# Fileset

[NRIMSR-94-02.pdf](https://mdr.nims.go.jp/filesets/7aced1f9-5ca9-4b1f-9529-30c10d96f067/download)

## Creator

YASUNAKA, Takashi

## Rights

In Copyright[In Copyright](http://rightsstatements.org/vocab/InC/1.0/)

## Other metadata

[Dynamic Fracture Toughness and Its Evaluation in a Heavy-Sectioned Ferritic Nodular Cast Iron NRIM Special Report(Research Report)No.94-02](https://mdr.nims.go.jp/datasets/688f4127-2972-4761-a944-f77313ca9d1a)

## Fulltext

Dynamic Fracture Toughness and Its Evaluation in a Heavy-Sectioned Ferritic Nodular Cast Iron NRIM Special Report(Research RepNRIM SR-94-02 1; o ~' ~ ~ o ~ ~' o c c o ~: o = ~ o CL a) O~ ,,)G) o CL = ~ c o li: ,D o = 1' a) ID c ,D J: o ,D G) G) o ~' ~ G) ~ o ~ ")a, ")E o :~ ~ ,D ~ CO :~ Z H-o o ,,)= a) J: ~ Dynamic Fracture Toughness and Its a Heavy-Sectioned Fenitic Nodular by Takashi YASUNAKA NRIM Special Report (Research Report) No. 94~2 Evaluation Cast lron 1994 National Research Institute for Metals 2-3-12, Nakameguro, Meguro-ku, Tokyo, Japan in NRIM　SR－94－02Dynamic　Fracture　Tbug㎞ess　and　Its　Evaluation　in　　aHeavy－Sectioned　Fe㎡tic　Nodular　Cast　Iron　　　　　　　　　　　　　　　　　　　　by　　　　　　　　　　　　　Takashi　YA．SUNAKANRIM　Special　Report　（Research　Report）　　　No．94－02　　　　　　　　　　至994　　Natlonal　Research　Institute　for　Me芝als2－3－12，Nakameguro，　Meguro－ku，　Tokyo，∫apanDynamic　Fracture　Tbughness　and　Its　Evaluation　in　　　aHeavy－Sectioned　Fe㎡tic　Nodular　Cast　Lon　　　　　　　　　　　　　　　　　　　　　　　　　by　　　　　　　　　　　　　　　Takashi　YASUNAKA．NRIM　Special　Report　　（Research　Report）　　　　　　No．94－02ContentsAbstract。．。．。＿＿＿　．．＿＿　．＿　．＿　，＿＿　．9．．．＿＿＿　．．＿＿　．．＿＿　．＿＿　．．．．＿　．．．．．＿＿＿　．．．．＿＿　．．．．．＿　．＿　．　e．．＿　．1．Introduction．．．．．．．．．．．．．．．．．．9．．9Q．。．◎．◎，◎．◎9．．．邑．．．．．∴．．．6．．．．．．．．．◎．．．．．．．9．◎．．．．．．．．．．．．．．．冒．．．『．．．．．『．．．．．．響．響．，．，．．．．．．．　　L1　0切ective　ofτhis　Study＿＿＿＿＿＿．．＿．＿＿＿＿＿．．＿．＿．＿＿＿＿．＿＿＿＿＿．．＿．＿．．．＿　　L2　Historical　Review．．．．＿．＿．．．＿，．＿．．＿＿．＿．＿。＿．＿．＿．．．．．．．．．．。．＿＿．．＿．＿＿．．．．．＿．＿．＿＿＿　　　1．2．1Static　Fracture　Toughness　and　Specimen　Thickness．．＿＿＿＿．＿．．＿＿＿＿＿．．＿．＿＿＿　　　1．2．2　Chemical　Composition．．・・・・・・・・…　．．．．．．．．．．．．・．・．．・…　一一一・・・・・・・…　●●’●●●●●●’．．．．．◎伽．’，”．◎．◎．．．．．．．．．’．’”　　　1．2．3　Pearlite．．．．9．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．唖．．9．．．。．．．．．．．．．．．．響．．．．．．．．．．．．．．．．．．，．．．．．．い．．．9．．◎．．9．9．．．．．．。．．．．　　　1．2．4Graphite　Nodularity＿＿＿＿＿＿．＿＿＿＿．．＿＿．＿＿＿＿．＿＿．．．．＿＿＿＿＿＿．＿＿＿＿　　　1．2．51nternodule　Spacing　and　Nodule　Size　of　Graphite＿＿＿＿＿．＿．＿＿．＿．＿．＿．＿．＿＿．．　　　1．2．6Dynamic　Fracture　Toughness．＿＿＿．＿．＿＿＿＿＿．．＿＿＿＿．＿＿＿．＿＿．＿．．＿．＿．＿．＿2。Development　of　a　Drop－weight王mpact驚sting　Machine＿＿＿＿＿．．＿＿＿，＿，＿．＿．＿．＿．＿．3．Experimental　Procedure．．＿＿＿＿＿＿．＿＿．．．＿．＿，＿，＿＿＿＿＿＿．＿。．。＿．＿＿．＿．＿．＿．＿．＿．　　3．1Materials．＿．＿．＿．＿．＿．．＿．＿，＿．＿．＿．＿．＿．＿．＿．＿．＿．＿．＿．＿，＿＿．．＿＿．＿．＿．＿。＿．＿＿　　3．2Mechanical　Property驚sting＿＿＿＿＿＿．＿．＿＿＿＿…………………・…・…・…………・…・…・　　3．3　Frac蝕re　Toughness　Testing＿＿＿．．＿．＿．．＿．．＿＿＿＿＿＿＿＿＿＿＿．＿＿．．＿．＿．．＿．．＿．＿．4．Mechanical　Properties＿。＿＿＿＿＿．．＿．＿＿＿＿＿．。＿．＿＿一＿＿．＿＿＿．＿＿．＿．＿＿＿＿＿＿　　4．1Tensile　and　Chaq）y　三mpact　Properties＿＿＿＿＿＿．．．．．＿＿＿＿＿＿．．．＿＿＿＿＿．．＿＿＿＿　　4．2Y6ung’s　Modulus．＿＿＿＿＿．＿＿．＿＿＿＿．＿．＿＿＿＿．＿，＿＿＿＿．＿＿．＿＿＿＿＿。＿．．95．Fracture　Toughness．．，＿＿＿＿＿＿＿．＿＿＿．＿．＿＿．＿＿＿．．＿＿＿．＿＿＿＿＿＿＿．．＿＿＿＿＿＿　　5．l　Ef£ect　of茎nternodule　Spacing　＿＿＿．．＿．．＿＿．＿．．＿＿＿＿＿＿．＿＿＿．＿．＿．＿．＿．＿＿＿．．．　　5．2Effect　of　Stress　Intensity　Rate＿＿＿．＿＿．＿．＿．＿．＿．＿．＿．＿＿．＿＿＿．＿．＿＿＿．＿。＿．＿．　　5．3Fracture駄）ughness　Parameter＿＿．＿．＿．＿．＿．＿．＿．．．．，＿．＿．．＿＿。．＿．＿．．．．＿．＿．＿．＿．＿．．　　5。4Estimation　of　Fracture　Toughness　from　Small　Size　Specimen＿＿．．．．＿．＿．＿．＿．＿＿．＿．．．6。Evaluation　of　Frac亡ure　Toughness＿＿．＿．＿＿＿．．＿＿＿．．＿＿＿＿＿＿．．．．．．＿．．．．．＿．．＿．．．．＿．＿．7．Concluding　Remarks．＿．＿＿＿＿．＿．＿．＿．＿＿．＿＿＿．．＿＿＿＿＿．＿．＿．＿．＿．＿．＿．＿．＿．．．＿．Acknowledgements．＿＿＿＿＿．＿＿＿＿＿＿．＿．＿．．＿＿＿＿＿＿。＿＿＿＿＿．＿．＿．＿．＿．．＿＿．．＿．References◎．。．．．。．．9．．◆，◆．。．．．．．．．．．．．．．．．．．．．．．．5．．．．．．．『．『．．．噛．．．．．◆．．．．．．．．．．．．．．．．．．．『．『．．．．．．．．．『．『．．．．．．．．．．．．．．．．．．．．．．．．．．．　1　2　2　2　2　2　3　3　4　4　5　6　6　6　6　9　9　9　9　911111213141515NRIM　SR－94－02Dymmic　Fracture　To皿ghness　and　Its　Ev紐1uation　in　　　　aHeavy－Sectioned：Ferritic　Nodular　C紐st　IronbyTakashi　YASUNAKA＊Abstract　　　　Recently，　heavy－sectioned　fbrritic　nodular　cast　iron　has　become　to　be　made　and　the　use　ofthis　material　is　noticed．　Ferritic　nodular　cast　iron，　however，　has　low　toughness　because　it　containsalarge　amount　of　graphite．　Furthemlore，　embrittlement　occurs　at　low　temperature．　Therefbre，　touse　this　material　fbr　large　structural　components．，　the　evaluation　of　upper　shelf　f士acture　toughnessas　well　as　ductile－brittle　transition　temperature　is　necessary．　　　　Static食acture　toughness　increases　with　increase　in　internodule　spacing，　whereas　dependenceof　dynamic丘acture　toughness　on　internodule　spacing　is　small．　The　upper　shelf　ffacture　toughnesslncreases　wlth　decrease　in　temperature　and　with　increase　in　stress　intensity　rate．　This　increase　in丘acture　toughness　is　mainly　attributed　to　the　increase　in　strength．　Fracture　toughness　transitiontemperature　is　linearly　related　to　the　logarithm　of　stress　intensity　rate．　　　　In　the　upPer　shelf　region，　plane　strain　ffacture　toughness　divided　by　yield　stress　is　con．stant，and　it　is　proposed　as　a　matefial　constant　that　is　independent　of　stress　intensity　rate　andtemperature．　Evaluation　of　the　critical　Haw　size　fbr　unstable　f㌃acture　can　be　simplified　with　this丘acture　toughness　parameter．馳ツWOπ護ε’ノ診rr’∫’Cη04闘」αr　cO5∫’mη，のηαη2’C〃α（）薦形　‘0那8んη（ヲ∬，10W　’6叩θrα魏泥励翻例・ηち・舵∬∫・’・・吻・α畝8・・卿∫・∫・伽・ぬ1・脚。’η9，・枷。磁1∫η’・8吻＊Senior　Researcher，　Environmental　Perfbrmance　Division，　NRIM2 Takashi　YASUNAKA1．　1臓troduC重韮onL］LO煽ective　of　This　Study　　　　Ferritic　nodular　cast　iron　has　great　f沁idity，　littleshrinkage　and　good　machinability　in　comparisoロwithcast　stee1，　an（萱has　been　deve正oped　as　a　material　fbr　aductile　cast　iron．　One　of　the　disa（ivantages　of　thismaterial　is　low　toughness　because　of　l我rge　amount　ofgraphite．　Furthermore，　embrittlement　occurs　at　low　tem－perature．　This　embrittlement　was　studied　by　impact　test－ing　such　as　Charpy　i田pact　testing．　Afterward，　f治acturemechanics　became　to　be　applied　to　the　evaluation　of　thetoughness　of　this　material，　and　the　effbct　of　metallurgica豆factors　on　fracture　toughness　has　been　studied（1）．　　　　Because　the　control　of　cooling　rate　is　required　in　theprocess　of　production，　thickness　of　castings　was　Hmitedin　the　early　stages．　However，　heavy－sectioned　castingshave　become　to　be　made，　an（i　they　are　used　as　Iargestructural　components（2）．　In　recent　years，　from　the　view－point　of　economy，　the　nuclear　spent　fuel　shipping　cask　of艶rdtic　nodular　cast　iro鍛　which　a｝so　serves　as　an　inter一　　　　　　　　　　　　　　　　　　　　　，mediate　storage　cask，　has　attracted　particular　attentionand　has　been　developed　in　Germany（3）。　Such　casks　havebeen　also　developed　in　Japan，　and　thickl－walled　fbrriticnodular　cast　iron　castings　have　become　to　beproduced（ヰ‘6）．　The　Japan　Industrial　S‡andard　of　thick：一walled　fbrritic　nodular　cast　iron　castings　for　iow　tempera－ture　service（JIS　FCD　300　H’）has　been　established（7）．　Forthe　application　to　casks，　it　is　required　to　evaluate　dynamicffacture　toughness　for　preventing　unstabl．e　ffacture　againstimpact　load　at　low　temperatures。　　　　王n’the　case　of　thick－wa豆led艶πitic　nodular　cast　ironcast童ngs，　the　microstructure　varies　with　the　location　ofwall　thickness．　The　mid－thickness　portion　of　castings　atwhich　final　solidificatio籠　occurs　has正arge　graphitenodules　because　oHow　coollng　rate．　The　mid－thlckηessportion　is　often　infbrior　to　the　other　portion　in　strengthand　elongatio簸（5），　and　is　regarded　as　the　weakest　portio簸of　castings．1t　is　inconsistent，　however，　with　the　datareported　previously　on　the　ftacture　toughness　of　the　mid－thick：ness　portion　of　castings．　It　is　worthwhile　to　clari牽ythe　effbct　of　graphite　distribution　on　ffacture　toughnessbecause　the　results　relate　to　the　sampling　of　test　couponsand　the　method　of　quality　assurance．　Furthermore，　it　ispreferable　to　est｛mate　fracture　toughness　using　smallspeclmens　because　of伽e　di鎚erent　microstructuresdepending　on　the　location　in　castings．　　　　0切ectives　of　this　study　are　to　characterize　thebehavior　of　ffacture　toughness　with　respect　to　graphitedistribution　and　loading　rate　and　to　evaluate　fracturetoughness　of　this　materia正．1．2Historic劉l　ReviewL2。1．　St飢ic　Fracture　Tbugkness　and　Specimen　　　　　Thickness　　　　In　the　e盆rly　period，　the　measurements　of　plane　strain倉acture　toughness　were　attempted　on　the　basis　of　linearelastic　fracture　mechanics．　These　attempts　were　not　suc－cessfUl　except　at　low　temperature　where　brittle　ffactureoccurred．　　　Nanstad　et　a1．〔8－lD　measured　plane　strain　fracturetoughness　Kl［c　of　fbrritic　nodular　cast　irons　using　thecompact　tension（CT）specimens　of　21　mm　in　thickness，but　valid　Klc　value　was　not　obtaiHed．　They　regarded　KQor　Kmax　that　waS　calculated　frOm　the　5％seCant　load　orthe　maximum　load，　respectively，　as　f｝acωre　toughness．　　　　In　the　upper　shelf　region，　it　was　dif丘cult　to　obtainvahd　Kic　even　if　the　specimen　with　large　thickness　wasused．　Ostensson（i2）used　CT　specimen　of　100　mm　inthickness　and　reported　that　Kic　was　not　valid．　Recently，Arai　e亡a玉．（】3）fbund　that　valid　KKc　was　not　obtained　in亡heupper　self　region　even　with　CT　specimens　of　200　mm　inthickneSS．　　　　玉n　the　uPPe「　shelf　region，　therefbre，　elastic＿plastic倉acture　toughness　Jlc　was　measured．　Here，　Klc　estimated丘om∫夏。　is　termed　Klc（J）．　Bradley　and　Mead，　JL（14）meas－ured　Jlc　at　room　temperature　using　CT　specimens．　Theyobtained　Klc（J）of　41　to　54　MPami12．　Another　Klc（」）of　25to　67　MPami／2　was　reported　by　Bradley（15＞．1．2．2．Chemica塁Compos姫on　　　　Ef艶ct　of　the　amount　of　carbon　on　critical　crackopening　dispiacement（COD）was　studied　by　Holdsworthand　5011ey（16＞．　Increase　in　volume　ffaction　of　graphite　ledto　a　decrease　in　critical　COD　value　in　the　upper　shelfregion　an（i　to　reduction　in　COD　transition　temperature．　　　　Bradley　and　Mead，　JLo4）fbund　that∫lc　of　nodularcast　iron　containing　3％Si　was　small撒der　some　temper－ature　conditions．　Maezono　et　al．（i7）showed　that　Jlcincreased　with　increasing　the　amount　of　Si　in　the　rangeDynamic　Fracture　Tbughness　and　Its　Evaluatlon　in　a　Heavy－Sectioned　Ferritic　Nodular　Cast　Iron　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　目oistSpecimen（CT）／LQad　ce113Weight　releasedeviceWeightStopperLiquld　nitrogenStOrage　COntainer8　　■1撮；；l11目1．一1 1一一Cooling　boxLoad　cellSafety　pln／ShockabsorberF二9．1Schematic　il匡ustration　of　tbe　droP－weight　testing　machine症om　1。87　to　3．12％but　decreased　when　the　amount　of　Siwas　beyond　this　range。　Komatsu　et　al．（ヨ8＞studied　theef艶ct　of　the　amount　of　Si　on　Jlc．　They　fbund　that∫lc　inthe　upPer　shelf　region　increased　with　increasing　theamount　of　Si　up　to　2．9％，　and　further　increase　in　theamount　of　Si　led　to　a　decrease　in　Jlc．　On　the　other　hand，at　123　K　in　the　lower　shelf　region　Jlc　decreased　withincreasing　the　amount　of　Si．　Ductile－brittle　transitiontemperature　of　Jlc　was　Iowered　with　increasi簸g　theamount　of　Si．　　　　Komatsu　et　aLG9）showed　that　the　resistance　to　crackpropagation　decreased　with　increasing　the　amount　of　Pand　the　amount　of　P　above　O．05％resulted　in　rapidincrease　in　the　ductile－brittle　transition　temperature．However，　in　the　upper　shelf　region　a　little　decrease　in∫lcwas　observed　with　increase　in　the　amount　of　P　Moreover，Krasowsky　et　al，（20）reported　that　Klc　decreased　withincreasing　the　amount　of　Mn．1．2．3．Pearhte　　　　Matrix　is　composed　of驚rrite，　pearlite　or　mix£ure　ofthe　two　phases．　Kobayashi（21）showed　that　JIc　increasedwhen　the　amount　of　pearlite　was　less　than　10％．　Bradleyand　Mead，　Jr（正4）also　obtained　the　similar　conclusion．Transition　temperature　increased　with　increase　in　theamou盛of　pearlite（22）．　It　was　con倉rmed　by　Komatsu　etal．（23）that　Jlc　at　room　temperature　decreased　with　increas－ing　volume　ffaction　of　pearlite　up　to　92％and　transitiontemperature　increased．　Krasowsky　et　al．（20）also　reportedthat　KIc　decreased　as　pearlite　percentage　increased．L2．4．　Graphite　Nodu蓋ar藍ty　　　　Kuribayashi　et　al．（24）studied　the　ef陀ct　of　graphitenodularity　using　the　samples　with　nodularity　up　to　84％，They　showed　that∫夏。　increased　with　increasing　nodularity，and　the　relationship　between　nodularity　n　and　J｛c　was4 勲kashi　YASUNAKAI　l戟@Ih　lh　II　Ih　I戟@I戟@lF韮g．2Stopper　h紐vlng　an　inner　ol匪damper1．5　望　　1．0ε99の婁望Φ　　0．500 1 　　　　2　　　　　　3Speed　of　fal擁ng　weight（朗／s）4 5Fig．3Re且ationship　between　tensile　speed　and　speed　of㎏1藍ing　weightportion　where負nal　solidi行catlon　occurred　was　smaller　inmagni重ude　than宅hat　at　the　other　portions．represen£ed　by　the　followi歎g　equation：」IC－62－1・・7（n－1／2一・3／2） （kJm『2）…．（1）Komatsu　et　aL（25）also　reported　that　Jlc　in　the　upper　shelfregion　increased　with　increasing難odu重arity．L2．5．　hternodu塁e　Spacing　and　Nodule　Size　of　　　　　Graphite　　　　Holdswor£h　and　Jol玉ey（16）showed　the　effbct　of簸odu至ecou熱t　on　COD　when　the　amoun重of　carbon　was　constant，Increase　in　nodule　cou韮t，　that　is，（iecrease　in　in£ernodulespacing　reduced　COD　value　in　the　upper　she璽f　reg圭on　andreduced　s至ightly　COD　transi£ion　temperature．　Sorensonand　Salzbrenner（26）showed　uslng　many　data毛hat　KQincreased　with　increase　in　i煎ernodule　spacing　and　nodulesize．　Maezono　et　al．（27）carried　out　the　fracture　toughnesstests　using　3　point　bend　specimens．　They　reported　that　Jlcin　the　upPer　shelf　region　increased，　but　Jlc　in　the　britt至e倉acture　region　decreased　with　increasing　graphite燕odulesize．　Salzbrenner（28）　fbund　large　dependence　of　∫ic　onin£ernodu至e　s茎）acing　or　nodule　size．　　　　Regardi慧g　fbrritic　nodular　cast　iron負）r　th童ck－walledcastings，　Iwabuchi　et　aL（29＞reported　that倉acture　tough－ness　i蕪creased　with韮ncreasing　noduie　size．　However，Nakamura　et　al．（4＞found　that　Jlc　at　the　mid－thickness1。2．6．Dynamic　Frac加re　1＆｝ughness　　　　Measurement　of　dynamic　plane　straln　ffacture　tough－ness　KIごby　dynamic£ear　test　was　made　by　Cheng　andWorzala（ll＞and　a　shift　of　KI♂temperature　curve　to　highertemperature　was　shown．　Kobayashi　and　Nishi（3。）reportedthe　characteristics　of　f士acture　inciuding　Kld．　Luyen面kand　Nieswaag（31）measured　dynamic　fねcture　toughnessusing　an　instrumented　Charpy　testing　machine．　Theyfbund　little　variation　in　fracture　toughness　at　the　impactspeed倉om　2．2　to　5。4　ms｝1．　Kobayashi　et　aL（32）made　theinstrumented　Charpy　impact　tes亡us重ng　fゑ亡igue　precrackedspecimens　taken　ffom　thick－walled　fbrritic　nodular　castiron　castings。　They　reveaied　the　ef艶αof　specimen　sizeand　the　behavior　of　Kld（J）．　Transitlon　temperature　ofKId（J）aξthe　lmpact　speed　of　1．34　ms一；was　about　90　K歪ower亡han　that　of　Klc（J）．　　　　Loading　rate　in　the　instrumented　Charpy　test　hasbeen　represented　by　the　impact　speed　of　the　hammer．Behavior　of　fねcture　may　depend　on　strain　rate　in　theregion　a（寿acent　to　the　cra6k　tip．　There負）re，　stress　intensityrate（dK／dt　or　K）is　pre琵rable　fbr　loading　rate．　　　　Quality　assurance　committee　on　ductile　cast　ironcasks（正986　to　i　990）was　organized　by　Central　ResearchInstitute　of　Electric　Power　I鍛dustry　in∫apan．　The　thick－wa璽led　castings　of　f6rritic益odular　cast　iron　were　suppliedby　several　casting　manufacturers，　and　characteristics　ofthese　materials　were　studied．　Dynamic　ffacture　toughnessDy礁amlc　Fracture　Tbughness　and　hs　Eval騰at量on　i鶏aHeavy－Sectioned　Ferrit圭。　Nodular　Cast　Iron 5膠：幽しoad　ce畦AMPDi＄piacernent　meter　　　　（COD＞Displacemen生meterしoad　ceil　AMPStrain　AMPしlitrasonic　detectOr遍　巳■1・　9OscilloscopeTransientconverterComputerPlotterVolt　meter↓↓Fig．481Gck　d童agram　of　meas腿ring　systemwas　measured　using　CT　specimens　up　to　150　mm　inthickness　and　the　leve玉of丘acture　toughness　values　wascon丘rmed（33）．2．Deve藍opment　of　a　Drop－weight　Impact「恥st韮ng　　　Machi賑e（47）　　　For　the　measurement　of　dynamic　ffacture　toughness，many　kinds　of　testing　machines，　such　as　hydraulic　andgas－actuated　tensile　machine（34），　pendulum－type　impactmachine（35），　drop－weight　testing　machine（36・37），　and　spHtHopkinson－baぼ（38），　have　been　used．　　　　Although　the　drop－weight　testing　machine　has　a　dis－advantage　of　relatively　na：rrow　range　of　Ioading　rate，　ithas　many　advantages　in　economy，　operation，　and　main－tenance　by　virtue　of　the　simple　structure．　Mostly，　it　hasbeen　used　as　an　impact　bending　machine（36・37）．夏t，however，　would　be　available　as　an　impact　tensile　testingmachine．　For　measuring　dynamic　ffacture　toughness，　animpact　tensile　machine　with　a　falling　weight　wasdesigned　and　made．　　　　Figure　l　shows　schematically　the　specially　designedtesting　machine．　A　loading　rod　is　suspended　under　the　CTspecimen．　A　stopper　is　mounted　at　the　lower　end　of　theloading　rod．　In　this　study，　the　weight　of　200　kg　was　used．The　weight　is　lifted　to　a　predetermined　height　by　a　hoistand　then　released．玉mpact　test　is　made　by　forcing　downthe　loading　rod　by　the　free　falling　weight．　The　specimenscan　be　automatically　cooled　in　a　cooling　box　and　held　atapredetermined　temperature　by且ne　spray　of　liquid　nitro－gen．　Specimens　are　shiel．ded　ffom　liquid　nitrogen　byshielding　Plate．　Two　load　cells　are　installed　above　andbelow　the　specimen．　　　　In　the　case　of　iron　block　stoPPer，　the　tensile　speedwas　ascertained　to　be　equivalent　to　the　falhng　speed　ofthe　weight　Dropping　the　weight　from　the　height　thatexerted　enough　energy　to　fracture　the　specimen　led　to　theoccurrence　of　elastic　waves．7rhe　data　obtained　were　notable　to　be　treated　similarly　as　those　in　static　test，　Further－more，　time　Iag　of　the　output　of　transducers　became　large．The　stopper　that　had　an　inner　oil　damper　as　shown　in　Fig．2was　used　to　dampen　the　in亘ial　shock　wave　and　toobta1n　lower　tensile　speed．　The　relationship　between　cal－culated　falling　speed　of　weight　and　tensile　speed　is　shownin　Fig．3．　　　　The　block　diagram　of　the　measuring　system　is　shownin　Fig．4．　For　the　measurement　of　COD，　the　displacementbetween　two　targets　marked　on　the　f塗ont　face　of　the　CTspecimen　was　measured　by　an　electro－optical　displace一6顎ble　l　Chemica且composition（wt％）Takashi　YASUNAKAMaterial C Si Mn P S MgA 356 2．02 0．18 0，O19 0，002 0．04．B 3．56 2．00 0．18 0，019 0，002 0．04C 3．56 2．04 0．19 α002 σ，002 0．04ment　meter．　Tensile　speed　was　measured　from　the　dis－placement　of　a　light－emitting　diode　located　at　the　upperend　of　the　loading　rod，　using　an　optical　displacementmeter．　The．　data　were　stored　into　a　transie．nt　converter　andsaved　on　fioppy　disks．and　then　analyzed　with　a　computer．3．Experimental　Procedure3．1Materi訊］s（50）　　　Tbst　materials　were　taken　f｝om　an　as－cast　cylindricalcasting　of　500　mm　in　wall　thic㎞ess　and　l350　mm　inoutside　diameter．　The　casting　was　cut　into　many　60×125×150min3　b．lock．s．　The　test　blocks　ffom　15，30　and　45％thick　position　in　the　direction　to　the　inner　surface　aredenoted　as　Materials　A，　B　and　C，　respectively」n　thosematerials，　the　graphite　distributions　diff巳r　each　other　dueto　di」自陀rent　cooling　rates．　　　The　chemical　compositions　are　listed．　in　Table　1．There　was　little　dif石erenee　between　the　chemical　compo－sitions　of　Materials　A，　B　and　C．　Microstruc加res　areshown　in　Photo．1．　The　graphite　nodularity　w．as　high　andlittl．e　pearlite　was　obs．erved．　Graphite　nodule　sizeincreased　and　nodule　count　decreased　in　the　sequence　ofMaterials　A，　B　and　C．　The　grain　size　in　Material　C　was　alittle　large　in　comparison　with　the　others．　Graphite　nodulediameter　was　measured　with　an．image　analyzer；each　areaof　graphite　was　converted　to　the　correspondiロg　diameterof　circle．　Figure　5　shows　the　distribution　of　graphitenodule　on　a　test　plane．　Metallographic　details．　such　asgraphite　nodule　and　grain　size　are　listed　in　Table　2．　Thede且nition　of　inteq；）article　spacing　depends　on　interactionbetween　a　particle　and　the　phenomenon　under　considera－tion．　Figure　6．shQws　the　relationship　between　the　anglejust　ahead　of　the　precrack　tip　and　the　expected　value　ofdistanceξA　fナom　the　precrack：tip　to　the　center　of　thenearest　neighbors　of　nodules　within　the　angle．　W6　useNA－1〆2　as　a　parameter　of　internodule　sp．acing．3．2Mech勘nical　property　Ib．sting　　　Tbnsile　specimens　with　a　diameter　of　4　mm　and　agage　length　of　20　mm　were　tested　in　an　Instron　machine◎．集♂．9　　　　　　　のヘ　　　　　へ　　　・’．、＼ン～1　　　　　　　　．◎●1・・ I，）一＼●．＼・．／　　　昌’（b）L』●．1　　　　　　●）・1　　　　◎ゾ1　　　　　　　　画ジ＿＼．／蜜’》曵Photo．10ptical　microstr皿ctures　of　Materials　A（a）β（b）and　C（c）and　a　high　rate　tensile　testing　machine．　Impact　tests　werecarried　out　on　the　standard　V－notched　Charpy　specimensand　on　the　bend　specimens　with　machine　notch　of　O．06mln　in　root　radius　and　of　2　mm　in　notch　depth　instead　ofVnotch．3．3Frε巳cture　Toughness　Testing（48・49）　　　Static　and　dynamic　elastic－plastic　f止acture　toughness，Jlc　and　Jld，　respectiv．ely，　were　measured　by　the　multiple－specimen　and　single－specimen　techniques　according　toASTM　E813－810r　JSME　S　OO1．　The　specimen．s　werefatigue　precracked　CT　specimens　of　25　mm　in　thicknessas　shown　in　Fig．7．　The　testing　machines　used　were　the　㍗εε　δρ慈⊆皇君8黛でユ鰻σ20で61284o12840840Dynam｛c　Fracture］翫）“ghness　and　Its　Evaluatlon　in　a　Heavy－Sectioned　Ferri£ic　Nodular　Cast　Iron0t57睾1くζが1．00．50Precrackθ　50　　　　　　　　　　100Graphite　nodule　diameter（μm）唯50Fig。5　Distribution　of　gr日ph批e　nodu萱e　diameter　on　a　test　plane勲ble　2　Metallographic　detailsMateria1 A B CGraphite　nodule　numbermA（mm｝2） 9α6 67．2 49．5Mean　graph｛te　diameternn　a　teSt　plane（mm）α0391 α0443 0．05匪3Mean　graphite　dlameter奄氏@volU鶏e（mm）0．0479 0．0542 0．0628Internodule　spacing垂≠窒≠高?ｔｅｒ　NA－112（mm） 0，105 0，122 0，142Degree　of　nodularity（％） 98 97 98Grain　slze　number 5．6 5．1． 4．745 90θ（degree）435 180Fig。6Reiationship　between　ang垂e　near　the　precrack　tip　a脳l　dis・　　　　　tance　from　the　precrack　tip重。重he　nearest賑eighbor　graphite　　　　　nodu且e　fbund　w虻hin　its　a賑g藍e（a）の”1一十一1の。 γ一一一一一一1一十一 8014．050 13Target　forCODdrop－weight　type，　an　electrohydraulic　type　and　Instronte】ユsile　ones．　　　　脳）detect　the　onset　of　crack　propagation　in　dynamictest，　the　ultrasonic　method（39）used　in　static　tests　isnoteworthy　though　electrical　potential　method（40）andcaustics　method（4D　are　o衰en　used．　An　ultrasonic　methodwas　applied　in　this　study．　Ultrasonic　method　makes　itpossible　to　detect　the　onset　of　crack　propagation　in　themid－thickness　portion　of　several　square　mil玉imete「in　a「eawhere　the　crack　can　be　propagated　most　easily　in　thespecimen，　As　shown　in　Fig．8two　disks　of　piezoelectricceramics　were　cemented　at　the　location　that　correspondedto　the　crack　tip　on　the　upPer　and　lower　faces　of　the25（b）σり1卑012．5響50 138Con最gurations　of重he　compact　tensiondroP鱒weight　type　tensi且e　machi無e（a）a甑dl無StrOn　tenSi塵e　maChineS（b）．25Fig．7」specime無s　fbr　a，e且ectrohydra岨。　typespecimen・Variation　in　ultrasonic　reHection　and　throughtransmission　pulse　heights　was　measured．　Figure　g　shows8PiezoelectriC　Ceramic％kashi　YASUNAKALoadρ　噂f　　　、霧層＼’◎聖　、　、 、　、@、Utrasonic餐aw　　　　、@　　　、@　　ρ　・@　爵　ρノ陶7孕で、ｺCジ・detector　　　夢　， 、　　　　　　　　、、ヤ 　　　　　　1A、　　　　　　　　　　　■、　　　　　　　　　　1、　　　、、　　　　　　　　　　　暉’　　吻、　　　層uらξ，@、@　、@　、、、　　　　　，　飼@ヤ　　　　　“　“C“覗幽　　　　　幽馬　　　、Eσ’l　F　崖　　　　　　　　　　　、@　　　　　　　　　　　、@、　　　　　　　　　　　　　　　　　　　　　　　　　、@　1A』’Fig．8Schemat莫。　illustraIio鳳of　measuri離g　the　onset　of　crack　propa－　　　　　gatlon　by　aR　u匡trasonic　method　z潔　ユ℃o」35302520霊510500．0 0．2 0．4 0．6 0．8Load一｝ine　displacement（mm）tO 1．2≧］口＝⊃だ。’盃＝Φ勲コα．9ぢ父b≧1ト⊃三〇’6＝Φ皿コα⊆．9霧’姦2響各コ9二←typica正changes　in　pulse　height　under　static　test．　Theonset　of　crack　propagation　is　shown　by　the　arrows．　Thesimilar　change　under　dynamic　test　in　the　drop－weighttype　testing　machine　is　shown　in　Fig．10．　Re丘ectionmethod　was　ef艶ctive　in　this　case．　Re員ection　pulse　heightbegan　to　i難crease　at　the　moment　of　impacしThis　wasfollowed　by　another　abrupt　and　disco誼inuous　changewith　crack：propagation．　The　onset　of　crack　propagationoccurred　befbre　reaching　the　maximum　load．　　　　至nthe　range　of　tensile　speed　below　l　ms　1，　the　loadmeasured　by　the　upper　load　cell　was　in　good　agreementwith　that　estimated　by　back－face　strain．　Therefbre，　Jld　canbe　calculated　by　the　fbllowing　equation（42）applied　instatlc　test：」IC，∫1d；Af（・／w）／（Bb）．＿．＿＿＿．＿＿．＿（2）where　A　is　the　area　under　Ioad　and　load－line　displacementrecord，　a　is　the　precrack　length，　b　is　the　initial　uncrackedligament，　W　is　the　width　of　specimen，　B　is　the　thicknessof　specimen，　andFig．9男yp長ca翌variation　i鳳裂oad　and　u夏trasonic　pulse　height　showing　　　　　the　onset　of　crack　propagatio曲y　the　arrows　under　stat養。　test　　　　　in　an　Instron　machine4030　Z邑。－20忌210oPOnset　of　crackpropagationu1．5　　　　　　　　〉　　　　　　　　コ1．0　　　　葛　　　　璽　　　墜　　　　＆　　　碧　　　　霧　　　　田。，5蓄・（・／W）・・（1・αソ（1・α2）α一o（・・／・）・＋・（・・／・）・・｝1／2一｛（・・／・）・1｝”（3）0 0．4　　　　　　　0．8Load－liPe　disp［acement（mm）1。20Fig・豆0野pica震variation　in　load　and岨trasonic　reHection　pulse　　　　　　he董ght　showing　the　onset　of　crack　propaga重ion　by　tke　arrow　　　　　　under　dynam童。重est　i融adroP・weight　type　Iensile　machine600Dynamic　Fracture］殴）ughness　and　ks　Evaluation　ln　a｝leavy－Sectioned　Ferritlc　Nodular　Cast　Iron　（　飢　Σ　　　£」＝　σ｝6＄5窃栃働て）’?糞唇〉ト（　に．9罵02田50040030020020鷹00●O　A△　Bロ　C臼●●　　　TSO　　　YS○○ ○詮○厓A　　　　　〔］　　　△　50　　　　　　壌00　　　　　　150　　　　　　200　　　　　　250　　　　　　300　　　　　　　　　　　　　　　　Temperature（K＞Fig・11　E旋Ct　Of　tempera重Ure　On　tenSi藍e　prOper重藍eS25　　20會ねユ9）15ゆ5冨毬10馨　　50150 200 　　　　250Temperature（K）300Fig．13　Charphy　V－notch　i㎜pact　energy　curve350400350（窪芝　　300ζり250　　　　200　　　　　　1r5　104　10－3　10－2　10”1　100　灌01　102　　　　　　　　　　　　　　　　　　　　　　Strain　rate（S－1＞F養9．12E縦lct　of　strain　rate　on　yie藍d　stre卿h　at　room　tempera加re，　　　　　Material　A4．Mech段n韮。段韮Properties（49・51）4．笠Tensi韮e段勲d　Ch紐rpy　Impact　Properties　　　The　effbct　of　temperature　on　tensile　properties　isshown　in　Fig．11．　Strength　was　insensitive　to　graphitedistribution．　Figure　12　shows　the　dependence　of　O．2％proof　yield　strength　on　strain　rate　at　room　temperature．The　result　of　V－notch　Charpy　impact　test　is　shown　in　Fig．13．The　absorbed　energy　in　the　upper　shelf　region　was　l　7∫and　the　absorbed　energy　transition　temperature　was　253K．The　data　indicate　a　small　scatter　even　in　the　transition「eglon・94．2Ybung，s　Modu1HS　　　Ybung’s　modulus　E　can　be　calculated　ffom　the　com－pliance　C　of　CT　specimen　by　the　fbllowing　equation（43）：E一（1／CB）｛（W・・）／（W一・）｝2　　　　2．1630＋正2．219（a／W）　　　　　　　　　　　　　　　）3＋20．609（・／Wプ一〇。9925（a／W－99314（a／W）5一20．065（a／W）2（4）　　　The　results　are　illustrated　by　Fig．14．　Relationshipbetween　Ybung’s　modulus　E　and　temperature　T（K）wasrepresented　by　the　fbllowing　equation：E＝1965－0．066T5．Fractureτbughness（49－51）（GPa）．．．＿＿＿．．　（5）5・1　】Ef色ct　of】［nternodu璽e　Spacing　　　Figure　l　5　shows　the∫一R　curves　obtained　by　static食acture　toughness　tests　at　room　temperature．茎n　thesespecimens　tested，　stretched　zone　was　not　observed，　andcrack　blunting　line　was　not　able　to　be　measured　clearly．Therefbre，　the　intercept　between　the　regression　Iine　andthe　crack　blunting　line　given　by　the　equation　J＝2σf△awas　identified　as∫lc．　Here，σf　is　the　effbctive　yield10200190雪5uoE　で80ρoコ。＞170160　　100Takashi　YASUNAKA灌50 　　200　　　　　　250Temperature（K＞300Flg．14　Y6ung，s　mod蜘s　as　a　function　of垂empera加re，　Material　A，1005040f∈も邑寄3020O　RT△　　233K100ぐEもこ「110 　120　　　　130NA－1尼（μm）Fig．16　Rela重lonship　between　Jlc　and　Nパ112　　　　　　and　233　K，　Ma重erla監A140 150806040200O　A△　Bロ　CJ＝2σ1△a60at　room重emperature5040多重30モ「20100O　A△　B〔］　C　　βφノ’．（為＿3　　ノ406ノ蟹100 1500 0．5 　　tO△a（mm＞t5 2．0200　　　　　250Temperature（K）300 350Fig．17　E焼ct　of　temperatロre　and　g即hite　distributio鷺on　JlcFig・15」・R　curve　under　stat藍。　test　at　room　temperature，　in藍重ial　　　　　　stress　in重ensity　rate　of　O．42　MPam豆！2s4strength　and△a　is　the孟ncrement　of　crad（歪ength．　The　JKcincreased　in　the　sequence　of　Materials　A，　B　and　C；∫夏cwas　infhenced　by　graphite　d重stribution．　The　reiationshipbetween　parameter　of　intemodule　spacing　and　Jlc　in　theupper　shelf　region　is　shown　in　Fig．豆6．　Fracture　toughnessJlc　increased　w歪th　increasing　Paramαer　of　internodu】．espaClng・　　　　The　ef色ct　of　temperature　on　Jlc　is　shown　in　Fig．17．The　data　plotted　by　the　solid　symbols　in　the丘gureindica£e　those　obtained　by　the　multiple－specimen　tech－nique．　UpPer　she歪f　region　existed　until　temperaturedecreased　to　about　200　K．　In　the　upPer　shelf　region，　inwhlch含aαure　apPearance　was　ductile　one，　Jlc　increasedwith　decreasing　temperature．　Further　decrease　in　tempera－ture　led　to　an　increase　in　cleavage　ffacture　resulting　indecrease　in　Jic；∫lc　decreased　with　increase　in　percentageof　brittle　ffac加re　appearance．　At　ear至y　stage，　cleavage倉acture　surねce　o負he　order　of　o簸e　grain　size　was　isolatedeach　other．　In　the　ductile－brittle　transition　region，　dif飴rLence　between　JIc　values　in　Materials　A，　B　and　C　was　notobserved。　This　may　be　attributed　to　little（iiff6rence　i簸grain　size．　Transition　temperature　was　188　Kl。　Crackpropagation　in　the　upPer　shelf　region　occurredDynamic　Fracture　Tbughness　and　Its　Evaluation　in　a　Heavy－Sectioned　Ferritic　Nodular　Cast　Iron60118060｛隻40ζ200　MateriaIO　A△　Bロ　CJ＝2σ　∠a50　　40f鳶エv30　9つ　自「　　20100　　　　K（MPa・m112・s㎜bO　　4．2x10滞1△　　t2×103ロ歪．0×105　　　　　　　　　　　　　　　0△△△t5100 150 　　200　　　　　　250Temperature（K）300Fig．19　Variation　in　fracture　toughness　as　a負mct量on　of　temperature　　　　　　With　VariOUS　StreSS　in重enSity　raIeS0 0．5 tO∠a（mm）F19。18」・R　curve　u麟der　dynamlc重est　at　room　temperature，　in麓iaI　　　　　　stress　intensity　rate　of　6．7　MPam112s－1predominantly　by　void　growth　around　graphite　noduleand　coalescence．　　　Figure　18　shows　the∫一R　curves　at　dK／dt　of　6。7MPam112s－1。　Dynamic　elastic－plastic　ffacture　toughness　Jldwas　influenced　to　a　less　extent　by　graphite　nodule　spac－ing．　In　the　case　of　larger　dK／dt，　experime凪al　accuracywas　reduced，　so　that　the　diff6rence　betwee簸Jld　values　ofMaterials　A，　B　and　C　was　not　found　experimenta豊ly．5．2Effbct　of　Stress　Inte鰍sity　Rate　　　The　eff6ct　of　dK：／dt　on倉acture　toughness　is　shown　inFig．19．　Upper　shelf　ffacture　toughness　was　increasedwith　increasing　dK／dt．　Ductile－brittle　transition　tempera－ture　was　naturally　raised　with　increasing　dK／dt．　Depen－dence　of　f士acture　toughness　and　critical　crack　tip　openingdisplacement　on　dK／dt　is　demonstrated　in　Fig．20．　Withinthis　range　of　dK／dt，　specimens　were　fractured　in　ductilemode　at　room　temperature．　UpPer　shelf　f士acture　tough－ness　was　Iinearly　related　to　the　logarithm　of　dK／dt．　Cracktip　opening　displacement（CTOD）can　be　calculated　fromthe　following　equation　specified　by　British　standard（44）：CTOD・K2 iレv2）／2σ．E　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　＿．（6）　　　　　　＋0．46（W－a）Vp／（0．46W＋054a）where　K　is　the　stress　intensity　factor，σY　is　the　yieidstrength，　v　is　Poissoゴs　ratio，　and　Vp　is　the　plastic　compo－nent　of　load－line　displacement．　The　second　term　of　theright　side　of　this　equation　is　the　plastic　cornponent　ofCTOD　and　is　denoted　by　CTODp亘、、ti、　distinguishing　it倉om　total　CTOD（CTOD吐。t、1）．　Especially，　the　dependenceof　CTODpl、、ti、　on　dK／dt　waS　Small．　Therefbre，　the　increasein食acture　toughness　with　increasing　dK／dt　is　mainlyattributed　to　the　increase　in　strength．　　　　The　effbct　of　dK／dt　on　transition　temperature　isshown　in　Fig．21．There　was　a　linear　relation　between　thetransition　temperature　and　the　logarithm　of　dK／dt．　Inaddition，　it　should　be　noted　that　the　transition　temperatureof　this　material　was　equivalent　to　that　of　a　low　carbonsteel，5．3Fractureτbughness　Parameter　　　　Plane　strain　fracture　toughness　divided　by　yieldstrength　as　a　function　of　temperature　is　shown　in　Fig22．Static　and　dynamic　plane　strain　fracture　toughness，　Klcσ）and　Kld（J）were　estimated　from　J夏。　and　Jid，　respectively．The　K　values　were　converted　by　the　fbllowing　equation：K一｛EV（1一・2）｝i／2．…．．……………＿．……（・）Yield　strength　was　estimated　using　the　equation　of　nomi－nal　limit　load　of　CT　specimen（43）on　the　assumption　ofconstant　yield　ratio．　In　the　upPer　shelf　region，　Klc（J）／σYand　Kld（∫）／σY　were　kept　constant　regardless　of　tempera一12 肱kashi　YASUNAKAf∈誘邑　2コ　重つ5040302040Oo　　　CTODtotal△　　CTODplastic ○○○○○101　歪02　歪03　104　　k（MPa・mI／2・s－1）10080　　　倉60　、∋　　　040　0　　　ト　　　02010－1 歪oo 105 1060　寒ε匠　　〉・o、ど〉：：o、ゴ151050　　　　K（MPa・m112・s司）0　　4，2x10－1△　　1．2×103ΩtOx105○□歪00 150 　　200　　　　　　250Temperature（K）300罫ig。22　Fracture　toughness　norma叢ized　to　yie置d　strength　as　a　fロ賑。－　　　　　　tion　of重empera加re　w醗h　var董ous　s重ress　intensity　raIesFig．20　冠挽ct　of　stress　intensity　ra重e　on　fracture　toughness　and　　　　　　CTOD　at　room　tempera加re　in　the　upper　shelf　region240　邑Φ220当筍おユε2002ぼ・9’あ岳歪80㍍160歪0－2 100 　　　　　102K（MPa　mv2・S昌）104 106Fig・21　Ef艶cI　of　stress　intensity　rate　on　fracture　toughness　trans童幽　　　　　　tio賑temperatureture　and　dK／dt。　This　fねcture　toughness　parameter　may　beregarded　as　a　material　constant　of　this　material．5．4Estimation　of　Fracture　lbughness　from　Sma韮l　S量ze　　　Specimen　　　　There　are　various　dif丘culties　in　measuring　dynamicfracture　toughness　of　f6rritic　nodular　cast　irons．　It　isprefbrable　to　estimate　the　behavior　of倉acture　toughnessat　the　predetermined　dK／dt　ffom　the　results　of　static倉acture　toughness　tests　and　impact　tests　of　small　speci－mens．　The　results　of　a鍛lmpact　bend£est　using　smallspecimens　with　machined　notches　are　shown　in　Fig．23．The　dK／dt　in　the　impact　test　can　be　estimated　from　theimpact　speed　and　the　dK／dt　in　a　static　bend　test．　Whenthe　root　radius　of　machined　notch　is　in由e　same　order　ofmagniωde　as　internodule　spacing，　the　sharpness　ofmachined　notch　may　be　equivaient　to　that　of　fatigue－cracked　notch　because　of　the　existence　of　graphitenodule．　In　such　a　case，　the　estimation　of　the　behavior　of倉acture　toughness　might　be　possible由rough　percentageof　brittle倉acture　apPearance　in倉ont　of　the　precrack　tip．　　　　In　this　study，　increase　in　percentage　of　brittle　ffactureappearance　of　more　than　50％led　to　rapid　decrease　in　J正d．The　co簸ditions　of　50％br隻ttle　ffacture　appearance　cor－responded　to　that　of　the　lower　limit　of　temperature　in　theupper　shelf　region．　It　was　fbund　that　the　temperature　atthe　beginning　of　embrittlement　in　fracture　toughness　andthe　corresponding　temperature　of　50％brittle　ffactureapPearance　in　CT　or　machine－notched　specimens　wereiinearly　related　to　the　Iogarithm　of　dK／dt．　　　　Figure　24　shows　a　method　to　estimate　the　behaviorof　ffacture　toughness　at　the　predetermined　dK／dt．　Point　Ais　obtained　by　static　test　usingσr　specimens，　whereasPoint　B　is　obta玉ned　by　impact　test　usi敷g　small　bendspecimens　w貢h　machi簸ed　notches．　In　the　figure，　Ti　is　thelowest　temperature　of　upper　shelf　region　and　Ts　is　the50％brittle倉acture　transition　temperature．　Point　D　isdetermined　ffom　Point　C　that　corresponds　to　the　predeter－mined　dK／dt　o簸the　line　AB。　Thus，　the　relationshipbetween　Kld／σY　and　temperature　ca簸be　estimated，20Dynamic　Fractuごe　Tbughness　a艮d　Its　Eva垂uat重on　i江aHeavy－SectiQned　Ferri毛ic　Nodular　Cas毛Iron　、　あ9の89£8ミ151050oAbsorbed　energy●Brittle　frac室ure　surface　　　　　　　　　　　　　　　　’　　　　　　　　　　　　　　　　’　　　　　　　　　　　　　　　’　　　　　　　　　　　　　　　！　　　　　　　　　　　　　　ノ　　　　　　　　　　　　　！　　　　　　　　　　　　　！　　　　　　　　　　　　！　　　　　　　　　　　∠　　　　　　　　　　1　　＿一一一一’　！！’！！　　　一一一岬一欄糟哺　ノ’ノ歪00755025150 200 　　　　250　　　　　300Temperature（K）35002ヨむ要い盈甚お芒Φ9①巳F韮g。231mpact　energy　a識d　percentage　of　bri棚e　fracture　appearance　　　　　　of重he　mach韮ne・notched　bend　specimen　as　a　fhnction　of　　　　　　tempera加re．　The　broken蓋韮籍e　shows　Charpy　V・notch　　　　　　impact　energy　c装rve，2ヨ頸①皿ΦトKI、ノσYDこ　　う　　ヨ　　　　　　　　り　　れ　　わ　　ゆ　　は　　む　　コ　鞠　鞠　贈　喝　　　　馬鱒、馬　　　　　　　、、、　　　　　　　　、　　　　　　　　　o　　　　　　　　　　一．一C　　　　　　　　　　　　　l　　　　　　　　　　　　　l　od　　　　　　　l　峯3　の　　　　　　　　　　　　　1　短トー　のO　　　　　　　l＿一@　一一一一一一一一一一一一トー　　IA　　　　　　　l　　l　　　　　　　：　　l　　　　　　　l8　　．．　　　；　　　ld　ol。o　咀創建　贈8δ踊　　　1岳　　　l　　　l　　　コ　　　　　コ　　ト　　　　　コ　　　　　　l　　　l　　　lTsTI2ヨ歪ΦΩ．∈…Φ　　　　　　　　Klc／σY　　　　　　　　　　　　　　　　　log　KFig．24　Estimation　of　dynamic　fracture　toughness　be血av量or　by　static　　　　　　fracture　toughness　tests　and　impact　tests　with　sma監監　　　　　　machine・no重che“bend　specimen　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　13dK／dt，　the　refbrence　curve　relating　to　Kld　and　relativetemperature　may　be　established．　In　general，　however，　theprediction　of　unstable　frac加re　on　the　basis　of　crackinitiation　criterion　seems　to　be　reasonable．　Here，　theanalysis　based　on　Iinear　elastic　ffacture　mechanics　isemployed．　　　　Variation　in　K玉d（J）with　temperature　is　shown　in　Fig．25．The　shaded　area　in　the　figure　shows　the　range　of食acture　toughness　values　reported　by　Arai　et　al．（26）In　theregion　shown　by　a　lighter　shade，　plane　strain　f士acturetoughness　was　valid．　On　the　other　hand，　in　the　regionshown　by　a　darker　shade，　namely，　in　the　upper　shelfregion，　plane　strain　ffacture　toughness　was　invalid　regard－Iess　of　using　CT　specimens　of　150　and　200　mm　in‘thickness．　The　upper　shelf　ffacture　toughness　in　this　studywas　smaller　than　that　reported　by　Arai　et　al．　The　criticalallowable　f【aw　size　can　be　calculated　by　these　data．Analysis　of　unstable　ffacture，　however，　can　be　simpli負edby　the　above－mentioned倉acture　toughness　parameteL　　　　For　example，　it　is　assumed　that　a　semi－ellipticalsurface　crack　exists　perpendicular　to　the　surface　of　thickplate　as　shown　in　Fig．26．　A　load　that　is　superimposed　bytension　and　bending　is　applied　to　the　plate．　The　stressintensity　factor　becomes　maximum　value　at　Point　Aunder　the　condition　described　later　There　are　manypapers　that　give　the　calcul飢ion　of　the　stress　inte駐sityfactor　in　such　a　case．　According　to　Ishida　et　al．（45），　thestress　intensity　factor　at蔓）oint　A　is　represented　by　thefbllowing　equation：K一二）1／2（σM馬・σ、恥）＿．＿＿＿＿（8）6．Ev31uat韮on　of　Fracture　Toughness（51・52）　　　　In　the　f6n陰itic　nodular　cast　iron，　it　量s　necessary　toevaluate　upPer　shelf　倉acture　toughness　fbr　preventingunstable　ffacture　in　the　upPer　shelf　region　as　well　asductile－brittle　transition　temperature．　An　attempt　has　beenmade　to　recommend　the　design　criterion　by　the　conven－tional　method　on　the　basis　of　strength　and　static　ffacturetoughness（45）．　The　resistance　of　crack　propagation　in　thismaterial　is　relatively　small．　It　is　difθcult　to　establish　thePellini－type食acture　toughness　ref6rence　curve　because　ofthe　difHculties　in　measuring　nil　ductility　transition　tem－perature　by　dynamic　tear　test．　Within　the　limited　range　ofwhere　both　FM　and　FB　are　the　functions　of　b／a　and　b／t，which　are　given　by　polynomials．　The　total　stress　on出esurface　of　plate　is　expressed　by：σ＝σM＋σB．＿．＿．＿．．．．．＿．＿．＿＿＿，＿．．　（9）WhenσB／σM＝3，　the　stress　intensity　factor　is　given　by：K・σM（痴）1／2低，＋碑、）．＿．＿＿＿＿＿＿（10）Relationship　between　stress　on　the　plate　sur飴ce　and14　　　　　　　　　　　　　　　Temperature（℃）一120　　一で00　　　－80　　　－60　　　－40　　　－20 0笛akashi　YASUNAKA20歪50f琴100窪｝500　　　　　k（MPa・m112・s　1）●　Kld（J）　　　　　1．2×103▲　Kld（J）　　　　　tO　x　105　　Klc㈹ASTM　＞300纏懇Klc（J）　　　　　＞300く一一一一一く一一一ロ冒　く一　唱　一　冒　一　”　桶　脚　　　　　　　　’　　　　　　　’冊　　腰　　隔　　一　　幽　一　　　　　　’　　　　　　’　　　　　’　　　　’　　　　’一　　　　　　　’　’　’’　：”崖冊刷　　n　　堂，　”罐　　A螺　：　：　匡　　　　　→　…→→　　＿＿＿レ　じ冊ノ騨　曽　一　一　一　一　一　冒　盟　曜　彌’　＿＿＿＿＿→o〕oΣo一一一「一rrr一一幽L、 、、」剰　　　　　150　　　　　　　　　　　　200　　　　　　　　　　　　250　　　　　　　　　　　　300　　　　　　　　　　　　　　　　　　　　　Temperature（K）Fig．25　Dynamic　p匪ane　stra産n　fracture　tough賑ess　as　a　hmction　of　　　　　　tempera加re・「critical　depth　of　surface　Haw　bc　can　be　determined　from：σ／σY一（K・d／σ・）（1・βソ｛（励・）’／2（恥鮒｝（11）Here，　b／a　ahd　t　are　assumed　to　be　1／4　and　400　mm，respectively．　In　the　upper　shelf　region　of　this　material，Kld（J）／σY　is　regarded　as　a　constant　as　mentioned　bef6re。When　stress　on　the　plate　surface　is　constant，　uni負）rmtension（β＝0）results　in　the　minimum　value　of　criticaIHaw　depth．　Relationship　between　stress　under　uni｛brmtensionσT　and　critical　naw　depth　bc（T）is　shown童n　Fig．27，The　solid　line　in　the丘gure　shows　the　relation　fbr　thematerial　used　in　this　study．　Ifσis　constant，　the　criticalHaw　depth　bc（Mβ）increases　with　increasing　bending　stressσB．Relationship　betweenσB／σM　and　bc（M，B）A）c（T）is　shownin　Fig．28．7．Conduding　Remarks　　　　Heavy－sectioned　castings　of　fbrritic　nodular　cast　ironhave　become　to　be　produced　recently，　and　this　materialhas　attracted　particular　attention　as　an　econornic　one。硬bughness　of　thls　material　is　relatively　low　because　oflarge　amount　of　graphi£e．　Careful　evaluation　of　tough－ness，　therefbre，　is　necessary　in　the　use　of　this　material．However，11ttle　sys£ema書ic　study　of　dynamic　fracturetoughness　has　been　made。　In　the　productio簸of　castings　of　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　L，∴Fig．26　Tension　and　ben（韮ing　of　a　f童n虻e一酌ickness　p蓋ate　with　a　　　　　　semi－el騒p重ica藍s磁r鉛ce　crack60＿　　50∈豊　ε8　40言9壷30葱馨0　　　2010●　　　　　　　●　　　　　　　●　　　　　　　o　　　　　　　●コ　　　　　　　コ　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　コ　　　　　　　　　　　　　　　　曳　　」　　㌔　　、　　●・・．、瓦之瓦’㍉㍉路、、・、，、1ミ　　　　　＼〉誉緊：：：ト、　　　　　　　　　　　’●・。．　　・．B　　●●。．．　’。・・。．　　　　　　　　　　　6、％％、％、、…．．．i；’Kld…〆…mm1／2・”・・…．．こ．・・・…．．：＝　　　　　　　　　　　　　　　　　　　　　　　　　・Oo・．　　　　　　　　　　　　　　　　　　　　　　　　　　　　●・●●●●●　　　　　0．3　　　　0．4　　　　0．5　　　　0．6　　　　0．7　　　　0．8　　　　0．9　　　　1．0　　　　　　　　　　　　　　　　　　　　　　　　σT／σYF孟g。27Cr託ica置sur飴ce　f藍aw　depth　as　a危mction　of　unifbrm£ens丘藍e　　　　　　stress　in　the　upper　shelf　region　with　normalized　fracture　　　　　　toughneSSthis　material，　relatively　much　know－how　is　required，　andthis　enables　manufacturers　to　produce　castings　wi芒hsomewhat　diffαent　quality．　The　microstructure　of　f6rritenodular　cast　iron　used　in　this　study　was　excel玉ent　onehaving　low　volume　fraction　of　pearlite　and　high　nodular－ity。　Behavior　of　fねcture　tough賎ess　in　regard　to　tempera－ture　and　歪oading　rate　has　been　stud童ed。　In　this　study，elastic－plastic丘acture　toughness　was　adop！ed　as倉acturetoughness　because　plane　strai11倉acture　tough簸ess　wasprobably　not　effbctive　in　t鼓e　upPer　shelf　region　even　if出ick　specimens　were　used．　　　Embri亡tlemen亡occ膿red　adowとemperatαre　and　frac－ture　toughness　transition　temperature　was　linear圭y　relatedε8＼璽三〇．Q1．10Dynamlc　Fracture］文）ughness　and　Its　Evaluation　in　a　Heavy－Sectioned　Ferritic　Nodular　Cast三ron1．08tO6tO41，021，00♂塩O　　　　　　0．0　　　　　　0」　　　　　　0．2　　　　　　0．3　　　　　　0．4　　　　　　0．5　　　　　　　　　　　　　　　　　　　　　　　σ8ノσMFig．28　Correction　valロe　of　cr辻ic3艮sur飴ce　Oaw　dep重h　under　tensio駐　　　　　and　bendingto　the　logarithm　of　stress　intensity　rate．　The　upper　shelffracture　toughness　increased　with　increasing　stress　inten－sity　rate．　Graphite　internodule　spacing　or　nodule　size　hadIittle　ef陀ct　on　dynamic　fracture　toughness　although　static倉acture　toughness　was　inf玉uenced　by　these　factors．　　　Although　unstable　fヤacture　can　be　readily　estimate　onthe　basis　of　linear　fracture　mechanics　using　conve宜edplane　strain　fracture　toughness，　fbr　more　accurate　analy－sis，　it　is　prefbrable　to　estimate　the　propagation　of　crack　byacomputer　on　the　basis　of　nonlinear貸acture　mechanics・5）6）7）8）Acknowledgements　　　Apart　of　this　study　was　supported　by　the　atomicresearch　fund　ffom　Science　and馳chnology　Agency　inJapan．　The　author　wishes　to　express　his　thanks　to　allthose　concerned．｝｛e　also　wishes　to　express　his　thanks　tohis　colleagues，　especially　Messrs．　K．　Nakano，　N．　Iwaoand　N．　Furuya，　who　contributed　to　the　per拓㎜ance　of　theexperimental　work，　The　author　is　much　indebted　toKubota　Ltd．　fbr　the　help　in　preparing　the　material．1）2）3）4）Re艶rencesWL．　Bradley　and　MN．　Srinivasan：Int　Mater．　Rev．，35　（1990），　129－161．H．Mayer：AFS至nt．　Cast　Met　J．，1（1976），21－27．W．L．　Bradley：J．　Met．，37，　No．1（1985），74－76．S，Nakamura，　N．　Sakamoto，　K．　Inoue，　K．　Ogi　andK．Matsuda：∫．　Jpn．　Found．　Soc．，59（1987），664－669．9）10）11）12）13）14）15）16）17）18）19）20）21）22）23）24）25）　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　15T．Yanaka，　H．　Saito，　D．　Sakurai　and｝｛．　Arata：J．　Jpn．Found．　Soc，，60　（1988），20－25．YIwabuchi，　H．　Narita，　M．　Murata，　S．　Shimizu　andO．Tsumura：J．　Jpn．　Found．　Soc。，60（1988），167－172．JIS　G　5504，　Heavy－waHed　Ferritic　SpheroidalGraphite玉ron　Castings　fbr　Low艶mperature　ServiceR．K．　Nanstad，　FJ．　Worzala　and　C．R．　Loper，　Jr．：Int．Symp．　Met．　Cast　Iron，2（1975），789－807，RK．　Nanstad，　F．J．　Worzala　and　C．R．　Loper，　Jr：Trans，　Am．　Foundrymen　Soc．，83（1975），245－256．F．」．Worzala，　R’v．　Heine　and　Yi　W6n　Cheng：Trans．Am．　Foundrymen　Soc．，84（1976），675－682．昭一W．Cheng　and　FJ，　Worzala：茎nt．　Conf，　Mech．Behav．　Mater．，2（1976），1223－1227．B．Ostensson：Scand．　J．　Met．，2（1973），194－196．T．Arai，　T．　Onchi，　T　Saegusa　and　G　Yagawa：Proceedings　the　6th　Sympo．　Fract。　and　Fract．　Mech．，Soc．　Mat．　SciJpn，（1991），99－104．W．L，　Bradley　and　H．E．　Mead，　Jr．：ASME　MPC，11（1979），69－87．W，L．　Bradley：Trans．　Am．　Foundryme駐Soc．，89（1981），837－848．S．R．｝loldsworth　and　G　Jolley：Int．　Symp．　MeしCast　Iron，　（1975），　809－825．T．Maezono，　Y　Ohtsuka，　R．肱kahashi，　M．　Yano　andY．王shihara：J．∫pn．　Found，　Soc．，58　（1986），195－199．S．Komatsu，　T．　Shiota　and　K．　Nakamur縦：J．　Jpn．Found．　Soc．，60　（1988），442－447．S．Komatsu，　T．　Shiota　and　K　Nakamura：J．　Jpn．Foun（至．　Soc．，59　（1987），554－559．AJ．　Krasowsky，　LV　Kramarenko　and　VV　Kalaida：Fatigue　Fract．　Eng，　Mater．　Struct．，10（1987），223－237。T．Kobayashi：Casting＆Forging，30，　No．7（1977），5－12．F．Henke：Giesserei　Prax。，　No．9／10　（1976），131－139．S．Komatsu，　T．　Shiota　and　K．　Nakamura：J．∫pn．Found．　Soc．，60　（1988），643－648．K．Kuribayashi，　T．　Kishi，　P．　Bhandhubanyo駐g，　M．Ito，　T．　Umeda　and　Y　Kimura：τbtsu－to－Hagame，60（1983），663－670．S．Komatsu，　T．　Shiota　and　K．　Nakamura：JJpn．Found．　Soc．，59　（1987），159－163．1626）27）28）29）30）31）32）33）34）35）36）37）38）　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　セ　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　肱1（ashiK．B．　Sorenson　and　R．　Salzbrenner＝“The　Mechan－ism　of　Fracture，”ed．　by　Goel，　ASM，（1986），415＿419．T．Maezono，　R．　Takahashi　and　M．　Suenaga：J。　Jpn．Found。　Soc．，60　（1988），578－583．R．Salzbrenner：J．　Mater．　ScL，22（豆987），2135－2147．Y　Iwabuchi，　H．　Narita　and　O。Tsumura：J．　Jpn．Fou韮d．　Soc．，60　（1988），215－220．T．Kobayashi　and　S．　Nishi：J．　Jpn．　Found．　Soc．，52（1980），76－87，（Communication　paper　fbr　the　46thInt．　Foundry　CongL，　Madrid，　Spain）．T．Luye簸d麺k　and　H．　Nieswaag：至nt．　Foundry　CongL，（1982），　至一正3．T。Kobayashi，　H．　Yamamoto　and　K．　Matsuo：Eng．Fract．　Mech．，30（1988），397－407．‘‘qesearch　on　Quality　Assurance　of　Ductile　castlron　Casks，”Quality　Assurance　Committee　on　Duc－tile　Cast　Iron　Casks，　CR肥PI　Report，（1990）．E．F両ii，　Y　Sakai　and　Y　Ando：J．　Tbs．　EvaL，14（豆986），　181－190．T．Kobayashi，1．　Yamamoto　and　M，　Niinomi：Eng．Fract．　Mech，，24（1986），773－782．A．Kl．　Shoemaker　and　S。T．　Rolf6：Eng．　Fract．　Mech．，2（1971），319－339．WN．　Shaq）e，∫L　and　JM　Shapiro：∫．艶st．　EvaL，18（1990），　38－44．MR　Bayoumi・∫・R・Klepaczko　and　M耳・Bassi甲：J・Test．　Eval．，王2（1984）316－323．YASUNAKA　　　39）　K．Hirano，　H．　Kobayashi　and　H．　Nakazawa：」．艶st．　　　　　　　EvaL，13（1985），356－362．　　　40）　J．R．∫oyce　and　C．S．　Schneider：J．　Tbst．　Eval．，豆6　　　　　　　（1988），257－270．　　　41）　J．F．　Kalthoff：至nt．∫．　Fract．，27（1985），277－298，　　　42）ASTM　Standard　E813－81，　Standard騰st　fbr　Jlc　　　43）ASTM　Standard　E813－87，　Standard　Tbst　fbr　Jlc44）45）46）47）48）49）50）51）52）British　Standard　Method　7448（1991），　FractureMechanics，　Tbughness艶sts，　Part　1．M．W．　Schwartz　and　L．　Boyce：“Ductile　and　BrittleFailure　Design　Criteria　fbr　Nodular　Cast　Iron　Spent－Fuel　Shipping　Containers，”US　DOE　Rep．，UCRL－53046，（1983）．M．Ishida，　H．　Noguchi　and　T．　Ybshida：IntJ．　Fract．，26　（1984），　157－188．T．Yasunaka，　K，　Nakano　and　T．　Saito：1SIJ　Interna－tional，31　（1991），　298－303．T．Yasunaka，　N。　Iwao，　N．　Furuya，　H．　Yamawaki，　S．Matsumoto　and　Kl．　Kimura：Tbtsu－to－Hagane，71（1985），　1398－1404．T。Yasunaka，　N．　Iwao，　N．　Furuya，　Kl．　Kimura　and　H．Yamawaki：J．　Soc．　Mater．　Sci．　Jpn．，33（1984），1336－1341．K．Nakano　and　T．　Yasunaka：丑）tsu－to－Hagane，78（1992），926－993．T，Yasunaka　and　K．　Nakano：Proc．　the　lOth　Int．Symp．　Packaging　and　Transportation　of　RadioactiveMaterials，（1992），1304－1310．Kl．　Nakano　and　T．　Yasunaka：Tetsu－to－Hagane，80（1994），330－335．「Dynamic　Fracture　Toughness　and　Its　Evaluation　in　　aHeavy－Sectioned　Fe㎡tic　Nodular　Cast　IronbyTakashi　YASUNAKANRIM　Special　Repo直　（Research　Report）　　　　No，94－02Date　of　issue：31March，1994．r・ノ　　　　　Editodal　Committee：Kazuh廿。　YOSHIHARA．＿．Ch謡㎜anSaburo　MATSUOKA＿Co－cha㎞1an　　　　　　Hirohisa皿ZUKA　　　　　Kazuo　KADOWAKI　　　　　Hideyuki　OHTSUKA　　　　　　　Yoshio　SAKKA　　　　　　Kohei　YAGISAWA，　　　　　　　　　　　Publisher，　Contact：　　　　　　　　　　　　Toshikazu　ISHII　　　PIanl盛ng　Section，　AdministraUon　Division　　　　　National　Research　hlstitute　fbr　Metals2－3－12，Nakameguro，　Meguro－ku，　Tokyo　l　53，　Japan　Phone：＋81－3－3719－2271　Fax：＋81－3－3792－3337グ　　ヒを：！、　　　　　　Copyrigbt◎1994　　　　　　　　　’　byNational　Research　Institute　fbr　MetalsDhector－General　Dr．　Kazuyoshi　NHTypeset　using　the　SGML　by　Uniscope，　Inc．，Tokyo8Dyn㎜ic　Fracture　Tbug㎞ess　and　Its　Evaluation　hl　　aHeavy－Sectioned　Fenitic　Nodular　Cast　Iron　　　　　　　　　　　　　　　　　　　　　　　by　　　　　　　　　　　　　　T瓠（ashi　YASUNAKA㎜Special　Repo丘　　（Research　Report）　　　　　No．94－02ContentsAbsαact・．．．．．．．・．．．．．．．●．．．●”．”6’”9，’．’6．’●’．’●◎．．．．．．’”●’6．．”●”6”．’’”．’’’’”．”，．”．．．’．．．．’’”．．9．．．”．’．．膠’．．．．願．．’●．．．．1．Introduction．．9．．．．．．．．．．．．．．．．．．．．，．．．9．．．．．．．．．噛．，．．．．9，．，．，．．．．．．．．．．。．畢．．．．．，．．．．．．．9．．．．．．．．．．．．．．．，9．．．．．．．．．．．9幽3．．．．．．．　　f．10切ective　of　This　Study＿．．＿．＿．．．．．．＿．＿．．．．．．．＿．．．．．．．．．．．＿．．．．．．．．＿，．，．．．．．．．＿．＿．．＿＿．。．．，．．　　1．2Historical　Review＿＿＿＿＿＿＿．．＿＿＿．＿＿＿＿．．＿＿＿．＿＿＿．＿＿＿．＿，＿．＿．．＿＿．＿．　　　1．2．1Static　Fracture　1㌃）ughness　and　Specimen　Thic㎞ess．．＿．．＿．．．．＿＿．＿．＿．．．。．＿．．＿＿．．．．．．　　　1．2．2Chemical　Composition．．＿．＿、．＿．．．．．．＿．＿．．．．．．。．．．．．．＿．．．．．．．＿＿．，．＿．．．＿．．＿．．．＿．＿＿＿．　　　1．2．3　Pearlite．，．．＿．．＿．．．『．．．．．．．．．．．．．．．，．＿．＿．＿．．．．．．．．．．．．．9．9．．．．．．．．．，．．．9．．，．．．．．＿．．．．6，．．．．＿．＿．．．曹．．．．．．．　　　1．2．4Graphite　Nodularity＿＿＿＿＿＿．．．．＿．．．＿。．．．．．．．．．，．．＿．＿．．．．．．．．＿．＿．．＿＿．．．．＿．．．．＿．．。．＿　　　1．251ntemodule　Spacing　and　Nodule　Size　of　Grap1オte．．＿＿＿＿．＿＿．＿．．．．．＿．．．．．．．．．．．．．．．．＿　　　1．2．6Dynamic　Fracture　Tbughness．。．．．，．．．，．．．．．．．．．．＿．．．．．．．．＿．＿．．＿．．．．．．．．．．＿＿．．．．．．．．．＿．．．＿．＿2．Development　of　a　Drop－weight　Impact　Tbsting　Machine＿＿＿＿．＿．．＿＿＿＿＿．．＿．＿＿＿＿3．Experimental　Procedure＿＿＿＿＿＿・．＿＿＿．＿．＿．＿＿．．．＿．．．，＿＿．．．＿．．．．．．．．．．．．．．．．．．．．．．＿．．＿．．．　　3．1Materials＿＿＿，＿．＿＿＿＿＿．＿＿．．．．．＿＿．＿＿＿．＿．．．．＿＿＿＿＿．＿＿＿．＿．＿＿＿＿＿＿．　　3．2Mechanical　Propertyτもsdng．．＿＿．＿＿．．．．．＿．．＿．。＿．．．．．＿．．．．＿．．．．．．．．．．．．．＿．．．．．．．．．．．．．＿＿＿．　　3．3　Frac田re　Tbughness　Testing＿＿＿＿＿．＿＿＿＿＿．＿．．．＿＿．＿＿＿＿．．＿＿＿，．，．＿．．．．．＿，＿．4．Mechanical　Prope質ies＿＿＿＿＿＿．．．．．＿．，，＿．．．．．。＿．．．．．．．．．．＿．．＿．＿，．．．．．．＿．．．＿．．．．＿．．．．．，．＿．．＿　　4．l　Tbnsile　and　Charpy　Impact　Properties＿．＿．，。．．＿．．＿．．．＿．．．．．＿．＿．，．．．．．＿．＿，＿．．＿．，．．＿．＿　　4．2Ybung’s　Modulus＿＿＿＿＿．．．．．．．．．．＿．＿．，＿．．．．＿．＿．＿＿．．＿，＿＿，＿＿．＿，＿．．．＿＿．．．．＿＿．5．Fracture　Tbughness＿．＿．＿＿＿＿．＿．．．．．．．．．．＿．．＿．，．＿，．．．，．，＿．．．．．．，．．．．．．．．＿．．．．＿．．＿．．．＿．．．．．＿．．．．　　5．1Ef色ct　of　Intemodule　Spacing＿．．＿．．＿＿＿＿＿＿．．＿＿．＿＿＿＿．．＿＿．＿．．＿．＿．．．．＿．＿，．∴　　5．2Effbct　of　Stress　Intensity　Rate＿＿＿＿．＿＿．＿＿．＿＿．．＿＿＿＿＿＿．．．＿＿＿．．．．＿＿．＿．．．．　　5．3Fracture　Tbughness　Parameter＿＿＿＿＿＿．＿＿＿＿．＿＿．＿＿＿＿＿＿．＿．．．．．．．＿．．．．＿，．＿．　　5．4Estimation　of　Fracture　Tbughness丘om　Small　Size　Specimen＿＿．＿＿＿．．．＿．．．＿．．．．．．．．．．．6．Evaluation　of　Frac1血re　Toughness＿＿．＿＿．＿，＿，＿＿＿＿．＿＿＿．＿＿．＿．．．＿．．．，．＿．．．．．＿．．．＿．7．℃oncluding　Remarks．．．．＿．．．．．．．．．＿．．．＿．．＿．＿．＿．．，＿．＿．＿＿＿．＿．。＿．．．，＿，．．＿．．＿．＿．＿．．．．＿．．．Ac㎞owledgements＿＿．＿＿＿．．＿＿＿．．．＿＿＿＿＿＿．＿＿＿＿．．＿＿．＿＿．．＿＿＿＿。＿＿＿＿＿＿Refbrences．．．．＿．，．．．．．．．．．．．．．＿．．＿．．＿．．．＿．．．．＿．．＿．．．．．．．＿．．．．．。．．．．．．．．．．＿．．＿．．＿．＿．．．＿．．．．．．．．．．．．．．．．．．．　1　2　2　2　2　2　3　3　44　5　66669　9　9　9　911111213141515・ピ．．（ぞ，、’