Cooling and heating rate dependence of precipitation in an Al-Cu alloy

,,lhm,tml,~ ,~', . ' m ' e m , d F ngmccntu, t, ,41 '~h ( I ~'-~21 I N t - 1~-~4 I N

Cooling and heating rate dependence of precipitation in an AI-Cu alloy

M..I.

S t a r i n k a n d

P.

van M o u r i k

L.ahr ~1 tilt ~tl' r~/ I~ h,ltllhlrkq, ' [ )elfl I Illll','r.~'lll' O/ '[i', '/Ill, ~/t ~kn', Rt ~llt,rthlm.s'e I'l'eg 13 7, 2h2,~' ,.4 L. l h'lll (Nethel hmd~) tRcct'nvcd N~werrnher I,-I, I~4~41 Ill revn,,ed h~im FL'L~llhJl9 IN, 1~4~4,- ~,

Abstract

DnHerenlnal ,,¢annnng eah~rnmelry and X ray dnHuaclnun were u,'~ed I1= 'dudv Ihe o~olnng arid healing hale dL'perlderlce .~1 precnpnlal.m nn an AI-I (~h al % ('u alhw Al-let h, miogernznng, oH)Innl, al a tale ~I 2~ K ram-~ (SC22)ns ,,uffncnenl I~ relann all e~pper nr= s~dnd seduln, m (.iP z,~ne lormal.m dunng su[v~equenl heal Irealmenl ns hnndered, lhns n,, ascnhed h~ an nnsulhcnenl numher ~I (excessl vaeancne.~ Aller a waler quench (W()) a lar/.;e number ~I (_iP .',,~nes are h~rmed durnng ,,uhsequenl ,,l~rag, e al r ~ m lemperalure h~r I h The heal c~mlenl ~I lhe (_iP z,~me dnss~duln~n eHecl can quanlnlal,vely he do, crnl-wd m lerms ~I lhe heal ~I pre~np~laln~n ~I (.iP [ z~me'~ and lhe s~dnd ,.duhnlnlnes as dernved In~m lhe (_iF' I z,~nc .,,dvu,,, The heal omlenl ~I lhe o)mhnned O' //9 phaw preelpllall~m eHe¢l appeared h~ he pr~p~rl.mal I~ lhe number ~I c~pper al~m', preenp~laled, yneldnn~, an averaE, e value h~r lhe heal ~I copper preenpnlal.~n ~I "lh kl m~l ~ c~pper 'The acl~val,)rl energy h~r O' pha,,e h)rmalu~n n~. () 7"~ ('V h~r S(.'22 ,,pecnmens and I I()-I-() I(I eV h~r W(.,) ,,p~.'enmen!, T'he dnHetenees Irl lepqHled a~l~vallq~n enetgnen h~r #' phase h~rmal.~n are d~,~L'lJ'~'d Irl lerlTl~ q)l [rl~d~nlnly ~1 dls,.dved al¢~m,, (related h)lhe vacancy coneerllr,.lllq~n), mlerlacnal ~.'m.'rgv and dileCll~)rl ~)I ~.~,rq~wlh (m~rmal ~r l:wrpendncular h)ll-le O' pha~,e plah')

I. Inlrodu¢lion

Recenlly nl has been shuwn lhal lhe heal ~I' precipn l a l ~ n m a suhd quenched A I - C u allay uennl'~rced wilh snluc'~m parlncles can quanlnlalnvelv he descrnlwd nn terms ~1 the heats ol preopnlaln~n ~1 holh alluynng elemenls and ol s,~lnd solubnlnlies ,~1' Ihe cemslnlulnng h n n a r y a l h w s l l ] Further, u l a p p e a r e d l h a l e,~pper pre- cnpnlaliem was dependenl on Ihe healnng rate; al I¢~w healing rales (2() K mnn-i ,~r less)~.'¢~pper preeipilaled mannly as Ihe melaslable O' phase, whereas at hJgh healing rules (.-I()K mnn-n or =nore) ~.'~pper precipi- tated mannly as the equilibrium 0-phase

AIIhough lhe preenpilaln,~n Ir¢~m a supersaluraled AI rneh phase in binary A I - C u alloys during m~n ~s~- Ihermal agenng has often been sltndned ]2-4], Ihe relaln, m between precnpnlal,_m eHeels and Ihe healing rales applied has, h~ ~ur knowledge, nell been nnvesH galed unlnl n~w T h e Iull preenpnlal~on sequem.'e nn quenched A I - C u alh~yn us usually given as Iolh~ws' supersaluraled s~lnd s~lknln¢~n ---," (_;P I ---,. (._;1:' II

--- 0 ' - - . o

(l)

where GP 1/11 ,,.,lands I'or Gumner-Preshm 'z,~mes, O' us a Iransnln~m phase havnng a slruclure whnch us a lelrag~mal distorlnon ot the CaF, slruclure and whnch has a e~mposnlnon AI,C'u and 0 ns Ihe equnlnbrnum phase

havnng a body cenlred lelragtmal slruclure wnlh the same chemncal eompusitnon as #' (LJnlnl recenl pknhliculn~ms, dnseussnon ahuul the nalknre ~I GP I and G P II zones has persisted

I I.

GP II zones are o m - sndered h~ be a slighlly in~dnhed (mullnlayered)h~rm ul (predomnnantlv mLm~dayered)(.;F' I zones, ipn ti~ be a dnfferenl phase (then Ihe nndicaln~m /-/"-phase nnslead ul' G P II zunes us usually prelerred). In Ihns paper Ihe nndn- caln~m G P II us used In any ease, G P II us subsequent h~ (.iP I) According I~ Nakamtnra cl ,11 Ihe sLy-called X phase w~uld ~wcur helween ( j P II and O' T'hns phase shl~uld be responnnhle h~r the peak hardness durnng agenng i~l quenched A I - C u alh~ys T h e X phase has, however, nol been ndenlnlned by X ray t~r Irans- mnss,~)n elecl run mncrosc~ ~py ("rE M ) analysis

In Ihm paper the results ~l'a sludy ~n Ihe nnlluence ~t Ihe healing rale tIn Ihe precnpnlall~n eHecls nn a c~n- venlnonally casl A I - C u alh~v are presenled T h e quench sensflnwly ~1' Ihe plecipilaln~n reactn~ms was inveslngaled by varnaln4~n ~1 lhe c~dung e~ndilnons aller lhe h~mogennznng Ireatmenl, which c~msisled ol waler quenching and c~uhng at a rale JUSl lasl enough I~ keep copper nn solid solulnun. As c~ppeu precnpnlatn~m Ir~m an A I - n e h matrix ns assocnaled wHh enlhalpy changes large em~tugh h~ alh~w dnfl'erenlnal ncannnng cah~rnmelry ([)SC'), II'us lechmque was applied. In addfln~m, X-ray dil:t'racln~n was used h_Ir lhe ndenldlcalion ol E)S(_ ~ elJecls and I~ ohlam, via Ihe delerminalnon ~t Ihe AI

I,~,4 M ,I .~'tmml~, P ~'mt ,~hmn~ / [%'~ tpmmml re.n,41-( 'u ,dlm' rich phase lalllce parameler, m l o r m a l . m al.~til Ihe

average copper c~mlenl ol Ihe Al-rieh phase

2. Experimental procedures

? I. .%'pecmu'n l u e p a t a t i o n a n d heat treatment

A binary A I - l e , e, a~ '~.,Cu

alh~v

wan pr~duced bv convenli~mal caslmg. Atler cuslmg, the AI-Cu alhw was h~mogemzed al 7'73_ '~ K h~r ,-18 h. C'hemieal analvs~s sh~wed Ihe hflh~wmg c~mp~siti,~n ol Ihe alh~y I h¢~ al 'Y,,Cu, () 111 al °/,,Si, I)I)12 al %,Fe, () ()()1 al.'~,,'T'l, t) ()()1 al 'Y,,Zn, ().()()~ aI.'Y,,Mg, balance alunnmum

Aller homogenizing, cylindrical specimens (helghl ab~)ul () S mm, d~ameler ab~ul q~ mm) were machined Irt~m Ihe mgols These specimens were subjeeled I~ a s~dul,m heal Irealmenl, omslsling ~)t' h~m~genizing al '7~-)'~ +_ 2 K I~r 1(1 mln I'~)lh~wed by a quench in wale~ al r ~ m lempeialure. The specimens were slored al r~)oin

lemperalure I'~r I h beh)re lurlher mvesligal,~n

(m~slly by E S(. ). These speomens will be referred h~ an W ( ) ( w a l e r quenched)speomens S~)me~l Ihe WO ,specimens were rehealed Io '7c~() _ ,~ K inside Ihe E)SC apparalus, kepl al Ihls lemper,'dure h)r 2 mm and sub- sequenlly c~)~fled Io room lempe~alure Inside Ihe E)SC apparalus. Fi~r one sel ~1 specimens lhe c ~ d m g was Iree and hence delermmed by Ihe heal conduel.m Irom lhe E S(. cell I~ lhe suri~undings', Ihe c~,dmg ot each speomen m ll'us sel was ~denl~cal and nearly exponenlial (Ihe c~ohng tale ~n passing lhe o~pper solvus al '7~() K was 22 K min-~ and ~n passing 5~() K ~l wan 6 ~ K ram- ~) These speomens will be referred h~ as Ihe SC22 (sh~wly c~oled al 22 K mm-~) speo mens Fi_~r olher specimens, Ihe milial e~l~ng rule was sel at Itxed values: 12, 8, ¢-~, '~, 4, 2, I and ().5 K mm - However, Ihe heal e~mdtiel.m t'r~)m Ihe DSC cell Io Ihe surv~undm.gs was in all cases I ~ low Io mamlam Ihe sel t'~l~n~:.,, rule over Ihe wh~de lempe~alure range. Fi~r each sel value, ~1 appeared Ihal from a eerlam lempera lure d~wnwards, Ihe c~ohng was ~denlical h.~ Ihe free eoohng,~l Ihe SC22 specimen The s p e o m e n s with a ,set inslial c~ohng rule will be referred I~ as Ihe SCI2, SC~,, elc., specimens

II appeared Ihal h~l Ihe SC22 speomens nearly all o~pper was kepl in solid s~)luli~n (see Seclion ,.1 1 ) In ~rder I~ sludy possible GP 'zone or X-phase formallon, Ihe S( ... specimens were aged al lemperalures ' " ~ belween ,-1'23 and .S23 K (?~dmg ~1 Ihese specimens l~ok place inside Ihe I-)SC apparalus. Subsequenlly, Ihese specimens were mvesligaled by DSC.

WO and SC2'2 specimens healed al a eonslanl heal- ing rule I~ lhe end lemperalure ol the muin precipila- I . m peak (see Secli~m t.I.) were also prepaled. Fi_)r Ih~s, E S(. scans al healing rules (} .~, 2, 5, 21) and 4() K ram- ~ were mterrupled al the end lemperalures ot Ihe

main preop~lal,~n peaks Atler mlerrupl,_m ~l Ihe scan, the speomens were immedmlely (wilhin I()s) rem~wed Irom Ihe DSC apparalus and quenched m waler al r~om lemperalure. 'T'hese specimens will be indwaled an WQL~,, and S(.'t~,,~ speomens, h.dh~wed by Ihe healing rale

.~. 2 D~ljere~mal w , m m n g ,'ahu'tmettv

'T'he DSC apparalus used wan a E)up~mt Ivpe t~l(). A p~oleclive gas alm~sphere ol pure argon was emph~yed. The cahbrat,~n and baseline c~rreel,m melh~ds are described elsewhere [g] For Ihe WC) speomens Ihe hdh,wmg heating tales were apphed () 5, 2, ~, 2(t and ,4() K r a m - ' , h,r Ihe SC22 specimens () 5, 2.5. I(), 2(). 4(), O() and ,~() K min-~ were apphed Fi,r Ihe SC speomens c ~ l e d al tales lower than ?'2 K mm- ~, only Ihe healing tale 5 K ram-~ was used Fi. Ihe SC22 speomens aged al lemperalures belween 423 and 523 K, [,S(. runs were ped~,rmed al ,.11)K ram- Runs were recorded belween 2t~8 and 7c~1 K Unless slaled ,,Iherwise, Ihe accuracy ~1 Ihe measuremenl of peak lemperalures is w . h m aboul _+ 2 K, Ihe accuracy ~d' Ihe measuremenl ~,1 ,)lher characlerislie lempera- lutes ~s wllhm ab~ul _+3 K, and Ihe accuracy ol Ihe delermmalion ol heal omlenls is wHhm ah~,ul +_ 5%

2 3 X t . v d t l ~ . c l u m

'T'o sludy Ihe e~mpl~silion ~t Ihe AI rwh phase at the end lemperalure ~t the mann p r e o p l l a l . m peak (,see Secl.~n ~1), Ihe lallwe pramelers of WQt~s, and SC't,s~ speemwns were delermined by X-ray diHrac- l,_m These expenmenls were performed ~)n hhngs ~1' Ihe [)SC' speeunens by Ihe 15)ebqe-Seherrer (DbS) melhod (nee, I'~r mslanee, rel. ~). X-ray dd-lraet,m slarted w , h i n 3 h atlev quenching of Ihe original DSC speeHnen. F'.xp~sure time was t~ h. Copper r a d m l , m Idlered by a mekel liller was used The lemperalute inside Ihe DbS camera .'11 a pore! ch~se Io Ihe specvmen wan measured anti ree~rded Temperalures were belween 2~-~4 and 2~47 K. Temperalure variations during single expetlmenls were lypwally q~l Ihe order ~1 () ~ K F'flm Iype and him development procedures were idenlieal I~r all expenmenls 'The AI-rich phase laltwe parameler was delermmed by using the so called Nels~m-Riley exlrapolali~m (nee ref q) The lattice parameters were corrected Ior the average lem- pelalure during Ihe measurement, ad~pling the hnear expans.m e<wl~ewnt <d pure alummium (2+ '~ x II)-<': see ret. 1t)) All lallwe paramelers presented m Ihis w4~rk are corrected Io 2~X K

T'o invesllgale lhe influence ~1 I'.hng on Ihe AI rich phase lalllee paramelet Ihe hdhlwmg experlmenl was perh~rmed Fffslly, an SC22 specimen was healed Io ~-~73 K, kepl is~lhermal Ior 2 mm and Ihen c ~ l e d very sh~wly (al rules belween ().1 and () ~ K ram-~ and inter

.,11 .I .~,'latml~. [' v,m ,Ihuml~ / r',,., ff~ttal, m m m~ ..I1-(.'~ ,film' I N"~ rupled I'~r i s l ~ l h e r m a l a g e i n g a l , 4 7 3 a n d ,-123 K h~r

I ( i h ) d ~ w n h ~ r o o m lemperature From Inhngs(~l Ihns specimen, Ihe AI ruth phase lallnce parameter was delermmed, yieldnng IL.-1II4NN __. 11.1tllt1112 nm. 'These lilnngs were Ihen sublecled Io exaelly Ihe same heal Irealmenl as descrnhed above, and again Ihe lallnce parameler wag delermmed, now guying (t 4(),4~() +_ ().()()()()2 nm. In b~lh cases ~mly the A I - r i c h phase and Ihe ~-phase were delecled

F'~r phase idenlll'ical.m, an addntnonal X ray diHrac- I . m experlmenl was perlormed on a WC) specimen using, a high lemperature (Juinier camera. 'The diLIrac- lion pallern ~11 C'u Kee~ r a d i a l . m Iransmilled Ihrough a thin (apprl~ximalely '7() pm) specimen was recorded on a waler c~.fled I'ilm during healing lhe specnmen Irom 2~-/~, h~ 77 ~, K allt.S Kmnn -~

2 4. ( ~ p m ' a l m , v o s c o p v

,lena Ne(~ph,~l upllcal microscopes were used h~ characternze lhe m w r o s l r u c l u r e s ~)1' W()t,,,~ and S(.'L,,~I

specnmens P~l=shed specimens were etched wilh a I. I mtxlure ~1 Keller and Wnle~)x reagenl and ~'Y,, Nnlal

3. Results

e o

I. D r / ] e r e n l m l .s','mmmg c a h , ' m w l t ~ ,

Figure 1 shows a DSC run , f l u W ( ) specimen taken ,.~" al 5 K mm-n Five e H e c l s ( t w o e x o l h e r m i c and lhree endolhermnc) are dnscerned T h e ell eels are numbered

I l~) V In line with resulls tram literalure ]2-,.1, I I, I 2], ~,

Ihe cHem.'Is observed in Ihe DSC curves (11 Ihe WC) run ,)1 Ihe A I - C u alhw ;.ire a l l r i b u l e d h~ Ihe Iolh_)wing re- actions (see also Sectiuns 3.2. and 4.2) (i) (_;l:)-z(~ne t,~r .

O mali,~n, (nn) (.3P-zone dnssolul,)n, (=in) (--Y //-/-phase prec=pilalion (main precnpntalion el-lecl), (w) (/'-phase • dissolutn(m; (v) e phase diss(fluli, m

¢,1,/ V e j L . ~D , . ~ ill 5 K/mnn 5 ,J IK.mol / \ \ ] ,, ,, ,,' v 4 I.)ll , r t l l l l 61111 7 ( 1 0 Temperat.u re (K) ....

In Fig, 2 a E)SC curve ~)t an SC22 specnmen laken al K mnn n us sh,~wn 'T'he eLlecls are ,nlerpreled anah)gously I,~ lhe el]eels nn Ihe W(.) specimen (F'ng, I) and are numbered acc~rdnnglv Comparns(m ol Fig ~ wnlh Fig I yields Ihe I(~llowing ~bservat.~ns'

(n) Ihe (-;P 'z,¢me Iormalnon eflecl (effecl Inn F'ng I)ns absenl in lhe SC22 specnmen,

(hi) lhe GP-zone dns,,,,flulnon effecl (ell'eel II nn Fng. I ) S( ... specnmen,

us very small for lhe " "'~'

(nun) lhe magnntude ~I lhe mann precnpntalion effect (ell'eel III)us smaller h~r lhe S(.'22 specnmen lh:in h~r

I h e W ( J s p e c i m e n ;

(iv,) the end lemperalure (~1 Ihe mann precipnlal.m ell'ect us hJgher I',~r Ihe SC22 specimen than Ior the W(.) specimen,

(v) Ihe 61-d ss~)lul (~n eHecl (ellecl IV in Fig, I)us S( ... specimen lhan h~r Ihe less pr, m~)unced Ior Ihe " ' ~ ~

W(,) specnmen

'Table I summartzes Ihe expernmenlal dala ~blanned lor effecls I and II (except h~r the SC specimens) A l l characlernslic lemperatures nnerea,se wnlh nncreasnn~, healing, uate, whereas lhe heal o~nlenls generally decreasewnlh nncreasnnghealnngrate. Fr(~m Fig I ulns ~)bserved Ihal eHecl II seems t~ ~'~nslsl ~1 lwo sub

5 ,J IK.mol II 5 Klmnn III v !.'im ~-;[m '71.~ Temperat.ure (K) 4 0 0 F'q,. 2 DSC run ~1 a SC22 '..p¢¢llTlerl

'TABLE I The endlemperalure('T,'),peaklemwralure(/.)and heal ci~nlenl ifl eHeel ] and II-i¢ heal c~nlenl i~l ellecl II tar W() ,,pecamen,, a,, a hJrleluorl ol E)NC healing rale

Healing[ale E',Hetll E,ltt'cl II

,',K m , n -I )

/i, (' Healc~mle. nl HeulO)nlerll

(K) (K) (,l mlfl -L') (,J m~l -I )

(I '~ - " ~7g -t-_ '~ - , , 14-~(1 _+ 4 0

2 1,47 t g g - " 1,4(1 _+ t(1

", ltb3 1~1 I 7 I ~(1 ± 20

211 ~72 ~,qN IN I~,() 1 I'~

,411 ~Xq ,4 I'~ I 7 Illl) ± I1)

I ~¢~

'TABLF. 2. Theend

DSC' h e a l i n g rule

M .I ,h_'mrmA, I' i,¢lH ,4,h~ttrll~ / I ' r c c t p t l . m m m m t A I - ( '1~ ~lll,~r

l l ' m p e r u l u r e (/, ). peak l e m p e r a l u l e I It,i, a n d heal ¢¢mlerll i~l eHeel III h~r W(.) arid S(.'22 s p e c i m e n 4 a,, a hJn~.'lll)n I>l

Healing, rate Peak l e m p e r a l u r e E n d h ' m p e r a l u r e H e a l o ) n l e n l

(K mln -I) (K) (K) ()m~fl "J)

W(.) S(.'2 2 W(.) 5(.'2 2 W(J S(.'2 2

() '~ 4.4t~ '~) 1 '~78 hi)l) _+ 4 4 2 0 _+ 4() ,4 t() _+ h(I

'~ "~Sf~ (fl)l t~18 t, t8 "~7() +_ '~() ]()() -I- ~,()

"~ t d l 3 t,2'* t~ ~8 t,t~8 ~12tt _-t- I '~ 211(l __. I()

I() 6.48 e,8'4 I t1) -I- t~

2() t, ~,8 t)t,() t~7 ~, f,'-~8 I 7() _--/- I() X() +_ ,4

.4() ~'~8 ~-~ 7() (~q ~, 7117 8() -I-.4 It~ -I- I

t,() - " 71 ~ ± 4 ,.I _+ 4

8(I - " 718 +_ '~ t _+ ~i

"(. ouhJ nl~l he e.,.,limaled dLIc hi h~w heal I],)w

eHeels. Fi~r higher healing rate,.; (2() K m i n ' ~r m~re), two endulhermte peak.`' are discerned in the tempera- lure range ot efteel I1.

In Table 2 Ihe experlmenlal dala ,~blamed h~r ell'eel III h)r Ihe W(.) and S(.•22 speetmens are summarized Clearly, all eharaelerislie lemperalures increase with increasing DSC healmg tale, whereas Ihe heal e~mtenl ~1' efteel III det'rea.ses wilh mereasmg heatmg rale. Furlhet, HI is noled, Ihal h31 a eonslanl [)SC healing, rule, Ihe eharaelerl.'.;tie lemperalures I'~l Ihe SC22 specimens are alway.`' higher Ihan Iho.,,;e Ior Ihe WC)

specimens Fi~r a e~mslanl healing rule Ihe heal

e~nlenls ~)1' eft'eel III I'~r Ihe SC22 speeimens are generally Iowel than those h~r Ihe W(.) specimens.

T h e lemperaluie ol m m i m u m absolule heal tl~w between eft'eel IV and el-leel V in lhe W(¼) specimens increases wtlh mcreasmg healing rate from f17() K al () S K m i n - I to '7,4'~ K at ,4() K rain- I 'T'he peak lem peraltire ~)l eft'eel V IS almosl ident,.'al h)l S(." and WC) specimens II appeared h~ t.',e nearly msensilive h~ varia- I , m s m Ihe healmg rule, alth~)ugh Ior high heatmg rules (21) K i r o n - I or m ~ r e ) i l slatted h) slightly higher value.`'.

DSC run.',, un Ihe very .',,h~wly coaled specimens were pert,~rmed at ':, K m m - ~ T h e DSC" curves obtained did

) ~ •

m~l sh,~w any heal ell'eels beh~w '~21)K T h e E S(. curve.,,, h~r the S(.."~, S(_"O, SC'N and SL"I 2 specimens slr~mg, ly resemble Ihe DSC' curve h~r Ihe SC22 specimen (Fig 2) Fig, tire 3 shrews the E)SC curves ,fl' Ihe S(_" "4, S( TM .... -' "~ and SL•I specimens', I'¢~r comparison Ihe curve ol the SC'22 specimen is als~ .,.;hewn. It is cd',served Ihal Ihe heal e~mlenl ,~1' ell'eel Ill (Ihe rnam preclpilalion eft'eel)decreases wilh decreasing e~oling rate Fi_)r u t'(~flnng rate ~fl' () .'5 K mnn - ~ no precipitation el-leel Is ~bserved during, Ihe sub.,.;equenl DSC' expert- menl al a healing, rale ~1 5 K man ~ (ES(_ eurve ¢)1' S(_(I.. is m~l .,,h~wn). T h e end lemperalure uf eHecl III decreases with decreasing, ,.'~ohng rale

u A ¢J III ~:) K J m l n 804 ~ , ." SC'2 S('I J/K mule v vl b L)I.) ~'p( I( ) '71 )( ) 'T'emp~ral-ur¢, (K) .

Fig t r~s(.' curve,, ,,I SCI. S(.'2, SC.-I and S(.'22 ,,pe,.'lmen,,

800

Fur euoling rates ,4 K m m - i ~r less, Ihe magmlude ~)1 effeel V decreases wilh decreasing e~ultng rate. Fi~r lhese law cooling rules a new eftecl, stlualed m Ihe leml:~eralure rang, e 7,31)-7q()K, appear.`' (see F'tg. 3) ']"he magnitude ~fl' Itus new efte~.'l, denoled eft'eel VI, im:reases wilh deereusmg eoohng, rate.

T h e Iolal heal release durmg a DSC" scan (1~' Ihe inlegral of lhe heat fh~w Ir~m r ~ m lempetalure Io end lemperalure ifl Ihe .,,can), is dtreelly related Io Ihe number ol clIpper ah)ms preeiptlaled during, Ihe e~)ltng, beh~re Ihe scan. T h e lutal heat release during, a [')SC s c a n , A~.~hH , was oblamed by inlegrall~m ~fl' lhe

DS(? eurves between .~2() and '7,4q K (before 52() K no heal effeel occurs) In F'q.2, .4. AeJ,,,, (whleh is always end~flhermle) is given as a f u n e l , m ~i lhe e ~ f l l n g rale T h e accuracy/ ~fl'these measuremenls is wilhln _ 3 ( ) J m o l - ' II is observed lhal A(.:~,,,, Inereases slIghlly when lhe e~)oling rule is decreased from 2() h~ 2 K min i When lhe e~l~)Iing, rule is decreased lurlher. A(~),,,, Increases rapidly h~ ah~ul 5()(I J ml~l i 'T'hJ.~ is higher

,~1 ,I A'mtm~. f' van ,'ll, mtr/~ / I'n', IplhtllOH In m~ A/-( 'It alh~v I g 7 Ihan the values oblained t'~)r Ihe heal release d u , i n g the

main preetpilalion eft'eel (see Table 2).

SC22 specimens were isolhermally aged al ,42~I, .-14~, 4~3, 493, 503 and 523 K h~r fixed limes T h e

ROO

,.~ 400

0

B l u r t •

I IO

(2uohng Rat,e (K/nun) >

Fig ,-I 'Ti~lal heal c, msumed during Ihe [)St.' run ,~t S(_' spee~ men,,, A¢J,,, as a I u n c l i ~ n ot Ihe t'~)ltnlt, tale

¢U

V

4,;.

r,/.,

.,;.

7;,.

Temperat, ure (K) ':,

F~I..,, ", D_~C run ~1 a ~(."22 ',1:"-'~ tmen aged h~r 2()(1() mm al

42~K

E)SC' runs ~I SC22 specimens aged al 423 K h~r mare lhan ,-I,"I() min show an end~lhermie eflecl in lhe lem- peralure range ,-lhl)Io V7() K (see Fig. '~). Alter ageing l'~r 2(I()() rain lhe heal conlenl ~fl' lhls eftecl equals 22() 4- I() J m~l ~ E)SC runs ~I' SC22 spec,mens aged h)r up h~ 2()()() rain a! lemperalures helween ,-1,43 and '~2.'~ K do nol show any heat eflecl heh)w %8()K. Ab~we ~X()K either Ihe main preclpllal,m efl'ecl (eLtecl III), ~r the endolhermic (/'-diss~flulion eLlecl (eLtecl IV ) was observed.

,? 2 .~' t a l ' , l / ] f l ' u c l i m l

'T'he results ~t Ihe AI rich phase lalliee parameler d e l e r m m a l . m s are summarrzed m 'Table 3 T h e DbS experlmenls ~n Id.ngs (~1' s p e o m e n s quenched t'r~)m Ihe end lemperature ~1 eftevl Ill (see Secli~m 2 3 ) enabled Ihe idenlilicali~m or phases presenl al Ihis lemperalure. T h e phases delecled are als~ gwen in 'Table 3 As a minimum amounl ol a dittracling phase should he presenl in ~rde, t~ er~ss the d e l e c l . m Ihresh~dd, the absence ~1' dfffracled Imes tr~)m a certain phase d~e,~ n~l imply lhe absence ol lhat phase

From 'T'ahle t tl is ~hserved thai all measured AI rich phase lalltce paramelers a,e I~)wer than Ihe lalltce parameler h~r pure alummtum ((),4()4t4fl nm see [el

I ~I) Fi~r b~lh the W ( ) and Ihe SC specimens Ihe AI rich phase lalltee parameler increases wilh decreasmg L)SC healing rale.

Durm~.:, Ihe Intltal parl ~1 the healmg m lhe high lemperalure (-immer camera expertmenl ~m a W(.) spe- o m e n ~ m l y l h e A l - r , c h p h a s e ~ s d e l e c l e d L m e s r e l a l e d Io the melaslable t--)' phase appear al ,-1.()() K and dis- appear at h,-ll) K L.mes relaled h~ Ihe (-/-phase appear al 5'71) K and disappear al '7'~(I K (Ihe precisiem ()1' Ihese lemperalure delerminali~ms is wtlhm ahoul + I S K)

,3 ¢ lplz+'al mict+~sr'+~pv

F'tgures h, 7 and X sh<~w oplical mwr~graphs eft' Ihe S(."i,,,< 2, SCt,,, 2(1 and W(.)t,,,~ .~ specimens respectively

'T'ABLE 3 The re~,ull'~ ~1 Ihe X ray diHracli~m analy,~l,,

Spec=merl DS(.' end lernpcralure Pha~,e,, delecled AI lallice pararrwler h ,,

(KI (nm 4 111111111121 (al "A, -I- 11 I1,11

S( '22 u () ,41),-I 2 I I.~ 7

S(_"l,,,~ 41 '~ fl()() c~, if', ~ () ,4()4 7X () ~

S(-'l~,,c 2 t~ ~g . , 0', # () ,-1( ),4 fl ,-I () fl 7

N(.',,,,, ~ t m g a , 0', ~ l) ,-111,4,-I 7 I (I

S(." I ,,~ ,4() 7117 a 1) 4( 14 2 2 I "~ "~

WI.I rz I1 ,-111.-I I 7 I ed.l

W(.)l,,,~ 2 ~-~ I g a, H 11 ,-Ill.-I(') 7 (I fll

W(.)ll,,I "~ t) Ig If, H' 11 '411'-i %'~ (I 8t)

W(.)l,,, 211 f17 ~i ~t, O', O 114114 1,¢p i 2e~

INN .~1 I .~'nmm~, [' vml M m m l ~ / [ ' r e r t p m t m m m m~ ,41-(.'. a l h w

Fig t, (.)pineal m~cr~gruph ~1 a S(222 :,po.'~men h e a l e d al 2 I~ F'~p.. 7 (_)pm.al m~er~g, aph ,~I a S('22 specimen Ilt'aled al 211 K

mill -I II)¢)q~ I~,

All m~cr~graphs show Ihe presence ~1 /--1' phase and /l-phase prec~p~lales The ~' phase precipflales can be d~slingmshed b,¢ Ihe~r eharaclerist~e plale-like slrue- lure. II ~s apparenl that Ihe densil¥ ~t prec~pflales ~s smallesl m Ihe S(."t~s~ 2(1 specimen. II is turlher observed, Ihal Ihe lenglh ~1' lhe ()'-phase plales ~s ab(~ul l~m in lhe SCI),,, 2 aJid ab(~ul I) h l~m m the W(Jt,s~ specimen. The gram s~ze ~fl' lhe Al-r~eh phase m lhe S(.71,,,~ 2, S(_'L)~, ~ 211 and W(.)L,~,,.~ speennens ~s aboul 4()1)

4. Discussion

4 I the [r.('fiml ol dt.s'.s'oh'ed ( 'u t,,mm.~ ol'lhr A l . n r h

rnlltt,,,'t'

T'he Al-rlch phase la{lice parameler (~1' hmary A I - C u alh~ys shows a linear dependence ~n Ihe copper Ira¢lum, ~.,,, ~1 Ihal phase I~l . Llsing dala I r~)m rels. I-I - I h Ihe tollowmg relallon helween Ihe Al-rieh phase lattice parameler (at 2c)N K)and ~ ,, ~s obtained

a . ~ = a~,~ - ~, ,, (i (14'77 nm (2) where a~,~l is Ihe lall~¢e parameler uf pure aluminium. 'This equal,_m ~s vahd in eases where Ihe average macroslress m a specimen ~s zeros. AIIhough hhng causes hroademng ~1 d~Hra¢led hnes, lalm'e para- meters oblained I'n~m tihngs reflecl in go~d approx~ma lion Ihe "unslra,ned" lall,ce parameter I1 71 'The eh~se eurresp~mdence belween Ihe AI-nch phase lalliee parameler ~fl' Ihe ~veraged and suhsequenlly filed specimen and Ihe AI r~eh phase lallice parameler ~1 lhe hled and subsequenlly overaged specimen (see Secl,~n 2.3 ) I-urlher ev~dem.'es Ihis Thus, Ihe AI-neh phase lalliee paramelers ~)blained Ir~)m hlings can be

F~g N (_)pl,cal m w t ~ g r a p l l q~l a W O spec,men h e a l e d al h K mm -~ h ) t p ~ K

sately used in eqn (2)Io i~blam values ol h u, wtuch are summarized m 'T'ahle 3

4. 2. h/emlh~'um,1 oI' D3'(' efle~ 't~'

The idenliliealion of lhe heal effects (~ccurnng in [)S(." runs ~)I W(,) and S(."22 speclmens (see Seel,~n 3. I ) has been d i s c u s s e d previ~usly IxJ. For mosl ~I the DbS experimenls ~n W(.)D~. ., and S(_"Ds ~, speclmens as well as h~r lhe high lemperalure Gumler expenmenl on a W(.) speelmen heated al (1.5 K mm- i lines due to the U'- and 6/phase have been ~bserved al lhe end lemper- alure eHecl III (see Seel,m 3 2.). These ~bservalions c~mhrm lhat efleel III is caused hv lhe formali(m ol ~' and ~ phases. F(~r the SC22 specimen heated al 4() K mm- i, a signil,e:inl heal evoluli~m ( l h _ I J m~,l i see Table 2 ) i s ohserved during the main precipitation effecl However, lhe E)bS X-ray dfl-fraellon on lhe S(?r,,. 4() specimen deleeled only Ihe AI-r,eh phase

Clearly DSC can deleel smaller amounls ol preetp=lal- mg /-/' //-/-phase Ihan £)bS expertmenls can Hence, allhough h~r s~tme WC)r~s~ specimens n,~/-)'-(~r (-/phase m detected ('Table "~), Ihe presence ol lhese phases m these specimens cannol be ruled ~ul This is also mdiealed by Ihe amtmnl of c~pper dtss~tlved in Ihe AI- rich phase after compleled precipitation ( t~., m Table

3), which =s lower than the Cu conlenl ~1 Ihe alhw. Except h~r efteel VI, Ihe heal eft'eels ~bserved in Ihe DSC runs ,tt lhe SC specimens c~(~led at rates lower Ihan 22 K mm- ~ are atlrtbuled I~ Ihe same reactions as those observed m Ihe DSC run (~1 the S(."22 spec,men E',flecl Vl is discussed =n Sect,on 4.3

'The lemperature range ol Ihe end,)lherm.: eHeel between ,-Iht) and S7()K m DSC runs ~1 Ihe SC22 specimen aged for 20(I() mtn al .-123 K (see F'tg 5) c,~rresponds well wtlh Ihe end~lherm,c GP II-z(me dmsolut,m el-I'ecl h~r a GP II-z, me conlatnmg, A I - 1 7 at.%Cuspeetmen[IH] The lemperature range for Ihe dissolult~m of GP I ztmes is uboul S() K lower than the temperature range lor OP II-z~me d=ss~lul,m [2, IN I 'T'h=s indicates Ihal Ihe ~bserved endolherm=e eHecl is caused by the dmst~lu/ion t~l (_;P II zones In addil,_m =l ts m~led thal tn waler quenched AI- 1'7 al.% Cu spect mens aged for 2(.I()() mm al ,423 K lhe h~rmali, mol OP II zones and the diss~tlut,.m ~1 GP I zones are b~)lh ch~se I,~ being c(~mpleled, while Ihe am~unl ~1 tY-phase h~rmed =s slill n%,,I,g~ble ] I q[

Aee~trdmg h~ lhe metaslahle s~lvus ,~t' Ihe X-phase (see tel. 7), X phase d~ssolul,~m should occur =n a lem- perature range ahttul '~() K higher than GP II-z(me dis solul,m The end~dhermtL' eft'eel belween ,4hi) and

~7() K m lhe S(."22 spectmen aged hIt 2()()() m.n al 423 K is allrtl~uled h) (3P II z,~ne diss,~lul~on SC22 specimens aged Ior up h~ 21)I)() mm al lemperalures belween 443 and ~2 ~l K d,d n(~l show any dissolulton ell eels heh_)w 58()K. Thus =1 is c~mcluded Ihal no X-phase h~rmal,m had ~tccurred in the ~solhermally aged SC22 spee,mens E)uring Ihe X ray diftracl~on expe,tmenls n~ Cu-containing phases except h~r Ihe t/' and O-phases were observed S(~ the occurrence ~1 Ihe X-phase, which was reporled by Nakamura t't it/

171, canm~l be e~.nhrmed.

4,~ The t~!ll, tem'v . l r . ~ h / t ~ ralc . t / c~ppe/

,s'llpt'/;s'tlltttdltt tit it~Ill t lit ('tlppt'/ prt't'tplltllt¢ tit/di.~,s'~ t/ill/ell/ Since pree~pilal,m d u n n g cot~hng mvolves t/-phase h~rmal~tm (see, lur instance, 'T'T"T' diagrams ~1 quenched AI-C'u m ret 2), the amounl ot copper pre- ctptlaled during c~.~hng, A t , can be esltmaled In~m (see als~ nexl secl, m):

AQ,,,~ =/& t a l l .

(3)

,n which A H ' m the heal ~tt prectptlat,.m ~)1 t/ phase, 4,4 kJ mol-~ c~pper m the //-phase 'The am~unt ~)1

el)pper alarms prectpilaled during co(~lmg al a tale ol 22 K ram-~ can be esl~mated Irq~m e~lher AQ.... (using eqn ('~))(~r Ir~)m the AI neh phase lalltee parameler (using eqn (2)). The two delernunaltons agree wtlhtn experlmenlal err~r, and ,I h~lh~ws Ihal ab~ul "~'Y.. ~1 all eopper al~ms had pree=ptlaled S~ ,'ill ~tr nearly all e~pper remams m sohd s(tlul,m At c~)ohng rules beh~w 22 K rain -~, Ihe pr(~D~rltun ~tl' c~pper alarms remaining in s~lid solul~on decreases wHh decreasing cotdmg rule. AI ct,~hng rules heh~w 2 K mm-~ Ihe p r ~ p o r t . m ol c~tpper al~ms precipitated during e ~ h n g increases str~mgly. W,th eqn (3)tl tolh)ws thai durmg c~,flmg al ()5 K mm-~, atxml 7()'~,, ~1' Ihe copper al,)ms preeip,lale

The end~lhermw eft'eel Vl ts presenl after ct~ohng al very low rules, tor which Ihe amount ~)1 /-/-phase strongly increases (see ab~we) Thus Ihe eft eel VI ts altrthuled h.~ dtssolulion ~t (-/-phase preetpttales htrmed during Ihe sh~w c~,~ling pr=,~r h) lhe DSC analysts. The O-phase formed dunng slow eool,ng (al low underc~,~hngs) will be c~arser Ihan Ihe O phase h~rmed in efteel III during heating (al large undercool- ings) Since, due h)hm.ger d=Husion paths, dtss(tlul,m ot c~arse preeipilales lakes m~)re lime, diss~)lutton of Ihe c~)arse O-phase preciptlales h~rmed during e ~ l i n g will pr,~eeed bey~md effect V (d,ss~lul,m ol /-/phase prectp=lales lormed during el-lecl III) This ts indeed t~t~served (see F'tg 3).

For c~(~hng rules ol 22 K mm-~ ~r less, vtrluallv no Iraces ol (-;Pz~me I(~rmalion ,)r dissolull~m are ,~bserved tn Ihe subsequent I-)SC' runs. T'h=s ts explained as I'~lh~ws. Slow c,~oling Intm the hom, tgen tzing lemperalure resulls tn a reduction ,~1' the excess vacancy e~mcenlral,m. Smce vacancies are needed Ior a qmck Marl ~fl (-;P-zone tormalll~n, GP z,~)ne tormal,m durmgsuhsequenlheal Irealmenlisrelarded Fbr very h)w cording rules (I K mm-~ ,~r less) the reduced supersatural~(m due h~ (-)phase prectp~latton during coohng mlghl prevenl (-;P-z~me htrmalt(m.

4 4 Heal ~'~.llt'lll,~' r~lpn'~'iprhlll~)ll/,h,~'s~ dllllOll t'/~i'cl.~' 4.4 I llllr~du~'ll~m

'The heal pntdueed during, Ihe prectptlat.:m ~1 a eerlain phase can quantttalively be described in lerms l~l Ihe (melaslable) solid .',;~lubility aller preclpitalion and Ihe heal ~1 prectptlalton ~1' the phase mwtlved

III

II i'.; here assumed Ihal Ihe end ol preeipHallon is reached when the driving torte, Ihe difference between Ihe average proD~rl,m ot uloms dts,;~lved m lhe AI- rich phase and Ihe (melaslable) solubihly, becomes zero 'T'he tolal heal ol prectpilal,_m per mole alhw, A(;, is Ihen gwen by (see ret 1):

tl u - I l l u

(,4) A Q = A H I - ~ . . ,

lqll /11 .I .~'ImmA, f' ~'u~t/llntm/~ / /Yet~l,/lUl/~m m m~ .41-(_ u ~dlm' where A H is the heal ol precip~lall~)n per m()le c,)pper,

t~(~,~ is Ihe al(.)ni~e trael~(~n ()1 t a p p e r in=lially dissolved =n Ihe Al-r=eh phase (m ()ur ease Ihe gr,)ss c(~pper e()nlenl), ti,. =s Ihe (melaslable) s~)lid sulubd=ly al Ihe

end lemperalure uf Ihe preeipllallon elt'ecl As

~;'u" t~',,"' I , l h e { I - t ti'.l lerm can he =gni_)red 'The value,., ,,1' Ihe heal,,, ~)1 pree=pilal.)n ,_)1 (_iF) I zones, GP II zl)nes, fl'-phase and 0 phase can be ,~btained t'r(~m Ihe slope iff the slrai~hl line Ihr~)ugh lhe dala poinls ~1 Ihe ph_)l ~)1 Ihe hL~,arflhm ()1' Ihe (mela- slable) s~)lid s()lub=lity vs. Ihe ree~pnwal lempe=alure I I , 2(I I. Table 4 summar=zes values h)r the heal (~1' preeip=- lali~n (~bl;ilned from Ihe lilerature

Fur GP z, mes, Ihe mlertaciul energy and Ihe energy relaled t¢~ elaslie strains around Ihe zc.~ne,.; can have i~ s=gn.l'ieanl mlluenee i_)n lhe h_flul energy at' Ihe h)rma- li¢)n (~1' lhe z~nes 12 I] Since holh the eream)n ~)t' slrains and mlerl'aees require an ener.~,y mpul I,~ Ihe specimen. bulh slrain and inlerlacial ener~,y reduee Ihe I~)lal amuunl ()f energy ass(~eiuled wilh Ihe f u t m a t . ) n ~1 zunes T h e imp~rlanee ~)1 Ihese cunlribuli,)ns is deter- mined by Ihe average s~ze ~1 Ihe z~)nes' Ihe tulal Iorma- I=,m energy ~)1 small zunes cuntains relauvely large slrain and mlertaeial ener.~,y e(.mlr=buli(_ins F~r GP zunes h_~rmed al r,.)m lemperalure (the average d=ameter ~)t Ihese zones is ah~ul "~ nm: see rel. 21) Ihe I~)lal energy ()I Ihe z, unes and elastically def~rmed malrtx ar~und Ihem, is abutil 13 kJ mol-~ e~)pper m Ihe z~nesl211.

F'r(_~m Ihe ah()ve II tulh~ws Ihal afler h_)rmat,_~n (~1' (-iP zones during a DSC run has heen eumpleled, al leasl Iw~ rea~.'l~n,~ can (~ccu~ ~_~n euntmued healing LiP z.(~ne diss~)luli()n and GP zone c¢~ar,~enlng, resulling In end~- Iherm~e and ex~)lhermw eltecls respeclivelv F'urlher. il is expeeled thai c()arser GP zones d=ssulve al tu.~,her lemperalures and produce an increased heal ell'eel as e,~mpared I~ the hner z,(me.~ Thus, elt'ecl II =n DSC nJns un WQ spec=mens ~s dommaled by dissolul,.~ ul (3P zones hut =t als() e~mprise',; (iP-z~me e~arsenmg. '['his explains lhe c()mphealed, healing, tale dependenl s.,hape r~[ eHeel II (see al.,.,(~ rel. 21)

1'7 .I inol-~ alh~y see 'T~=ble I)~s ~ne (,rder ~fl'mugnilude h)wer than Ihe heal ul d=ss~)lut=lm uf Ihe z, unes (ahl)ul I ( ) ( ) - l q l ) J m~l-~ alloy). 'T'has rod=tales lhal ab(_)ul q()o/,, ~)I lhe LiP zl.)nes are h.)rmed during the .,.;Ii.)ra~e h_)r I h al r,,)m lemperalure helween quenching and lhe .'`'tart (_)1 Ihe DSC run II ~s Iheref(~re lUslihed I~) appr~)x~male lhe heal (~I h)rmalil.)n ol lhe z(.ines hy the heal ul (..;P z une h)rmali(m al r(,~m lemperalu=e 13 kJ m u l - ~ cup- per m lhe z~)nes (see Seel~,m 4.4.1 ). S,nee lhe heal ev¢)]uU~m during lhe n._~m lemperalure slorage was n,q measured, =I =s n(fl po.,,,s=hle h),.)blain lhe lolal heal evulved during lhe h)rmali(m (~I' lhe zones. However, since reversi,~n uf CiP z()nes m AI-C'u is c(,mplele

12 II,

lhe heal dlssipal~un during lhe GP-zune diss~luliun eHeel (heal elYe,.'l II) equals lhe lolal heal evulved during, lhe h)rmal~i,1 ~)I the z¢)nes II is now puss=ble lu apply eqn (,4) lu lhe heal dis.'.;ipal,'m durin~ lhe diss~l- uli(_m (.)f GP zones Fi)r lhe sulvus uf (-;P I zones lhe l'~)rmula g~ven by Nakamura el ul

1'71

,s used 'T'he resulls (~I' lhe ealeulal,)ns are ~,qven in 'T'ahle .'3 A g~)(.~d agreement belween eah.'ulaled and measured values =s ubserved

T h e endolhermle heal el-led =n lhe SC22 spec=men aged h~r 2()()(l mln al ,-123 K (see F'=~. 5)=s :lllrlhuled I,) lhe dissolul~)n ul GP II zunes (see Se~.'l~i)n ,-1.2.) Since GP II zones h)rm by c()arsening ol prew()usly f()rmed (-iP I z.unes (see ref. 22), lhe prupurli()n ol eupper alums in (_IP II z.(~nes is delermined hy lhe GP l s~)Ivus. In applying, cqn (4), we musl lhus take ~i ,~ equal Io the .,.;(~Ivus ol (-iP I z~)n~.'s al 423 K (abuul ().ql al %). rl.us leads It_) a value ()I 231) _-I- ,-I() J mt)l- ~ fur lhe dls.',;ulul,)n ~I' the (-iP II z(~nes, which corresp~mds well wlth lhe measured value(22(l-l- I l l J m o l i)

4.4 .? H¢'~I e m b l e m ql'elJi'c1111

C)n lhe bus=.,,; ,)I lhe assumpliuns given in Seelion 4. I I., lhe heal conlenl uf elleel III can he calculated. In this ease lwu p~is.,,;ible m~des ul precIp11atlon shuuld be c~n,sldered, and lw() values are i)blained h_)r lhe heal ev~lul=un during precip=laliun: (~ne h)r ~'-pl=ase

4 4 2 H e . I e m l l e m q/¢'~l'ec'r I m l , I I I

Fi)r Ihe W(.) specimens, lhe heal ev~)lulic)n during lhe DSC' run caused by lhe l'ormali~)n uf GP z~,nes (abuut 'T'A BL.E 4 Heal,, ol prL:,~.',pllal.m ~I (JP z,~ne,.,, ,.,lahle and rnela,.,lahlc pha'.,es m AI-(.'u a.lh pV.,.,

Pha,,e/z~,le AH (kJ mq~l- ') Relurence,, (3PI 22± 3 7,21 (JP II '~1 -t._ '~ 7 O' LI +_ ,4 I O ,.I,J _ I I, 7

TABLE '~ ( '~)rrlpafl,~(in ~,1 Ihe mea,,ured and Ihe,)rcllcal ht';ll ¢onlenl(~(),,, .,.. arid ,~ (.),i-,,.,,, ie',,pecllvely)~)I lhe LIP z,~ne di.,.,~., ) lull,an ellel.'l (eHv(.'l II) a'., uh',,erw:d ,rl IIi¢' DSC' run',; ,el lhe WL) AI-( 'u ulh)v a",. a [unellL)rl ,)IDS(." healln/.~ talc, lh¢' ~.', )pper ,.,~)luhlhly ul lhe ~nd Ivmperulurr ol lhe (.iF' z,~n¢' h)1'mali~)n peak ( l' c .) a,., ~)blalned In~m the (jP I z4~rle ',,i)Ivu~ i~ al',,() L..,Iven

Heal.ng rale ~i ,, /3 U,h,~,,, AC~ ...

(K mm -') al % IJ rr.)l- ') (l rnul i)

() '~ () ,4 ~ I~,() Iq(I ± ,4()

2 () '~ I I'~() I,J() _+ ;()

2() l l t ) l 14() I~()_+ I~

,~1 .I .%'hlllHk, I' r a . . 4 h ) , n k / I'~e, nT, ml~O, m a , A I - ( '~ allm' I q I precipnlaln, m (A~2"') and une h~r ~ phase preclpllatn~n

(A(.)"). T h e respeel,ve values (d ~'~ ,, are d e n v e d trom Ihe ~' a n d / / p h a s e s~dvus lines as .~.iven by M u r r a y I.~1 F'r,~m eqn. (4) il I',fllow,~ Ihal Ihe end l e m p e r a l u r e can be regarded an Ihe d e l e r m m m g variable I'~r the he:it o~nlenl ,fl' ell'eel III. Hence, F'l~. q gives Ihe t h e n r e l w a l estnmales h)r A(,.~ ~ and

A(2 'r,

as well as Ihe experl- menial values hit Ihe heat e~ntent ul effect III, A O, as a Iunel=un ol lhe experimental end lemperalure ~1 eHecl III 'The values el' A O are always mnlermedlate belween A O t' and A(.)"'. This agam c o n h r m s that el-leer III is caused by Ihe eumhined /-/' / # phase preclpflalion All data pomls m F'~g q are well represenled by a stogie line, ~.e Ihe end temperalure of effect III is, in lhis case, sulely related Io Ihe experimentally ~bseived heal conlenl of eft'eel III

In F'% I11 lhe average o~mpusm~n ot lhe matrix at lhe end lemperature ~1 lhe mum preelpmlamm eHeel is given as a tuneln.)n ~1 Ihal end lemperalure Again, as was also) ~bserved for the heat ewflved during Ihe main preclpilamm eflecl, there is wrlually no dflterence between the average malrix c~mpusflion atler Ihe main precipilamm efl'eel h~r Ihe SC' and Ihe WO specimens al a hxed end lemperalure ~1 Ihe preclpflall~m elieel

F'n~m the value ~)1 Ihe al(~mw c~pper I'raell~n ~1 lhe AI rich phase al Ihe end lemperalure, ~r,,.,,,~, and lhe ah)mw copper traction ot Ihe AI rwh phase al Ihe slarl ol Ihe DSC' run, tl~,~, the am¢~unl ~fl' copper al~ms pre clpflaled, A~, can be caleulaled as A~ = ~I ~,, _ ~ t,,,,,,,_,, In Fig I I , Ihe expenmenlallv ~d~served heat eonlenl ~1 effect III ~s given as a I-unemm ~1 A~. II is ~d'~served Ihal all Ihe dala p~fints, b~lh h~r We) and SC'22 specimens, are well represenled by a single slraight line lhr(~ugh Ihe origin, the shape (~1 whwh gwes the average heat of preclpflallon ,~1 Ihe c,~pper ah~ms whwh are presenl m

[] SC22 ;-, ~ preeipit,aLion ,-~ 4oo " - , , , "',, ... d ,~ . ... I , _{,, ,,,} '"., _{' %,,} • j _{[ "} %, "%" ,,,%, ~ ,N (-I' preenpitatn~m '...,, ,~, ill)iF 65(l 70(1 'T'e (KI ,

F'=~ q T'h¢' heal ewdved dunng the m a i n p r e c i p l l a l l ( m eHecl us a iunt'licm ~i iIs cm'l lemperalure (.'aleulaled value,, I u,~in).~ eqri (4)) h)r Ihe case q)[ ~ pfecipllallOn and I(ir Ihe case ¢)1 #t pteeipilaliori and measured values h)r Ihe main precnpnlalion eHc¢l in S(_' arid W() specimen,, are ~.iw.m

a precnpntale al Ihe end ()1 e r e c t III, Aft: A/~ = "4h + I kJ tool-~ e~pper 'This value is inlermedmle belween Ihe heals ,fl' preclpilam~n for ~' and /-/-phase ( 1,J and 4-1 k.l imfl-' respechvelv), and thus in agreemenl wllh the arlrlbulion ~fl' effecl III 1o (--/' and/-/-phase preeipfla- li,~n.

4 ') Actll'allOll {'tl{'l~rV OII(I/V,~'I,%'

T h e aclivaHon energy ot a reacll,m ¢~ecurnng d u r i n g heating al a constanl healing rale can m sc~me ca.,es be ~blained by a K=sslnger analysis ,~r vanunls Ihere, fl

1231.

Such an analysis can only he perl'ormed in cases

where heat eHecls are caused hy a single p r e e l p i l a t l o n pr()cess Fi~r m m - i s u l h e r m a l anneahng with a constant heating r i l e , ~|~, Ire, the ease ol E)SC e x p e n m e n l s ) i t can he sh~wn, that in go(~d appr~xlmati¢~n Ihe h_fllow-

0 II~() O 0 Ib ,A I.~ (I 111(~ (I (lllb • (I ()O0 . . . . [ ] _S(.,2~ - , - 1 , ,~ = W Q m ~PmP '" [ ] [ ] i1 [] , .... . .... , .... . - . . . - . . fiilll fib(I 7(l() "Pc ( K ) .

Fig I(I Average malnx c,)mp~sili,)rl al II'ie end Icmpewalure (~t Ih~. = main precipilali¢)n eHecl as a Iunclion ~1 Ihe end lemperalure ~fl Ihe main precipllamm elt'ecl $~flld ~,oluhililies h~r # arid h~r ~/' phases h~hlained tr~,m rel '~} and measured values h~r Ihe main preeqlilali(m p~.'ak in S(.' and W(.) ,,pecimen:, arc' p, lVerl

[] SC22 ,, ... • W Q ,,,~" 4 0 0 A ... : ... , ,, " ,,,,' 21jq) , ,' ," ,#' , , ,," (I . . h.!, . .;I. Ax ,

F'iL;,. I I T'I'w h e a l e v o l v e d d u r l n l ~ , l h e m a l r i F H e t ' i p l l a l l o n ei-leL'l as a h J l i C l l ~ f f i ~,i lh,2 a m o l J r l l u i c ~ p p e f p r e ~ : i p l l a i ~ d

Iq2 ..11 / .Yt, rml~. P vmt ,~lmm~ / Ptectpttat,;, m , , .41-1 ", ulhw

mg relalt~m belween the lemperalure h_)r a hxed slage of I r a n s h l r m u l r m , 1- I, and Ihe healing rale, ~-I-L holds

'T'I '~"a

In - + Infli (5)

where fl~ is the slate variable fully delermmmg Ihal fixed slale ill Iransl'tlrmali~m, E,'a is lhe aclwalum energy ill Ihe prlwess and kt~ ts B~llzmarm's eonslanl. 'Thus, Ihe aelwalltm energy can be oblamed trom Ihe slope ~il' slrmghl hnes Ihrough Ihe data pomls i~l

tn( r l,i> )

i l r,

li was decided Itt o b l a l n efleclive values for Ihe acli

V a l l O n energy ,rely Ior Ihe early sluges ,~t elteci III,

where iI can salely he assumed Ihal ,rely O' phase I l l r m a l l o n is lakln.~, place (see resulis (it' Ihe X ray dilfracllcln experlmenls, Seclic, n 3 2) II fillh_iws Irilm eqn. ( 3 ) l h a l a fixed am~lunl ol heal developed, Qi, cllrresponds Ill a tixed amounl cll precipilaled O'- phase 'T'w,~ iransh~rmalton slages were ch~sen. I',~t

Q l = 2 X J m~il-i and Q l = 5 ' ~ J m~l -~ In i~rde= lu

ensure Ihal Ihe lempetalure ol Ihe ttxed slage ol lrans Iormallon, 'T~, was always hlwer Ihan lhe peak lempera lure ~ll eHeel III, only values of 'T I equal I¢~ llr below 1~5(-~ K were c~mstdered T h e h~wesl lemperalure included in the analysis was 32t~ K T h e values Ihus oblamed dil m~l depend signd'wanlly ,in Ihe slage chllsen T h e average values h~r the apparenl aelivali, m energy ~1' 0' phase Iormalt~m are h~r lhe WQ spec= men, l . l ( ) _ ( l , l l l eV', h~r Ihe SC22 specimen: (I '7~ _+ () 15 eV.

In 'T':ible/-~ aclwal,,m energws h~r O'-phase formalion i~blamed from hieralure are summarized. M~lsl ,~1' Ihe dala c~mcern Ihe overall kinetics ol 0' phase hlrmuI,un. O n l y Aar,ms~m and L, aird 124] observed dtHerences m Ihe kmelws of Ihickenm~, and lenglhemng ill O' phase plales in a sohd quenched and aged A I - 1 . 7 al.'¼, Cu For Ihe ihickenmg, the rate dependent slep appeared ii.i

be ledge t o r m a l i o n ol the broad surface ol the U'-phase plales, whereas h_ir Ihe lenglhening, Ihe rale dependenl step was tound to be pipe dtl~st~m 'T'he Ihickemng was described Iw a linear lime dependent'y, whereas Ihe lenglhenmg was deseNbed by a square rolll lime dependency T h e a c l i v a l r m energy for Ihiekemng was h~und Io be I 20 + () 1.3 eV, and Ihe aclivalicm energy Ior lenglhenmg wan htund hi be 1.()2 _+ ().()2 eV. Chen and I-h~hetly i2'~l alsll sludted Ihe rale i l l Ihtckemng T h e y pr,ipilsed a square ro, iI lime dependency i~t Ihe Ihickening during tsolhermal ageing. (.hlmbinmg Ihelr d a l a ( a s p r e s e n l e d m F'tgs I a n d 2 ( a ) ~ f f r e l 2 ~ ) a n d l h e dala by Aar~lns, ln and Lmrd, and usm.~, Ihls square rolll dependency, results m an aelwaltem energy t'clr Ihteken m g o l l 2~ -I'(l()~-~eV

'The aelivali~m energy I'~lr O' phase preeipllalt, m in lhe SC22 specimen is .slgntltcanlly h~wer lhan aellva liim energws repllrled in Ihe quilled sludws (see 'T'ahle h) II is als¢l much hlwer Ihan Ihe aclwalion energy I'~lr O'-phase I ormalton in Ihe W ( ) ,.;pectmens 'This can be accllunled I'~lr by' (I) diHerence m lenglhenmg and Ihickenmg kinetics, (fi) excess vacancies, (=ti)mierlaclal energy 'These wdl be discussed beh_~w

4 ~ I L . e , g l l w , m g a , d t l m ' k e m , g kmetir.s'

T h e ~;plwal mwrographs ~;t pre¢lpllale sln..Iclules (F'tgs /l, 7 and X) sh, lw Ihal Ihe 0' phase plales m the

S(fDM,

specunens are generally rnuch longer Ihan Ihl~se

m the

W(¼)D!,,I,

specimens (ab~lul '~ and (I '~ tim respec Iwely). 'T'hls strongly suggesls that lenglhenmg r.; rela- lively m~lre imporlanl I~r #'-phase hlrmalton in SC'22 specimens lhan Ior /-/'-phase h~rmalion m W Q speci- mens. Since Ihe aclwallon energy hjr lenglhemng is about ()2 eV hlwer lhan Ihe at'lwaH~m energy I',lr thickening (see above), this can, al leasl parlly, explain Ihe i~bserved diHetenee helween Ihe aclivalton energy Ior /#'-phase preeipllal=on m SC'22 spee=mens and Ihe aelwalion energy for e'-phase preeipilalt~m m Ihe W ( ) specimens.

TABLE ~-~. Aellvalwprl energ,e', h~r 0' phase hlrnlallon (W(.) = watc'r quench. IW(J = nee water quench, I = ,s~llherrnal, NI = r,m

Isol he r ITI;.II )

H~mepgentzml.~ ' (_'~olmt., Heal Irealmenl (.'u f ,, KJneltt. s

lemperulu re ¢~ mlenl lee ) !,ludled

(K) I/NI Temp. rang.e (K) (al %)

Rt'lcrc'm.e

7qX W(J" I 47 ~,-'~.-IX I 7 I I Overall

;'421 W(J I 47 ~ - ', 7 t I ? I () 2 L engl hen,hi.,.

I 20 Thwkernng

XI ~, IW() I ',...I 1-(-~X 1 I ~ ii X~ Overall

7q 1, W(.) NI ~2¢~-I~,..I..I I 7 I I() C)w.'rall

7~I) S(.22 NI ",27-t~htl ] 7 II 7"; Overall 2¢1 24 24 27 T'hp, nludy T hi,'., '..ludy 'Folh~wed by ;t rever',,t~,n Irealmenl

,11 I ,~'mtm~, [' vmt ,,U~um~ / ['t¢l tl~tl~ll/ott llt m / ,41-( 'H alIov I q "~

,4:1 ~ Ek('('s,'~ I'O~'OII('H',~'

A l l apparenl aCllvall()n energws h)r Ihe (-/'-phase pree=pitalton (see Table h) are lower than the u c l w a t . m energy I',~r c,~pper d~H'us..m m a l u m i m u m and sell dd-has.m m alum=mum (holh ah~ul I 4 eV: see rel II)).

As e~pper ddtu,.;.m can ~mly proceed via a vacant'v mechanism, Ihis indicates lhal in all these specimens slgnlhcanl am~)unls (~.1 excess vaeaneles are pre,~enl.

4.5 .? hm'dilCl,I/em'L~,'v

Inlertaeial energy acls h~ increa'~e apparenl at'tlva- li~n energ.ws ,fl~lamed Ir~un A r r h e n l u s analy,~is

12g]

Aec~)rding h~ B e r k e n p a s

1281,

this eft'eel increase.'`' w i l h

deereasmg t e m p e r a l u r e 'T'his can explain w h y lhe ael~ vali~m energies ~ h l a i n e d f r o m studies al relalwelv h~w l e m p e r a l u r e s (ai~)und ~ ~() K) ale

hi.ghe=

lhan lhe ,~nes ~ h l a l n e d l'n~m sludies al hig, her l e m p e r a l u r e s (around hlll) K ) (',;ee 'T'ahle h )

5. C o n c l u s i o n s

Fi~r an A I - I hh at'~,,(.'u a l h w HI is I o u n d Ihal (=) D u r i n g t ' ~ h n g I r~m Ihe hom~L.,,enizing lelnpera lure al lales Iowe~ Ihan 22 K man -~, s~gnifieanl n u m b e r s ~1 c o p p e r ahem'.; precipilale F'~r c ~ u h n g rales higher Ihall 2 2 K m m - ~ , all c,~pper ah~ms remain in s~ did s~ dul,_m

/u) A l t e r ~.'~)l=l~g, al a rate ~t 22 K mm ~, Ihe c ~ p p e r al~m,',, are w r l u a l l v t'~mplelely relamed in s~hd s~dut.m G P - z o n e fi~rmali~m d u r i n g suhsequenl heal Irealmenl is hJndered. T'h~s is Ihuughl h~ be due I~ an meLt..'wnt n u m b e r ~1 (exeess) vacancies.

/ill) A f t e r a waler q u e n c h t i e m the presence ol a large n u m b e r ot excess vacancies) many G P 'z~nes are h~rlned d u r i n g ,suhsequenl ,~torage al i~.~m tempera- lure h , I h

(iv) 'l'he heat L'~)nlenl ~1 the d=,~solut.m ~1 (-;P z~mes lel-lecl II) can q u a n l l t a l i v e l v be described in terms ~1' Ihe heal ol p r e t ' i p I l a l . m (~1 G P I zones and Ihe s~hd s~lub~lille,., as der,ved from Ihe G P I zone s~dvus

(v) T h e heal ¢(~nlenl ~t Ihe c o m b i n e d

e'-/e

phase p r e c i p i l a l ~ m eltet'l appeared h~ be p r o p o r l , m a l I~ Ihe n u m b e r ol c o p p e r ah~lns precIpllaled, yielding an average value h~r Ihe heal ~1 c o p p e r precq.~llal,m ~)1 3h kJ tool-~ t'~pper

(vi) 'The appa~enl aeltval~,m energ.ie,~ h~r (-/'-phase h~rmal~on a~e much smalle~ aher sh~w c o ~ h n g than aller waler quench.ng. T h r e e lael¢~rs can inlluence Ihis apparenl at'tJval,m energy m o b d l l y ,~1 dlssulved atoms, mlerlaeml energy', d H e c l l o n ol gr~w¢lh (m~rmal ,~r p e r p e n d i c u l a r to Ihe (/' phase plale). T h e relatwe imp,~rtance ~)t Ihese lat'h~rs may d e p e n d ~n tempera- lure and ~.',, drag tale

A c k n o w l e d g m e n l s

T h e auth,~is are m d e h l e d I~ M r H K l e m l a n h~r pre- paring, the alhw, h_~ Ing N. M van der Per.,, and M i . J F. vail L,enl I(~r assislance wllh and dl,.,eLissions o m e e r n - mg Ihe X ray d = f l r a c l ~ n exper=mcnls, and h~ Mr. P .I van der Schaal I'~r e x p e r i m e n t a l ass~slance. Sl~mulalmg discus.,,,~m,~ wilh Pr~lesso~ B M K ~ e v a a r are grale- tully a c k n o w l e d g e d

'T'he Imancml supporl ~d Ihe F ~ u n d a l , m h_~l Funda- menial Research ~1 M a l l e t and h~r 'Techm~h~gwal Sewnces ( F'( ) M / S T W ) is graletully aekm ~wledged

Reference'`,

I M J Slaunkarld P.~,anMq~urlk,,~h'm// IrmLs A , ) ) ( I q q l ) h(~ '~ -h 7,.1

2 K Hiram~and H Iwa.,akl, l%m~ Ipn In~l ,lh'f, '~ (Iq~-~4) Ih2- I 70

A M Z, ahra, C Y Zahfa and M La~llt', Act hh'tall, ~ ( I~-~ 7'~) X 71;-XXh

,4 I L Polmeal and H K Hardy I I.sl ~lh'l, ,~,'t, ,',1~'~,4)

q 1, _ 3 ~-;,-I

'~ I L Murray. Ira ,~h,mll R,,I,, ttlllq~4"~)211-21"~

h V (Jer~ dd, ,b'~ i ,'1/e/till, 3J ( [ qXX ) q 27 - q '~ 2 arid ,,uh'~equenl paper,, m VlevVp~Hnl St'l

7 F N a k d m u r a , F F'urukawa, M h'anal and I Tukamura,

I'hflos ,~l.g ,4, '~4 ~ I qX¢~167- 7X

X M J ~la~mk and P van M o u ~ k , m T K h a n a n d (_; EHerd~eH..,,

"ed', ), ['tol ol I1.' /HI ( ~ml On ,'Idl' ,~l . o d I~lg ,4110V~, I . m ' ?0-.~2, I(~1, ,'~lll,~/t'rdlllll, ,ASH F,.Jl,~pc, Bu.J,.,,el,., I~01 pp hw',- 70,4

8 D (.'ulhly, ['h'//l~'ll/~ ~l A' toy I)lh~h.'llOll, Add~,s~m We'~lev

R~;adml;,, MA, 2nd edn, Iq7X

C I ,kmHhr'll.. /wl ), ~lh'l.ls I?c/cwm ~' B.~d~, Bullerva~rlh!,, L~md~m, "qh ¢dn, IqTh

(' Gart'~a (.'~rd~wdla and E L~tu,. ~h'mll [t.n,s ,4, I ~,,I, ( 1~;~,41 ~,,'4q- ~'.~ I

l M Papazmn, ,~h'mll f'mHs A, I.',4, ;IqXI)2h~4 - 2t~l) E I MHlemeqe~, P ear, M~ur~k and Th H de Keq,.cr, f'hth~ ,1lag ,4, ,4 t ,', I~X I ) I I ~, 7 - I It~-I

H I A ~ m and W Humc R~lher~,, ['~o, /? Ao, L m . h m ,

,41~¢ , ', 1~,-18) I-2,-I

l E' Dl~rn, P Fhelf~k~w.,,,ky and T E ]-Ielz, [t.tls Al,,llf, I-, (' E IIw~..:l anti J M Sd¢~t'k, ] Inst ,,lh't, 74 (Iq,4X)

-1 '~ 7 -,-Ih 7

I C N~wan and l B Cohen, Re,~'uh.fl .~'ltt',~'s /Ih'l.l~llt¢'lllL'/l/h~' /)Ih~H lll~ll ,tlld hllt'l, ptt,ltlfll~ll, ,";pnnl:,er Verlal:,, New Y~rk, Iq~7 F, P 12h-I~,()

A Z ahra Kuhk M L al-Illle, P Vlglcr and M Winh.qbeu!er, ,4hlllttlltllllt, '~2 ( I q 7h) 1 ~, 7 - ~,h(I

'T Sal~)and T Takahu'.,h~, .~,'{ t ,lh'mll, 2) ( I ~.~,'4X) ~,41 -q,4h M van R~.wen and E J MIIh-mel)er, ,~h'tall [t,m~ ,,I, .'q),,4 (IqSqi12()7-121,-1.

A M Z,ahra, M L,altdle P V~g~er and M W m l e n h e r g e r ,

,~ h'm ,~. ~ R~'~' ,,1 h;loll,, 74 ( I q 77) hhl - hh 7

J B (.'~ ~hen, ~'~ t ,,1/el.I/, 22 ( IqXg)q I ~-~ 'if,

F I Mnllemenler. Lnu Cheng. P .l van dee behaal. C M B r a k m ~ n and B M g,~revaar, ,,lh'lall ftmLs ,4, I~J,4 (IUXX)

q I() I 2 t 1,4 h 7 ,'4 q 2(I 21 22 2~

I q.-I Ill .I .%'l~t~ml~, [' I'~tt .4hpHHI~ / ['t~,~'~p~l=m~m m m~ ,.~l-I 'r~ ~ll~, 2,4 H I Aurq~n,,~n and C Laird. 'TmH~ .41,,111-'., 342 (I~~N)

1,4 t 7 - 1,4,4

2~ Y H Chun and R. H [)l~hcny.._%~r [ll¢'htfl, I I (Iq77) 72%- 72q

2t) P Murlc, F' Fi_~uqu¢'l and I Mcrhn, .%'~'t ,~h'tLtll, /"~ ( I q X l )

.~7~- ~7~

7 I Karlw arid W. Y Yl~udcll'.,, i~l~ll~'t .~_'m 1-~'~lm,I, 2(I~X~-p)

I I~t-I I,XX