:l tr`p zq lw`xtyt`x&wt st`x lwwzd^ · (fj dibjo h`o\ggpmbt9 j> h`^c\id^\g \ggjtdib9 mj...

$: :L TR`P ZQ LW`XTYT`X&WT ST`X LWWZd^ · (FJ dibjo hò\ggpmbt9 J> h`^c\id^\g \ggjtdib9 MJ kjr_`m hò\ggpmbt9 OPM m\kd_ njgd_dad^\odji kmj^`nndib, o>n jsd_`, q^fkfkd z 2 $ Kf() clo qoâfqflkî$
Fatigue of aluminium-lithium alloysK. T. Venkateswara Rao and R. O. Ritchie

Aluminium-lithium alloys are a class of lowdensity, high strength, high stiffness monolithicmetallic materials that have been identified asprime candidates for replacing 2000 and 7000series aluminium alloys currently used incommercial and military aircraft. In this review, thecyclic fatigue strength and fatigue crackpropagation characteristics of aluminium-lithiumalloys are reviewed in detail with emphasis on theunderlying micromechanisms associated withcrack advance and their implications to damagetolerant design and lifetime computations.Compared with traditional aerospace aluminiumalloys, results on the fatigue of binary AI-Li,experimental AI-Li-Cu, and near commercialAI-Li-Cu-Zr and AI-Li-Cu-Mg-Zr systemsindicate that alloying with Li degrades the low-cycle fatigue resistance, though high-cycle fatiguebehaviour remains comparable. The alloys,however, display superior (long crack) fatiguecrack growth properties, resulting from aprominent role of crack tip shielding, principallydue to deflected and tortuous crack pathmorphologies, induced by the shearable nature ofcoherent b' precipitates, crystallographic texture,and anisotropic grain structures. Environmentalfatigue resistance is comparable with 2000 seriesalloys and better than 7075-type alloys. Theaccelerated growth of small fatigue cracks, stronganisotropy, poor short-transverse properties, and asensitivity to compression overloads are theprincipal disadvantages of AI-Li alloys.

IMR/237

© 1992 The Institute of Materials and ASM International.Dr Venkateswara Rao and Professor Ritchie are with theCenter for Advanced Materials, Lawrence BerkeleyLaboratory, and the Department of Materials Science andMineral Engineering, University of California, Berkeley,CA,USA

IntroductionDesigners of modern commercial and military aero-space vehicles and space launch systems are con-stantly in search of new materials with lower density,and higher strength, stiffness, and stability at elevatedor cryogenic temperatures. To meet these challenges,much effort has been directed toward developingintermetallics, ceramics and composites as structuraland engine materials for future applications.1-6 How-ever, for structural airframes, age hardened 2000 and7000 series aluminium alloys have long been preferredfor civil and military aircraft by virtue of their highstrength/weight ratio, though the use of compositematerials, particularly for secondary structures, israpidly increasing. Nearly 750/0 of the structural

weight of the Boeing 757-200 aeroplane is comprisedof plates, sheets, extrusions, and forgings of 2024,2224,2324,7075, and 7150 aluminium alloys;7 typicalcomponents include, body frames, fuselage (frames,skins, and substructures), wings (stiffeners, ribs, string-ers, and skins), and horizontal stabilisers.

Today, conventional aluminium alloys face strongcompetition from emerging composite technologies,particularly in the structural aerospace market; hybridmaterials based on organic and metal matrices withfibre, whisker, or particulate ceramic reinforcementsoffer impressive combinations of strength, stiffness,and high temperature resistance.2~8-12 In addition,aramid polymer-reinforced aluminium alloy (ARALL)laminates,13-15 fabricated by resin bonding aramidfibres sandwiched between thin aluminium alloysheets, show exceptional promise as fatigue resistantmaterials. In the light of these advances, the alumin-ium industry has recently introduced a new generationof aluminium-lithium alloys, * by additionallyincorporating ultra-low density lithium intotraditional aluminium alloys; these alloys representa new class of light weight, high modulus, highstrength, monolithic structural materials, which arecost effective compared with the more expensivecomposites.16-28

Despite possible limitations regarding specificstiffness and high temperature stability, AI- Li alloysenjoy several advantages over composite materials.Economically, AI- Li alloys are typically only threetimes as expensive as conventional aluminium alloys,whereas competing hybrid materials can be up to10-30 times more expensive.28 Secondly, AI-Li alloyfabrication technology is generally compatible withexisting manufacturing methods such as extrusion,sheet forming, and forging to obtain finished products.Finally, they offer considerably higher ductility andfracture toughness properties compared with mostmetal matrix composites.9-12

Historically, efforts to design lithium containingaluminium alloys date from the early 1920s, thoughconcerns over their poor ductility and toughnesshalted production and use in the 1960s; Refs. 26, 27,29, and 30 provide an excellent review of thesedevelopments. Escalation in fuel costs during the1970s rekindled research on the AI- Li system withprospects of building more fuel efficient aircraft;design tradeoff studies had shown that reducingdensity was the optimal route.30-32 Each weightper cent of lithium added to aluminium lowers thedensity about 3% and improves the elastic modulusby nearly 6% for Li additions of up to 4%.21,33Accordingly, direct substitution of AI-Li alloys (con-

* Although this class of alloys contain alloying elements such ascopper, magnesium, and zirconium in addition to aluminium andlithium, the term aluminium-lithium (or Al- Li) alloys is generallyused to encompass all aluminium alloys containing greater than0·5 wt-% lithium.

International Materials Reviews 1992 Vol. 37 NO.4 153

154 Venkateswara Rao and Ritchie Fatigue of AI-Li alloys

Table 1 Chemical composition limits for major alloying elements in advanced aluminium-lithiumalloys and traditional aluminium alloys, wt-%

Alloy* li Cu Mg Zn Zr Mn ot Other AI

1M 2090 1,9-2'6 2·4-3'0 0·0-0·25 0'08-0'15 Bal.1M 2091 1·7-2,3 1'8-2·5 1·1-1,9 0·25 0·04-0·16 Sal.1M 8090 2·2-2·7 1'0-1'6 0'6-1'3 0·04-0·16 Bal.1M 8091 2·4-2·8 1·6-2·2 0·5-1·2 0·08-0'16 Bal.1M W049 1·3 5·0 0·4 0·1 Bal.

(Weldalite)1M 2020 1.1 0·5 0'2Cd Bal.MA IN-905XL 1·5 4·0 0·8 1·1 C Bal.RSP 644-B 3'0-3'2 0'8-1'1 0·4-0·6 0·4-0'6 Bal.

1M 2124 4·5 1·5 0·25 Bal.1M 7150 2'1 2·2 6·16 0·13 Bal.1M 7075 2·0 2·6 5·8 Bal.PM 7091 1·7 2·5 6·2 0·35 0'4Co Bal.

*IM ingot metallurgy; MA mechanical alloying; PM powder metallurgy; RSP rapid solidification processing.tAs oxide.

taining ~ 3% Li)* for traditional aluminium alloyscan yield ~ 11% weight savings; complete redesign ofaircraft utilising the full potential of Al- Li alloyscould save up to ~ 17% by weight. 30

Early studies by Sanders et al.21,34 on binary AI-Li alloys showed that problems of low ductility andtoughness could be traced to inhomogeneous slip andstrain localisation, resulting from coherent b' (AI3Li)particle hardening in the matrix, nucleation andgrowth of grain boundary b (AILi) precipitates, andthe formation of b'-precipitate free zones (PFZs).Subsequent developments in alloy design and pro-cessing have attempted to reduce this inhomogeneousdeformation mode by incorporating additionalalloying elements and modifying thermomechanicaltreatments. As a result, advanced commercial AI- Li-x-Y-Z alloys today exhibit strength-toughness com-binations which are comparable, and often superior,to traditional aluminium alloys 2024, 2124, 7075, 7150and 7475, particularly at cryogenic temperatures.35-42

Details of the principal alloys of commercial interest,their designations, and nominal chemical compositionlimits are summarised in Table 1.

Since advanced AI- Li alloys are principallyintended for safety critical structural aerospace appli-cations, their durability and damage tolerance per-formance are of considerable importance. Over thepast decade, a number of investigations have focusedon various aspects of fatigue crack initiation andcrack propagation in AI- Li alloys. It is therefore theobjective of the present paper to review these cyclicproperties with emphasis on the underlying micro-mechanisms; reviews on the crack growth mechanismsunder monotonic loading may be found in Refs. 43and 44. Crack initiation and crack growth behaviouris examined as a function of microstructural, mechan-ical, and environmental factors. It is concluded thatin general the fatigue behaviour of AI- Li alloys isnot significantly different from traditional aluminiumalloys, though crack propagation kinetics of longcracks are retarded by more pronounced crack tipshieldingt from such mechanisms as crack deflection,crack path tortuosity, and consequent roughness

* All compositions are in weight percent unless otherwise stated.t Crack tip shielding mechanisms act to impede crack advance bylowering the local stress intensity actually experienced at the cracktip.45.46 Such mechanisms, which act principally in the crack wake,

International Materials Reviews 1992 Vol. 37 NO.4

induced, crack closure mechanisms. In fact, theobserved variation in fatigue resistance of AI- Li alloyswith respect to microstructure, crack size, loadingsequence, and product form are largely a manifesta-tion of the degree to which shielding is promoted orrestricted in the different microstructures under vari-ous loading conditions.

~icrostructuralfeaturesStrengthening in AI-Li alloys is predominantly fromhardening by the nucleation and growth of one ormore second phase particle distributions, precipitatedfrom a supersaturated solid solution. This is achievedby aging naturally (at room temperature) or artificially(at an elevated temperature below the metastablesolvus line) following solution treatment and quench-ing the alloy from the single phase field. By controllingthe aging time and temperature, microstructures areoptimised for targeted mechanical properties in thevarious tempers. Generally, these temper designationsare referred to as (a) underaged (T3), synonymouswith low strength and fine uniform distributions ofcoherent hardening precipitates in the matrix, (b) peakaged (T6, T8), with maximum strength and partiallycoherent precipitates, and (c) overaged (T7), withslightly reduced strength and coarse incoherent pre-cipitates. Many published papers16-20,47-56 havereviewed extensively the solid state phase transform-ations and precipitation sequences in AI- Li alloys;salient microstructural features are summarisedbelow.

Binary AI-Li alloysBinary alloys of aluminium contaInIng less than~ l'7%Li are essentially random solid solutions ofaluminium and lithium. The elevated flow stress com-pared with pure aluminium is due to short rangeorder and dislocation interactions between soluteatoms and associated strain fields. For lithium con-tents greater than 1,7%, the alloys are hardened bylong range ordering through the precipitation of

include transformation and microcrack toughening in ceramics,crack bridging in composites, and crack closure during fatiguecrack growth. Closure mechanisms are described in detail in thesection on 'Fatigue crack propagation', below.

Venkateswara Rao and Ritchie Fatigue of AI-Li alloys 155

Table 2 Summary of chemical compositions, heat treatments, microstructures and mechanicalproperties of selected binary aluminium-lithium alloys (Ref. 57)

Tensile propertiest

Alloy, wt-%

AI-0.7Li

AI-2.5LiunderagedAI-2.5Lipeak aged

Heat treatment*

SHT+CWOnaturally agedSHT+CWOnaturally agedSHT+CWOaged 1 h at 473 K

Microstructural features

Li in solid solution with short range order

(j' (AI3Li) 2-5 mm in diameter

(j' (AI3Li) 15-20 nm in diameter, (j (AILi)

45

67

185

65

157

220

Elongation, %

26

33

2·6

*SHT solution heat treatment; CWO cold water quench.to"y and O"u are the yield and ultimate tensile strengths, respectively.

ordered (L12 structure), metastable, and spherical J'particles, in rx (fcc) AI- Li solid solution matrix. Owingto the low particle/matrix misfit strains ('" -0'120/0),the precipitates tend to remain coherent with thematrix and retain their spherical morphology, evenfor particle diameters as large as 300 nm.54 In thenaturally aged condition (or underaged tempers),microstructures show fine homogeneous distributionsof J' particles (2-5 nm in diameter) in the matrix andat grain boundaries. With increased aging time and/or temperature, the matrix J' precipitates coarsen,and equilibrium J particles nucleate heterogeneouslyalong grain boundaries, thereby resulting in theformation of Li-depleted J' -PFZs surrounding J.Improvements in strength in these alloys are primarilydue to the resistance of J' particles to dislocationmotion (from order hardening, elastic moduli differ-ences, and misfit coherency strains between the matrixand precipitates); the extent of hardening is pro-portional to both the volume fraction and size of J'precipitates. However, deformation in binary alloysis highly localised along narrow slip bands and withinsoft PFZs, which can promote premature cracknucleation, and contribute to poor ductility andtoughness of AI- Li alloys.34 Typical chemical com-positions, heat treatments, and mechanical propertiesof selected binary alloys are summarised in Table 2.

Ternary AI-Li-Cu and AI-Li-Mg alloysTo reduce the inhomogeneous mode of deformation,ternary elements have been incorporated in AI- Li

alloys to enable the nucleation of additional harden-ing precipitates and to modify the coherency strainsbetween J' particles and the matrix. For example,AI-Li alloys containing 2-3%Cu are strengthened bysemicoherent T1 (AI2CuLi) and f)' -like (AI2Cu) platesin the matrix. Precipitation of the equilibrium T2

(AI6CuLi3) phase is sometimes observed along grainand subgrain boundaries particularly in peak andoveraged tempers. Moreover, in addition to the com-monly observed spherical morphology, the orderedJ'-phase also forms surrounding the f)' plates. In AI-Li-Mg alloys containing less than 2%Mg, J' is theonly hardening phase; however, magnesium reducesthe solid solubility of lithium in aluminium and henceincreases the volume fraction of J' precipitates. Atlonger aging times and for alloys with Mg contentsgreater than 20/0, Al2MgLi precipitates in the matrixand along grain boundaries.

Commercial alloysCommercial AI- Li alloys are alloyed with Cu, Mg,and dispersoid forming elements such as Zr, Mn, andCr that form a fine dispersion of intermetallic phasesto reduce the tendency for localised deformation.Zirconium is generally preferred because it results infine, as cast grain structures and better corrosionproperties. In addition, to reduce the deleteriouseffects of J' -PFZs on ductility and toughness, wroughtAI- Li sheet and plate products are given a 2-6%permanent stretch before aging; this promotes uni-form intragranular precipitation and thus suppresses

Table 3 Experimental thermomechanical treatments utilised to obtain various microstructures incommercial aluminium-lithium and aluminium alloys (Refs. 16-20, 58, 59, 68)

Alloy

2090-T812091-T3512091-T88090-T3518090-T88091- T3518091-T8W049-T3W049-T82020-T651IN-905XL644-B2124-T3517150-T6517075-T651

7091-T7E69

Condition

Peak agedUnderagedPeak agedUnderagedPeak agedUnderagedPeak agedUnderagedPeak agedPeak agedPeak agedNear peak agedUnderagedPeak agedPeak agedOveraged

Heat treatment*

SH, CWO, 6% stretch, aged 24 h at 436 KSH, CWO, 2% stretch, naturally agedSH, CWO, 2% stretch, aged 10 h at 408 KSH, CWO, 3% stretch, naturally agedSH, CWO, 3% stretch, aged 16 h at 463 KSH, CWO, 3% stretch, naturally agedSH, CWO, 3% stretch, aged 16 h at 463 KSH, CWO, 3% stretch, naturally agedT3 condition + aged 24 h at 433 KSH, CWO, 1.5% stretch, aged 24 h at 422 KSH, CWO, aged 24 h at 443 KSH, HWO, aged 16 h at 405 KSH, CWO, 2% stretch, naturally agedSH, CWO, 2% stretch, aged 100 h at 394 KSH, CWO, 2% stretch, aged 24 h at 394 KSH, CWO, aged 24 h at 394 K + 4 h at 436 K

*ST solution treat; CWO cold water quench at 298 K; HWO hot water quench at 333 K.



a ingot metallurgy; b rapid solidification; c mechanical alloying; R/Dand E/D refer to the rolling and extrusion directions

1 Three dimensional optical micrographs ofgrain structures in commercial AI-Li alloys(Ref. 58)

heterogeneous grain boundary precipitationphenomena.

The resulting grain structures for most ingot metal-lurgy (IM)-based, high strength aluminium alloys


following such thermo mechanical processing are discshaped and elongated in the rolling direction; AI- Lialloy sheet and plate, however, show a greater degreeof anisotropy because of the small Zr additions whichcan effectively suppress recrystallisation and retardgrain growth. As shown in Fig. la, grain sizes in 1Malloys are fairly coarse, typically 500 Jlrn wide, 50 /lmthick, and several mm in length; the exception is the8091 alloy where grain sizes are finer. 59 Micro-structures in powder metallurgy (PM) processed AI-Li alloys are more refined; grain sizes are typicallybetween O'5 and 5 Jlm and the structures are lessanisotropic (Fig. lb,c). Strong deformation texturesare common in 1M alloys owing to prior thermomech-ani cal treatments. These textures are found to bepredominantly of the 'brass' type ({110}<112»), withevidence of weaker '5' ({123}<634») and 'copper'({112}<111») types.60-67 However, the various com-ponents of texture vary strongly across the thicknessof the plate with the intensity of each componentshowing a maximum at the centre;60 moreover,{001}(110) type recrystallisation textures areobserved in the near surface regions.63 Chemicalcompositions and thermomechanical treatments forthe principal commercial AI- Li alloys are listed,respectively, in Tables 2 and 3.

In underaged tempers, commercial AI-Li alloys arestrengthened by metastable {)' spheres, fJ' (AI3Zr, alsoreferred to as a') dispersoids, and composite precipi-tates of {)' surrounding /3' (Fig.2a). Al3Zr particlesare spherical, coherent, and ordered with an L 12superlattice structure. In addition, aging to peakstrength (T8 condition) in AI- Li-Cu-Zr (2090) sys-tems causes matrix precipitation of T1 and (}' plates,coarsening of matrix {)', the formation of relativelynarrow (-- 100 nm wide) {)'-PFZs at high angle grainboundaries, and fine precipitation of T1 plates alongsubgrain boundaries (Fig.2a,b). In Mg-containingalloys, T1 and 0' plates are replaced by S (AI2CuMg)laths in the matrix (Fig. 2c); precipitation of T1 occursin competition with S for available nucleation sitesand copper atoms. For example, precipitation of T1

plates has been reported for 8090 type alloys but notfor 8091 compositions.47 Despite the pre-aging defor-mation, grain boundary effects are prevalent, even inpeak aged tempers. In 8090- T8, small amounts ofgrain boundary precipitation result in '" 500 nm wide{)'-PFZs (Fig. 2d); in 8091- T8, heterogeneous precipi-tation is more extensive and PFZs are correspond-ingly wider ('" 1 Jlm).

With longer aging times (overaged tempers), coarseequilibrium Cu- and/or Mg-rich phases and T1 platesor T2 precipitates decorate grain and subgrain bound-aries, respectively, concurrently leading to the forma-tion of solute denuded PFZs. In addition, undissolvedimpurities during processing are also present as coarseFe- and Cu-rich intermetallic phases along the elon-gated high angle grain boundaries. Microstructuraldetails and room temperature mechanical propertiesof the principal 1M and PM AI- Li alloys qre summar-ised in Tables 4 and 5, respectively.

Specific AI- Li alloys can be '" 3qo/o higher instrength with similar elongation and toughness prop-erties when compared with traditional aluminiumalloys (Table 5). However, these properties are typical


a {/ (AI3Li) spheres and (,l'-like (AI2Cu) plates; b 7; (AI2CuLi) plates seen in AI-Li-Cu-Zr systems (2090-T81); c S (AI2CuMg) laths in Mg-bearing alloys (peak aged 8090 and 8091); d grain boundary precipitation with associated £5'-PFZs (8090- T8)

2 TEM micrographs showing predominant microstructral features and strengthening precipitates incommercial AI-Li alloys. Spherical P' (AI3Zr) dispersoids surrounded by 0', and 0' wetting the (}'platesmay also be noted in a; imaging in a, b was done under dark field conditions using 0' and T,superlattice reflections, repectively (Ref. 41)

of thermomechanically processed wrought productforms (namely prestretched and artificially aged sheetand plate) in the longitudinal orientation; generallyshort-transverse properties are considerably lessattractive. The unrecrystallised elongated grain struc-tures and deformation-texture variations ensure that

properties are strongly dependent on orientation,both in the rolling plane and through the platethickness.41,6s-67,70 Moreover, the excellent strengthand toughness properties, characteristic of sheet andplate products, are often compromised for forgings,extrusions, and thick section plates, where uniform

Table 4 Summary of grain size measurements and strengthening precipitates invarious advanced and conventional aerospace aluminium alloys (Refs.16-20,58, 59, 68)

Typical grain size*

Alloy L,mm T,llm S,llm Principal hardening precipitatest

2090- T81 2-3 500 50 <5'(AI3Li), 7; (AI2CuLi), (,l'(AI2Cu)P'(AI3Zr)

2091-T351 1-2 600 50 b', P'2091-T8 b'. P'8090-T351 1-2 350 40 b', P'8090-T8 b', 7;, S (AI2CuMg), P'8091-T351 0.3 65 25 b', P'8091-T8 <5',5, P'W049-T3 G-P zones, b'W049-T8 <5', T" e'2020-T651 O·g 75 35 <5',7;, e', T B

IN-905XL 0·0015 004 0·3 AI203, AI4 C3

644-B 0-05-0'1 1-2 1-2 <5', P', composite 1>' - P'2124-T351 0·7 350 50 G-P zones7150-T651 2-3 750 30 1]' (MgZn2-Mg(CuAI)2)7075-T651 0-125 16 10 G-P zones, 1]', AI'2Mg2Cr7091-T7E69 2-5 1-2 1-2 G-P zones, 1]', Co2AIs, AI2CuMg, oxides

*L, T, and S refer to measurements in the longitudinal, long-transverse, and short-transverse directions, respectively.tRefer to text for details on morphology and structure of the different phases.



quenching rates and pre-aging deformation are notfeasible. 65-67,70-75

Cyclic stress-strain and fatiguecrack initiation behaviourThe application of cyclic loads to metallic materialsat stress levels much below the yield or tensilestrengths can significantly alter their constitutivebehaviour. Depending on the initial wrought productform, heat treatment, and applied plastic-strain ampli-tude (~Bp/2), the material may undergo cyclic soften-ing, cyclic hardening, show mixed (hard;~ing andsoftening) response, or have no effect at all. Follow-ing transient hardening or softening, materials gener-ally exhibit saturation or stability at some equilibriumstress or strain value, before eventually softening dueto incipient damage or crack nucleation. Highstrength aluminium alloys are good examples of cyclichardening materials, whereas softening is observed inquenched and tempered steels; low carbon steels, highstrength low alloy (HSLA) steels, and titanium alloysare examples of materials which initially soften beforehardening.76 The cyclic stress-strain properties aretraditionally evaluated either in low cycle fatigue(LCF), where the strains are predominantly plastic,or in high cycle fatigue (HCF), where the macroscopicstrains are predominantly elastic. Typically, lifetimesunder LCF conditions are below", 103 cycles, whereasRCF pertains to lifetimes exceeding '" 103 cycles.

Low cycle fatigue behaviourAs with most high strength Al alloys, AI- Li alloys ingeneral initially harden under cyclic loading, particu-larly at low plastic-strain amplitudes (typically ~Bp/2below '" 10 - 3), followed by saturation before soften-ing to final fracture.57,77-79 At higher plastic strains(~B /2 above '" 10 - 3), however, the alloys continu-ously soften to failure with little cyclic stabil-ity. 57,79-81 Similar trends are apparent for mostexperimental and commercial AI- Li alloys, though

the extent of hardening, saturation, or softening andresultant fatigue lifetimes are a strong function ofalloy composition, microstructure, temperature, andenvironment. 34,57,77-94 Typical variations in the cyc-lic stress amplitude with applied strain, under fullyreversed, strain controlled loading for binary AI- Lialloys and commercial 8090- T6 alloy, are illustratedin Fig. 3; corresponding monotonic and stabilisedcyclic stress-strain curves are compared in Fig. 4.

Binary AI-Li alloysStrengthening by lithium in solid solution ('" 1'7%Li)yields microstructures which initially cyclically hardenat plastic strain amplitudes less than '" 10 - 3, beforesoftening and eventually attaining cyclic stability untilfracture. 57,78,84,85 Hardening can be permanent foralternating strains greater than'" 10-3 with no satu-ration before softening57 (Fig. 3); corresponding cyclicstress-strain curves show continuous hardening tofailure without any plateau region (Fig. 4a). Theincreased flow stress during fatigue can be attributedto an increase in the dislocation density, dislocation-dislocation interactions, and dislocations interactingwith solute-atom strain fields.77 Transmission electronmicroscopy (TEM) observations reveal the defor-mation to be largely homogeneous.34,57 Cellular dis-location structures, similar to those observed in singlephase, high stacking-fault energy fcc metals, are com-mon particularly at high plastic-strain amplitudes(Fig. 5a), though they are slightly more elongated andless polygonised compared with pure aluminium(wavy slip); no veins or ladder-like persistent slipbands (PSBs) are seen. At low plastic-strain ampli-tudes, slip is planar.34 With increased deformation,softening is associated with the nucleation of micro-cracks along slip band intrusions emerging on thesurface for ~Bp/2 < 10- 3, and along grain boundariesor where slip bands impinge at grain boundaries for~Bp/2 >10- 3. Propagation of the fatal crack occursby non-crystallographic, ductile striation type, crackgrowth mechanisms (Fig. 5b) that involve alternatingshear or successive blunting and resharpening of thecrack tip. 34,57,78

Table 5 Room temperature mechanical properties of selected experimental and commercialaluminium-lithium and aluminium alloys*

Young's Yieldmodulus strength UTS Elongation

Alloy E, GN m-2 ay, MN m-2 au, MN m-2 (on 25 mm gauge), %

2090-T8 76 552 589 112091-T351 75 369 451 102091-T8 75 425 481 88090-T351 77 226 352 178090-T8 77 482 534 68091-T351 309 417 118091-T8 537 581 6W049-T3 79 379 496 15W049-T8 79 683 703 52020-T651 77 534 567 5IN-905XL 559 596 2·3644-8 422 539 8

2124-T351 73 360 488 187150-T651 73 404 480 67075-T651 73 503 572 117091-T7E69 545 593 11

Strainhardeningcoefficient n

0'060·120·100'190·080'160·07

0'19

FracturetoughnessK,c' MN m-3/2

3633462736382037

211324

24212846

*Uniaxial tensile properties are obtained in the longitudinal' (L) direction and fracture toughness in the long-transverse (L- T) orientation. Valuestaken from Refs. 16-20, 58, 59, 68, 69.



o101 102 103 104 105 106 107

NUMBER OF REVERSALS3 Variation in cyclic stress amplitude with

number of reversals at various plastic-strainamplitudes in a AI-0·7Li solid solution, under-aged (UA) and peak aged (PA) AI-2·5Li alloy(precipitation hardened), and commercial8090- T6 alloy (Refs. 57079)

In contrast, aged microstructures with lithium con-tents exceeding '" 1'7% that are precipitation hard-ened by b' particles (e.g. AI-2·5Li alloy), showsignificant cyclic hardening (Figs. 3, 4a). Cyclic stress-strain curves show an initial increase in stress ampli-tude due to dislocation interactions, with small sat-uration periods (which diminish with artificial aging),followed by softening to failure from b' shearing andstrain localisation.34 Unlike dislocation cell structuresin solid solution hardened microstructures, fatiguedeformation is inhomogeneous and slip is essentiallyplanar for all plastic strain amplitudes;34,57 intenseslip bands are developed parallel to {Ill} planes(Fig.5c).

However, the strong cyclic hardening behaviour ofAI- Li alloys with higher Li content and higher b'volume fractions is offset by marked reductions inLCF resistance, as shown by the Coffin-Mansoncurves in Fig. 6a. Although solid solution hardenedand underaged microstructures experience only lim-ited losses in fatigue strength compared with purealuminium (an exception is underaged alloys at lowplastic strains where slip is planar), fatigue lives arereduced by as much as four orders of magnitude inartificially peak aged and overaged binary AI-Lialloys. 34,57,78,80-86

The detrimental effects of Li on the LCF resistanceof AI- Li alloys is principally due to the presence oflarge volume fractions of shearable, ordered b' precipi-tates; small Li additions which go into solid solutionhave almost no effect. 34,57During deformation, initialdislocation motion shears the b' precipitates andreduces order by creating antiphase domain bound-aries within the particles; following dislocations pref-erentially move on the same glide plane at a lowerresolved stress to restore order, thus resulting in pair-dislocation motion and a tendency toward planarslip. Additionally, planar slip is promoted by thereduced effective diameter of sheared b' particles andreductions in the stacking fault energy of aluminiumdue to Li additions.34 So deformation is local-ised along narrow PSBs parallel to the {Ill} glide

---

Binary AI-Li Alloys

R=-1 Loading

(b)

Commercial AI-Li AlloysR = -1 Loading

2090- T8l 8090- T8o monotonic

cyclic

Peak-aged AI-2.5Li

(cyclic) \

~

AI-O.7Li(mOnotoniC)

AI-O.7Li (cyclic)

\ "-- Pure AI (cyclic) ( a )

100

500

(\/

IE 400z2btf)' 300tf)wa:::II; 200

oo 0·01 0·02 0·03

TOTAL STRAIN,Ea binary microstructures; b commercial 2090- T81 and 8090- T8alloys

4 Monotonic and stabilised cyclic stress-straincurves for various AI-Li alloys (longitudinaldirection, R= -1) (Refs. 57, 70, 79)

800

10-4 10':'3 10-2

PLASTIC STRAIN AMPLITUDE, ~2ep

C}l

E~ 200

~I(\/

~ 150=>.....:i~ 100<{tf)tf)~ 50til

250

planes causing stress concentrations at their intersec-tion with grain boundaries. These effects become morepronounced in the peak and overaged tempersbecause of an increase in size and amount of b'particles;34,57 equilibrium b precipitates further accen-tuate this effect by aggravating the degree of strainlocalisation in the soft b' -PFZs at grain boundaries.Microcrack nucleation at dislocation pile-ups is com-mon for all plastic-strain amplitudes; coalescence ofthese microcracks to form a fatal flaw may occurcrystallographically along slip bands or at weakenedgrain boundaries, depending on the alloy temper. Asshown in Fig.5d, fatigue fracture in underagedAI-2·5Li is transgranular as microcracks link up byslip band cracking.34,57 Conversely, intergranularcracking is prevalent in the peak and overaged micro-structures; few secondary cracks initiate, and the firstintergranular crack to nucleate propagates cata-strophically to failure.

Commercial AI-Li alloysIn commercial AI- Li alloys, cyclic and monotonicflow stresses are substantially higher, and fatigue livescorrespondingly lower, compared with binary AI-Lialloys because of additional strengthening from T1


---_x '--W-W-__ l_:_,

l!.£p/2'" 8090- T6. 5.4 xlO-3v 8090- T6. 3.0 xlO-4

o AI-2.5Li PA. 5.0 xlO-4

CJ AI-2.5Li PA. 7.5 xlO-6

• AI-2.5Li UA. 5.5 xlO-3• AI-2.5Li UA. 6.5 xlO-6

+ AI-0.7Li. 1.2 xlO-3

x AI-0.7Li. 1.0 xlO-4__ -v 7 ~

-----

---~~----~ ~

--~ --~

C\I 500IEz2,400

~I(\/

w'Q~ 300:Ja..

~ 200tf)tf)w~ 100tf)


5 TEM micrographs of dislocation substructures generated during cyclic deformation a,e, andcorresponding SEM images of fatigue fracture surfaces b,d, in binary AI-Li alloys hardened by: a,bLi in solid solution (underaged AI-O·7Li alloy), and c,d ordered ~' (AI3Li) precipates (peak agedAI-2·5Li). Failures in solid solution occur by ductile striated type mechanism corresponding tohomogeneous, cellular deformation structures, whereas ~'-hardened structures fracture by slip bandcracking due to localised deformation along close packed {111} planes (Refs. 34, 57)

and S precipitates (Figs. 4b, 6b). T1 and S precipitatesare thought to favour a dispersion of slip and promoteplastic strain homogenisation at the subgrain levelthrough the activation of multiple slip systems withinthe grain. However, inhomogeneous fatigue defor-mation still occurs along PSBs or embrittled grainboundaries. In fact, the partially coherent S precipi-tates have been observed to shear during fatigue athigh plastic strains, presumably due to stress con-centrations at the matrix/S interfaces that favourdislocation nucleation and precipitate cleavage.88

Moreover, the highly textured grain structures incommercial alloys limit the number of slip systemsby reducing the effectiveness of grain boundaries asbarriers to dislocation motion due to the low degreeof misorientation. 78,79,81,86-91

Cyclic stress-strain behaviour in commercial alloysis similar to that in binary alloys strengthened by {)'particles (Figs. 3, 4b). Limited hardening and stability,followed by softening, is seen at low strain amplitudesdue to crack nucleation at slip bands emanating fromlarge ('" 1-10 )lm) cracked or uncracked constituentparticles; at high strains, the alloys soften continu-ously to failure. Crack nucleation occurs in coarseslip bands, possibly at intrusions or extrusions on the


specimen surface; hardening and softening are gener-ally associated with dislocation-particle interactionsand {)' shear, respectively.89,90 However, Coffin-Manson curves for commercial alloys show a trans-ition at low strain amplitudes below '" 10-3, whichapproximately corresponds to strains below whichthe cyclic stress response is stable (harden) and abovewhich the alloys soften (Fig. 6b). Most studies associ-ate this transition with changes in deformation mech-anism, or the degree of slip homogeneity.79,89,90Such changes have also been noted in 8090- T8 and2020- T651 AI- Li alloys; ductile intergranular frac-tures, characteristic of monotonic failure, are preva-lent at high strain amplitudes above the transitionstrain, whereas transgranular shear (slip band crack-ing) failures are observed below the transition strain,suggesting marked effects of slip planarity at lowstrain ranges. 78,79,81,86-88

Microstructural factors that enhance slip homogen-eity in AI- Li alloys are generally found to improveLCF resistance. For example, stretching (before aging)to increase the number density of the partially coher-ent T1 and S phases, the addition of disperoid formingelements, and fine recrystallised grain structures, areall considered beneficial to LCF.78,80,89-91 In con-


FATIGUE 150STRESS,MN m-2

7 Illustration of fatigue stress instabilitiesobserved on stabilised cyclic stress-strainhysteresis loop of AI-2·5Li alloy, aged for 5 hat 423 K. Results are for plastic strainamplitude ASp/2 = 2 x 10-3 (0'01 Hz) at roomtemperature (Ref. 92)

seen at specific temperatures above 233 K for AI- Limicrostructures hardened by [)' (as opposed to solidsolution strengthening), when accumulated plasticstrains (2N x l1epj2, where N is the number of cycles)exceed 1'-10·1 (independent of alloy composition andaging temper), and at frequencies below 0·2 Hz. More-over, increased cyclic flow strengths have been meas-ured by interrupting the test and room temperatureaging the material.

Gentzbittel and Fougeres92 suggest that this behav-iour is influenced by dynamic strain aging or thePortevin- Le Chatelier phenomenon; however, as thestress instabilities are not seen in AI- Li solid solutions(i.e. in AI-0'7Li alloy), this implies that the effect isassociated with [)' precipitates rather than dislo-cation-lithium (solute) atom interactions. It is moreprobable that the effects originate from the instabilityof small [)' particles (~ 1·5 nm) to localised shearingby moving dislocations; mechanical instabilities in thefatigue stress are then a result of successive reversionand reprecipitation of [)' in the matrix. In fact, TEMstudies by Brechet et al.93 have confirmed this obser-vation. In the early stages of deformation, spherical[)' particles are sheared by dislocations lead.ing to theformation of antiphase domain boundaries (Fig. 8a).On repeated shearing, the [)' precipitates becomeunstable from the antiphase boundary energy anddissolve in the matrix. Since deformation is concen-trated along narrow bands parallel to the {Ill} slipplanes, the resulting microstructure (Fig. 8b) consistsof precipitate free bands enriched in lithium, wheredislocation motion is relatively unimpeded comparedwith microstructure outside the bands. Finally,assisted by fatigue induced vacancies, lithium diffusesto the side of the band and reprecipitates as [)', thusenhancing the local coarsening kinetics and volumefraction of [)' particles (snow plough effect). Simplemodels proposed to explain the redissolution of [)'within the bands, band width, and band separation,based on an increase in the free energy from shearinduced, antiphase boundaries, are in reasonableagreement with experimental results.93,94

Although the precise relevance of such microscopicfatigue damage processes to macroscopic crack


(b)

o AI-O.7 Li• AI-2.5 Li Underagedo AI-2.5 Li aged lh at 473 K

• AI-Li 8090-T851o AI-Li 2020-T651o AI-2024-T4• AI-7075-T6v AI-RR58-TNH

4A

101 102 103 104

NUMBER OF REVERSALS TO FAILURE, 2Nf

a binary alloys; b commercial 8090 and 2020 alloys

6 Comparison of Coffin-Manson relationshipsfor different binary and commercial AI-Lialloys, compared with pure aluminium andtraditional 2000 and 7000 series AI alloys in thelongitudinal orientation. Note that Li presentas h' particles, is detrimental to LCF resistance.Data taken from Refs. 34, 57. 79

(a)

102 103 104 105 106 107 108

NUMBER OF REVERSALS TO FAILURE, 2Nf

trast, results in 2020 suggest that unrecrystallisedmicrostructures may exhibit better fatigue resistancethan recrystallised structures;80 this effect, however,seems to be associated with the suppression of grainboundary precipitation by the pre-aging deformation.

Finally, compared with traditional high srengthalloys, commercial AI- Li alloys possess comparablefatigue crack initiation properties (Fig 6b). In the highstrain region of the Coffin-Manson curve, however,crack initiation resistance may be somewhat lower inAI- Li alloys because of higher elastic cyclic stressesfor a given cyclic strain, resulting from their increasedYoung's modulus.70

Stress instabilities during cyclic hardeningRecent studies92-94 have shown that following cyclichardening, stabilised cyclic stress-strain curves (orhysteresis loops) for binary AI-Li alloys often exhibitstress instabilities, characterised by successive dropsin the fatigue stress (Fig. 7), depending on the plastic-strain amplitude, alloy temper, test frequency, andtemperature. For example, the instabilities are only


o104 105 106 107 108

NUMBER OF CYCLES TO FAILURE, Nf

9 High cycle fatigue curves for a binary AI-1·8Lialloy, following various thermomechanicaltreatments compared with pure aluminium.Results shown are for smooth, 2·3 mm thin-sheet specimens, tested under fully reversed(R= -1) bending fatigue parallel to rollingdirection, taken from Ref. 95. ST and CR referto solution treatment and 670/0cold rolling,respectively

Pure AI

Binary AI-1.8Li AlloyR=-1

~ ST+CR+Aged 167 h at 423 K

o",---·~ST + Aged 500 h at 423 K

150

C\J

IE 120zz:bv>' 90V>W0:::t-V> 60z::Jz:X 30«z:

Artificial aging further increases strength and inducesgreater inhomogeneity in fatigue deformation withincreasing 6'-particle size and distribution. Planardislocation arrays are seen in TEM studies along< 110> in the solution treated condition; similar fea-tures are evident in artificially aged tempers at lowcyclic stresses, and dislocation cell structures areevident at high stress levels. After thermomechanicaltreatments, AI- Li alloys show greater improvementsin fatigue strength because of higher monotonicstrength, coupled with a cell structure that alleviatesslip localisation and retards crack initiation.95

Smooth- and notched-specimen stress-lifetime(S-N) fatigue data (at R=O'I) for commercial AI-Lialloys 2020- T651, 2090- T81, 2091- T8, and 8090 arecomparable with results for 7075- T6 and 2124-T35198-101 (Fig. 10). Typical endurance strengths ofthe commercial alloys are roughly 40-50% of thetensile strength; stretching before aging can lead tofurther improvements.95,98 However, results on 2091-T8 alloy99 suggest that undissolved intermetallic con-stituent particles (rich in Cu, Fe, and Mg) can alsoinfluence RCF as they are preferential nucleation sitesfor microcracks. Moreover, the HCF properties ofcommercial AI- Li alloys are highly anisotropicbecause of marked deformation textures and discshaped grain structure that result in strength vari-ations with specimen orientation; in 2090- T83 AI- Lisheet, for example, S- N fatigue lives for the L + 45°orientation are significantly lower than for thelongitudinal (L) or long-transverse (L-T) orienta-tions, particularly in smooth as opposed to notchedsamples.70

Effect of surface treatmentSurface treatments using shot peening, ion implan-tation, nitriding, and laser glazing techniques arecommonly employed to improve fatigue crackinitiation resistance through the introduction of com-pressive residual stresses at the surface; work

1 [111]

8 Dark field TEM micrographs showinga shearing of ordered (j' particles during earlystages of cyclic deformation (as seen in peakaged 8090 AI-Li alloy, A8p/2 = 0·50/0, fromRef. 88) and b fatigue induced, (j' precipitatefree bands parallel to {111} slip bands in binaryAI-Li alloys (AI-2·5Li, aged for 8 h at 373 K;mean diameter of (j' particles '" 2 nm, fromRef. 93). Imaging performed using (j'-

superlattice reflections

High cycle or S-NfatigueIn contrast to LCF behaviour where Li in the formof 6' precipitates has an adverse effect on fatigue life,the high cycle fatigue properties of AI- Li alloys areexcellent.95-98 Compared with pure aluminium, thefatigue resistance of binary AI- Li alloys increaseswith Li content and aging time (Fig. 9), primarily dueto strengthening by Li in solid solution, and thegrowth of ordered 6' precipitates; cold work beforeartificial aging further enhances this effect. Similarresults have been reported for binary Mg-Li alloys,96where increases in fatigue strength and tensile strengthare again related.

Although the global strains that cause RCF failuresare nominally elastic, plastic strains are experiencedlocally at microstructural heterogeneities. Metallo-graphic observations on as quenched AI-I·8Li alloyshow very fine surface slip bands distributed homo-geneously within the grains during the early stages ofcycling at high stress levels; these coarsen and assistmicrocrack nucleation at sharp slip bands or grainboundaries before finally propagating to failure.95


advance mechanisms is as yet unclear, these obser-vations do confirm that cyclic deformation in AI- Lialloys proceeds by dislocation shearing of the ordered6' precipitates. Similar studies on the LCF behaviourof 8090 and 2020 AI- Li alloys have also shown that,at high plastic-strain levels, spherical 6', lath-like Sprecipitates and even Al3Zr dispersoids are shearableand may contribute to low fatigue lives.87-89


o103 104 105 106 107 108 109

NUMBER OF CYCLES TO FAILURE I Nt

10 Comparison of a smooth unnotched (stressconcentration factor, kt = 1), and b notched(kt= 3) fatigue crack initiation resistanceunder axial loading conditions (R= 0·1) forvarious commercial AI-Li and conventionalAI alloys. HCF resistance of AI-Li alloys iscomparable with 2000 and 7000 series AIalloys. Results compiled from Refs. 100, 101

hardening effects also play a minor role. Studies onpeak aged 8090102-104 have confirmed that the RCFresistance of AI- Li alloys can be improved using suchtechniques. However, crack initiation lifetimes(defined as the number of cycles required to grow a100 J-lmcrack) can be shorter in shot peened comparedwith untreated specimens, because of early cracknucleation at stress concentrations or fold like defectsgenerated by the peening; subsequent crack propa-gation is retarded such that total fatigue lifetimes areimproved.

Other treatments, such as acid pickling and anodis-ing (used as precursors to adhesive bonding and forcorrosion protection) can lead to reduced fatigue lifedue to premature crack nucleation from preferentialgrain boundary attack and the rupture of surfaceoxide films. Recent results102,l04 indicate, however,that 8090- T8 is less susceptible to problems fromchromic-sulphuric acid pickling than AI-Zn-Mg-Cualloys, because the reduced grain boundary precipi-tation decreases the electrochemical driving force fordissolution and crack formation. Moreover, unlikeother aluminium alloys, the fatigue strength can berestored by reanodising the pickled surfaces; anincreased fracture strain of the lithium bearing surface

Band for 7075- T6

oxide films is presumed to be the principal cause forthis behaviour.l02

Fatigue crack propagationIn recent years, damage tolerant design procedureshave become increasingly important in predicting lifeand ensuring the durability of many safety criticalstructures; the concept presumes that defect popu-lations pre-exist in components and that structuralintegrity is a function of the number of cycles topropagate the largest undetected flaw to fail-ure.10S-107 Lifetfines are conservatively predicted byintegration of a relationship describing the crackgrowth rate (da/dN) as a function of the mechanicalcrack driving force, from initial to final (critical) cracksize. For nominally linear elastic conditions, wherethe driving force can be defined in terms of the appliedstress intensity range, ~K = Kmax - Kmin (Kmax andKmin are the maximum and minimum stress intensit-ies), these relationships are generally based on theParis power-law relation, namely

da/dN = C~Km . (1)

where C and m are experimentally determined scalingconstants. Although equation (1) provides a reason-able description for crack growth rates in a rangetypically between "" 10 - 9 and 10 - 6 m/cycle, it doesnot adequately characterise behaviour at ~K levelsapproaching the fatigue threshold, ~KTH, below whichlong cracks are presumed dormant, or close to insta-bility, where limit load or catastrophic failure (e.g. atKmax = K1c, the fracture toughness) occurs; in theseinstances, more complex equations are curve fitted tothe crack growth data. Such crack propagation 'laws'are generally determined under constant amplitudeloading conditions, using fracture mechanics typespecimen geometries containing pre-existing, long(> 5 mm) through thickness cracks, via manual orcomputer-controlled load shedding schemes whichreduce ~K levels at a fixed load ratio (R = Kmin/ KmaJ.

In many materials crack propagation behaviour isstrongly affected by ·crack closure, paricularly at nearthreshold levels;4S,l08-114 in fact, significant effects ofmicrostructure, load ratio, and even environmenthave been attributed to variations in crack closurelevels. The closure phenomenon is a consequence ofthe crack not being fully open during the entireloading cycle, even under tension-tension loadingconditions, such that premature contact of the cracksurfaces occurs at the closure stress intensity, Kcbbefore minimum load is reached. Prominent mechan-isms include closure caused by residual plastic defor-mation - plasticity induced crack closure, interferenceof fracture surface asperities between mating fracturesurfaces in the crack wake - roughness induced crackclosure, and wedging of the crack from corrosion oroxide deposits on the fracture surface - oxide inducedcrack closure.108-112 The salient mechanisms perti-nent to high strength aluminium alloys and theirinfluence on crack growth rates are illustrated inFig. 11; measurement techniques are discussed in Refs.113,114.

Similar to other crack tip shielding mechan-isms,4s,46 the role of closure is to perturb the near- ,

International Materials Reviews 1992 Vol. 37 No.4

......

Longitudinal OrientationR' = 0.1, kt = 1.0

• SOgO-Medium strengtho 2020-T651• 2090-T81

Longitudinal OrientationR = 0.1, ~ = 2.5 - 3.0

• B090-Medium strengtho 2020-T651• 2090-T81

Band for 7075- T6

o103

300

600

C\I

'E500z~400b(/)

~3000::In2: 200::>2:~ 1002

~ 250Ez2b'200

If)(/)

~ 150t-(/)

2100::>2X<{ 502


I. CRACK DEFLECTION AND MEANDERING =====.::=:::'/~ ~~fi/4 K reduced. R con.llIllI 'U ~

AK

2. CONTACT SHIELDING

- wedoino:

corrosion debris-induced crock closure " • Jii *?crock surface rouOhness-induced closure ~

!V\f\~f/I4K reduced, Rinc:rea•••• 'ULR-

AK

- plasticity - induced crock closure

3. COMBINED ZONE AND CONTACT SHIELDING

11

IYJl)tlLAK

Schematic illustration of crack tip shielding mechanisms influencing fatigue crack propagation inhigh strength AI alloys, showing primary mechanisms and their consequences on crack growthrate behaviour (Ref.115)

tip stress or strain field by lowering the local drivingforce actually experienced at the crack tip. Thisdriving force can be defined as45

L1Keff = Kmax- KcI (KcI> Kmin)

L1Keff = L1K (Kcl < Kmin) • (2)

Where growth rates are correlated in terms of L1Keff,

the crack growth relationships have been termedintrinsic, i.e. devoid of mechanical crack-closureeffects. Alternatively, the intrinsic fatigue crack propa-gation resistance can be directly assessed using con-stant Kmax/increasing Kmin (variable R) load sheddingschemes which restrict closure effects.116

As noted previously, localised planar slip defor-mation in 6' -hardened AI- Li alloys has adverse effectson (LCF) crack initiation resistance; however, para-doxically, it can be beneficial to crack propagationresistance by inducing crystallographically tortuouscrack paths. Such geometrical changes in crack pathmorphology promote crack tip shielding, in particularfrom the wedging of fracture surface asperities, andretard crack ad vance.

Behaviour in experimental AI-Li alloysand 2020Early studies of fatigue-crack propagation in AI- Lialloys were performed on AI-3Li-l ,25Mn systems 77and peak aged 2020- T651101 in moist air, where itwas found that crack growth rates, between 10-9 and10-5 m/cycle, were significantly slower than in 7075-T6, particularly at low stress intensity levels. Suchobservations were attributed to the higher elasticmodulus of AI- Li alloys, which results in lower cracktip opening displacement (CTOD) at fixed L1K andconsequently lower crack growth rates per cycle,though at higher ~K levels approaching instability,growth rates were faster in the AI-Li alloys. Similarbehaviour was noted for AI- Li- Mg alloys 117 and


rapid solidification processed (RSP) AI- Li-Cu- Mnalloys118 and the improved fatigue properties wereassociated with slip band cracking; enhanced slipreversibility because of the planarity of slip from 6'-particle hardening was reasoned to diminish the cracktip damage (irreversible cyclic plastic strain) accumu-lated each cycle. 118Residual stresses developed duringquenching after solution treatment were also foundto have a significant influence on crack growth inRSP materials.

More recently, with the renewed interest in AI-Lisystems, the fatigue behaviour of peak aged 2020alloy has been re-examined over a wider range ofgrowth rates between 10-11 and 10-6 m/cycle.119Crack propagation data under constant amplitude(variable L1K/constant R = O'33) and spectrum loading(~ingle tensile overload every 8000 cycles with KoL/

Kmax= 1'8) are compared with corresponding resultsfor 7075- T651 and 7091- T7E69 in Fig. 12 (for anambient moist air environment). As reported in earlierstudies,101 2020 exhibits significantly slower crackgrowth rates, by more than two orders of magnitudeat near-threshold levels; however, this cannot beexplained purely on modulus differences since theimproved crack growth properties of 2020 are appar-ent even after normalising growth rate data withYoung's modulus.119 Of more importance here is thefact that crack advance in 2020 is highly crystallo-graphic at all L1K levels with substantial crack deflec-tion and branching; consequently, fracture surfacesare extremely rough and faceted, with little evidenceof ductile striations. In addition, oxide deposits havebeen detected on these surfaces, with excess oxidethicknesses estimated from secondary ion mass spec-troscopy to be between f<V 5 and 10 nm.

Based on these observations, the superior fatiguecrack propagation kinetics in AI- Li alloys, specifically2020, are attributed to the highly non-linear andtortuous crack path morphologies. This follows from


AKTH

~K,ksi \lin5

.,/~'~~o "••. " " .•••

,,/'" ~I'

/~;' ~,/'b "'..... ,I'...~... ,--

" /o " /., ,., ,, .! !I 'I' ,, '.' R = 0.05 0.5

o AI-l.1Li-4.6Cuo AI-2.9Li-1.1Cu

long Cracks in Vacuumv = 25 Hz

1(J123 5 10 15

STRESS INTENSITY RANGE, ~K, MN ni3/2

13 Growth rate behaviour of long fatigue cracksin experimental AI-li-Cu-Zr alloys underconstant amplitude loading for R= 0·05 and0·5 in vacuo at room temperature. Note thatincreasing li/Cu ratio leads to slower crackvelocities, particularly at low R (Ref. 124)

reversion and reprecipitation studies44 indicate thatT1 plates inhibit planar slip induced crack deflection,thereby resulting in faster crack growth rates. Growthrate behaviour was also found to be sensitive to loadratio; alloys with higher Li content showed thestrongest effect. 124

Behaviour in advanced commercialAI-Li alloysThe design of modern AI- Li alloys has principallyfocused on homogenising planar-slip deformation byadding solute elements that form incoherent precipi-tates and dispersoids; the rationale is to achieveacceptable ductility and toughness. The resultingmicrostructures are often unrecrystallised and highlytextured with a laminated grain structure. Such micro-structures can lead to exceptional fatigue crack pro-pagation resistance in specific orientations, yetconcurrently often result in poor short-transversetoughness and fatigue properties.68,70,115,123,128-138

Crack growth rate results, as a function of 11K, areplotted in Fig. 14a for commercial 1M AI-Li alloys,namely peak aged 2090, 8090, 8091,2091, and Welda-lite (L- T orientation, R = 0'1), and are compared withdata on 2124- T351 and 7150- T651.68,123 Corres-ponding crack closure levels, measured using back-face strain compliance methods and plotted as Kclvalues normalised by Kmax, are shown in Fig. 14b. Aswith experimental AI-Li-Cu-Zr alloys, the commer-cial alloys exhibit consiste~t1y slower crack velocitiesover the entire spectrum of growth rates (exceptcompared with 2124- T351 at near-threshold levels).As shown in Fig. 14b, this can be 'attributed primarilyto their higher crack closure levels which, unliketraditional 2124 and 7150 alloys, remain significantat high stress intensities, e.g. values of Kcl in 2090-T81 are 90% of Kmax (Kcl '" 2-3 MN m -3/2) at thethreshold I1KTH, and remain over 50% of Kmax(Kcl'" 3-4 MN m - 3/2)at 11K levels as high as 7-8 MNm -3/2(Ref. 123).

Such high crack closure levels can be associatedwith highly tortuous and deflected crack paths,

International Materials Reviews 1992 Vol. 37 No.4

.~

20

1 latticespacing 10-8per cycle

..

.....•.....-....... ........-...

7075-T651 Plate

• I·. .• .: •••-se. • 1 lattice

•• ~.... spacing•••• : per cycle..

~K,ksi \lin5

Periodic Tensile-Overload Spectrum LoadingOverload Ratio = 1.8

2020-T651

Constant-Amplitude loadingl-T Orientation. v=25 Hz• 2020-T651

5 10 25STRESS INTENSITY RANGE, ~K, MNm3/2

a constant amplitude; b periodic single-spike overload spectrumloading

12 Fatigue crack propagation behaviour in peakaged 2020-T651 alloy compared with 1M7075- T651 and PM 7091-T7E69 alloys. Notethat growth rates in 2020 are significantlyslower, particularly at near-threshold stressintensity levels and under tension dominatedspectrum loading (Ref. 101)

coherent particle (l5') hardening in the coarse grainedand textured microstructures which results in markedplanar slip deformation along close packed {Ill} slipplanes; this in turn promotes cracking along slipbands leading to a crystallographically faceted modeof crack extension.115,119-123 The resulting crackdeflection and branching locally reduce the mode Istress intensity at the crack tip and further promoteroughness induced crack closure via wedging of thefracture surface asperities (facets) in the crack wake.Despite the more heavily oxidised fracture surfacesin 2020 (compared with traditional 2000 series alloys)caused by the presence of lithium, additional contri-butions to crack closure from the wedging of oxidedebris are relatively insignificant because the excessoxide thicknesses remain small compared with theCTOD.

Subsequent results on high purity AI- Li-Cu-Zralloys, containing controlled variations of Cu andLi,35,124-127 provided further evidence that thesuperior crack propagation resistance of AI- Li alloysis associated with crack tip shielding mechanismsinduced by inhomogeneous planar slip deformation.Increasing the Li/Cu ratio, and hence the amount ofl5', was found to enhance crack path tortuosity, whichin turn resulted in lower crack extension rates andhigher fatigue thresholds (Fig. 13). Conversely, l5'


~K,ksi v'in5 10

of {Ill} slip band cracking preserves such high closurelevels into the intermediate and high growth rateregimes;l37,138 at near-threshold levels, where {100}or {110} pseudo cleavage cracking occurs, behaviourin AI- Li alloys is comparable with conventional Alalloys.134,135 Moreover, since {l11} cracking is pro-moted by strong deformation texture, early vintage2090 alloys, for example, tend to display the bestfatigue crack growth properties.115 However, theonset and relative proportion of {Ill} v. {IOO} or{110} type crystallographic fatigue cracking modeswith respect to 11K level can change in chlorideenvironments, as discussed in the section 'Behaviourin aqueous chloride environments', below.134,136

The effect of the higher elastic modulus, whichreduces the CTOD, and increased slip reversibility,which reduces the crack tip damage per cycle, act inconcert with mechanisms of crack tip shielding; how-ever, in light of the high crack closure levels andobserved effects of crack size, load ratio, and loadingseq uence on crack growth kinetics (as discussedbelow), the role of shielding appears to be moredominant. In essence, the tortuous crack paths in AI-Li alloys lead to slower (long crack) growth rates byincreasing the path length of the crack, by reducingthe local stress intensity levels due to deflection fromthe principal stress plane/39 and by reducing thenear-tip stress intensity range by enhancing rough-ness induced crack closure caused by asperitywedging109-112 (Fig. 16).

Crack path morphologyAs noted above, an essential feature of the fatiguebehaviour of AI- Li alloys is the shearable nature of{)' strengthening precipitates, which results in inhomo-geneous (planar-slip) deformation and unusually tor-tuous crack path morphologies. The nature of thedeflected crack paths takes several forms, as shownin Fig. 17: the crack may undergo macroscopicbranching, where the entire crack deflects at someangle to the principal stress plane because of defor-mation texture; alternatively, crack growth may befaceted ('zig-zagged') from crystallographic deflectionat boundaries because of the marked planarity of slip,or cracks may undergo (intergranular) short-transverse delamination because of the unrecrystal-lised, anisotropic 'laminated' grain structure.115

Additionally, pronounced deformation textures inAI- Li alloys promote deflection along the crack front.This is illustrated in Fig. 18 by a metallographicsection of a fatigue fracture surface in 2090- T81,taken across the specimen thickness perpendicular tothe crack plane and crack growth direction; the crackfront is highly deflected and faceted, with apparentsecondary crack branching beneath the surface. Tex-ture analysis shows that the crystallographic facets(with an included angle of "" 60°) result from a changein slip plane orientation between two components 'ofthe {110}<112) deformation texture, resulting incracking along two sets of intersecting {Ill} slipbands.62,63 However, precise modelling of the {Ill}slip band cracking mechanism occurring along planeswith the highest resolved deviatoric stress is currentlylacking. Combined with small mode III shear dis-

AK

• 2090-T81o B090-TBX• 8091·TBX6 2091-TBX

•••••• 2124- T351-'-7150-T651

o1

xroE

:: 0·8u~

~ 0·6:::>If)g 0-4U~~ 0-20:::U

5 10STRESS INTENSITY RANGE,~K, MNm3/2

14 Constant amplitude a fatigue crack growthrates and b corresponding crack closurelevels for long cracks in peak agedcommercial AI-Li and traditional AI alloys inplate form (L-T orientation, R=O'1, moistambient-air environment, using 6·4 mm thickspecimens, t/2 location). Growth ratebehaviour in AI-Li alloys is superior totraditional AI alloys consistent with highercrack closure levels (Refs. 68, 123)

Long Cracks, R=0.1

W • 2090-T81 • 00

~ 10-6 0 8090- T8X ~rJt,Cf'~ • 8091- T8X __ .~. oill;

If- 10-7 6 2091- T8X ./' .··~ffo W049-T8 #/ .··6t.~

3~ 8 ···· .. 2124-T351 ?" D·' ~~o ~ 1CJ -.-7150-T651 " DO 0

ffi ~ 0000 cr.. l:i

~ ~ 10-9 0 r;6b 1 latticeUz I.' 6"5~ 10"10 Do " J: ~~~c~~~le

"0 .( ~i.w .," &:o 10-11 .: •~ :/j;L.C 10-12

10-13

1·0 -t-l-----"--"""""'-----'---- .........•...........-------j

specifically for 2090, 8090, and 2091 AI- Li alloys(Fig. 15); fracture surface morphologies are thereforeunusually rough and covered with trans granularfacets.68,100,115,123,128-137At intermediate to high 11Klevels exceeding "" 6-8 MN m - 3/2, these facets areassociated with slip band cracking along {Ill} planes;however, at low 11K levels, 'pseudocleavage' typecracking along {100} or {110} planes has beenobserved.133-137 In fine grained 8091, however, whichexhibits the lowest closure levels and consequentlythe fastest growth rates, crack paths are essentiallylinear; in 2090- T81 conversely, which displays themost deflected crack paths, closure levels are thehighest and corresponding growth rates are the slow-est.123 This implies that tortuosity of the crack pathprovides a major contribution to crack tip shieldingin AI- Li alloys; additional shielding may arise fromclosure induced by cyclic plasticity108 and corrosiondebris,111,120-122,124,125 though the latter mechanismis relatively insignificant.41,135

Such shielding mechanisms are common to mostcoherent particle hardened aluminium alloys, thoughtheir influence is generally limited to near-thresholdstress intensity levels.120-122 What is particularlystriking about AI- Li alloys is that the predominance



a 2090- T81; b 8090- T8; c 8091- T8; d 2091- T8

15 SEM (left) and optical (right) micrographs of fatigue fracture surfaces and crack path morphologies(L-T orientation, R=0'1, t/2 location) at ilK levels between 6 and 8 MN m-3/2• Note rough fracturesurfaces and deflected crack paths in 2090, 8090, and 2091, compared with relatively flat surfaceand linear crack path in fine grained 8091. Arrow indicates general direction of crack growth(Ref. 123)

placements, this cracking morphology promotesroughness induced closure through the specimenthickness, in addition to mode II shear inducedwedging of asperities along the crack growth direc-tion. The crystallographic facets, however, becomeless sharp (the included angle approaches 180°) closeto the specimen surface, consistent with the throughthickness variations in textural components forwrought AI-Li alloys.63

Microstructural effectsThe influence of microstructure on fatigue crackgrowth in AI-Li alloys is essentially not unlike other

precipitation strengthened aluminium alloys. Resist-ance to crack growth is generally superior in under-aged structures, because the marked planarity of slipfrom coherent (shearable) precipitate distributionspromotes deflected crack paths and hence high closurelevels.35,68,115,123-128 With artificial aging, defor-mation becomes more homogeneous concurrent withloss of coherency of the precipitates; crack growthresistance is now diminished somewhat as the degreeof slip reversibility, crack path tortuosity, andresulting crack closure levels all are reduced. In Al-Li alloys, however, fJ' precipitates remain coherentup to large particle diameters; superior fatigue crack



16 Meandering fatigue crack paths and resultantwedging of fracture surface asperities(termed roughness induced crack closure) inthe plane of loading for AI-Li alloys.Micrograph obtained from specimen surfaceof 2090- T81 in L-T orientation; horizontalarrow indicates general direction of crackadvance ( Ref. 123)

propagation resistance can be maintained in coarsegrai~ed micr.ostructures even in peak aged tempers,provIded graIn boundary precipitation and associatedPFZ formation are largely suppressed.68,1l5,123,127,128

The close correlation between optimum crackgrowth resistance and high closure levels in Al- Lialloys, as illustrated by the T8 tempers of 2090, 8090,8091 and 2091 in Figs. 14 and 15, implies that if (longcrack) growth rates are compared as a function ofL1Keff (after correcting for closure), behaviouraldifferen~e.s with. respect to microstructure and alloycomposItIon WIll be less apparent (Fig. 19). This

. 123'notIon IS corroborated by a comparison of growthrates. determined under constant Kmax/variable RloadIng conditions* (Fig. 20). Al- Li-Cu alloys nowexhibit similar growth rate behaviour to 2000 seriesaluminium alloys, though faster growth rates areobserved for 7000 series alloys, presumably becauseof their increased environmental sensitivity.135-137

This should not be construed as an indication thatmicrostru:cture and alloy composition are unimport-ant to fatIgue crack propagation in Al- Li alloys sincethe microstructure clearly affects crack deflectionwhich in turn influences closure. In fact, coherent fJ' ~particle strengthening, strong deformation textures,coarse and unrecrystallised (elongated) grain struc-t~r~s, ~bsence of heterogeneous grain boundary pre-cIpItatIon, and PFZs can all be identified as theprominent microstructural sources of crack deflectionand branching.

* Constant Kmax/increasing Kmin procedures permit an estimationof the intrinsic crack growth behaviour by minimising closureeffects caused by crack wedging. As such, they result in crackgrowth data similar to that characterised in terms of ilKeff•

However, these techniques do not evaluate the influence of anyother shielding mechanisms, such as crack deflection or bridging,and moreover, require prior knowledge of the variation in KcI withilK level and R.


Load ratio effectsIt is well known that increasing the load ratio at aspecific stress intensity range can lead to faster crackpropagation rates and lower fatigue thresholds. Atnear-threshold levels, the effect is primarily associatedwith .a diminishe.d role of crack closure induced bywedgIng mechanIsms from the larger CaDs at highmean stresses. In Al- Li alloys, growth rates (plottedas a function of L1K) are particularly sensitive to theload ratio because closure levels are exceptionallyhI·gh·68,1l5,123-128 h bIt' f', owever, y p 0 tIng as a unctIonof L1Keff, after allowing for closure, the effect of loadratio becomes less apparent (Fig. 21). Moreover,because of this limited role of closure at high loadratios, the superior (long crack) fatigue crack growthproperties of Al- Li alloys are less evident at highR 123,135-137S· '1 1 h .. .'. ImI ar y, as s own In FIg. 20, If growthrates In Al- Li alloys are measured under constantKmax/variable R conditions which minimise crackclosure, behaviour is equivalent to that obtained for2000 series alloys.135-137 The superiority of AI-Lialloys can also be compromised under completelyr~v.ersed or tension-compression loading (R < 0) con-dItIons, ~h~re the compressive cycles can similarlyact to hmlt closure by crushing fracture surfaceasperities.138

Plate orientation effectsAs fatigue crack growth rates in Al- Li alloys are astrong function of crack path morphology, behaviourin commercially rolled sheet and plate products isstrongly anisotropic, 63,67.70,115,140both in the rollingplane and through the thickness. Growth rates forlong cracks (at fixed L1K) in 13 mm thick 2090- T81rolled plate can vary by up to four orders of magni-tude between various orientations, concomitant withlarge differences in measured crack closure levels 115(Fig. 22). Specimen orientations in the rolling plane,namely L- T and T-L, develop the highest closurelevels from highly deflected transgranular crack pathmorphologies (Kc1 values approach 90% of Kmax);

these orientations correspondingly show the slowestgrowth rates and highest thresholds (,....,3-4 MNm - 3/2). Similarly, crack growth resistance is high forthe T -S orientation where crack growth is oftendeflected through ,....,90° by delamination along weak-ened short-transverse grain boundaries.67,115 How-ever, intrinsic crack growth rates for the L-T, T-L,and T-S orientations are all identical as the mechan-ism of crack propagation remains essentially trans-granular;133-135 differences are associated with thevariations in shielding. Conversely, crack paths forthe S-Land S-T orientations are highly linear andinvolve delamination along grain boundaries, oftenweakened by the presence of Fe- and Cu-rich interme-tallics, fJ' - PFZs, and possible Li, Na, or K segregation;growth rates in these orientations are accordingly farfaster than for the L-T, T-L, and T-S orientations.lls

In addition to the short-transverse orientations, theL+45° (LT-TL) orientation also shows high growthrates67,70,1l5 and low thresholds (L1KTH = 2·4 MNm - 3/2), behaviour which appears to be associatedwith pronounced deformation texture in wroughtproducts. Slip band cracking is favoured along close

c


~25 1J~

a macroscopic crack branching; b crack meandering (crystallographic slip band cracking); c intergranular delamination (short-transverse)cracking

17 Various modes of non-linear cracking morphologies observed during fatigue crack propagation incommercial AI-Li alloys. Micrographs taken for 2090- T81 plate; arrow indicates crack growthdirection (Ref.115)

packed {lll} planes which are aligned. closer to theprincipal stress plane in the L +45° configuration;consequently, crack paths show a lower tendency fordeflection and meandering.115 Fatigue cracking mech-anisms in the L +45° orientation have not been fullyexamined, though reduced modulus (....,6%

) and lowerstrength ('" 16%

) in this configuration can additionallycontribute to faster crack growth rates.

Fatigue crack growth rates in Al- Li alloys alsotend to be substantially slower (at fixed L1K) in themid-thickness locations of wrought thick section platecompared with the surface regions,63,67 again associ-ated with marked variations in grain structure andcrystallographic texture. For example, in 2090- T81plate, surface regions are often partially recrystallisedand grains exhibit a weak {001}(110) texture; con-versely, at mid-thickness, the microstructure remainsunrecrystallised with strong brass, copper, and weakS type texture components. Since the growth rate

kinetics are a strong function of crack path tortuosity,which is promoted by unrecrystallised and texturedmicrostructures, growth rates are retarded at mid-thickness locations by the increased effect of crackclosure.63

Small crack effectsCurrent design methodologies often compute the life-times of structural components in terms of the cyclesto propagate a pre-existing flaw to failure throughintegration of experimental crack growth relation-ships (e.g. equation (1)) derived from standard speci-men geometries containing long cracks, typically inexcess of 5-10 mm in size. Although presumed to beconservative, this approach can potentially over-estimate lifetimes in the presence of small cracks,which are small compared with microstructuraldimensions or the plastic-zone size, or are simply



2~m

18 Crystallographic crack path tortuositythrough specimen thickness observed duringfatigue crack growth in AI-Li alloys inducedby pronounced deformation texture. Sharpfacets with an included angle of '" 60° are aresult of shearing along intersecting (111)planes across several high angle grainboundaries. Micrograph obtained for 2090-T81 plates (L-T orientation) at mid-sectionthickness normal to the crack growthdirection and crack growth plane; ilK=8-10 MN m-3/2: crack growth direction is intopage

physically small (i.e. < 1 mm);* such cracks are knownto propagate faster than long cracks at equivalentstress intensity levels, in part owing to a restrictedrole of crack tip shielding (e.g. crack closure) withcracks of limited wake.141-143 This effect has partic-ular significance for AI- Li alloys because theirsuperior (long crack) fatigue crack growth resistancearises primarily from (roughness induced) crack clos-ure, a phenomenon which is strongly crack sizedependent.

Small crack behaviour is generally characterised bymonitoring the early growth of naturally occurring

* Mechanistically, a crack is considered 'small' when its size is lessthan, or comparable with, the size of the steady state shieldingzone in the wake of the crack tip.

High Strength Aluminum AfloysMoist Air, L-T OrientationLong Cracks, High R>0.75

2090~K

R=0.1

1 5 10

STRESS INTENSITY RANGE , ~K, MN m-3/2

20 Intrinsic fatigue crack propagation rates incommercial AI-Li, and 2000 and 7000 serieshigh strength AI alloys in moist air,obtained by using constant Kmax/variable Rloading techiques (R increases from 0'1 at thehighest ilK to about 0·9 at ilKTH) or bymeasuring crack growth rates at high R. Allalloys show comparable fatigue resistance;7000 series alloys are slightly more sensitiveto environment (Refs. 135-137)

surface flaws (typically in the range 2-1000 Jlm insize) in unnotched bend or tension samples usingoptical, replication, or electrical potential techniques(see Ref. 143 for details). Using such procedures, stud-ies in commercial AI-Li alloys 2090, 2091, 8090, and8091 have indicated that small cracks initiate alongslip bands, predominantly at cracked Fe- or eu-richinclusions or constituent particles on the surface,polishing pits, and uncracked matrix/particleinterfaces.123,144-151 During early stages of small-crack growth, fracture surfaces exhibit large 'cleavage-like' features,145 rese"mbling those observed at near-threshold 8.K levels for long cracks. Subsequent crack

30

1 latticespacingper cycle

10ilK, ksi v in

5

-1210 0-5 1 5 10

STRESS INTENSITY RANGE, LlK,MNni3/2

Influence of load ratio R on (long) crackextension rates in 2090- T81, as function. ofilK and ilKeff; results'obtained using constantKmax/variable R schemes (Kmax=15MN m-·3/2,

R= 0·1-0'9) are also compared. Note thatdifferences in growth rates with respect toR are reduced when data are characterisedusing ilKeff or determined using the constantKmax/variable R approach (Refs. 115, 133-137)

21

ilK,ksi vin50-5

10-5Commercial AI-Li alloys

~ 106 L-T Orientation, R = 0.1

~I 10-71-(1)~~ 10-8~u(9 E 10.9~ ~ Long Cracks 1 latticeUZO° 2090-T81 spacing~ ~ 1d'1 0 8090-T8X per cycle

U"O + 80S1-T8XW 10-11 x 2091-T8X~ <) 80S0-T351i=; 1012 'l 8091-T351~ • 2091-T351 10-11

10-13 I AK ,

0·5 1 5 10 30STRESS INTENSITY RANGE, LlKeff, MN ni3/2

19 Intrinsic fatigue crack propagation resistanceof various commercial AI-Li alloy micro-structures in moist air environments obtainedusing variable ilK/constant R( = 0·1) schemesand plotted in terms of ilKeff, after correctingfor closure. Note that gtowth rate differencesbetween alloys are reduced compared withresults presented in Fig. 14 (Ref. 123)



SEM micrographs illustrating growthmorphology of naturally occurring, micro-structurally small (2-1000 J.lm) surfacefatigue cracks in AI-Li alloys. As shown ina they initiate at local heterogeneities inmicrostructure, propagate by periodicallydeflecting b along intersecting slip bands,and c at grain bounda-ries. Micrographs wereobtained from gold coated cellulose-acetatereplicas of specimen surface for 2090- T81:horizontal arrow represents direction ofloading (Ref. 123)

Since shielding mechanisms act primarily on the crackwake, small cracks generally are unable to developthe same degree of closure as a crack of a 'lengthlarger than the equilibrium shielding zone (where, 'forexample, fracture surface asperity contact is achieved);thus, at the same nominal t1K, the small crack may


10-3

1(j4

10-5

10-6 <llU

10-7 ~c

10-8 ;.'1J

10-9 m1J

10-10

(a) 10-11

,,

2090-TB1Long Cracks. R=O.1

a l-To T-Ld T-S

• S-L• S-Tv L+45

ilK, ksi v in5 10

2090-T81Long cracks. R=O.1

a l-To T-Ld T-S• S-L

• S-Tv L+45

o1

~ 06~~d 004~u~ 0'2u

."\~\••••• .~~ %

• d. wN•• d.~ ~• ~~ 0

• ~vvvvvv-: • ~ Vv ";,.'1 Vvv

•••• .'19 ttl Od v•• ··";".,.f 00o e.oo

o O."'<llJ.OO 0

(b)

5 10 • 30STRESS INTENSITY RANGE, ilK, MN m3/2

22 Variation in a fatigue crack extension ratesand b crack closure levels for long cracks in2090- T81 alloy at R= 0·1, as function ofspecimen orientation. Note how growth ratesparallel to rolling plane (S-L, S-T) are fastest,whereas growth rates normal to rolling plane(L-T, T-L, T-S) are slowest, consistent withnon-linearity of crack path morphologies andresulting levels of crack closure (Ref. 115)

advance is marked by frequent deflection of the crackpath at intersecting slip bands and high angle grainboundaries (Fig.23).144,145,151 At higher cyclic stresslevels, several cracks may nucleate along coarsenedslip bands, which tend to link up rapidly to form anetwork. 145

The growth rate response of such microcracks iscompared with conventional long crack data inFig. 24, as a function of t1K (the use of K here isconsidered appropriate as computed maximum andcyclic plastic-zone sizes remain small (typically 4%)compared with crack size). Despite the large scatterin experimental results, it is clearly apparent thatsmall cracks propagate at rates as much as 2-3 ordersof magnitude faster than long cracks at equivalentapplied stress-intensity levels/44,151 however, uncer-tainties do exist in the computation of t1K for smallcracks, owing to the stress concentrating effect ofnucleating cracks near inclusions and to irregularcrack geometries.141-143 Nevertheless, small crackgrowth is observed at t1K levels as low as 0·8 MNm - 3/2, far below the threshold t1KTH for longcracks.144-151

A major reason for the discrepancy between longand small crack behaviour is the restricted crack tipshielding associated with flaws of limited wake.144-152


10~K, ksi v'in

5

o 7150o 2124A 2090-T81• 2091-T351• 8091-T351

Influence of tensile overloadsExperiments to examine the role of single tensileoverloads are generally performed on fatigue crackspropagating under constant L1K conditions by moni-

Behaviour under variable amplitudeloadingThe design and selection of materials for aircraftstructures is increasingly based on fatigue crackgrowth response under complex loading spectra whichsimulate in-service operating conditions. This is par-ticularly important since the fatigue performance ofa material can vary widely with the specific nature ofthese spectra. For example, tensile overload cyclesof sufficient magnitude can act to retard crack growthand hence prolong fatigue life; compression overloads,on the other hand, tend to diminish this effect. Thebehaviour of AI-Li alloys in this respect is discussedbelow.

(crack deflection and closure). This ,is consistent withthe fact that microstructural effects become less evi-dent where the shielding effects are suppressed, i.e. athigh load ratios, with small cracks, or where resultsare characterised in terms of L1Keff or using constantKmax/variable R procedures.152 Accordingly, in con-trast to long crack behaviour, it is to be expected,and indeed is the case, that small crack growth ratesare relatively insensitive to load ratio and specimenorientation.135,144,151 However, since the long crackbehaviour of AI- Li alloys is, in general, superior toconventional aluminium alloys (because of extensiveshielding), differences between long and small crackresults are most significant in these materials. Thus,AI- Li alloys can be considered to display the bestlong crack properties of all high strength aluminiumalloys yet, by comparison, the worst small crackproperties, paradoxically for the same reason, i.e.crack tip shielding. 151

1611

1012

0-5 1 5 10 30

STRESS INTENSITY RANGE , ~K, MNrri3/2

25 Growth rate response of naturally occurring,microstructurally small surface cracks incommercial AI-Li and traditional alloys. Notethat results show no apparent influence ofmicrostructure and composition; small crackresistance of AI-Li alloys remains comparablewith other high strength AI alloys (Ref. 1513)

10

LiK, ksi v'in

5

2090-T81Small Long cracks

III L-T •o T-Li> T-S

experience a larger local stress intensity and henceextend at a faster rate. In fact, in situ compliancemeasurements in 8090 and 2090 confirm that closurelevels for small cracks are smaller than for correspond-ing long cracks.145,147 Accordingly, if long and smallcrack results are characterised in terms of L1Keff (aftercorrecting for closure), or if long crack growth ratesare evaluated using constant Kmax/variable R pro-cedures (which minimise the closure effect), theresulting data display a much closer correspondence,as shown for 2090 in Fig. 24. With increasing cracksize an equilibrium shielding zone is established inthe' crack wake, such that closure levels are similar,and small and long crack data merge (even whencharacterised in terms of L1K); in commercial AI- Lialloys, this occurs typically at crack sizes above~ 700-1000 Jlm and nominal L1K levels above~ 7 MN m - 3/2.144,146However, it should be notedthat there is not a perfect match between long andsmall crack data at near-threshold levels. This mayresult from scatter, uncertainties in L1K computation,and/or the fact that naturally occuring small cracksinitiate at local heterogeneities in the microstructurewhich are not necessarily simulated in a long-cracktest involving a pre-existing flaw and artificial loadshedding schemes.

In contrast to long crack results where microstruc-tural effects strongly influence crack growth behaviour(Figs. 12-14), a minimal influence of alloy chemist:yand microstructure is apparent for small cracks InAI-Li alloys/23,135,137,151 in fact, despite extensivescatter their small crack behaviour is comparablewith that of traditional alloys123,150,151 (Fig. 25). Thisimplies that the salient role of microstructure duri.ngfatigue crack propagation in these alloys occurs pnn-cipally through the development of crack tip shielding


1012

0·5 1 5 ., 10 30

STRESS INTENSITY RANGE, LiK, MNni3/2

24 Growth rate behaviour of microstructurallysmall (2-1 000 J.lm) surface, and long(>10mm) fatigue cracks in peak aged 2090AI-Li alloy. Note that at ilK levels below7-8 MN m-3/2, small crack velocities arenearly 2-3 orders of magnitude faster thanlong crack da/d N at R= 0·1. Closercorrespondence between long and smallcrack data is seen when long crack resultsare characterised using ilKeff or measuredusing constant Kmax/variable R schemes(Refs, 144, 151)


toring the transient crack growth response, afterresuming cycling at baseline stress intensity levels.Typical data for 2090- T81, showing the effect of 50,100, and 150°/0 single tensile overloads on crackpropagation rates at a constant baseline 11K of 8 MNm - 3/2, are plotted in Fig. 26 (Ref. 153). It is evidentthat 50°10 overloads have little effect; overloads ofhigher magnitude, however, result in an immediateacceleration followed by a period of prolonguedretardation before growth rates return to baselinelevels. The extent of the delay increases with themagnitude of the overload, and at near-thresholdlevels can completely arrest crack advance.153

Mechanistically, studies on AI- Li and titaniumalloys153-156 have indicated that the post-overload retardation is associated with residual com-pressive stresses in the overload plastic zone, andwith crack closure which results in the immediate

a before; b at; c after

27 SEM images of surface fatigue crackmorphology, showing crack tip duringapplication of a 1000/0 single peak tensileoverload in 2090- T81 at baseline AK level of8 MN m-3/2 (R=0·1). Imaging was performedafter unloading; vertical arrows indicateposition of overload event (Ref. 153)

vicinity of the crack tip as the crack grows into thezone; this effect is enhanced by lower CTO Ds in theoverload zone which promote wedging of the matingfracture surfaces. Crack deflection and branchingalong flow bands (Fig. 27), though predominantly asurface effect, may also contribute to the delay.153

Because AI- Li alloys develop such high shieldinglevels, they generally show a greater degree of post-overload retardation than traditional aluminiumalloys.10l,124,153-158 This behaviour is illustrated inFig. 28a, where fatigue crack extension following an80°10 tensile overload is plotted as a function of thenumber of cycles for three experimental AI- Li-Cu-Zr and 7075- T6 alloys.124 Although the AI-Li alloyswith low LijCu ratio show only marginally increaseddelays compared with 7075, alloys with higher LijCuratio clearly exhibit far superior variable amplitude,fatigue crack propagation resistance. Mechanistically,the large improvement in variable amplitude fatigueproperties with increasing lithium content is consist-ent with the larger volume fraction of b' precipitatesin these micro~tructures; this induces crystallographiccrack growth which in turn promotes crack deflection


2·5

( b)

2090- T81Baseline AK = 8 MN m3/2

Single Tensile Overloads

50%6. 100%• 150%

2090- T81 -3/2Baseline AK = 8 MN m

Single Tensile Overloads50%

6. 100%• 150%

~/•.•••..•----•....•..

(a)

o 2 4 6 8 10NUMBER OF POST-OVERLOAD CYCLES,X106

E 3·0Em-<] 2·5CI<{

9 2·00::W

e50:: 1·5w~1.L<{ 1·0zaVi~ 0·5....xw~ 0~0::U -0.5

-2

w~ 108

~:I:

~0:: 109(!)~~0::U 1010w::>(!)

6 11~ 10 I

-0·5 0 0·5 1·0 1·5 2·0DISTANCE FROM THE OVERLOAD EVENT, mm

a crack growth; b growth rate

26 Transient fatigue response, as function ofcrack extension following tensile overloadsin 2090- T81 (L-T orientation) at a constantbaseline AK of 8 MN m-3/2• Retardationdistances following overload are typically 2-3times computed maximum overload plastic-zone size, rol' as indicated by vertical arrows(Ref. 153)


30o 10 20 30 40 50 60

NUMBER OF POST-OVERLOAD CYCLES,X106

when compared with traditional 2124- T351 and7150- T651 alloys (Fig. 28b and c).

Compression overloadsAlthough compressive overloads are commonlyexperienced by numerous aircraft components in ser-vice, their influence on fatigue crack propagation hasreceived surprisingly little attention. As the crack ispresumed to remain closed during the negative por-tion of the loading cycle, the effect of compressioncycles has long been considered minimal. Recentresults on high strength steels, Al alloys, and AI- Lialloys, however, have indicated that periodic com-pression overloads can significantly accelerate nearthreshold crack growth rates, specifically by reinitiat-ing the growth of long fatigue cracks previouslyarrested at ~KTH, or by reducing the fatigue thresh-0Id.138,159,16o This behaviour has been associatedwith diminished levels of crack tip shielding, resultingfrom a reduction in residual compressive stresses inthe cyclic plastic zone and, more importantly, fromthe compaction of the fracture-surface asperities dur-ing the overload cycle. In view of their increaseddependence on crack deflection and asperity wedgingclosure mechanisms, AI- Li alloys are far more sensi-tive to compressive overloads than traditional alumin-ium alloys; for example, it takes 400-500%compressive overloads to reinitiate the growth ofarrested fatigue cracks in traditional 2124 and 7150alloys, whereas a mere 200% overload is sufficient tocause the same effect in AI- Li alloy 2090- T81(Ref. 138).

c

(a)

• 7075-T651o AI-1.1Li-4.6Cuo AI-2.1Li-2.9CuA AI-2.9Li-1.1Cu

80% Single Tensile OverloadBaseline 11K = 7 M N m-3/2

80

EE 70roJ:~~ 60l1J..J

~U« 50a::Ul1J:J(!)~ 40Lt

150% Single Tensile OverloadBeseline 11K - 8 M N rii3/2

• 2090-TalIa 7150-T651• 2124-T351 (b)

and roughness induced closure. Similar crack grow~hand crack closure behavioural trends have beenreported 153 for commercial 2090- T81 AI- Li alloy


o-0·5 0 0·5 1·0 1·5 2·0 2·5

DISTANCE FROM THE OVERLOAD EVENT, mma crack extension; b crack growth rates; c far field, crack closurelevels.

28 Post-overload, fatigue crack propagationbehaviour following single tensile overloadsin AI-Li alloys compared with other AI alloys.Note greater delay periods and retardationsin higher Li-containing alloys, and specifically2090, consistent with higher closure levels(Refs. 124, 153)

* These spectra represent, respectively, the lower root wing andhorizontal tail hinge moment load histories of the F-18 fighter.

Spectrum loadingComprehensive studies at Northrop161,162 on thespectrum fatigue resistance of several 2000 and 7000series aluminium alloys and experimental AI- Li alloysprovide further corroboration of the differing responseof AI- Li alloys to tension and compression overloads.Specifically, where behaviour was compared for twoload spectra, dominated either by tension overloads(TD)* or by tension and compression overloads ofapproximately equal magnitude (TC)*, AI- Li alloyscan show a nearly fivefold improvement in spectrumfatigue lives compared with 2024 and 7050 under TDloading, consistent with the large crack growth retard-ations observed following the tensile overloads(Fig. 29). However, alloys with the highest fatiguelives under such TD spectra also show the largestdifferences between TD and TC spectra, consistentwith the diminished role of shielding following com-pression cycles. In fact, using a modified TC spectrumwhich elimates all compressive loads (TCZ), the spec-trum fatigue life of all alloys was increased relativeto either TD or TC spectra. For example; an 'under-aged AI-3Cu-2Li-0·12Zr alloy, which exhipited life-times greater than 100000 flight hours under the TDspectrum, showed lifetimes less than 25 000 flighthours under the TC spectrum, a mere 250/0 of theTD life; under the modified TCZ spectrum, life wasincreased to nearly 110 000 flight hours.162

•

150% Single Tensile OverloadBaseline AK = 8 MNrrr3/2

• 2090-TalA 7150-T651

• 2124- T351 ( c)

•••

A A A A

• •

A

•

- . . ..----.

.. ~••• /V'

•

1·0

)(ro

~E 0.8-U~~ 0·6:JV>o..JU 0.4o..Jl1Jii:0:: 0'2~


It is clear that as with behaviour under constantamplitude loading, microstructures that promoteshielding, i.e. coarse grained unrecrystallised andunderaged coherent particle hardened structures thatdeform by planar slip and induce deflected crackpaths, are preferred for optimal performance undervariable amplitude loading conditions.

10-5

106Q.l

. U

10-7 ~c

10-8~"0ro

10-9 "0

llatt~spacingper cycle

• IN-905XL• 644-Bo 2090-T81~

.\. 644B

~.:/• =.;. IN-905XL 'b •

\ /

00Ql •

••••• o~~o

•••• • 000

(b)

30

L-T Orientation R=O.l

oo

o• P/M IN-905XL 000

• P/M 644-8 0

o 11M 2090-T81 / ••• # 00

/" ' ~",aa:/: :

LONG CRACKS, R=O.l 104L- T Orientation

xO'SCllE~-60'6~

w~ 004(/)

o...JU 0-2

o1 5 10

STRESS INTENSITY RANGE, L\K, MNm-3/2

30 Constant amplitude a fatigue crack pro-pagation and b crack-closure behaviour inPM AI-Li extrusions processed via rapidsolidification (Allied Signal 644-8 alloy) andmechanical alloying (IN-905XL). Growthrates in 1M 2090- T81 plate are also plottedfor comparison (Ref. 58)

~K, ksi "in5 10

the ductility and toughness of AI- Li forgings andextrusions, since pre-aging stretching procedures arenot feasible for these product forms.

Prominent among the experimental PM alloys arerapid solidification processed (RSP) AI-3'OLi-0'9Cu-0'4Mg-0'4Zr alloy 644-B, developed by AlliedSignal, and the mechanically alloyed (MA)AI-I'5Li-4Mg alloy IN-905XL, developed by IncoAlloys Int. The latter alloy also contains substantialamounts of carbon and oxygen in the form of oxide(AI203) and carbide (AI4C3) dispersions, which areformed during ball milling of elemental powders;carbon is added as an organic agent to preventagglomeration. Microstructural details and mechan-ical properties are summarised in Tables 2-5.

Such near commercial PM alloys have roughly100/0 lower density than traditional 7075 type alloys,and often exhibit far higher strength with moderateductility and fracture toughness. The HCF and LCFresistance of mechanically alloyed PM extrusions isgenerally superior to corresponding 1M alloys9o,169because of the more homogeneous mode of defor-mation in PM alloys. Moreover, no change indeformation mode is observed for PM AI- Li alloysbetween the high and low strain amplitudes consistentwith the observation of a single slope on the Coffin-Manson plot. 90 However, crack propagation prop-erties are invariably inferior to 1M alloys (Fig. 30a);


C1J co ali) I Ln I ..- ~ ~ ~2: U LOLO @co ~~ 2: u ~~ ~ R~ ~1- @1- ~ ~ (Y)

3, 2 and 1 refer to alloy composition, namely,AI-3Cu-2Li-1 Mg; C and M refer to coarse and mixed grainstructures, respectively; U and 8 denote underaged andpeak aged tempers

29 Comparison of spectrum fatigue lifetimes (insimulated flight hours) for experimental AI-Li-Cu, 2020- T651, 2024- T351, and 7050- T7551AI alloys, under tension dominated (TO),tension-compression dominated (TC), andmodified tension-compression dominated(TCZ) loading. Estimated lifetimes are basedon growth of 6 mm flaw to final failure, formaximum peak stresses below 145 MN m-2•

Note that AI-Li alloys show remarkableimprovements in lifetimes under TO spectrumloading over 2024 and 7050; behaviour underTC spectrum loading remains comparable(Refs. 161,162)

Powder metallurgy alloysIn addition to the 1M alloys discussed so far, severalpromising AI-Li alloys have been processed bypowder metallurgy. The PM route offers flexibility indeveloping alloys with unique compositions andhomogeneous microstructures, via such techniques asrapid solidification and mechanical alloying,163-168as problems of segregation encountered during castinglargely confine the chemical compositions of 1Malloys to the maximum solid solubility of the alloyingelements in aluminium. With PM techniques, AI-Lialloys can potentially be developed with furtherreduced density and improved strength and stiffnessby incorporating higher Li contents exceeding 30/0.In addition, high temperature characteristics of AI-Li alloys can be improved by alloying with largeconcentrations of transition or rare earth elements,such as Zr, Hf, Ti, Ce, or Co. The latter approach,which utilises dispersion hardening for resisting dislo-cation motion, is also the most effective means ofreducing planarity of slip, and hence in improving

+145MNm2 -------

o O~~~I~~~I-145MNm2L ~

+145MNm21 I~.120000 • ~~-~

~100000 -145 MNm ~------i80000 +14~MNm~I-~JU-jlti-iIJN-~L-ll"MiN-)J-.-~-~-JlI

[2 60000 -145 MNm2- Tez ,Cl~ 40000<t:...J~ 20000

(7j 0


20 porn~

31 Comparison of fatigue crack path and fracture surface morphologies in a, b MA IN-905XL, c,d RSP 644-8 extrusions at AK levels between 6 and 8 MN m-3/2• Crack paths in MA 905-XL areunusually linear, without features of crystallographic crack growth: arrow indicates direction ofcrack advance (Ref. 58)

fatigue crack growth thresholds are generally lowerand crack propagation rates up to one or two ordersof magnitude faster.58,170-172 The prime reason forthis can be traced to a limited role of crack tipshielding. Essentially, the fine microstructures (grainsize typically below 5 Jlm) and more homogeneousdeformation of MA AI- Li alloys, which results fromdispersion hardening by oxides and carbides ratherthan by ordered {)' precipitates (Li remains in solidsolution), yields unusually linear crack profiles andexceptionally smooth fracture surfaces (Fig. 31a, b),with little sign of slip band cracking.58,172 As a result,MA microstructures, unlike coarser grained 1Malloys, do not develop high shielding levels fromdeflection and related crack wedging closure mechan-isms; in fact, measured Kcl values at near-thresholdlevels are a mere 35% of Kmax in the AI- Li- Mg 905-XL alloy, compared with a corresponding value of~ 90% of Kmax in 1M alloys (Fig. 30b). The beneficialeffects of lithium on fatigue crack growth are thusnot so apparent in MA AI- Li systems. 58

Rapid solidification processing, conversely, can leadto better fatigue resistance for PM alloys comparedwith MA microstructures. 58 For example, the crackgrowth and closure properties of RSP 644- Bextrusions are comparable with 1M 2090- T81 plate,particularly at lower 11K levels (;5 10 MN m - 3/2). Aswith 1M AI- Li alloys, the superior behaviour of RSPalloys can be attributed primarily to factors that


favour planar slip deformation, namely, precipitationhardening by shearable {)' particles and the presenceof fi' dispersoids that inhibit recrystallisation andimpart preferred grain orientations. 58 Accordingly,fracto graphic features in the RSP 644- B alloyresemble those seen in 1M AI-Li alloys; crackpaths are highly deflected with rough transgranularfracture surfaces 3-4 Jlm sized facets and evidenceof local slip band cracking (Fig. 31c, d). In fact, theRSP AI- Li alloy displays superior crack growthproperties to most traditional RSP PM aluminiumalloys.58,164

In accordance with these arguments, crack growthrates and fatigue thresholds in MA alloys are oftenfound to be less sensitive to load ratio 58,171,173loading sequence, 58 specimen orientation. 58,~pecimengeometry,171 and crack size174 because of the limitedextent of shielding. For example, nominal I1KTHvalues in IN-905XL are equivalent to I1Keff,TH andare independent of load ratio for either tension-tension or tension-compression cycling. Similarly,growth rates are similar for small surface cracks« 1 mm) and long through thickness cracks in thesealloys; in fact, small cracks have been reportedto propagate slower.174 Conversely, faceted crackgrowth, tortuous crack paths, and high closure levelsare all prominent in the RSP 644-B alloy, such thatcrack growth behaviour in this material becomessensitive to load ratio and specimen orientation. 58



High temperature propertiesThe use of aluminium alloys at moderately elevatedtemperatures (350-450 K) in advanced aerospaceapplications can be limited by severe overaging whichdegrades strength, toughness, and fatigue resistance.Accordingly, there has been a considerable motivationto develop higher temperature alloys, such as theAl- Fe- V-Si, Al- Fe-Ce, and Al- Fe-Co systems, toreplace the more expensive titanium alloys for aero-space applications above 400 K. Naturally, there hasbeen much interest in using ultra-low density Al- Lialloys for such applications as well.

Current results on the high temperature perform-ance of Al- Li alloys suggest that despite a loss inproperties from overaging, behaviour is generallysuperior to traditional aluminium alloys.98,101,169,175Specifically, the crack initiation resistance of 2020-T651, 8090, and IN-905XL alloys has been found tobe relatively stable up to temperatures of ~ 450 K;mean values of the HCF strength for 2020 and 905-XL are above the scatter band for 7075- T6.101,169This improved thermal stability is presumed to resultfrom slower coarsening kinetics of strengthening pre-cipitates and dispersoids in these alloys.101 However,stretching before aging can diminish the high temper-ature fatigue resistance; for example, the endurancestrength of 8090 has been found to decrease at highertemperatures because of dynamic recovery in thecold worked (stretched) materia1.98 In addition, local-ised oxidation and grain boundary embrittlement atelevated temperatures can contribute to reductionsin LCF resistance, particularly at low strainamplitudes.86

Following prolonged exposures to high temper-atures of up to 1000 h at 436 K, peak aged 2090 and8090 AI-Li alloys have been shown to suffer onlymarginal' reductions (~15%

) in strength and tough-ness,175-177 even though such temperature excursionslead to grain boundary precipitation, an increase inthe width of (j'-PFZs, and coarsening of matrix pre-cipitates. However, 1000 h exposures at 533 K canlead to drastic reductions in strength, similar tobehaviour in 2124- T3, 7075- T6, and 2219- T8 alu-minium alloys.176,177 Fatigue crack growth resistancein AI-Li alloys can also be degraded by high temper-ature exposure, as crack paths become more linearleading to diminished crack tip shielding andenhanced crack growth rates.175,176 The increase ingrowth rates, however, is confined to high 11K levels(where crack velocities exceed ~ 10 - 9 m/cycle); near-threshold behaviour is generally unchanged. However,the fatigue properties of thermally exposed AI- Lialloys are still comparable with traditional 2124-T351, 7150- T6 and 7150- T7 alloys (Fig. 32).

Environmental effectsEnvironments such as moist air, water vapour, andaqueous halide solutions are known to acceleratecrack propagation rates in most 2000 and 7000 seriesaluminium alloys under both monotonic and cyclicloading.178-200 For example, high strength aluminiumalloys, including AI- Li alloys, are susceptible to inter-granular stress corrosion cracking (SCC) under mono-

L\K, ksi v in5 10 20

105

~ 166 Long Cracks, R=O.l • 0

~ : :~~~~~~ •••• _••• ~ •• 0000

~ 107 0 2090-T81 •••••••• ~. -; tJP 000

(1J • 2090-0A ••• •.• ~ °3: u 8 2124-T351 ••••; ••• ~.~ ~ 10 ' .. '" 7150-T651 •••.••.•1t.~<.9 -E -9 e••# ."

~ .10 - ~.Uz .'.0;

~~ 1610 / ~/:~

~ ~ 1011 i ;0° i : D~ ~~ 1612

11 -13 AK 10-11

10 1 5 10 20 30STRESS INTENSITY RANGE, L\K,MNm-3/2

32 Influence of high temperature exposure of1000 h at 436 K (overaging) on fatigue crackpropagation rates in 2090- T81, comparedwith traditional alloys. Although overagingdegrades crack growth resistance of AI-Lialloy, it still remains comparable, if notsuperior, with traditional alloys (Refs. 175,176)

tonic loading, particularly in the short-transverse(S-T, S-L) orientations.188-192 Under cyclic loads,environmentally assisted crack growth may beobserved at stress intensities either above or belowthe sustained-load threshold for environmentallyassisted cracking (KIscc)' In the latter case, environ-mental damage mechanisms are associated with thesynergistic effects of plastic deformation and chemical/electrochemical reactions.133-135, 178-187

Because of the highly reactive nature of lithium,AI-Li alloys are often incorrectly perceived to bemore sensitive to corrosive environments thantraditional aluminium alloys. However, in manyrespects Al- Li-Cu alloys behave like 2000series alloys, and are superior to 7000 series alloys,in moist air, water vapour, and aqeuous chlorideenvironments.133-137, 187

Behaviour in gaseous environmentsAs with 2000 and 7000 series alloys, water vapouraccelerates fatigue crack propagation rates in AI-Lialloys, 1-33-135,179,180as shown for 2090- T81 in Fig. 33(crack closure effects were minimised in these data bymeasuring growth rates using constant Kmax/variableR techniques or with physically short « 1 mm)cracks.135 At equivalent 11K levels, crack growth ratesin pure water vapour are comparable with those inmoist air, but are accelerated relative to behaviour inpure helium, oxygen, and vacuum. Crack velocitiesare independent of water vapour pressure for PH

20

above ~ 0·2 Pa (at a frequency of 5 Hz); below this'saturation pressure', growth rates decrease withdecreasing partial pressure to approach inert environ-ment crack velocities. Environmental effects are mostpronounced at near-threshold levels and diminishwith increasing 11K, showing a plateau behaviourcoincident with the transition; in both regimes, growthrates display al1K4 power law relationship.133

In common with other high strength aluminiumalloys, hydrogen embrittlement has been suggested as



30

2010~K,ksi v in

5

•••••••••••••• ~~iP

........ .af'" ;8~,.,.,. ,i'" ~ •• t;, .D~D ••

'P"~'.··~!/JDft,~" 00

00

'.•....••...-f ••:::•••.. ... :

Apart from the intrinsic damage mechanismsdescribed above, many extrinsic factors can affectenvironmentally assisted cracking in AI-Li alloys.For example, in vacuo testing can result in moretortuous crack paths with extensive faceting comparedwith laboratory air, possibly because of the greaterease of reversible slip in inert environments.35Additionally, the presence of corrosion products oncrack surfaces can promote (oxide induced) crackclosure and retard crack advance, particularly at near-threshold stress intensity levels.111,119 However, lim-ited studies115 show that the role of such oxidationdebris in commercial AI-Li alloys is generally minimal(at least in moist air environments), as experimentallymeasured oxide thicknesses are considerably smallerthan the relevant crack tip opening displacements.115Results of AI- Li-Cu alloys in gaseous oxygen andsaline environments, on the other hand, suggest oxideinduced closure effects may be important.135

Behaviour in aqueous chloride environmentsAluminium-lithium alloys are particularly suscept-ible to accelerated fatigue crack propagation in aque-ous chloride solutions.133,178-187 Typical growth ratedata for commercial 2090 and 8090 AI- Li alloys in3'5% aqueous NaCI (measured using conventionalload shedding schemes at R = 0'1) are illustrated inFig. 34. Crack velocities in 2090- T81 are typicallyone to three orders of magnitude faster in salt watercompared with moist air at low stress intensities, witha ~ 50% reduction in fatigue threshold. The acceler-ation in growth rates has been proposed to resultfrom a diminished contribution from roughnessinduced closure, arising from the combined action ofcorrosive attack and mechanical fretting in flatteningfracture surface asperities. 186 Alternatively, thereduced faceting and closure may result from anintrinsic change in cracking mechanism in chlorideenvironments.135 However, at equivalent 11K levels,growth rates in AI-Li alloys in sea water environ-ments are still an order of magnitude slower than in7075- T6 (Fig. 34). Studies on the role of micro-structure in influencing corrosion fatigue of AI- Li

164

W Long Cracks. R=O.l~ 10-5 NaCI Air0:: 0 2090-18

6 0 8090-18I 16 7075-16~~~~167(9-~ E168U .~~169urnW "01610

:J~ 1d1~u:: 1612

1 5 10STRESS INTENSITY RANGE , ~K, MN m3/2

34 Comparison of fatigue crack propagationrates in 2090- T81 and 8090- T8 AI-Li alloys(L-T orientation, R= 0·1, 50 Hz) comparedwith traditional 7075- T6 AI alloy in aqueous3'50/0 NaCI environments (Refs. 186,193)

10

10-6 ~U~u

.10-7 ~Z

1 lattice ~spacing 10-8 ~

a Oxygen (21 kPa) per cycle

: ~:~~~~7~:~-7pa)•__~_ ~~?s~ ~~~a), in He (2 kPa) 10-9

~K,ksi v in5

••••••••••••••••••••••••••• c

../"

/ .•;;.,/ .....

/'~.,/

2090-181. L-T Orientation, v=5 HzConstant Kma/AK. Step Changed RKmax= 17 MNm-3/2

1611

1 5 10 20STRESS INTENSITY RANGE, ~K, MNni3/2

33 Influence of environment on intrinsic fatiguecrack growth rates in 2090- T81 alloy (L-Torientation, 5 Hz), obtained from shortthrough-thickness cracks or long cracksunder constant Kmax/variable R conditions.Note that crack velocities are significantlyenhanced only by water vapour, particularlyclose to AKTH; behaviour in O2 is comparablewith that in inert environments (Ref. 133)

the prime mechanism of crack tip damage involvedin the accelerated cracking of AI- Li alloys in moistair relative to vacuum.133,179-181,186 Mechanistically,this occurs in gaseous and aqueous environmentswhich produce atomic hydrogen via reaction withfresh metal surface exposed at the crack tip by thecyclic deformation. Hydrogen is then adsorbed andtransported into the plastic zone ahead of the cracktip by diffusion or moving dislocations,179 leading tofaster crack growth rates; cracking may be aidedfurther by reduced slip reversibility in the presence ofan aggressive environment. 35 Because nearly identicalcrack growth rates are observed in oxygen and inertenvironments (Fig. 33), neither oxygen nor oxide filmformation on crack surfaces is considered to influencethe morphology and reversibility of plastic defor-mation.133 Moreover, the measured dependence ofgrowth rates on H20 partial pressure and dependenceof saturation pressure on load ratio and frequencyare consistent with an impeded molecular transportmodel of hydrogen embrittlement.133,18o

Fractographically, the embrittlement is reflected indifferences in the crack path morphology. In inertenvironments, fatigue cracking in AI- Li alloys pro-ceeds along intersecting {Ill} slip planes, particularlyat moderate to high 11K levels exceeding ~ 6 MNm - 3/2 (Refs. 135, 186), resulting in large crystallo-graphic facets and unusually rough fracture surfaces(Fig. 18). Conversely, a flat trans granular crackingmorphology (resembling cleavage) along {100} or{l10} planes, combined with regions of faceted slipband cracking within subgrains, is seen at near-threshold levels. Hydrogen environments, on the otherhand, lead to increased intersubboundary cracking athigh I1Ks,133-135 and promote the cleavage like crack-ing at progressively higher 11K levels,132-134 bothfactors that enhance crack growth rates; no evidencefor ductile or brittle straitions has been reported inAI- Li alloys.



2090-TB1, L-T Orientation, v = 5 HzConstant Kmax = 17 MNm-3/2

• 1% NaCI, -0.840 V (SCE)01% NaCI, +0.37% Li2 C03 ,-0.840 V (SCE)~Moi5tair<> 1% NaCI, -1.240 V (SCE)o 7075-T651 (L-S), 1% NaCI, -0.840 V (SCE)• Helium (2 kPa) •

•R=O.7R = 0.05--r--

0.60 ~ R ~ 0.93

I2 5 ill 20

STRESS INTENSITY RANGE I D.K,M N m-3/2

35 Intrinsic corrosion fatigue crack growthrates for 2090-T81 and 7075-T6 alloys (L-Torientation, 5 Hz) in aqueous 1% NaCI underimpressed anodic (- 0·84 V) and cathodic(-1·24 V) potentials, compared with heliumand moist air. Anodic polarisation producesthe largest corrosion fatigue effect, which isreduced by Li2C03 additions (Refs. 133-135)

alloys in chloride solutions are scarce; limited resultssuggest that overaging does not significantly lowerfatigue resistance as crack paths remain predomin-antly transgranular.193

More detailed studies 133-135on the intrinsic fatiguecrack growth kinetics show that aqueous NaCIenvironments accelerate crack propagation in AI- Lialloys under both anodic and cathodic polarisation(Fig. 35). The magnitude of the effect is a function offrequency and Li2C03 additions, and is most pro-nounced at near-threshold levels under anodic polar-isation; in 2090- T81 for example, growth rates arealmost two orders of magnitude faster than in inertor water vapour containing environments. In contrast,cathodic potentials reduce fatigue crack growth ratesin 2090 to levels approaching those observed in moistair, as with other aluminium alloys. 178,182In addition,passivating ions (0'4% Li2C03) can also retard crackgrowth rates in NaCI (Fig. 35) by forming a protectivefilm on the crack surface.133,187

Mechanistically, the intrinsic environmental dam-age of AI- Li alloys in aqueous NaCI environmentsis associated with combined effects of hydrogen andsurface films.133,187 Cathodic polarisation studies arenot conclusive in ruling out hydrogen embrittlementmechanisms because interpretations are complicatedby lack of knowledge regarding the crack tip solutionchemistry.133, 179The observed effect of frequency oncrack growth rates, however, does strongly suggestan active role of surface films. In most 7000 seriesaluminium alloys (at 11K levels of '" 5-8 MN m - 3/2),decreasing frequency results in increasing growthrates;194-199 hydrogen embrittlement models gener-ally relate this to enhanced surface reactions or masstransport during the longer time period of the cycle.Fatigue crack propagation rates in 2090- T8I, con-versely, decrease by a factor of two as the frequencyis reduced from 5 to 0·1 Hz at 11K levels 1"'.1 10 MN

m - 3/2 or above, though little effect of frequency isapparent close to I1KTH.133,193Based on these obser-vations, it has been suggested that mass transport isnot the rate limiting step in AI- Li alloys; rather,increased crack tip strain rates at higher frequenciesare reasoned to destabilise (protective) surface films,resulting in enhanced anodic dissolution or crackadvance by film rupture. This effect may be coupledwith time based hydrogen production and entry intothe metal.133,179

Fracture surfaces in NaCI and moist air environ-ments are nominally similar though, because of themore aggressive nature of salt water, a greater degreeof intersubgrain boundary fracture and reduced slipband cracking is seen at high 11K levels.134-136 Atlow ~K levels, on the other hand, the crystallographicfracture morphology resembling transgranular cleav-age is more prevalent in NaCI than in moist airenvironments. The associated reduction in crack pathtortuosity caused by an environmentally inducedchange from faceted slip band cracking in benignenvironments to relatively linear intersubgranular and{IOO}or {IIO} type cracking may be the prime causeof the lower closure levels measured in conventionalcorrosion fatigue tests.186

Unlike behaviour reported for steels179,198,199where local crack tip electrochemistry effects greatlyenhance environmental effects when crack sizes aresmall ('" 150 Jlm to 5 mm), the corrosion fatigue behav-iour of short cracks ('" 300 Jlm to 4 mm) in AI-Lialloys is essentially identical to that of long (> 5 mm)cracks.133,135 In fact in 2090- T81 alloy, K -basedsimilitude adequately describes the growth kinetics offatigue cracks for all crack sizes above 300 Jlm inlength.

In summary, it may be concluded that the intrinsicfatigue crack growth resistance of AI- Li alloys inmoist air and sea water environments is comparablewith 2000 series alloys and in fact superior to 7000series alloys133-137,179 (Figs. 20, 35). In addition,limited studies indicate that AI- Li alloys showsuperior crack initiation and S-N fatigue resistancein these environments, at least compared with 7075type alloys, a fact attributed to their reduced suscepti-bility to hydrogen absorption and hydrogen assistedintergranular fracture.183-185

Cryogenic propertiesThe cryogenic properties of AI-Li alloys have recentlybeen the focus of much attention following reportsthat indicated remarkable improvements in theirstrength, ductility, and fracture toughness at liquidnitrogen (77 K) and liquid helium (4 K) temperaturescompared with room temperature.38-42 Specifically,studies for the Land L-T orientations on 2090- T81alloy plate showed a 45% increase in strength, 67%increase in tensile elongation, 12% increase in modu-lus, and most importantly 90% increase in fracturetoughness between 298 K and 4 K. * These properties

* The dramatic improvements in toughness at cryogenic temper-atures are not, however, typical of all AI-Li microstructures orproduct forms; thin sheet AI- Li alloys, such as 2090- T83 forexample, may actually exhibit lower toughness with decrease intemperature.69,200



10-3

1cr41cr5

(])

10-6 ~u

.10-7 ~

10-8 ~

10-9~

10-10

10-11

\ 12.7 mm PI",

,/~\,

o •••••0°0 •• , ..

0000 •••

.", '

1.6 mm Sheet

o~o

00 °00

:

6.K, ksi V in5 10 20

• 2090-T8• 7075-T6

• 2090- T8. R=0.1L-T Orientation

L-T Orientation, R=O.1Sheet Plate

Xrtl

~EO'8--u~w-0'60:::=>(/)

gOAU~~ 0·20:::u

o (b)

1 5 10 30STRESS INTENSITY RANGE, 6.K, MN rri3/2

a fatigue crack propagation; b crack closure behaviour

37 Fatigue behaviour in 1·6 mm thin 2090- T83sheet and 12·7 mm thick T81 plate, as afu nction of ilK (L- T orientation, R= 0·1, 50 Hz,moist air), compared with 7075- T6.Properties in plate were measured on 6·4 mmthick samples (t12 location); sheet behaviourwas examined in as received thickness(Refs. 70, 207)

* St~ess state .effects may be expected to affect crack growth ratesat hIgher .1K1levels as deformation conditions in the sheet alloysapproach that of plane stress.207

Influence of wrought product formThe ~predomi~ant r~le of crack tip shielding in pro-motIng supenor fatIgue crack growth resistance inAI-Li alloys via crack deflection mechanisms impliesthat the effect may not necessarily be seen in allproduct forms.70,207 For example, at the same 11Kfatigu~ cr.acks in 1·6 mm thick 2090- T83 sheet propa~gate SIgnIficantly faster than in 12·7 mm thick 2090-T81 plate, particularly as 11K levels (at R = 0'1)approach the fatigue threshold (Fig. 37a), though both~~lled sheet and plate have nominally similar compos-ItIon and hardening precipitates.207 Such faster' crackgrowth rates in the sheet are concurrent with mark-edly lower crack closure levels compared to plate(Fig. 37b). This is in contrast to traditional aluminiumalloys, such as 7075- T6, which rarely show suchdifferences in behaviour between sheet and plate, atleast for 11K levels below rov 20 MN m - 3/2.*

Such differences in fatigue behaviour in AI- Li alloysheet and plate are principally associated with micro-structurally induced variations in crack path deflec-tion and resulting crack closure effects (Figs. 16-18,~7, ~8). In the T81 plate material, marked slip planar-Ity Induced by coherent b' and deformation texture

Do

20

oo

oo


10

o 0

6.K,ksi V in

5

T-L Orientation, R=0.1

298K 17Ko 0 2090-T81

___ ----- 7475-T651

2

5 10

STRESS INTENSITY RANGE, 6.K, MNm3/2

36 ·,Fatigue crack propagation behaviour in 2090-T~1 and 7475- T6 alloys at ambient and liquidnitrogen temperatures (T-L orientation, R=0·1, 50 Hz) (Ref. 38)

c0t,nbined .with lower specific modulus and superior~atIgue resIstance have made AI- Li alloys very attract-Ive for many cryogenic structural applications' theseinclude liquid hydrogen and liquid oxygen fuel tankson ~uture hypers?nic and transatmospheric aerospacevehIcles, supportIng structures for magnets in particleaccelerators, and high energy fusion devices.

As with. many high str~ngth aluminium alloys, theHCF resIstance of AI- Li alloys is substantiallyimproved at cryogenic temperatures.201-203 Forexample, the smooth bar axial fatigue strength of2090- T81 (at a life of 105 cycles, R = 0·1) increasesfrom rov 290 MN m - 2 at 298 K to rov 700 MN m - 2 at4 K (rov 90% of the tensile strength), behaviour whichis second only to 7075- T6; in addition, the AI-Lialloy is less notch sensitive.201 Studies on the AI- Li-Cu-Mg-Zr system203 indicate similar trends, andfurther show that though the fatigue strength at 298 Kcan be improved :by increasing the Zr content, thetemperature dependence of fatigue strength iscompromised.

Although data are limited, fatigue crack growthrates in AI- Li alloys such as 2090 show a decreaseat lower temperatures, typical of pure aluminium andmany aluminium alloys.38,204-206 Moreover theproperties are comparable with traditional aluminiumalloys at cryogenic temperatures (Fig. 36). The gener-ally improved fatigue resistance of aluminium alloysat cryogenic temperatures has been attributed to aninc~eas~ in the monotonic or cyclic yield strength,WhICh In turn increases the plastic work required forfracture; this has been associated with more homo-geneous plastic .deformation,204,205 or a reduction inrates of thermally activated processes at low temper-atures.203 Alternatively, the reduced low temperaturecra~k velocities in 2090- T8 AI-Li alloy may beattnbuted. to the s1?aller CTODs per cycle, resultingfrom an Increase In strength, elastic modulus andstrain hardening rate, and to limited environ~ental

..effects at cryogenic temperatures. However, the expla-nations are not definitive because of the lack ofintrinsic low temperature, crack propagation data ininert environments.



Concluding remarksThe resurgence of interest in AI- Li alloys has resultedin the development of a series of low density, mono-lithic aerospace materials with strength-toughnessproperties comparable with traditional aluminiumalloys and a considerable price advantage over com-


o103 104 105 106 107 108

NUMBER OF CYCLES TO FAILURE, Nt

39 Influence of wrought product form on HCFresistance of 2090- T8 (kt= 1, or 3, R= 0·1,longitudinal orientation) in ambient air. Notethat plates an9 extrusions show betterfatigue resistance compared with sheet(Ref. 70)

o 1.6mm 183 Sheeto 12.7 mm 181 Plateb. 19.0 mm 186 Extrusion

2090-T8X, R = 0.1

oIkt = 3.01

Sheet/ g

600

higher growth rates (Fig. 37a), presumably because ofan increased modulus and the benefical role of [)' inpromoting planar slip deformation and crackdeflection.

The influence of load ratio and specimen orient-ation on the crack growth behayiotir of AI-Li sheetis similar to that described above for plate alloys.Higher load ratios induce faster growth rates, particu-larly at near-threshold levels, and the lowest crackgrowth resistance is found for the L +45° orien-tation.7o,207 However, owing to the limited role ofcrack closure in sheet, the magnitude of the load ratioand orientation effects is less pronounced than inplate alloys.207 In fact, the larger CTOD levels associ-ated with higher load ratios minimise :the closureeffect to such an extent that, at R >0'75, sheet andplate material show essentially similar behaviour.

Differences in the S-N fatigue behaviour of com-mercial AI-Li alloys have also been noted withdiffering wrought product form (Fig. 39). Specifically,for the three principal product variants of 2090- T8Xalloy, T86 extrusions exhibit the highest smoothspecimen fatigue strength, followed by T81 plate andT83 sheet; behaviour, as in plate alloys, is aniso-tropic.70 For notched tests (with a stress concentrationfactor kt of 3), extrusions and plate show equivalentbehaviour, with the fatigue strength of the sheetmaterial being slightly inferior. Such results have beenattributed to delamination induced reduction inlateral constraint in the thicker product forms,70though they are more likely to be associated with thelower strength of the 2090 sheet.

%500z2~400J(/)

~ 300crl-(/)

<i! 200z~~ 100

Comparison of crack path and fracturesurface morphologies during fatigue crackgrowth in 2090- T83 sheet, showing crackpaths: a along crack growth direction,b specimen thickness, and c on fracturesurface; arrow indicates general direction ofcrack advance (Ref. 207)

promotes crack deflection through crystallograph-ically faceted slip band cracking (Figs. 16-18). Sucheffects are not as prominent in sheet207 becauserecrystallisation occurs readily in thin section prod-ucts, below 1"0o.l 4 mm in thickness, despite the presenceof Zr. The finer, partially recrystallised grain structureand; reduced texture induce essentially linear crackpaths and hence far smoother fracture surfaces(Fig. 38). It should be noted, however, that evenwithout significant crack closure effects, AI- Lisheet alloys still display superior fatigue cr4.ck growthresistance than traditional alloys, particularly at


AcknowledgmentsThis work was supported by the Director, Office ofEnergy Research, Office of Basic Energy Sciences,Materials Sciences Division of the US Department ofEnergy under Contract No. DE-AC03-76SF00098.The authors would particularly like to thankDr R. J. Bucci of Alcoa for his help and encourage-ment throughout this, work. In addition, they thankDrs P. E. Bretz and R. R. Sawtell, also from Alcoa,P. O. Wakeling, formerly of Alcan; Drs G. R. Chan-ani, G. V. Scarich, and K. M. Bresnahan of Northrop;Dr W. E. Quist of Boeing; G. J. Petrak of WRDC;Dr S. Das of Allied Signal Inc.; Prof. R. P. Gangloffof the University of Virginia; Prof. N. J. Kim ofPohang Institute of Technology (Korea); Dr K. V.Jata of the University of Dayton Research Institute;Dr R. S. Piascik of NASA Langley; and Prof. P. P.Pizzo of San Jose State University, for their collabor-ation. The assistance of the late Dr W. Yu, H. F.Hayashigatani, J. E. Miles, D. Kovar, and J. C.McNulty with the experiments," Ms Madeleine Pentonwith preparation of the manuscript, and the keyreader and referees for their critical comments, is alsogratefully acknowledged.

composition and microstructure, principally by reduc-ing through-thickness crack path meandering. Simi-larly, centre cracked, rather than compact tension,bend, or double cantilever beam, specimen geometriesmay also limit the fatigue crack growth propertyadvantage of AI- Li alloys by reducing off-angle crack-ing and inhibiting crack deflection. A similar suppres-sion of shielding in AI- Li alloys has been reportedby testing single edged notched, tension specimenswith fixed grips compared with freely rotatinggrips.133

Finally, it should be noted that primary sourcesfor crack deflection and crack path tortuosity, whichconfer the excellent fatigue crack growth propertiesin commercial AI- Li plate alloys, are marked slipplanarity, strong deformation texture, and weakshort-transverse properties, all microstructural fea-tures which limit toughness and which may be con-sidered undesirable for many service applications.Although new alloys can, and have been, developedwith less anisotropy to solve this problem, theirfatigue crack growth properties are likely to be inferiorto those of the early commercial alloys.

These factors suggest that the use of AI- Li alloys,or the direct substitution for traditional alloys, infatigue critical designs requires a greater mechanisticunderstanding and more detailed analysis of thecorrelation of laboratory scale coupon tests to theperformance of structures in service; with AI- Li alloys,it is by no means certain that the superior fatigueproperties measured in laboratory tests will alwaysbe realised in actual structures.

References1. F. H. FROES: Mater. Des., 1989, 10, (3), 111.2. M. A. STEINBURG: Sci. Am., 1986, 255, (4), 66. '3. D. R.TENNEY, B. LISAGOR, and s. DIXON: J. Aircraft, 1989,26,

953.4. P. CANNON: J. Met., 1989, 40, (5), 10.

~10-4

10-5<1l

.10-6 ~

. ~10-7 .~-. Z: 10-8~

"0

.1cr9

20

/~SiCp/Al x

x,j'!-

.' ~-Lixxx~

~K,ksi v in5 10

10-10-12

10 2 5 10 20 30STRESS INTENSITY RANGE, ~K, MNm-3/2

40 Comparison of fatigue crack growthresistance of advanced structural aerospacematerials, namely, AI-Li alloys (2090- T81),metal matrix composites (SiC-particulatereinforced 7091 PM aluminium alloy) andARALL-2 laminates (From Refs. 11,14,115)

peting composites. Moreover, they offer the designera material with improved fatigue crack growth prop-erties over many composites (Fig. 40) and traditionalAl alloys, provided in-plane (e.g. L-T, T-L) orient-ations are used. For other orientations such as S-Tand S-L, delamination cracking can lead to poortoughness and crack growth properties in thick platesections. Furthermore, short-transverse delaminationfailures are responsible for the low fatigue lives meas-ured under anticlastic bending fatigue. lOB Under thesecircumstances, the performance of AI- Li alloys isinferior to traditional alloys.

Since the beneficial fatigue properties can be tracedprimarily to extrinsic crack tip shielding mechanisms,designers must appreciate the implications of thisphenomenon on life prediction. For example, manyof the contrasting observations on the fatigue crackgrowth behaviour of AI- Li alloys can be explainedas consequences of crack tip shielding, resulting fromhighly tortuous crack paths induced by planar slip.This leads to superior long crack properties, underconstant amplitude loading at positive, low loadratios, and under tension dominated spectrum load-ing, all instances where crack closure due to wedgingphenomena is not restricted. However, factors whichlimit crack path tortuosity and asperity wedging cancompromise the superiority of these alloys. This canoccur at high load ratios, under compression domin-ated spectrum loading, with microstructurally smallcracks, and with refined and overaged micro-structures; in these instances, the fatigue behaviour ofAI- Li alloys is essentially similar to 2000 and 7000series alloys.

A further consequence of the role of crack tipshielding in AI- Li alloys is complications with theassessment of fatigue crack growth resistance. Inaddition to loading spectra, crack size, and specimenorientation, crack growth behaviour tends to bestrongly dependent on wrought product form andspecimen geometry. For example, thin specimens gen-erally will exhibit significantly faster crack growthrates than thick specimens of nominally identical


2165~

w _ ARALL-2 (90°)

~ 166.• ARALL-2 10°)+ SiCp/AI composite

~ -7 x AI-Li 2090-T81

3~ 10 :aUlli ~1C58:::t= E ARAll-2U - -9' (900

),,.-

« B 10: :.0:::_u~1dO :w !::Jt9 1d1-

~u::


5. J. J. DE LUCCIA, R. E. TRABOCCO, J. WALDMAN, and J. F COLLINS:Adv. Mater. Proc., 1989, 137, (5), 39.

6. T. M. F. RONALD: Adv. Mater. Proc., 1989, 137, (5), 29.7. w. E. QUIST, A. L. WINGERT, and G. H. NARAYANAN: in

'Aluminum-lithium: Development, application and super-plastic forming', (ed. S. P. Agrawal and R. J. Kar), 227-279;1986, Metals Park, OR, ASM International.

8. T. W. CHOU, R. L McCULLOUGH, and R~ B. PIPES: Sci. Am.,1986, 225, (4), 192.

9. s. v NAIR, J. K TIEN, and R. C. BATES: Int. Met. Rev., 1985,30,275.

10. A. L GEIGER and M. JACKSON: Adv. Mater. Proc., 1989, 137,(7),23.

11. J.-K. SHANG, W. YU, and R. O. RITCHIE: Mater. Sci. Eng., 1988,102A, 181.

12. s. v KAMAT, J. P. HIRTH, and R. MEHRABIAN: Acta Metall.,1989, 37, 2395.

13. R. J. BUCCI, L. N. MUELLER, L. B. VOGELSANG andJ. W. GUNNICK: 'Aluminum alloys: Contemporary researchand applications', (ed A. K. Vasudevan and R. D. Doherty),Treatise on Materials Science and Technology, Vol. 31,295-322; 1990, New York, Academic Press.

14. R. O. RITCHIE, W. YU, and R. J. BUCCI: Eng. Fract. Mech.,1989, 32, 361.

15. R. MARISSEN: PhD thesis, Delft University of Technology,1988.

*16. T. H. SANDERS, Jr and E. A. STARKE, Jr (eds): 'Aluminum-lithium alloys'; 1981, Warrendale, PA, The MetallurgicalSociety of AIME.

*17. T. H. SANDERS, Jr and E. A. STARKE, Jr (eds.): 'Aluminum-lithium alloys II'; 1984, Warrendale, PA, The MetallurgicalSociety of AIME.

*18. C. BAKER, P. 1. GREGSON, S. J HARRIS, and c. J. PEEL (eds.):'Aluminium-lithium alloys III'; 1986, London, the Instituteof Metals.

*19. c. CHAMPIER, B. DUBOST, D. MIANNAY, and L. SABETY (eds.):'Proc. 4th int. conf. on aluminium-lithium alloys', J. Phys.,C3, (9), 1987, Paris, Les Editions de Physique.

*20. T. H. SANDERS, Jr and E. A. STARKE, Jr (eds.): in 'Proc. 5th int.conf. on aluminium-lithium alloys'; 1989, Edgbaston, UK,Materials and Component Engineering Publications.

21. E. A. STARKE, Jr, T. H. SANDERS, Jr, and I. G. PALMER: J. Met.,1981, 33, (8), 24.

22. R. GRIMES, A. J CORNISH, W. S. MILLER, and M. A. REYNOLDS:Met. Mater., 1988, 4, (7), 357.

23. c. A. STUBBINGTON: Met. Mater., 1988,4, (7), 424.24. T. H. SANDERS, Jr, and E. s. BALMUTH: Met. Prog., 1978, 121,

(3),32.25. R. C. DORWARD and T. R. PRITCHETT: Mater. Des., 1988, 9,

(2),63.26. R. ~. JAMES: in 'Metals handbook', 10 edn, Vol. 2, 178-199;

1990, Materials Park, OR, ASM International.27. w. E. QUIST and G. H. NARAYANAN: in 'Aluminum alloys:

Contemporary research and applications', (ed.A. K. Vasudevan and R. D. Doherty), Treatise on MaterialsScience and Technology, Vol. 31, 219-254; 1990, New York,Academic Press.

28. c. J. PEEL: Mater. Sci. Technol., 1986, 2, 1169.29. E. s. BALMUTH and R. SCHMIDT: in Ref. 16, p. 69.30. w. E. QUIST, G. H. NARAYANAN, and A. L. WINGERT: in Ref. 16,

p.313.31. J. c. EKVALL, J. E. RHODES, and G. G. WALD: In 'Design of

fatigue and fracture resistant structures', STP 761, 328-341;1982, Philadelphia, PA, American Society for Testing andMaterials.

32. R. E LEWIS, D. WEBSTER, and I. G. PALMER: 'A feasibility studyfor development of structural aluminum alloys from rapidlysolidified powders for aerospace structural applications',Final Report No. AFML-TR-78-102, Lockheed Palo AltoResearch Laboratory, Palo Alto, CA, 1978.

33. B. NOBLE, S. J. HARRIS, and K. DINSDALE: J. Mater. Sci., 1982,17, 461.

*34. T. H. SANDERS, Jr and E. A STARKE, Jr: Acta Metall. 1982, 30,927.

*35. K. V. JATA and E. A. STARKE, Jr: Metall. Trans., 1986, 17A,1011.

* Important references on fatigue of AI-Li alloys.

36. w. s. MILLER, M. P. THOMAS, D. J LLOYD, and D. CREBER:Mater. Sci. Technol., 1986,2, 1210.

37. D. DEW-HUGHES, E. CREED, and w. S. MILLER: Mater. Sci.Technol., 1988, 4, 106.

38. J. GLAZER, S. L. VERZASCONI, E. N c. DALDER, W. YU, R. A.EMIGH, R. O. RITCHIE, and J. W. MORRIS: Adv. Cryogen. Eng.,1986, 32, 397.

39. D. WEBSTER: Met. Prog., 1984, 125, (4), 33.40. R. C. DORWARD: Scr. Metall., 1986, 20, 1379.41. K. T. VENKATESWARA RAO, W. YU, and R. O. RITCHIE: Metall.

Trans., 1989, 20A, 485.42. K. V. JATA and E. A. STARKE, Jr: Scr. Metall., 1988, 22, 1553.43. s. SURESH, A. K. VASUDEVAN, M. TOSTEN, and P. R. HOWELL:

Acta Metall., 1987, 35, 25.*44. A. K. VASUDEVAN, R. D. DOHERTY, and s. SURESH: in 'Aluminum

alloys: Contemporary research and applications', (ed.A. K. Vasudevan and R. D. Doherty), Treatise on MaterialsScience and Technology, Vol. 31,445-462; 1990, New York,Academic Press.

45. R. O. RITCHIE: Mater. Sci. Eng., 1988, AI03, 15.46. A. G. EVANS: in 'Fracture mechanics: Perspectives and direc-

tions (20th symp.)', (ed. R. P Wei and R. P Gangloff), STP1020,267-291; 1989, Philadelphia, PA, American Society forTesting and Materials.

47. H. M. FLOWER and P. J GREGSON: Mater. Sci. Technol., 1987,3,81.

48. s. F. BAUMANN and D. B. WILLIAMS: Metall. Trans., 1985, 16A,1203.

49. T. H. SANDERS, Jr, and E. A. STARKE, Jr: in Ref. 20, p. 1.50. D. B. WILLIAMS: in Ref. 20, p. 551.51. M. H. TOSTEN, A. K. VASUDEVAN, and P. R. HOWELL: Metall.

Trans., 1988, 19A, 51.52. R. J. RIOJA, P. E. BRETZ, R. R. SAWTELL, W. H. HUNT, and

E. A. LUDWICZAK: in 'Aluminium alloys: Their physical andmechanical properties', (ed. T. R. Sanders, Jr andE. A. Starke, Jr), 1781-1797; 1986, Warley, UK, EngineeringMaterials Advisory Services.

53. J. C. HUANG and A. J. ARDELL: in Ref. 18, p. 455.54. D. B. WILLIAMS and P. R. HOWELL: in 'Aluminum alloys:

Contemporary research and applications', (ed. A. K. Vasude-van and R. D. Doherty), Treatise on Materials Science andTechnology, Vol. 31, 365-388; 1990, New York, AcademicPress.

55. J. M. PAPAZIAN and P. S. GILMAN: Mater. Sci. Eng., 1990,A125, 121.

56. F. W. GAYLE and B. VANDERSANDE: Acta Metall., 1989, 37,1033.

*57. J. DHERS, J. DRIVER, and A. FOURDEUX: in Ref. 18, p. 233.*58. K. T. VENKATESWARA RAO and R. O. RITCHIE: Metall. Trans.,

1991, 22A, 191.59. K. T. VENKATESWARA, RAO and R. O. RITCHIE: Mater. Sci.

Technol., 1989, 5, 882.60. A. K. VASUDEVAN, W. G FRICKE, R. C. MALCOLM, R. J. BUCCI,

M. A. PRZYSTUPA, and F. BARLAT: Metall. Trans., 1988, 19A,731.

61. J. HIRSCH, O. ENGLER, K. LUCKE, M. PETERS, and K. WELPMANN:in Ref. 19, p. 605.

62. G. R. YODER, P. S. PAO, M. A. IMAM, and L. A. COOLEY: Scr.Metall., 1988,22, 1241.

63. G. R. YODER, P. S. PAD, M. A. IMAM, and L. A. COOLEY: Scr.Metall., 1989, 23, 1455.

64. c. J. WARREN and R. J. RIOJA: in Ref. 20, p. 417.65. M. PETERS, J. ESCHWEILER, and K. WELPMANN: Scr. Metall.,

1986, 20, 259.66. A. K. VASUDEVAN, M. A. PRZYSTUPA, and w. G. FRICKE: Scr.

Metall., 1990,24, 1429.67. s. J. HARRIS, B'- NOBLE, and A. DODD: in Ref. 20, p. 1061.68. c. P. BLANKENSHIP and E. A. STARKE, Jr: Fat. Fract. Eng.

Mater. Struct., 1991, 14, 103.69. K. T. VENKATESWARA RAO and R. O. RITCHIE: Acta Metall.

Mater., 1990, '38, 2309.*70. R. J. BUCCI, R. C. MALCOLM, E. L. COLVIN, S. MURTHA, and

R. S. JAMES: 'Cooperative test program for the evaluation ofengineering properties of AI- Li alloy 2090-T8X sheet, plateand extrusion products', Report NSWC TR 89-106, AlcoaTechnical Center, Pittsburgh, PA, 1989.

71. P. MEYER, Y. CANS. D. FERTON and M. REBOUL: in Ref. 19,p. 131.

72. M. A. REYNOLDS and E. CREED: in Ref. 19, p. 195.



73. w. S. MILLER, 1. WHITE, M. A. REYNOLDS, D. S. McDARMAID,and G. M. STARR: in Ref. 17, p. 151.

74. A. F. SMITH: in Ref. 20, p. 1587.75. N. J. KIM, D. RAYBOULD, R. L. BYE, and s. K. DAS: in Ref. 20,

p.123.76. M. R. MITCHELL: in 'Fatigue and microstructure', 385-437;

1979, Metals Park, OH, American Society for Metals.77. E. 1. COYNE, T. H. SANDERS, Jr, and E. A. STARKE, Jr: in Ref. 16,

p.293.78. Y. XIAO and P. BOMPARD: in Ref. 17, p. 737.

*79. N. ESWARA PRASAD, G. MALAKONDAIAH, K. N. RAJU andP. RAMA RAO: in 'Advances in fracture research', Proc. 7thlnt. Conf. of Fracture, (ed. K. Salama et al.), 1103-1112;1989, London, Pergamon Press.

80. T. s. SRIVATSAN, K. YAMAGUCHI, and E.A. STARKE, Jr: Mater.Sci. Eng., 1986, 83, 87.

81. T. S. SRIVATSAN and E. J. COYNE: Mater. Sci. Technol., 1989,5,548.

82. Y. BRECHET, F. LOUCHET, and J. L. VERGER-GAUGRY: in Ref. 19,p.709.

*83. Y. BRECHET, F. LOUCHET, and T. MAGNIN: J. Mater. Sci., 1990,25,3053.

84. S. HAN, A. TOURABI, and B. WACK: in Ref. 19, p. 701.85. J. SCHNEIDER, H.-J. GUDLADT, and v. GEROLD: in Ref. 19,

p.745.86. T. S. SRIVATSAN and E. J. COYNE, Jr: Int. J. Fatigue, 1988, 10,

91.87. D. KHIREDOINE, R. RAHOUADJ, and M. CLAVEL: Scr. Metall"

1988,22,167.*88. D. KHIREDOINE, R. RAHOUADJ, and M. CLAVEL: Acta Metall.,

1989,37, 191.*89. w. X. FENG, F. S. LIN, and E. A. STARKE, Jr: Metall Trans.,

1984, 15A 1209.*90. R. T. CHEN and E. A. STARKE, Jr: Mater. Sci. Eng., 1984, 67,

229.91. B. VELTEN, A. K. VASUDEVAN, and E. HORNBOGEN: Z. Metallkd.,

1989, 80, 21.*92. J. M. GENTZBITTEL and F. FOUGERES: Scr. Metall., 1987, 21,

1411.*93. Y. BRECHET, F. LOUCHET, c. MARCHIONNI, and 1. L. VERGER-

GAUGRY: Philos. Mag., 1987, 56, 353.94. J. LENOVAI, H.-J. GUDLADT, and v. GEROLD: Scr. Metall.,

1988, 22, 1755.*95. z. 01, S. SAJI, and s. HORt in Ref. 19, p. 753.96. R. E. LEE and w. J. D. JONES: J. Mater. Sci., 1974, 9, 476.97. s. H. CHEN: MS thesis, China, National Tsing Hua University,

1985.*98. P. J. E. BISCHLER and J. W. MARTIN: in Ref. 19, p. 761.99. L. FARCY, C. CARRE, M. CLAVEL, Y. BARBAUX, and D. ALIAGA:

in Ref. 19, p. 769.*100. M. PETERS, K. WELPMANN, W. ZINK, and T. H. SANDERS: in

Ref. 18, p. 239.*101. P. E. BRETZ, L. N. MUELLER, and A. K. VASUDEVAN: in Ref. 17,

p.543.102. P. 1. GREGSON, 1. NEWMAN, and A. GRAY: Mater. Sci. Technol.,

1989,5,65.103. Y. MUTOH, G. H. FAIR, B. NOBLE, and R. B. WATERHOUSE: Fat.

Fract. Eng. Mater. Struct., 1987, 10, 261.104. v. J. BOLAM, P. J. GREGSON, and A. GRAY: in Ref. 20, p. 1097.105.. R. W. HERTZBERG: 'Deformation and fracture mechanics of

engineering materials', 3rd edn; 1989, New York, Wiley.106. R. O. RITCHIE: Int. Met. Rev., 1979, 24, 205.107. s. SURESH: 'Fatigue of materials'; 1991,New York, Cambridge

University Press.108. w. ELBER: in 'Damage tolerance in aircraft structures', STP

486, 230-242; 1971, Philadelphia, PA, American Society forTesting and Materials.

109. N. WALKER and c. J. BEEVERS: Fat. Eng. Mater. Struct., 1979,1, 135.

110. K. MINAKAWA and A. J. McEVILY: Scr. Metall., 1981, 15, 633.Ill. S. SURESH, G. F. ZAMISKI, and R. O. RITCHIE: Metall. Trans.,

1981, 12A, 1435.112. R. O. RITCHIE and s. SURESH: Metall. Trans., 1982, 13A, 937.113. J. C. NEWMAN and w. ELBER (eds.): 'Mechanics of fatigue

crack closure', STP 982; 1988, Philadelphia, PA, AmericanSociety for Testing and Materials.

114. 1. E. ALLISON: 'Fracture mechanics: Eighteenth symp.', STP945, 913-933; 1988, Philadelphia, PA, American Society forTesting and Materials.


*115. K. T. VENKATESWARA RAG, W. YU, and R. O. RITCHIE: Metall.Trans., 1988, 19A, 549.

116. w. A. HERMAN, R. W. HERTZBERG, and R. JACCARD: Fat. Eng.Mater. Struct., 1988, 10, 303.

*117. S. J. HARRIS, B. NOBLE, and K. DINSDALE: in Ref. 17, p. 219.l18. P. S. PAO, K. K. SANKARAN, and J. E. NEAL: in Ref. 16, p. 307.

*119. A. K VASUDEVAN, P. E. BRETZ, A. C. MILLER, and s. SURESH:Mater. Sci. Eng., 1984, 64, 113.

120. S. SURESH, A. K. VASUDEVAN, and P. E. BRETZ: Metall. Trans.,1984, 15A, 369.

121. E. ZAIKEN and R. O. RITCHIE: Mater. Sci. Eng., 1985,70, 151.122. R. D. CARTER, E. W. LEE, E. A. STARKE, Jr, and c. J. BEEVERS:

Metall. Trans, 1984, 15A, 555.*123. K. T. VENKATESWARA RAG and R. O. RITCHIE: Mater. Sci.

Technol, 1989, 5, 896.*124. 1. PETIT, S. SURESH, A. K. VASUDEVAN, and R. C. MALCOLM: in

Ref. 18. p. 257.*125. A. K. VASUDEVAN and s. SURESH: Metall. Trans., 1985, 16A,

475.126. c. Q. CHEN and H. x. LI: in Ref. 20, p. 973.127. R. TINTILLIER, H. J. GUDLADT, V. GEROLD, and J. PETIT: in

Ref. 20, p. 1135.128. R. TINTILLIER, H. S. YANG, N. RANGANATHAN, and J. PETIT: in

Ref. 19, P. 777.129. M. PETERS, V. BACHMANN, and K. WELPMANN: in Ref. 19,

p.785.130. D. S. McDARMAID and c. J. PEEL: in Ref. 20, p. 993.131. H. D. PEACOCK and J. W. MARTIN: in Ref. 20, p. 1013:132. J. M. NEWMAN, P. J. GREGSON, P. D. PITCHER, and P. J. E.

FORSYTH: in Ref. 20, p. 1043.*133. R. S. PIASCIK and R. P. GANGLOFF: Metall. Trans., 1991, 22A,

2415.*134. R. S. PIASCIK and R. P. GANGLOFF: Metall. Trans. 1992,234,

in press.135. R. S. PIASCIK: PhD thesis, University of Virginia, 1989.136. R. P. GANGLOFF, R. S. PIASCIK, D. L. DICUS, and J. C. NEWMAN:

in 'Proc. 17th lCAS Cong.', London, 1990, Royal Aeronaut-ical Society.

*137. K. T. VENKATESWARA RAO, R. S. PIASCIK, R. P. GANGLOFF, andR. O. RITCHIE: in Ref. 20, p. 955.

~'138. w. YU, and R. O. RITCHIE: J. Eng. Mater. Technol., 1987,109,81.

139. s. SURESH: Metall. Trans., 1983, 14A, 2375.140. F. L. HADDLETON, S. MURPHY, and T. J. GRIFFIN: in Ref. 19, p.

809.141. R. O. RITCHIE and 1. LANKFORD (eds): 'Small fatigue cracks';

1986, Warrendale, PA, The Metallurgical Society of AlME.142. K. J. MILLER and E. R. DE LOS RIOS (eds). 'The behaviour of

short fatigue cracks'; 1986,London Institution of MechanicalEngineers.

143. s. SURESH and R. G. RITCHIE: Int. Met. Rev., 1984, 29, 445.1<144.K. T. VENKATESWARA RAO, W. YU, and R. O. RITCHIE: Metall.

Trans., 1988, 19A, 563.*145. M. R. JAMES: Scr. Metall., 1987, 21, 783.146. Z. CHEN, A. ZEGHLOUL, and J. PETIT: Scr. Metall., 1989, 23,

1005.*147. x. SU, J. M. LARSEN, and J. R. JIRA: in 'Fatigue '90', Proc. of

4th Int. Conf on Fatigue and Fatigue Thresholds, (ed.H. Kitagawa and T. Tanaka), 1049-1054; 1990, Edgbaston,UK, Materials and Component Engineering Publications.

*148. J. M. NEWMAN, P. J. GREGSON, P. J. E. FORSYTH, and P. D.PITCHER: Scr. Metall., 1989, 23, 1157.

149. D. J. NICHOLLS and J. W. MARTIN: Fat. Fract. Eng. Mater.Struct., 1990, 13, 83.

150. D. J. NICHOLLS and J. W. MARTIN: Mater. Sci. Eng., 1990,A128, 141.

*151. K. T. VENKATESWARA RAO, W. YU, and R. O. RITCHIE: Eng.Fract. Mech., 1988, 31, 623. "

*152. R. W. HERTZBERG, W. A. HERMAN, and R. O. RITCHIE: Scr.Metall., 1987, 21, 1541.

*153. K. T. VENKATESWARA RAO and R. o. RITCHI~: Acta Metall.,1988,26,2949. f

154. R. P. WEI and R. I. STEPHENS (eds.): 'Fatigue crack growthunder spectrum loads', STP 595; 1976, Philadelphia, PA,American Society for Testing and Materials.

155. c. M. WARD-CLOSE, A. F. BLOM, and R. O. RITCHIE: Eng. Fract.Mech., 1989, 32, 613.

156. D. J. ALEXANDER and J. F. KNOTT: in 'Fatigue '87', Proc. of3rd Int. Conf. on Fatigue and Fatigue Thresholds, (ed.


R. O. Ritchie and E. A. Starke, Jr), 395-406; 1987, Warley,UK, Engineering Materials Advisory Services.

157. c. ESPINASSE and c. BATHIAS: in Ref. 19, p. 793.*158. N. OHRLOFF, A. GYSLER, and G. LUTJERING: in Ref. 19, p. 801.

159. M. T. YU and T. H. TOPPER: J. Eng. Mater. Technol., 1985,107, 19.

160. E. ZAIKEN and R. O. RITCHIE: Eng. Fract. Mech., 1985,22, 35.161. G. V. SCARICH and P. E. BRETZ: 'Fatigue crack growth

resistance of aluminum alloys under spectrum loading: Vol. I- Commercial 2XXX and 7XXX alloys.' Technical ReportNo. NOR 85-141, Northrop Corporation, Hawthorne, CA,1985.

*162. G. V. SCARICH, K. M. BRESNAHAN, and P. E. BRETZ: 'Fatiguecrack growth resistance of aluminum alloys under spectrumloading: Vol. II - Aluminum-lithium alloys', Technicalreport No. NOR -85-141, Northrop Corporation, Haw-thorne, CA 1985.

163. w. E. QUIST, G. H. NARAYANAN, A. L. WINGERT, and T. M. F.RONALD: in Ref. 18, p. 625.

164. N. J. KIM, R. L. BYE, and s. K. DAS: in Ref. 19, p. 309.165. W. WANG and N. J. GRANT: in Ref. 17, p. 447.166. R. D. SCHELLENG: JOM, 1989,41, (1),32.167. P. S. GILMAN: in Ref. 17, p. 485.168. F. W. GAYLE, N. F. LEVOY, and J. B. VANDERSANDE: J. Met.,

1987, 39, (5), 33.169. H. KEMPER, B. WEISS, and R. STICKLER: in Proc. 8th Light

Metals Cong., Leoben, June 1987, Institut fUr Metallkundeund Werkstoffprufung, Montanuniversitat.

170. w. RUCH and E. A. STARKE, JR: in Ref. 18, p. 121.*171. H. KEMPER, B. WEISS, and R. STICKLER: in 'Fatigue 87', Proc.

of 3rd Int. Conf. on Fatigue and Fatigue Thresholds, (ed.R. O. Ritchie and E. A. Starke, Jr), 789-800; 1987, Warley,UK, Engineering Materials Advisory Services.

172. K. V. JATA, W. RUCH, and E. A. STARKE, Jr: in 'Light metals',55-62; 1985, Europe, Materials Research Society.

173. H. KEMPER, B. WEISS, and R. STICKLER: Eng. Fract. Mech.,1989, 32, 591.

174. M. D. WRIGHT and c. J. BEEVERS: in Ref. 20, p. 1087.*175. K. T. VENKATESWARA RAO and R. O. RITCHIE: Mater. Sci. Eng.,

1988, 100, 23.176. K. VENKATESWARA RAO, J. C. McNULTY, and R. O. RITCHIE:

unpublished work.177. R. C. MALCOLM, F. J. CORDIER, and R. J. BUCCI: 'The effects

of various thermal conditions on the tensile properties ofaluminum alloy 2090- T8E41 plate', Report No. 56-86-AH429, Alcoa Technical Center, Pittsburgh, PA 1986.

178. R. M. PELLOUX, R. E. STOLTZ, and J. A. MOSKOWITZ: Mater.Sci. Eng., 1976, 25, 193.

179. R. P. GANGLOFF: in 'Environment induced cracking of metals',(ed. R. P. Gangloff and M. B. Ives), 55-109; 1990, Houston,TX, National Association of Corrosion Engineers.

180. M. GAO, P. S. PAO, and R. P. WEI: Metall. Trans., 1988, 19A,1739.

181. R. P. GANGLOFF and D. J. DUQUETTE: in 'Chemistry andphysics of fracture', (ed. R. M. Latanision and R. H. Jones),612-645; 1987, The Netherlands, Martinus Nijhoff.

182. R. E. STOLTZ and R. M. PELLOUX: Metall. Trans., 1972, 3,2433.

183. c. P. DERVENIS, E. I. MELETIS, and R. F. HOCHMAN: Mater. Sci.Eng., 1988, AI02, 151.

184. M. REBIERE and T. MAGNIN: Mater. Sci. Eng., 1990, AI2S, 99.185. R. E. RICKER: PhD thesis, Troy, NY, Rensselaer Polytechnic

Institute, 1982.*186. P. S. PAO, M. A. IMAM, L. A. COOLEY, and G. R. YODER:

Corrosion, 1989, 45, 530.*187. R. S. PIASCIK and R. P. GANGLOFF: in 'Environment induced

cracking of metals', (ed. R. P. Gangloff and M. B. Ives),233-240; 1990, Houston, TX, National Association of Cor-rosion Engineers.

188. M. o. SPIEDEL: Metall. Trans., 1975, 6A, 631.189. A. GRAY, N. J. H. HOLROYD, and J. WHITE: in Ref. 20, p. 1175.190. J. G. RINKER, M. MAREK, and T. H. SANDERS, Jr: Mater. Sci.

Eng., 1984, 64, 20l191. R. C. DORWARD and K. R. HASSE: Corrosion, 1988, 44, 932.192. N. J. H. HOLROYD, A. GRAY, G. M. SCAMANS, and R. HERMANN:

in Ref. 16, p. 310.193. K. T. VENKATESWARA RAO, D. KOVAR, and R. O. RITCHIE:

unpublished work, 1990.194. N. J. H. HOLROYD and D. HARDIE: Corros. Sci., 1983, 23, 527.195. A. M. GREEN and J. F. KNOTT: in 'Advances in fracture research',

Proc. of 7th Int. Conf. of Fracture, (ed. K. Salama et al.),1747-1756; 1989, London, Pergamon Press.

196. F. J. BRADSHAW and c. WHEELER: Int. J. Fract. Mech., 1969,5,255.

197. R. J. SELINES and R. M. PELLOUX: Metall. Trans., 1972, 3,2525.

198. R. P. WEI and R. P. GANGLOFF: in 'Fracture mechanics:Perspectives and directions', STP 1020, (ed. R. P. Wei andR. P. Gangloff), 233-264; 1989, Philadelphia, PA, AmericanSociety for Testing and Materials.

199. R. P. GANGLOFF and R. O. RITCHIE: in 'Fundamentals ofdeformation and fracture', (ed. B. A. Bilby et al.), 529-558;1985, Cambridge, UK, Cambridge University Press.

200. K. T. VENKATESWARA RAO and R. O. RITCHIE: Scr. Metall.,1989, 23, 1129.

201. L. M. MA, J. K. HAN, R. L. TOBLER, R. P. WALSH, and R. P. REED:Adv. Cryogen. Eng., 1990, 34, 1143.

202. R. L. TOBLER, J. K. HAN, L. MA, R. P. WALSH, and R. P. REED:in Ref. 20, p. 1115.

203. Y. B. XU, Z. G. WANG, Y. ZHANG, H. H. ZHAO, and z. Q. HU:in Ref. 20, p. 1147.

204. P. K. LIAW and M. E. FINE: Metall. Trans., 1981, 12A, 1927.205. J. McKITTRICK, P. K. LIAW, S. I. KWUN, and M. E. FINE: Metall.

Trans., 1981, 12A, 1535.206. P. K. LIAW and w. A. LOGDSON: Acta Metall., 1988,36, 1731.

*207. K. T. VENKATESWARA RAO, R. J. BUCCI, K. V. JATA, andR. O. RITCHIE: Mater. Sci. Eng., 1991, A141, 39.

208. R. CROOKS and M. R. MITCHELL: in Ref. 20, p. 1023.


THE PREHISTORY O'F METALLURGYIN THIE

BRITISH IISlESR. F. Tyle(:ote

Based upon the author's 'Metallurgy in .Archaeology' (published in 1962),which was on'e of the few books on this subject and became a minor classicin its time, this revised text incorporates the results of work done in thescientific investigations of archaeology between 1960 ond 1982.

Since this is a subject of interest to metcJllurgists as well as archaeologists,the work has not been treated as a textbook on metallurgy for thearchaeologist, but is addressed to both (Jrchaeologists and metallurgists.

CONTENTSThe native metalsCopper and copper alloys

, ;;;.;~UU:::I' The production ond properties of tin and its alloys;".;~;:... Lead, silver and clntimony

Fabrication....,<i The coming of iron

. The Roman Iron A.geIron in the early medieval period

Water-powered bloomeries: the end of an eraThe charcoal blast furnace and the finery

Fuels and their ashesAppendixesIndex

Book 319 (1986) 297 x 210 mm 256pp ISBN0 904357 72 4 Casebound

UK £25.00 (Institute of Materials and Historical Metallurgy Society membersdiscount price £20,00)Overseas US$50,OO(Institute of Materials and Historical Metallurgy Societymembers discount price US$40.00)

Please send orders with remittance, quoting membership number (whereapplicable) to:

The Institute of MaterialsSales & Marketing Department1 Carlton House TerraceLondonSW1Y5DBTel: 071-839 4071; Fax: 071-839 2078; Tlx:8814813

:l tr`p zq lw`xtyt`x&wt st`x lwwzd^ · (fj dibjo h`o\ggpmbt9 j> h`^c\id^\g \ggjtdib9 mj...

Documents