lower tolerancelimit represents the limit above which at least 99% of all datawould fall with 95% confidence.

COto

B

•a

PG^GraplfitR= 1-1.0

g 0.9

T& 0.8

0.7

0.6

0.510"1 10° 101 102 103 104 1O5

Number of Cycles to Failure NFig.lO(b) Fatigue test data with R=-l,0 on PGX graphite (T-direction), the

S-N curves obtained by Price method. A 99/95% lower tolerancelimit represents the limit above which at least 99% of all datawould fall with 95% confidence.

i i mini i t i mnl i i iniiil i i i mill i i 11 mil i i 11 IIHI t i I utii

- 25 -


I K 0.9ft to U.o

0.7

0.6

0.5

R=ASH-ORB Garbon £L)

P.O

...i

10- l 10° 101 102 103 104 105

Number of Cycles to Failure N10fc

Fig. l l (a) Fatigue test d a t a with R=0.0 o n ASR-ORB carbon (L-direction), t heS-N curves ob ta ined by Price method . A 9 9 / 9 5 % lower tolerancelimit represents the limit above which at least 99% of all datawould fall with 95% confidence.

<C

I

fl.st 1

& 0.9

3 0.8<1>

d 0 J

S 0.6

—

_ o _

o

, . . . . . . .

O

!

1:; •xsoeo- ka>

-O-4-G0-QQ.0 ooI!

: i i •

ASJl-ORB CR= -1.0

i i | n n m

1 "7 1I

arbon (L){

i

-

t > 1 1 I I III I 1 1 1 I l l l

~

o

DO

0

-

10-i10fc10° 101 102 103 104 105


Fig . l l (b) Fatigue test d a t a with R=-1.0 on ASR-ORB carbon (L-direction),t h e S-N curves obta ined by Price me thod . A 9 9 / 9 5 % lowertolerance limit represents the limit above which a t least 99% ofall d a t a would fall with 95% confidence.

- 26 -

3 0.8

S 0.6

0.5


]..„ 1 _

AS|l-0RB CarbonR=D.O

T)

i i i imil i t 11 inil I I I Illl

10"1 10° 101 102 103

Number of Cycles to Failure N105 10fc

Fig. 12(a) Fatigue test data with R=0.0 on ASR-ORB carbon (T-direction), theS-N curves obtained by Price method. A 99/95% lower tolerancelimit represents the limit above which at least 99% of all datawould fall with 95% confidence.

to

& 0.9

I-5J 0.8

0.7

0.6

0.5i i 111 in I i i i i n nl i i i mill i i 11 11 ill t i 11 mil i i i in it

10'1 10° 101 102 103 104 105 106


Fig.l2(b) Fatigue test data with R—1.0 on ASR-ORB carbon (T-direction),the S-N curves obtained by Price method. A 99/95% lowertolerance limit represents the limit above which at least 99% ofall data would fall with 95% confidence.

- 27 -


coCO

1

10.9

3o 0.8

I °'7o

0.6

0.5 I I I 11 in I I i i mill

10e10'1 10° 101 102 103 104 105


Fig. 13 Analysis by the homologous stress method: comparison of thebest fit lines on the basis of the normal and Weibull distribution,(a) IG-11 (L), R= -1,

COCOs

C/3toO

o

Hom

ol

10.9

0.8

0.7

0.6

(b)

-.

-

1 1 f 1 1 III . . ,

PGX (i • " • " "

L) R=C)

•

1 • " " " i "

"WoiNO

---Wermal 1libull 1

-

-

t 11111 it0.510"1 10° 101 102 103 104 105 106


(b) PGX (L), R= 0,

- 28 -


COto

0.9

1 0.7I0.6

0.510"1

1 • " " • ' •

1 (c)1

PGX

;

1 ' • " " '

(L) R=Ki i

Normal 1WeibuU |

•

i i 11 mi I i i i I I I ill i i 11 mil i i i I I ml i i 11 I I til i i i i it ill i i i nt II


(c) PGX (L), R= - 1 ,

10° lO1 102 103 104 105 106

COCOs

M

COOtJJOO

£

I I f i t til I i i t t n t i l t I i l l m l

10-1 10° 101 102 103 104 105 106


(d) PGX, (L), R- -3,

- 29 -


Stre

ss

sno3<

Hom

olc

1

0.9

0.8

0.7

0.6

n =;

i t 111in

(e)

-

i ^ i n | III

1 1

1 I 111II

1 1 M i . . .

PGX•

i ^ f K n i 1

. . . . . . . .

(T) R= 0' " " I '

:

" ~ — ^

-Normal 1- Weibull 1

1O'1 10° 1O1 102 103 104 105 106


(e) PGX (T), R- 0,

(O(O

Str

CO

logo

uH

omol

1

0.9

0.8

0.7

0.6

0.5

(0 I NormalWeibull

10'1 10° 101 102 103 104 105 106


(f) PGX (T), R= - 1 .

- 30 -


a,Vl

4

3

2

1

0

—• - i

~ -2

-3

-4

!_ IGHlliLjtirectibn)

1.0 of

-0.8ac

l l t I l i i

-2 0 8 10 12 14hi N.

Fig. 14 Plot of fatigue test data with R=-1.0 on IG-11 graphite (In-direction) for P-S-N analysis.

°r

! • !

PGX (L dii-ectioii)

-2 14In N

Fig.l5(a) Plot of fatigue test data with R=0.0 on PGX graphite (L-direction)for P-S-N analysis.

- 31 -


4

3

_ 2

5 o

B -2

-3

-4

' ' ' I ' " ' I " ™ I I: PGX (L direction) I !— -i ?• : * * : •

1.0 of

-0.9o f

-0.8af

-2 0 8 10 12 14In N

Fig.l5(b) Plot of fatigue test data with R=-1.0 on PGX graphite (L-direction)for P-S-N analysis.

3

2

<=H 1

5 o5 -i- -2

-3

-4

PGJC(L direction)R=

r 1 id w

-3.0 i

©• / • !

nt J

-AA!1f •»•

/A: \ • !

£? 7

. . . i . . . i . . . i , . . i . . . j . • • i . . . i . . . :

-2 0 8 10 12 14In N

Fig.l5(c) Plot of fatigue test data with R=-3.0 on PGX graphite (L-direction)for P-S-N analysis.

- 32 -


L l . l J , ! . . , , i , , , i . , , , , .

-2 8 10 12 14In N

Fig.l6(a) Plot of fatigue test data vsdth R=0.0 on PGX graphite (T-direction)for P-S-N analysis.

4

3

_ 2

5 o

^ -2

-3

-4

I ! " ' " IPGX (T direction)T?= -1.0

-2 0 8 10 12 14In N

Fig.l6(b) Plot of fatigue test data with R=-1.0 on PGX graphite (T-direction)for P-S-N analysis.

- 33 -


(a)Rooin tempjeraturd

10"1 10° 101 10 103 10s 10e

Fatigue life (cycles)

Fig.l7(a) Constant fatigue life diagram for PGX graphite (L-direction),fatigue lines represent 99% survival probability with 95%confidence.

CDCO

! x

I 10.9•^^0.8

^ ^ 0 . 7

I^ 0.6

0.5

(b)Rdom terliperatiire

lOg(aa/S mean)=-0.

••

ago oo <$> o

0166-0.00770 log Nf

;I I t I I III I I I I 111 III I I I I ! M i l 1 | I I l l l i l I i l l n nl

10,-1 10° 101 10z 104 10fc

Fatigue l ife (cycles)

Fig.l7(b) Constant fatigue life diagram for PGX graphite (T-direction),fatigue lines represent 99% survival probabili ty with 95%confidence.

- 34 -

COen

-26 -24 -22 -20 -18 -16

enen\-

X

•10 -8 -6T

MINIMUM STRESS / MPa4 - 2 0 2 4 6 8 10

r

0-0.5 0 Q.5

MINIMUM STRESSMEAN TENSILE STRENGTH

1.0

00I

oto

Hg.l8(a) Constant fatigue life diagram for ASR-ORB carbon (L-direction),fatigue lines represent 99% survival probability with 95%confidence.


MINIMUM STRESS / MPa

-10 - 8 - 6 - 4 - 2 0 2 4 6 8 10

CO

en

-1.0 -0 .5 0 0.5MINIMUM STRESS

MEAN TENSILE STRENGTH

1.0

Fig.l8(b) Constant fatigue life diagram for ASR-ORB carbon (T-direction),fatigue lines represent 99% survival probability with 95%confidence.

36 -


MINIMUM STRESS / MPa- 2 0 2 4 8

oo-enLLJ

cc•fc

x

-1.0 -0.5 0 0.5MINIMUM STRESS


00ooUJ

OO

UJeni—

00

IxJ

00

UJ

MINIMUM STRESS / MPa• 8 - 6 - 4 - 2 0 2 4

-1.0 -0.5 0 0.5MINIMUM STRESS


i.O

Fig. 19 Fatigue strength of IG-11 graphite at 980*0 in vacuo,(a) applied stress normalized to the mean tensile strength at980*C and (b) applied stress normalized to the mean tensilestrength at RT in vacuo.

- 37 -


Appendix: Digital Data on Static Strength and Fatigue Strength

Table Al Tensile strength of PGX graphite (T)

No. Tensile strength (MPa)5mm Dx 10mm GL 10mm DX 10mm GL 20mm DXlOmm GL

9.347.688.418.187.519.038.998.238.518.70

123456789

10

8.869.509.118.499.148.908.999.449.30

8.159.178.948.148.80

10.039.808.649.479.10

MeanStd. Dev.

9.080.32

9.020.63

8.460.59

Table A2 Tensile strength of PGX graphite (L)

No. Tensile strength (MPa)5mm DXlOmm GL 10mm DXlOmm GL 20mm DXlOmm GL

12345678910

8.189.968.515.297.436.947.727.638.05

8.847.268.177.616.435.786.648.316.908.65

8.498.918.837.096.694.646.298.608.47

MeanStd. Dev.

7.751.25

7.461.02

7.561.47

- 38 -


Table A3 Compressive strength of PGX graphite (T)

No.

123456789

1011121314151617181920

MeanStd.Dev.

5mm DXlOmm L

29.0830.9830.5530.9728.0427.6928.4028.7425.2725.6631.2130.1629.8830.2128.7227.2128.9528.7731.0828.4929.00

1.69

Compressive strength (MPa)10mm D

x20mm L31.3831.7631.1531.1129.4230.0930.8831.5326.6326.9032.8832.4830.8530.1930.0727.8929.6830.0630.0729.7430.24

1.63

15mm Dx30mm L

31.8432.0530.9531.3130.4230.8531.5531.8627.4427.1932.7732.3831.8431.8130.6230.6030.4429.8730.3330.2930.82

1.43

30mm Dx60mm L

31.8731.9130.5732.1128.2632.9632.2430.7730.6031.1929.8129.2829.9129.9830.1530.4332.2030.5331.6530.7030.86

1.16

50mm DxlOOmmL

30.9131.9928.1130.8530.4632.9331.9527.8830.6930.32

30.611.60

- 39 -


Table A4 Compressive strength of PGX graphite (L)

No.

123456789

1011121314151617181920

MeanStd. Dev.

5mm DxlOmm L

30.0728.8129.7730.3128.8129.4728.9430.6530.4528.3927.1829.0330.5430.3129.4329.5229.4729.3429.5329.0129.45

0.84

Compressive strength (MPa)10mm D

X 20mm L31.2829.9930.3231.3829.8329.5529.8030.2730.6830.4730.7929.6132.6432.7029.3830.2330.8130.3630.7930.8130.59

0.90

15mm Dx30mm L

31.0231.1231.0632.0931.4531.0431.0130.7530.6830.7731.5731.0832.9432.1730.8330.7930.6830.3931.8330.7131.20

0.63

30mm Dx60mm L

31.4431.2131.3731.0131.0730.6931.0731.9031.8331.6332.8433.9632.5532.4529.5333.0228.9128.5030.9731.8831.39

1.33

50mm DxlOOmmL

30.9132.6330.1829.8030.9330.5530.2331.8631.9332.03

31.110.95

- 40 -


Table A5

No.

123456789

10111213141516171819202122232425

Average

Results of tensile strength test(L direction)

Diameter(mm)

10.01310.02610.0279.988

10.00010.01810.0269.985

10.01210.0299.990

10.02010.02210.01410.01910.04210.02610.02110.00410.00010.4509.9849.990

10.01410.024

Standard deviation

Young's modulus(GPa)10.139.879.089.85

10.489.85

10.2510.1310.609.74

10.979.74

10.679.74

10.309.969.79

10.4210.2910.3410.7310.139.63

10.3010.07

10.120.41

on IG-11 graphite

Tensile strength(MPa)28.9627.5826.0827.2230.1326.6029.1927.0928.8126.8530.8927.8329.2127.5529.2325.0727.3328.6026.5128.5729.7028.1026.2727.5524.85

27.831.51

- 41 -


Table A6(a) Results of tensile strength test on PGX graphite(L direction)

No.

123456789

10111213141516171819202122232425262728293031323334

Diameter(mm)

10.02310.22010.00410.02310.01110.02710.01310.01910.01710.02910.0259.998

10.0159.965

10.02310.0409.984

10.02510.01610.02010.02110.00910.02510.01910.00610.01410.02210.0019.9729.9979.987

10.01610.01910.060

Young's modulus(GPa)6.656.646.616.716.586.446.616.496.616.776.586.656.686.836.806.616.766.496.616.586.806.806.646.736.836.526.676.676.676.426.556.326.526.59

Tensile strength(MPa)8.899.078.258.428.578.018.138.188.907.237.958.688.788.089.128.528.557.887.249.056.627.777.087.837.097.037.437.888.337.959.507.508.608.83

- 42 -


No. Diameter Young's modulus Tensile strength(mm) (GPa) (MPa)

35363738394041424344

10.06810.02510.01710.04410.01310.01910.00010.02210.03810.043

6.466.646.406.526.216.776.646.676.526.71

8.929.098.469.268.239.198.589.178.438.39

Average 6.61 8.29Standard deviation 0.14 0.70

- 43 -


Table A6(b) Results of tensile strength test on PGX graphite(T direction)


1234567891011121314151617181920212223242526272829303132333435

10.0099.9919.9889.98810.00510.00610.0319.97910.0099.9819.9669.98910.0049.9769.98810.01410.01610.00010.0259.98810.0039.9809.9719.9599.9759.9859.9909.9729.96910.0009.96010.0279.970

10.01710.030

8.708.488.398.618.478.568.388.618.478.608.528.348.348.488.308.348.568.568.178.518.528.438.478.438.138.748.608.528.608.518.438.528.618.528.70

10.289.198.829.018.799.109.069.539.669.349.249.579.119.918.518.968.908.497.957.829.559.159.809.138.4110.219.389.799.999.119.699.389.617.969.00


- 44 -


Table A7(a) Results of tensile strength test on ASR-ORB carbon(L direction)


1234567891011121314151617181920

9.9909.98010.00010.00010.0009.9809.98010.0009.9809.9709.99010.01010.01010.00010.01010.0109.9709.99010.00010.000

9.739.739.469.0610.019.789.729.459.568.959.159.639.209.789.729.728.959.299.059.09

5.007.086.775.537.126.587.217.026.836.346.766.675.926.997.175.896.416.576.026.37


- 45 -


Table A7(b) Results of tensile strength test on ASR-ORB carbon(T direction)

No.

123456789

10111213141516171819202122

Average

Diameter(mm)9.980

10.00010.02010.0009.980

10.00010.0009.974

10.0009.9809.990

10.0109.980

10.00010.01010.0109.990

10.00010.0109.9809.9809.970

Standard deviation

Young's modulus(GPa)10.6310.7510.2610.7610.8210.9410.8810.7410.6810.1810.8811.0110.6710.8710.8710.3710.8110.7410.8110.6110.8810.74

10.720.21

Tensile strength(MPa)7.156.125.667.595.747.527.847.695.216.427.416.296.686.437.796.576.357.097.607.215.867.00

6.780.77

- 46 -


Table A8 Fatigue life data on IG-11 graphite (L direction)

R (Stress ratio) =-1.0Mean tensile strength : 27.83 MPa

No.

123456789

10

123456789

1011121314151617

Applied stress(MPa)27.8327.8327.8327.8327.8327.8327.8327.8327.8327.83

25.0525.0525.0525.0525.0525.0525.0525.0525.0525.0525.0525.0525.0525.0525.0525.0525.05

Number ofcycles

2222237

101022

20404555636580

100106114157164170176190547816

No.

123456789

101112

123456789

10111213

12345

Applied stress(MPa)22.2622.2622.2622.2622.2622.2622.2622.2622.2622.2622.2622.26

20.8720.8720.8720.8720.8720.8720.8720.8720.8720.8720.8720.8720.87

19.4819.4819.4819.4819.48

Number ofcycles

10451098115415361677197741117392

157103106056900

>103100

98441159015040193502540056690274303844099050

>105000>110800>112000>121600

>101800>105300>192600>217500>322200

- 47 -


Table A9(a) Fatigue life data on PGX graphite (L direction)

R (Stress ratio) =0.0Mean tensile strength : 8.294 MPa

No.

123456789

1011121314151617

123456789

10

Applied stress(MPa)8.2948.2948.2948.2948.2948.2948.2948.2948.2948.2948.2948.2948.2948.2948.2948.2948.294

7.8797.8797.8797.8797.8797.8797.8797.8797.8797.879

Number ofcycles

0.50.526

1017192028373845597090

172359

0.525

1124394894

33367396

No.

123456789

101112

i-i

23456

1234

Applied stress(MPa)7.4657.4657.4657.4657.4657.4657.4657.4657.4657.4657.4657.465

7.0507.0507.0507.0507.0507.050

6.6356.6356.6356.635

Number ofcycles

49304307482503632

832813911846826185019383

437108303029099970

>178100>221600

>126000>130880>191800> 207500

- 48 -


Table A9(b) Fatigue life data on PGX graphite (L direction)


No.

123456789

101112131415

123456789

1011121314151617

Applied stress(MPa)8.2948.2948.2948.2948.2948.2948.2948.2948.2948.2948.2948.2948.2948.2948.294

7.4657.4657.4657.4657.4657.4657.4657.4657.4657.4657.4657.4657.4657.4657.4657.4657.465

Number ofcycles

4489

1011121213141417192160

910114552739596

105136147197271350622968

2062

No.

123456789

101112131415

12345

1


5.8065.8065.8065.8065.806

4.976

Number ofcycles

10197320706

1146138419822035215126535898

104101494039720

>143500

612918080

>144100>146100>147200

>140600

- 49 -


Table A9(c) Fatigue life data on PGX graphite (L direction)


No.

12345

1234567

1234567

Applied stress(MPa)8.2948.2948.2948.2948.294

7.4657.4657.4657.4657.4657.4657.465

6.6356.6356.6356.6356.6356.6356.635

Number ofcycles

22336

58

1518242736

648788

110148154161

No.

12345678

1234567

Applied stress(MPa)5.8065.8065.8065.8065.8065.8065.8065.806

5.3915.3915.3915.3915.3915.3915.391

Number ofcycles

715835

10292050217758396698

24180

1410079330

>102500>108200>133700>135900>161500

50 -


Table A10(a) Fatigue life data on PGX graphite (T direction)


No.

123456789

101112131415

12345678

12345678


8.7248.7248.7248.7248.7248.7248.7248.724

8.2658.2658.2658.2658.2658.2658.2658.265

Number ofcycles

0.50.50.50.50.50.50.53456

14153953

224

193538

165369

811405675

137159225

No.

9101112131415161718

123456789

10111213

123


7.8067.8067.8067.8067.8067.8067.8067.8067.8067.8067.8067.8067.806

7.3467.3467.346

Number ofcycles

233311332360414621627664715

1993

4246

378379515

1268166848548525

1423025470

>123500>168900

>122800>153300>253500

- 51 -


Table A10(b) Fatigue life data on PGX graphite (T direction)


No.

123456789

10111213

123456789

10111213

Applied stress(MPa)9.1839.1839.1839.1839.1839.1839.1839.1839.1839.1839.1839.1839.183

8.2658.2658.2658.2658.2658.2658.2658.2658.2658.2658.2658.2658.265

Number ofcycles

0.250.250.2534

1010101520283245

2426486367748792

110114200207

1467

No.

123456789

101112131415161718

12345

123

Applied stress(MPa)7.3467.3467.3467.3467.3467.3467.3467.3467.3467.3467.3467.3467.3467.3467.3467.3467.3467.346

6.8876.8876.8876.8876.887

6.4286.4286.428

Number ofcycles

262267278365377512680894940

102511281723204023424497

136601954024970

138602312052800

>121600>169400

>118300>169100>289600

- 52 -


Table Al l(a) Fatigue life data on ASR-ORB carbon (L direction)


No.123456789

10111213

123456

123456789

Applied stress (MPa)6.5126.5126.5126.5126.5126.5126.5126.5126.5126.5126.5126.5126.512

6.1866.1866.1866.1866.1866.186

5.8615.8615.8615.8615.8615.8615.8615.8615.861

Number of cycles0.50.50.52334

101461

138700736

71034

382557

54110

60675

388542546117

2393050990

>333900>385900

- 53 -


Table Al l(b) Fatigue life data on ASR-ORB carbon (L direction)

R (Stress ratio) =-1.0Mean tensile strength: 6.512 MPa

No.

123456789

10

123456789

1011


6.1866.1866.1866.1866.1866.1866.1866.1866.1866.1866.186

Number ofcycles

0.250.254

1016202840

124145

0.250.250.250.254

5256

152195423615

No.

123456789

10

1234

1


5.2105.2105.2105.210

4.558

Number ofcycles

0.2594

283400429

1300148802149043190

>124800

>111800>114100>115700>196300

>117300

- 54 -


Table Al 2(a) Fatigue life data on ASR-ORB carbon (T direction)

R (Stress ratio) =0.0Mean tensile strength.: 6.783 MPa

No.123456789

123456789

10

12345678

Applied stress(MPa)6.7836.7836.7836.7836.7836.7836.7836.7836.783

6.4446.4446.4446.4446.4446.4446.4446.4446.4446.444

6.1056.1056.1056.1056.1056.1056.1056.105

Number of cycles0.50.50.59

20213539

256

0.5101875

1654781

169003153035280

>354800

5531943043130

>108200>119400>127400>146100>150000

- 55 -


Table A12(b) Fatigue life data on ASR-ORB carbon (T direction)


No. Applied stress(MPa) Number of cycles12345678

12345678

12345678

123456

6.7836.7836.7836.7836.7836.7836.7836.783

6.1056.1056.1056.1056.1056.1056.1056.105

5.7765.7765.7665.7665.7665.7665.7665.766

5.4265.4265.4265.4265.4265.426

0.250.250.250.253320

423

12523810651474266031906651

>126700

723572530670123180

>100500>115200>116200

2854>100000>117100>119900>125000>157100

- 56 -


Table,

SpecimenNo.

KA)2(A)3(A)4(A)5(A)6(A)7(B)8(B)9(B)

1O(B)11(B)

12(B)

13(C)14(C)15(C)

16(C)

17(C)

18(C)19(4)20(5)21(1)22(2)23(5')24(4')25(3)

26(8)

27(7)28(6)29(13)

\13 Results of

Tensilestrength

(MPa)43.543.142.7

42.3*

43.1

41.9

29.9*

43.7

32.047.139.935.445.538.337.142.3

39.9*

39.7*39.9*40.7*

Appliedstress(MPa)

33.233.233.233.233.233.233.233.2

33.233.2

33.2

35.6

33.633.637.2

the tensile

Normalizedstress

(oa/Smean)

0.7750.7750.7750.7750.7750.7750.7750.775

0.7750.775

0.775

0.832

0.7850.7850.869

and fatigue tests of IG-11 graphite

Fatiguelife

(cycles)

700029880

80909240

1407044650

400>114980

87900>321100

>100000

>100000

>100000>100000>100000

Test condition

980t: in vacuodittoditto980t fatigue in vacuodittodittodittodittodittoditto980t fatigue,RT in vacuo200^:^1 vacuo X2h,RT in vacuo980X: fatigue in vacuoditto200^ in vacuo x2h,RT in vacuo980T: fatigue in vacuo,RT in air200°C in vacuo X2h,RT in vacuoRTinair980°C in vacuodittoRT in airRT in vacuodittoditto200t) in vacuo,RT in vacuo9 8 0 t fatigue,RT in vacuodittodittoditto

- 57 -

SpecimenNo.

30(10)

31(C9)32(D2)33(C2)34(D3)35(D5)36(D6)37(C5)38(C7)39(D12)40(D14)41(D7)42(D8)43(D9)44(D10)45(C10)46(D13)47(D15)48(C1)49(D1)50(C8)51(D6)52(12)53(11)54(C4)55(D4)

Tensilestrength

(MPa)40.7*

40.3*40.5*40.3*40.8*41.9*40.4*43.1*42.3*42.0*42.9*

41.141.543.137.2*

Appliedstress(MPa)39.9

36.036.035.636.037.237.238.738.038.038.538.038.738.738.738.738.538.6

37.237.237.236.037.2

JAERI-Research

Normalizedstress

(oa/Smean)0.932

0.8410.8410.8320.8410.8690.8690.9040.888 .0.8880.8990.8880.9040.9040.9040.9040.8990.902

0.8690.8690.8690.8410.869

98-024

Fatiguelife

(cycles)>100000

>103570>104920>101010>104560>100090>100440>100050>104430>1OO32O>102840

23160980330

50270

88300

>1004401170

1020019501000

Test condition

980°C fatigue,RT in vacuodittodittodittodittodittodittodittodittodittoditto980°C fatiguedittodittodittodittoditto

980°C in vacuodittodittoditto980°C fatiguedittodittoditto

k) Specimens tensile-tested after fatigue cycles more than 105.

- 58 -

(SI) t

ft

M

ftft¥•

m.

•Pal

ft a

3D ftft ft

*

T

7

-

ify

A*

y

T 7

h'' 7

J>

t'

T

T

A-

A

ry

A>

7

y

ie ^m

kgsAK

molcd

radsr

313

J

® ij

E t) ,x;jwu+*-,

i m ,m n mSl *W 9.a y ?

W

•i y ¥-fe >v -y >

ftM

^ m

t

SS ftf S, « ffiE» i^^^J

S •ffi fit

? * y xIK

? 9 y x

) x g |

K

IS a

—i?

7

aiz

•o•f

^

•fe

IV

IV

«

0'•y

* *

i - h

x Aa -

— D

A/

7 7-

- / y

x -

X

y UJI/ -y <? x

- ^

9

9 \yV

-*,\,

•y

y

yi/

; i /

hy

KA

X

s<

7

-

y

X

A-

-fh

15^

HzNPaJW

cVF

as

WbTH

tlmlx

Bq

GySv

it 1 1 l ls "m-kg/s2

N/m2

N-mJ/sA-sW/AC/VV/AA/VV-sWb/m2

Wb/A

cd'srlm/m2

s -J/kgJ/kg

ij v |. ^

h y

«•? *' ^ h

15

min,

I, Lt

eVu

h, d

leV=1.60218xl0-'9J

lu=1.66054xl0"wkg

$k 4

/ s '

if

L y7

!. (. p - A

- y

- ;l/

•J ~

KA

15 ^

Ab

barGalCiR

radrem

1 A=0.1 nm=10-'0m

lb=100fm2=10-28m2

1 bar=0.1MPa=10sPa

l-Ci=3.7xl010Bq

1 R=2.58xlO-<C/kg

lrad = lcGy=10"2Gy

1 rem=lcSv = 10"2Sv

10"1015

10'2

109

10'103

10!

10'

10"'

io-2

io-3

10"'io-9

10-2

10""10""

mmx 9

T

* *

J

+' N ^7

f

f-b y

v^ 9•f

f7x A

T

E3

?7

ififD

1-*

•>

uD

/

|~

b

IB -tEPTGMkhda

dcm

ft

nPfa

1. a 1 - 5 I* r

K * « S lgss^TOTlciSo fcfc'L, leV

fc J; a' 1 u ©ffl li CODATA © 1986 ¥ } | ^

2.

3. barti,

, barnteJ;

N(=105dyn)

1

9.80665

4.44822

kgf

0.101972

1

0.453592

Ibf

0.224809

2.20462

1

tt £ lPa-s(N-s/m2)=10P(#TX')(g/(cm.s))

1 m2/s= 10'StC x b - 0 x ) (Cm2/s)

?J

MPa( = 10bar)

1

0.0980665

0.101325

1.33322 x 10"'

6.89476 x 10"3

kgf/cm2

10.1972

1

1.03323

1.35951 x 10"3

7.03070 x 10-2

atm

9.86923

0.967841

1

1.31579 xlO'3

6.80460 x 10" *

mmHg(Torr)

7.50062 x 103

735.559

760

1

51.7149

Ibf/in2(psi)

145.038

14.2233

14.6959

1.93368 xlO"2

1

X

*

+"1

ft9

m

J(=10'erg)

1

9.80665

3.6 x 106

4.18605

1055.06

1.35582

1.60218x10-"

kgf-m

0.101972

1

3.67098 x 10s

0.426858

107.586

0.138255

1.63377 x lO"20

kW-h

2.77778x10-'

2.72407 x 10"'

1

1.16279x10-'

2.93072 x 10"'

3.76616 x 10"'

4.45050 x 10-2'

ca! (itil'S)0.238889

2.34270

8.59999 x 10s

1

252.042

0.323890

3.82743 x 1Q-20

Btu

9.47813 x 10"4

9.29487 x 10"3

3412.13

3.96759 x 10-3

1

1.28506 x 10"3

1.51857xl0-22

ft • Ibf

0.737562

7.23301

2.65522 x 10s

3.08747

778.172

1

1.18171 x l Q - "

eV

6.24150x10"

6.12082 x 10l9

2.24694xl025

2.61272x10"

6.58515 xlO21

8.46233x10"

1

1 cal = 4.18605 J ( l t f l i f t )

= 4.184J miW)

= 4.1855 J (15 °C)

= 4.1868 J(

= 75kgf-m/s

= 735.499 W

Bq

1

3.7 x 1010

Ci

2.70270 x 10"'

1

Gy

1

0.01

rad

100

1

C/kg

1

2.58 x 10"'

3876

1

Sv

1

0.01

100

1

THE FATIGUE STRENGTH OF GRAPHITE AND CARBON MATERIALS FOR HTTR CORE COMPONENTS

Documents

THE FATIGUE STRENGTH OF GRAPHITE AND CARBON ...Room temperature fatigue tests were carried out on graphite and carbon meterials, which are used for the components in the core region