Upload
others
View
3
Download
0
Embed Size (px)
Citation preview
//U
A Progress ReportGrant No. NAG-1-745-2
April 1, 1987 - April 30, 1989
ENVIRONMENT ASSISTED DEGRADATION MECHANISMS INADVANCED LIGHT METALS
Submitted to:
Mr. Dennis DicusM/S 188A, Metallic Materials Branch
National Aeronautics and Space AdministrationLangley Research Center
Hampton, VA 23665
Submitted by:
Richard P. GangloffAssociate Professor
Glenn E. StonerProfessor
Department of Materials ScienceUniversity of Virginia
Robert E. SwansonAssistant Professor r
Department of Materials Engineering /Virginia Polytechnic Institute and State University
Report No. UVA/528266/MS88/102
June 1988
(^ASA-CB-182307) E K S I B C N M M ASSISTEDB1GBACATJCK EICfi-UlSHS IK AE?OC*i>K|1AI£ Progress Eepcr t (Virginiaj.te c CSCL
N88-24746
G3/26 0115912
SCHOOL OF ENGINEERING AND
APPLIED SCIENCE
DEPARTMENT OF MATERIALS SCIENCE
UNIVERSITY OF VIRGINIA
CHARLOTTESVILLE, VIRGINIA 2290I
https://ntrs.nasa.gov/search.jsp?R=19880015362 2020-05-30T02:59:08+00:00Z
A Progress ReportGrant No. NAG-1-745-2
April 1, 1987 - April 30, 1989
ENVIRONMENT ASSISTED DEGRADATION MECHANISMS INADVANCED LIGHT METALS
Submitted to:
Mr. Dennis DicusM/S 188A, Metallic Materials Branch
National Aeronautics and Space AdministrationLangley Research CenterHampton, VA 23665
Submitted by:
Richard P. GangloffAssociate Professor
Glenn E. StonerProfessor
Department of Materials ScienceUniversity of Virginia
Robert E. SwansonAssistant Professor
Department of Materials EngineeringVirginia Polytechnic Institute and State University
Department of Materials Science
SCHOOL OF ENGINEERING AND APPLIED SCIENCE
UNIVERSITY OF VIRGINIA
CHARLOTTESVILLE, VIRGINIA ,
Report No. UVA/528266/MS88/102 Copy No.
June 1988
ENVIRONMENT ASSISTED DEGRADATION MECHANISMS INADVANCED LIGHT METALS
INTRODUCTION
This report summarizes the research progress which has beenachieved on NASA-LaRC Grant NAG-1-745 during the period fromDecember 1, 1987 to May 31, 1988.
A multifaceted research program on environmental failuremechanisms in advanced light metallic alloys was initiated inApril of 1987 [1]. The current participants in this NASA-sponsored work include two faculty and three PhD level graduatestudents in the Department of Materials Science at the Universityof Virginia, one faculty and an MS graduate student in theDepartment of Materials Engineering at the Virginia PolytechnicInstitute and State University, and a NASA/LaRC based employeeenrolled in the PhD program at UVa. The NASA grant monitor forthis program is D.L. Dicus of the Metallic Materials Branch atthe Langley Research Center.
This program has been recently expanded to include twoadditional faculty and two graduate students in Materials Scienceat UVa [2]. This new work on the NASA-UVa Light Alloy TechnologyProgram is about to initiate and will not be reported here.
Problem
Problems of long-term environmental degradation and failurecould impede reliable applications of advanced light metallicalloys and composites in aerospace structures. Materials ofparticular interest include wrought Al-Li-X alloys, powdermetallurgy Al-Fe ternary and quaternary alloys, and titaniumaluminides. Environments include terrestrial aqueouselectrolytes, cryogenic and elevated temperatures, and gaseoushydrogen. The challenge at hand is to examine alloys; which havebeen developed based on short term, benign environment strengthand ductility; in terms of resistance to localized corrosion,hydrogen embrittlement, stress corrosion cracking and corrosionfatigue during prolonged environmental exposures.
Several deficiencies are generically relevant to theperformance of advanced light metallics.
+ Predictions of long terra durability and reliability,based on short term laboratory data, are hindered by alack of experimental observations, fundamentalunderstanding, and predictive models of timedependencies.
+ Laboratory measurements have not separated andcharacterized the localized processes which arerelevant to complex failure modes. Data are notscalable to predict performance.
+ Hydrogen contributions to localized corrosion,deformation and brittle fracture have not been studiedsystematically.
+ Fatigue-environment interactions have not been studied,particularly with regard to transgranular crackpropagation relevant to thin sheet.
+ The effects of defects common to advanced PM alloys andservice have not been characterized.
Program Objectives and Output
The general goals of the research program are tocharacterize alloy behavior quantitatively and to developpredictive mechanisms for environmental failure modes. Successesin this regard will provide the basis for metallurgicaloptimization of alloy performance, for chemical control ofaggressive environments and for engineering life prediction withdamage tolerance and long term reliability.
Current Projects and Personnel
1. DAMAGE LOCALIZATION MECHANISMS IN AQUEOUS CHLORIDE CORROSIONFATIGUE OF ALUMINUM-LITHIUM ALLOYS
R.P. Gangloff and Robert S. Piascik, PhD candidate
2. MEASUREMENTS AND MECHANISMS OF LOCALIZED AQUEOUS CORROSIONIN ALUMINUM-LITHIUM ALLOYS
G.E. Stoner and Rudolph G. Buchheit, Jr., PhD candidate
3. DEFORMATION AND FRACTURE OF ALUMINUM-LITHIUM ALLOYS:THEEFFECT OF DISSOLVED HYDROGEN
R.E. Swanson and graduate student to be recruited
4. DEFORMATION AND FRACTURE OF ALUMINUM LITHIUM ALLOYS: THEEFFECT OF CRYOGENIC TEMPERATURES
Richard P. Gangloff and John A. Wagner, PhD candidate
5. ELEVATED TEMPERATURE CRACK GROWTH IN ADVANCED PM ALUMINUMALLOYS
Richard P. Gangloff and William C. Porr, Jr. , a PhDcandidate
Research Status
Research progress for the period from April 1, 1987 toNovember 31, 1987 was reported previously [3]. Work conductedduring the past six months is summarized below.
Appendix I contains publications which were issued undergrant sponsorship. Appendix II summarizes grant travel andconference participation.
References
1. R.P. Gangloff, G.E. Stoner and M.R. Louthan, Jr.,"Environment Assisted Degradation Mechanisms in Al-LiAlloys", University of Virginia, Proposal No. MS-NASA/LaRC-3545-87, October, 1986.
2. T.H. Courtney, R.P. Gangloff, G.E. Stoner and H.G.F.Wilsdorf, "The NASA-UVA Light Alloy Technology Program",University of Virginia, Proposal No. MS_NASA/LaRC-3937-88,March, 1988.
3. R.P. Gangloff, G.E. Stoner and R.E. Swanson, "EnvironmentAssisted Degradation Mechanisms in Al-Li Alloys", Universityof Virginia, Report No. UVA/528266/MS88/101, 'January, 1988.
SUMMARY OF RESEARCH (December 1. 1987 to May 31. 1988) and PLANS
Program 1 DAMAGE LOCALIZATION MECHANISMS IN AQUEOUS CHLORIDECORROSION FATIGUE OF ALUMINUM-LITHIUM ALLOYS
Robert S. Piascik and R.P. Gangloff
Objective. The objective of this PhD study is tocharacterize and understand intrinsic fatigue crack propagationin the class of new Al-Li-X alloys, with emphasis on the damagemechanisms for transgranular environmentally assisted cracking.
Approach. Experimental design and analysis were presentedin the last progress report [1]. Work on corrosion fatigue inaluminum-lithium-copper alloy 2090 is progressing on threefronts; specifically: 1) intrinsic fatigue crack propagationkinetics, 2) gaseous environmental effects and 3) aqueouselectrolyte environmental effects.
Results. Generally, the data presented previously have beenconfirmed and expanded during this reporting period. Resultssummarized in Figs. 1 through 5 are, perhaps, the most completeand rigorous set of Al-Li corrosion fatigue crack propagationkinetics available worldwide. Results are, as yet, insufficientto quantitatively define the contributions of hydrogen,dissolution/film rupture and film-plastic deformation mechanismsto corrosion fatigue crack propagation. Progress has, however,been recorded in developing qualitative understanding of thesecomplex mechanisms. This work is presented in an appendixedpaper [2] and extended abstract, and is summarized here.
Intrinsic Fatigue Crack Propagation Kinetics. Twosignificant findings have been recorded during the past sixmonths. Firstly, constant stress intensity experiments provethat stress intensity range (AK) can be decreased substantiallywithout delay retardation, if the maximum stress intensity ismaintained constant. This result supports the use of the step-decreased AK/increased R/constant Kmax technique, which is thebasis for the current work.
Secondly, we have found that near threshold fatigue crackgrowth rates are increased significantly and the threshold stressintensity range is decreased when the single edge crackedspecimen is rigidly gripped to prevent free, in plane rotation.This result is illustrated in Fig. 1. The two dashed lines andthe data points represent the intrinsic FCP behavior for twoorientations of alloy 2090 as freely rotating SEN specimens andfor the increasing stress ratio methods described previously [1].These data were recently confirmed by a standardized compacttension experiment which included measurement of crack closure.
The shaded band represents the behavior of the same materialunder fixed grip conditions. Enhanced fatigue crack growth isobserved for each orientation, LT (upper line) and ST (lowerline). These new data approximate the very high fatigue crackgrowth rates reported by Ritchie and coworkers for small surfacecracks.
There are currently two explanations for the effect shown inFig. 1. While we employed a fixed grip stress intensity solutionwhich has been confirmed by extensive 3-dimensional finiteelement analyses (by the mechanics group at the General ElectricAircraft Engine laboratories), the gripping system may not havebeen sufficiently rigid to eliminate rotation. The actual stressintensity could therefore be higher than that plotted in Fig.land perhaps in agreement with the rotating specimen case.Alternately, the effect in Fig. 1 may be real and due to fixedgrip suppression of Mode II crack opening displacements. Theeffect of pure Mode I opening versus a mixed Mode I/II opening onfatigue crack tip damage is not known, particularly for the casewhere crack closure is unimportant.
If real, the result in Fig. 1 is important tointerpretations of small surface crack fatigue data which havehere-to-fore been ascribed to microstructural effects on Mode Iopening strain. Understanding in this regard is basic to lifeprediction models of surface fatigue and corrosion fatigue crackgrowth, and to relevant laboratory experiments for evaluations ofenvironmental and metallurgical effects. Additional experimentsare in progress to determine the importance of the two possibleexplanations and the generality of the effect.
Aqueous Electrolyte Environmental Effects. The aqueousenvironment effects introduced in the previous progress report[1], have been confirmed by additional corrosion fatigueexperimentation and fractographic analyses. The data arecompiled in Fig. 2 and several conclusions are supported.
Firstly, Alloy 2090 (peak aged) is susceptible to corrosionfatigue crack propagation in aqueous 1* NaCl (full immersion)under constant anodic polarization at -760 mV (SCE). At lowAK/high R, near threshold rates are significantly increasedrelative to cracking in moist air, with growth occurring by acleavage process. High AK/low R growth rates are increased byabout a factor of 3, with highly deflected slip band cracking thedominant microscopic mode.
Cathodic polarization and loading frequency effects are notindicative of hydrogen embrittlement, at least based on classicalarguments. As shown in Fig. 2, corrosion fatigue crack growth inNaCl is decreased by an applied cathodic potential of -1160 mV(SCE), with resulting rates about equal to those observed formoist air. Constant stress intensity experiments show that
crack propagation at the anodic potential is arrested by cathodicpolarization.
Results in Fig. 2 were obtained at constant frequency, 5 Hz.As shown in Fig. 3, corrosion fatigue crack growth (1% NaCl,anodic potential) is reduced by decreased loading frequency forhigh AK/low R, and is insensitive to frequency for nearthreshold cracking. The scatter bands represent the maximumvariation in point by point, secant-based growth rates observedduring constant AK experimentation at each frequency. Theclassical frequency effect reported for high strength aluminumalloys and characteristic of hydrogen environment embrittlementis not observed. Rather, the trends in Fig. 3 suggest that timedependent chemical reactions and mass transport, a requisite forhydrogen environment embrittlement, do not rate limit corrosionfatigue. Considering fractographic evidence and the effect ofcathodic polarization, it is unlikely that the beneficial effectof low loading frequency is due to crack tip blunting by longterm corrosion.
We currently believe that the aqueous environment corrosionfatigue response of alloy 2090 is rationalized based on surfacefilm effects on crack tip damage. Those environmental conditionswhich favor the presence of a protective film; namely low strainrate (frequency), cathodic polarization, and passivating ionssuch as lithium (see the LiaCOa experiment, Fig. 2); mitigatecorrosion fatigue crack propagation. Corrosion product inducedclosure is not an important factor here, given the high stressratios examined. Surface films may alter crack tip dislocationprocesses or affect the kinetics of hydrogen production andsubsequent uptake into the metal. These arguments are highlyspeculative; experiments are being devised to further ourunderstanding in this regard.
Gaseous Environmental Effects. All corrosion fatiguecrack growth rate data for the LT orientation of alloy 2090,stressed in highly purified gaseous environments, are presentedin Fig. 4. These results were reported previously; no newexperiments have been conducted in the past six months.
The data in Fig. 4 support the hypothesis that surface filmsmitigate environmental fatigue crack propagation rates, apartform crack closure effects. The data for water vapor alsosuggest that hydrogen may embrittle aluminum-lithium basedalloys, at least at high stress intensities.
The helium and oxygen environmental data shown in Fig. 4 areparticularly important from both the aerospace design and damagemechanism perspectives. During this reporting period, we haveconsidered the possibility that these data are influenced bytrace impurities, particularly water and hydrocarbons, in the
ultra-high vacuum system described previously [1]. We areparticularly concerned about the purity of nitrogen cold trapped,high purity (99.998%) He, and about contamination during verylong term fatigue experimentation in the static gas environment.The question at hand is whether very low levels of water vaporcan exacerbate environmental cracking. As described for Program5, we have assembled a gas purification system to examine theseconcerns. The data contained in Fig. 4 will be confirmed byexperiments during the next reporting period.
General Trends. Environment affects fatigue crackpropagation in peak aged alloy 2090. At high stress intensity,the order of growth rate ranking from fastest to slowest isaqueous chloride (anodic), aqueous chloride (cathodic), watervapor, purified helium, moist air and oxygen. Oxygenparticularly retards crack growth, while unexpectedly high ratesare observed for pure helium. At near-threshold stressintensities, chloride and helium environments promote fatigue,with water vapor, moist air, cathodic polarization and pureoxygen progressively decreasing environmental crack growth rates.
We have also characterized the intrinsic corrosion fatiguecrack propagation of peak aged alloy 7075. The data compiled inFig. 5 show that the resistance of the advanced aluminum-lithiumalloy to transgranular environmental cracking is better than thatof the current aerospace alloy 7075. The high chemicalreactivity associated with lithium does not result in abnormallyhigh rates of corrosion fatigue for the conditions examined.
Conclusions.
1. The constant Kmax/step increased constant AK and R methodaccurately characterizes the intrinsic fatigue crack growthbehavior of Al-Li alloys.
2. Near threshold, intrinsic fatigue crack growth rates may beenhanced substantially by fixed grip loading, due tosuppressed Mode II crack opening displacements.
3. Alloy 2090 is susceptible to corrosion fatigue crackpropagation in aqueous NaCl under anodic polarization, withthe effect particularly significant near threshold.
4. Classical hydrogen environment embrittlement is notsupported by experimental observations. Rather, crack tipfilms play a role in corrosion fatigue damage as evidencedby 1) increased da/dN with increasing frequency, 2) reducedda/dN upon cathodic polarization and 3) reduced da/dN due toLi2C03 additions.
5. Fatigue crack growth rates in gaseous environments support
8
the concept that surface films are critical to environmentalfatigue damage.
6. Alloy 2090 is intrinsically more resistant to corrosionfatigue crack propagation compared to conventional alloy7075.
Future Work.
1. Prove conclusion #2.
2. Devise experiments to substantiate conclusion #4.
3. Establish the effect of trace environment impurities onfatigue crack propagation in high purity gaseousenvironments.
4. Develop correlations between environmental crack growthrates and high resolution fractographic observations.
It is anticipated that Mr. Piascik's research project willbe completed by the end of year 2 in April, 1989.
References
1. R.P. Gangloff, G.E. Stoner and R.E. Swanson, "EnvironmentAssisted Degradation Mechanisms in Al-Li Alloys", Universityof Virginia, Report No. UVA/528266/MS88/101, January, 1988.
2. R.S. Piascik and R.P. Gangloff, "Aqueous Environment Effectson Intrinsic Corrosion Fatigue Crack Propagation in an Al-Li-Cu Alloy", submitted, Environmental Cracking of Metals,M.B. Ives and R.P. Gangloff, eds., NACE, Houston, TX inreview.
-3
ALLOY 2090Peakaged, Air. 5 HzConstant AK, step increased RA-L-T, 0-L-S
I /10
Stress Intensity Range (MPa\/m)10*
Figure 1 - The effect of gripping mode on near-threshold, intrinsicfatigue crack growth rates in alloy 2090 in moist air. The shadedband represents data for fixed grip, rotation free loading, while thedata and dashed lines relate to freely rotating SEN and CT specimens.
oo
EE
O
u0v_
CJ
cn
10 -*-
10 -'-
10
ALLOY 2090 Peakaged, L-T, f=5 Hz*—D — 1x — 1
-0.760 V (SCE)+ 0.05M Li2C03 -0.760 V (SCE)
NaCI,N°CINACI, -1.16 V (SCE). A—Moist Air
T T
10
Stress Intensity Range (MPaVrrT)10
Figure 2 - Intrinsic corrosion fatigue crack growth in alloy 2090(peak aged, LT) in aqueous 1% NaCI at constant anodic and cathodicpotentials, and in 1% NaCl+0.05M LizCOa. Air data are from Fig. 1.
.5?oo
oO
o
*-o io-5-
10
ALLOY 2090. Packaged, L-T1.0* NaCI, deaerated. -760 V SCEA-AK=9.9 MPaVm". R=0.1+ -AK=2.2 MParrT R=0.8
11
T
J.2
10 1 0Cyclic Frequency (f)
10 1 0 2
Figure 3 - The effect of cyclic load frequency on corrosion fatiguecrack growth in alloy 2090 exposed to 1* NaCI at constant anodicpotential for two stress intensity/stress ratio regimes.
10 •3-
oo
O
oD
O
1 o -s
10
A HEUUM
OXYGEN X
WATER VAPOR O
ALLOY 2090, Packaged. L-T5 Hz, Constant AK, step increased R1* NaCI (-760 mV SCE), 2 kPa Helium2 kPa Water Vapor, 20 kPa Oxygen
-i—i—r
10~r-|10 2I I U
Stress Intensity Range (MPavrn)Figure 4 - Intrinsic corrosion fatigue crack growth in alloy 2090(peak aged, LT) in highly purified gaseous environments includingoxygen, helium and water vapor. Moist air data are from Fig. 1.
0RIGB7AI: PAGE ISOE POOR QUALITY
o
oQ.2
<uCTco
0:0
< -S.2"£u "^SOT**Q < <]G E c co o o
7o
To
oQ.
V
y>
To
•O0)bo
0}0)
«H (Q4->
0) CO 0)•H S4-> C<u oc s*•H -H
A: >c
c <uo
•H m4J 3co obo 0)co ena. coo bo
oCL
;o
a0.S
oin
cM COCJCO 0)i- 30 o
0)<U 33 0<bo CO•H•P {QCO 3
«H O•H
CJ Li•H COn >c•H C^ -H•Pc toM C-
o1 t^
cCO CO'*3 ObOO)•H Ota CM
(3:0X0/0101)
o o
iu) LIJMOJO
o
3n6ijoj
12
Program 2. MEASUREMENTS AND MECHANISMS OF LOCALIZED AQUEOUSCORROSION IN ALUMINUM-LITHIUM ALLOYS
Rudolph G. Buchheit, Jr. and Glenn E. Stoner
Summary of Project Results:December 1, 1987 through June 1, 1988
Objectives
The objectives of this work are:
1.) the development of experimental techniques whichsurmount the difficulties associated with monitoring solutionchemistries that develop in occluded environments;
2.) the use of these experimental techniques tomonitor the processes that contribute to the local corrosion andembrittlement of aluminum-lithium alloys.
Introduction
The crevice corrosion behavior of 2090 in 3.5 w/o NaClsolution has been examined using artificial crevice experiments.The results of artificial crevice experiments using 2090 havebeen compared to results obtained from similar experimentsperformed with 2024 and 7075. The technique used for artificiallyinducing crevice corrosion has been described in a previousreport [1].
Crevice corrosion behavior has been monitored by measuringthe pH and free corrosion potential inside and outside thecrevice as a function of time. When the environmental conditionsoutside the crevice are varied, the crevice corrosion behavior isobserved to change. The conditions varied were degree of bulksolution aeration and the degree of bulk solution agitation. Thebehavior of an isolated crevice has also been compared to that ofa crevice coupled to an external surface cathode.
Scanning electron microscopy and X-ray energy analyses havebeen used to characterize the morphology and composition of thesurfaces of specimens used in artificial crevice experiments.
Experimental Results
pH vs. Time Experiments. Two methods for simulating acrevice environment have been used to monitor pH vs. timeresponse. In the first technique a hole slightly larger than the
13
measuring pH electrode is bored in a specimen [2,3]. Theelectrode is inserted into the bore. The small gap between theelectrode and the specimen simulates the geometry of the creviceand allows pH measurements to be made. In the second technique,machine shavings of the test alloy are immersed in a small volumeof test solution [4,5]. This simulates a high metal surface areato solution volume ratio characteristic of crevice corrosionsituations. The pH of the solution is measured with acombination pH electrode. The two situations are not completelyequivalent as evidenced by the fact that the pH response isdifferent in each case. The pH that develops in 2090 shavingsexperiments is 2 to 3 pH units more alkaline than that developedin artificial crevice experiments. The differences in theresponses of the two experiments is attributed to the highermetal surface area to solution volume ratio that exists in theshavings experiments.
Results of the pH vs. time experiments indicate severaltrends. Isolated crevices (crevices which are holes filled withsolution, contain a pH electrode and are sealed) in conventionalhigh strength aluminum alloys develop acidic crevices [6]. Incontrast, isolated crevices in solution heat treated (SHT) andpeak aged 2090 become mildly alkaline. This trend is commonlyobserved in lithium bearing aluminum alloys [6]. Experimentsperformed in this study have confirmed results reported in theliterature. Alloy 2024-T3 quickly achieved a pH of 4.5 whilepeak aged 2090 developed a pH of 8.5 (Fig. 1). However, when thecrevice in 2090 is coupled to an external surface cathode(achieved by immersing the specimen in bulk solution) the crevicewill become acidic reaching a pH of 4.5.
The pH of isolated crevices in 2090 depends on alloy temper.Isolated crevices in peak aged alloys become more alkaline thanthose in under aged or SHT alloys. The solution in artificialcrevices in peak aged 2090 specimens achieve a maximum pH of 8.5to 9 compared to a pH of 7.5 to 8 for crevice solutions in theSHT alloy. A temper dependence in pH versus time response alsoexists in the shavings experiments. Solution containing peakaged shavings develop pH values between 10 and 11, while the pHof solutions containing SHT shavings generally ranges between 9and 10.
Experiments monitoring the effects of aeration and agitationof the bulk solution on the crevice pH in 2090, 2024 and 7075 arecurrently in progress.
Potential vs. Time Experiments. Potential versus timemeasurements in shavings experiments are not practical since themeasurement requires that the specimen be connected to themeasuring circuit. In these experiments the specimen is finelydivided metallic shavings. Therefore potential versus time
14
experiments can be made for artificial crevices only.
In these experiments two free corrosion potentials aremeasured. The first potential is the crevice corrosionpotential measured between a AgjjAgCl reference electrodeinserted into a crevice and the metal specimen. The secondpotential is the free corrosion potential measured between acalomel reference electrode in the bulk solution and the metalspecimen external surface.
Potential versus time response depends strongly on alloycomposition, degree of bulk solution agitation and degree ofsolution aeration. Experiments are being performed to determinethe effect of alloy temper on crevice potential versus timeresponse.
Normally, free corrosion potentials for peak aged 2090 inaerated 3.5 w/o NaCl solutions achieve a steady state value ofapproximately -710 raVsce after 100 minutes of immersion insolution. However, in artificial crevice experiments with peakaged 2090, steady state crevice and bulk potentials were achievedonly after 300 minutes. Further, in the first 300 minutes ofthese experiments, a potential spike of as much as 150 raV in thenegative direction is observed in quiescent bulk solutions (Fig.2). After the potential transient, a steady state bulk potentialof about -710 mV is then recovered. Corresponding crevicepotentials are generally 20 to 50 niV more negative than the bulkpotentials after this potential transient.
Degree of aeration and agitation of the bulk solutionaltered the transient potential response in 2090. Deaerating thebulk solution before and during immersion of the specimeneliminated the transient response; that is , the more noblepotential was never recovered (Fig. 3). Both the crevice andbulk potentials achieved a steady state value of -950 rnVsce. Thecrevice potential was no more negative than the bulk potentialsuggesting that artificially stimulated crevice corrosion wassuppressed due to the reduction in the dissolved oxygen contentin solution.
Active aeration was accomplished by bubbling air through thebulk solution. Aeration reduced the magnitude of the potentialtransient to 50 mV in the negative direction (Fig. 4). The steadystate bulk potential recovered to -730 mV after 400 minutes. Thecrevice potential was 15 mV more negative than the bulk corrosionpotential in the steady state regime.
The reduction in the size of the potential transient whenthe bulk solution was actively aerated appeared to be caused bythe agitation introduced by the bubbling action. When the bulksolution was agitated (but not actively aerated) by stirring witha magnetic stirring bar, transient response was eliminated
15
entirely (Fig. 5). Steady state potentials were achieved after1000 minutes. In this case, the crevice potential was 75 mV morenegative than the bulk potential in the steady state regime.
Potential versus time experiments have been performed on2024-T3 and 7075-T651 in quiescent bulk solutions. The creviceand bulk potential response of 2024 were similar to those of peakaged 2090 in that a 150 mV potential transient in the negativedirection before 300 minutes had elapsed (Fig. 6). However, thebulk corrosion potential eventually became more negative than thecrevice potential. This behavior was attributed to severecrevice corrosion which occurred under the stop lacquer appliedto the exterior of the specimen over an electrical connection.
In 7075, no potential transient was observed (Fig. 7).Initially, the crevice potential was 100 mV more negative thanthe bulk potential, but the difference decreased with time untilthe potentials were nearly equal after 4000 minutes.
Scanning Electron Microscopy and X-Ray Analysis. Alloy2090 specimens used in artificial crevice experiments exhibitedcorrosion morphologies with several common characteristics.Corrosion was not homogeneous over the wall of the crevice. Increvices with longer aspect ratios the regions of most aggressivecorrosion were concentrated near the crevice mouth.
Surrounding the mouth of the crevice on the specimen surfacewas a region where corrosion product build up was greater thanat other points on the specimen surface. X-ray mapping showed adramatic increase in copper at the lip of the crevice. Furtheraway from the mouth of the crevice, small amounts of copper weredeposited on surfaces in between larger areas of corrosionproduct build up. Under an optical microscope the depositsappeared to be pure, metallic copper. Copper deposits have alsobeen observed at the mouths of crevices in 2024 specimens used inartificial crevice experiments.
X-ray analyses performed on corroded surfaces were doneusing a energy dispersive spectrometer attached to a scanningelectron microscope. The solid state detector on thespectrometer is not capable of detecting elements below oxygen.As a result, detection of lithium was not feasible using thistechnique.
Conclusions
1. An alkaline pH (8.5 to 9) develops in isolatedsolutions in 2090 crevices while an acidic pH (4.5)develops in isolated solutions contained in 2024crevices.
16
2. Coupling the crevice in a 2090 specimen to an externalsurface cathode causes the pH in the crevice solutionto become acidic (pH = 4.5).
3. Crevice potential response in 2090 can be affected bychanging the degree of aeration and agitation of thebulk solution in artificial crevice experiments.
4. Copper dissolution and reduction results in theformation of copper deposits at the mouths ofartificial crevices.
Future Work
Planned research activities for 6/1/88 through 1/1/89 areaimed at understanding the following:
1.) Are the potential transients observed in artificialcrevice experiments related to the phenomenon of copperdissolution and reduction observed in Al-Cu-X ternary alloys?
2.) Does copper dissolution and reduction control localizedcorrosion in Al-Li alloys?
3.) Does the crevice chemistry (e.g. potential, pH, [Cl~],dissolved oxygen) change with depth, and if so can these changesbe monitored and used to understand the stress corrosion behaviorof Al-Li alloys?
Specific experiments to be performed will include:
1.) An attempt to determine if the potential transientobserved in copper bearing aluminum is related to the phenomenonof metallic copper deposition. Efforts will focus on corrosionactivity which occurs in the first 300 to 400 minutes of anartificial crevice immersion experiment. Initial attempts willinclude:
a.) measurement of Cu2+ ion with an ion selectiveelectrode as a function of time in solutionperiodically decanted from shavings experiments.
b.) interrupting artificial crevice experiments toanalyze copper redistribution using scanningelectron microscopy and X-ray analysis.
2.) Artificial crevice experiments will be performed on Al-Cu binary systems to determine if the copper depositionphenomenon is limited to ternary (or multicomponent) aluminumalloy systems.
17
4. J.G. Craig, R.C. Newman, M.R. Jarrett and N.J.H. Holroyd,"Local Chemistry of Stress Corrosion Cracking in Al-Li-Cu-MgAlloys," Conf. Proc. 4th International Conference on Aluminum-Lithium Alloys, to be published (1987).
5. N.J.H. Holroyd, A. Gray, G.M. Scamans and R. Hermann,"Environment Sensitive Fracture of Al-Li-Cu-Mg Alloys," Aluminum-Lithium Alloys III, C. Baker, P.J. Gregson, S.J. Harris and C.J.Peel, eds., Institute of Metals, Oxford, UK, pp. 273-281 (1986).
6. N.J.H. Holroyd, G.M. Scamans and R. Hermann, "EnvironmentalInteraction with the Crack Tip Region During EnvironmentSensitive Fracture of Aluminum Alloys," Embrittlement by theLocalized Crack Environment. R.P. Gangloff, ed., IMS, Warrendale,PA, pp. 327-347 (1984).
^RECEDING PAGE BLANK NOT FILMED
19
X(X
10
ga7
6
5
4
3
peak aged 20902024-T861
500 1000 1500
Time (minutes)
2000 2500
Figure 1. Plot of crevice solution pH versus time for peak aged2090 and 2024-T651 in an isolated crevice experiment.
"o
R03jjC0)o
(X
-700
-800
-900
1000
-
? * * * * * * * « * * **t4• • bulk
* crevice-
• 1 . 1 . 1 • 1 .2000 4000 6000
Time (minutes)
8000 10000
Figure 2. Plot of bulk and crevice corrosion potentials versustime for peak aged 2090 immersed in quiescent 3.5 w/o NaClsolution.
^_^01^j"o• f*
J^r <̂0• t-t+JG0)
a,
-600
-700
-800
-900
1 f\f\f\— 1000
+**
•+
^»
^ **** • bulkt * crevice
i +
~~ +
*
-X+
+ + + i i i i i i i i i i t 1 * 1 ' * ' '+****»i, , . • • i*+******* ' ' '***
I . 1 . 1 • ' •« A A ^^«^« &f\f\ of\n in200 400
Time (minutes)
1000
Figure 3. Plot of bulk and crevice corrosion potentials versustime for peak aged 2090 immersed in deaerated 3.5 w/o NaClsolution.
20
o>
do>.»joIX
— OUU
-700
-800
-900
mnn
L
V~ ,
__
—
• I . I
* • bulk• crevice
1 . 1 .2000 4000 6000
Time (minutes)
8000 10000
Figure 4. Plot of bulk and crevice corrosion potential versustime for peak aged 2090 immersed in actively aerated 3.5 w/o NaClsolution.
o
au*->oa.
— ouu
-700
—800
-900
mnn
.̂_
W • • • • • • • • • • •A<-» . . „ . . . * * * . . « . * « •
— • bulk» crevice
, 1 , 1 . 1 , 1 .2000 4000 6000
Time (minutes)
8000 10000
Figure 5. Plot of bulk and crevice corrosion potential versustime for peak aged 2090 immersed in agitated 3.5 w/o NaClsolution.
o
j5
*24->
c0)
ocu
— OUU
-700
-800
-900
•mnn
f • bulkit » crevice
5^\ wy/^-
1 . 1 , 1 , 1 .
2000 4000 6000
Time (minutes)
8000 10000
Figure 6. Plot of bulk and crevice corrosion potential versustime for 2024-T3 immersed in quiescent 3.5 w/o NaCl solution.
21
o
Cu
«->oa.
-600
-700
-800
-900
-1000
bulkcrevice
2000 4000 6000
Time (minutes)
8000 10000
Figure 7. Plot of bulk and crevice potential versus time for7075-T651 immersed in quiescent 3.5 w/o NaCl solution.
22
Program 3. DEFORMATION AND FRACTURE OF ALUMINUM-LITHIUMALLOYS: THE EFFECT OF DISSOLVED HYDROGEN
R.E. Swanson
Background
A new program on "Environmental Degradation of Al-Li Alloys:The Effect of Hydrogen" has been approved by NASA for startup inyear 2. The detailed proposal is presented elsewhere anddescribes a 24-month study to characterize the effects ofhydrogen on the mechanical behavior and microstructure of Alloy2090 [1].
The goal of this work is to increase our understanding ofthe role of microstructure and stress state on the susceptibilityto hydrogen damage in Al-Li alloys. Microstructuralcharacterizations will include: optical microscopy, scanningelectron microscopy, transmission electron microscopy anddifferential thermal analysis or scanning calorimetry.Mechanical property characterizations will include uniaxialtension tests, biaxial loading tests, Charpy impact tests, andmicrohardness indention tests.
Progress has been achieved in anticipation of the programstart and with the preliminary funding provided to VPI under theyear 1 proposal. Specifically, a pressurized disk biaxialstressing apparatus has been modified for hydrogen embrittlementtesting of alloy 2090. A significant portion of the year 1(1987-88) funding was not spent due to Louthan's departure, andhas been carried over to support the newly proposed program. Amasters degree graduate student has been recruited and shouldbegin work in July of 1988.
References
R.P. Gangloff, G.E. Stoner and R.E. Swanson, "EnvironmentAssisted Degradation Mechanisms in Al-Li Alloys", Universityof Virginia, Report No. UVA/528266/MS88/101, January, 1988.
23
Program 4. DEFORMATION AND FRACTURE OF ALUMINUM LITHIUMALLOYS: THE EFFECT OF CRYOGENIC TEMPERATURES
John A. Wagner and Richard P. Gangloff
Background
Cryogenic tanks are responsible for a large portion of theweight of current launch vehicles. It is envisioned that, byemploying advanced alloy systems and advanced processingtechniques, a significant savings can be realized. Superplasticforming (SPF) of Al-Li alloys offers the potential of suchsavings over current tanks fabricated from 2219 aluminum alloy.Al-Li alloy 2090 not only offers lower density and higher elasticmodulus, but also the potentially desirable characteristic ofincreasing strength and toughness with decreasing temperature.
Currently 2090 + In is being investigated for application incryogenic tanks. Indium is added to the alloy in an attempt toincrease post-SPF properties without stretching. A program atNASA-LaRC has shown that by adding In to a basic 2090 alloychemistry, strength levels could be achieved which approach theT8 condition without stretching. This approach is attractive forSPF operations since formed parts are not likely to be stretchedto achieve high strength levels.
Northrop has recently demonstrated the superplasticformability of alloys based on the 2090, 8090 and 2090 + Incompositions. Northrop was, however, unable to achieve highstrength levels in post-SPF parts of indium modified 2090.
Objective and Approach
The goal of this research is to understand and optimizethe fracture resistance of aluminum-lithium-copper alloysprocessed for thin sheet, cryogenic tankage applications. Bothnovel alloy compositions developed under NASA programs, andconventionally stretched or superplastically formed,precipitation hardened microstructures will be investigated, withmaterials purchased from Reynolds Metals.
Four issues will be emphasized:
(1) Examination of the deformation and fracture resistanceof new Al-Li-Cu-In alloys and of superplastically formedmicrostructures.
24
(2) Development of quantitative fracture mechanics methodsfor characterizing the fracture initiation and stable growthresistance of thin sheet. Experiments will be designed todiscriminate the behavior of different microstructures in away which is both mechanistically interpretable and relevantto damage tolerant designs of cryotankage structures.
(3) Determination of the micromechanical mechanism(s) forthe significant improvements in fracture (initiation)toughness observed for thick section Al-Li based alloysstressed at cryogenic temperatures.
(4) Definition of the role of dissolved hydrogen in thefracture process at both cryogenic and ambient temperatures.
Results
The initial phase of this study is concerned withdetermining if indium additions to 2090 do indeed promoteincreased strength levels in the T6 condition. Differentialscanning calorimetry studies revealed a fairly narrow window forsolution heat treatment (SHT) of the alloy. In early work atLaRC, an SHT of 560 C was employed and resulted in superiorhardenability of the In bearing alloy, particularly for aging at160 C. This effect was attributed to increased number densityand homogeneous distribution of Tl precipitates. An SHT of 535 Cwas used in the Northrop study and the beneficial effect ofadding In to the alloy was not realized. It was hypothesizedthat an SHT temperature of 535 C did not permit completedissolution of T2 for subsequent optimal age hardening by Tl.
Specimens of the Northrop material were obtained, solutiontreated at 560 C and aged for 72 hours at 160 C For the 2090 +In composition, the higher solution treatment temperatureresulted in higher hardenability.
Because of the reasonable strain to fracture of the 2090 +In alloy and the potential to obtain T8 strength levels in SPFcomponents given T6 aging treatments, this alloy and a 2090baseline were scaled from 30 pound permanent mold castings to 400pound DC castings by Reynolds. The following specific materialshave been delivered to NASA-LaRC;
Plates: l/2"x47"x34"
l/8"xl7"x34"
Compositions: Al-2.65Cu-2.17Li-0.13Zr-0.058i-0.05Fe
Al-2.60Cu-2.34Li-0.16Zr-0.04Si-0.05Fe-0.17In
25
Processing: Solution heat treat at 556 CIce water quench3% stretch or 0* stretch for In alloy
Solution heat treat at 537 CIce water quenchAge 413 C for 16 hoursHot roll at 288 C to 1/8 thickness
(This is the so-called TMT C process)
The isothermal aging response of the following combinationsof these alloy and process conditions are being defined for 160and 190 C:
2090; 1/8" plate; SHT 556 C; stretch 3*2090; 1/8" plate; TMT C; SHT 556 C2090; 1/2" plate; SHT 556 C; stretch 3%
2090+In; 1/8" plate; SHT 556 C; stretch 3%2090+In; 1/8" plate; TMT C; SHT 556 C2090+In; 1/2" plate; SHT 556 C; stretch 3%2090+In; 1/2" plate; SHT 556 C
Additionally, 2090 and 2090+In specimens obtained from Northropwill be examined in terms of aging behavior. Solution heattreatment temperatures of 537 and 558 C will be evaluated.
This hardness study will enable us to establish theinfluence of casting practice, SHT temperature, alloy addition,product form and stretching on the hardenability of 2090 basedalloys. From this initial evaluation, alloy compositions andprocessing conditions will emerge for subsequent evaluations ofthe effects of crack/sheet geometry, cryogenic temperatures anddissolved hydrogen on deformation and fracture resistance.
We have also received six square feet of 0.081" thick, 2090sheet from the Alcoa Technical Center. This material is providedin the T8E53 temper, an experimental condition, after 1 to 3*stretch. While not firmly established, it is likely that thistemper involved 10 hours at between 150 and 163 C. The yieldstrength is specified as 467 MPa (longitudinal) and 459 MPa (longtransverse). The plane stress fracture toughness is 73 MPa-m1/2
for the LT orientation and 77 MPa-m1/2 for the TL orientation,with each value obtained for 16" wide and 44" high test panels.Apparently, this material was processed for optimal strength andtoughness, with the lower temperature underaging critical in thisregard.
26
Program 5. ELEVATED TEMPERATURE CRACK GROWTH IN ADVANCED PMALUMINUM ALLOYS
William C. Porr, Jr. and Richard P. Gangloff
Background
Advanced light metallic materials are typicallyevaluated in terms of simple microstructural, tensile, toughnessand corrosion behavior. As a second stage in this development,it is critical to assess environmental fatigue and fractureresistance in terms of modern fracture mechanics concepts, and toenable defect tolerant design for component durability in avariety of aggressive aerospace environments.
A new program was initiated in April of 1988 to examine theelevated temperature subcritical crack growth and fractureresistance of advanced, powder metallurgy Al-Fe based ternary andquaternary alloys. Over the next few months, the student willreview the literatures of elevated temperature ductility and"creep crack growth" in aluminum alloys, and of time dependentelastic-plastic fracture mechanics crack tip fields andexperimental techniques. In conjunction with NASA-LaRCresearchers in metals and fatigue/fracture, Mr. Porr will developa research plan involving alloy selection and experimentaldesign.
Mr. Porr has worked on NASA program areas since the Fall of1987. This effort was enabled by the fact that Mr. Piascik wasappointed and funded as a Teaching Assistant by the MaterialsScience Department for the Fall, 1987 term. Mr. Porr filled Mr.Piascik's grant slot and was funded for 6 months by NAG-1-745.Several accomplishments were recorded.
Research
Mr. Porr has designed, specified and participated in thepurchase of an environmental system for crack growth and fractureexperimentation in controlled pure gas or vacuum environments andas a function of cryogenic to elevated temperature. The fundingfor this capital equipment was a line item in the year 1 budget.
Specifically, we have taken delivery on an ATS box furnacewhich is attachable to our servoelectric Instron machine or ourservohydraulic MTS machine. This apparatus, in conjunction witha microprocessor based controller, will allow mechanical specimentemperature control from -170 C to +450 C. Porr has designed andis assembling an ultrahigh vacuum chamber to fit within thetemperature chamber, and to contain crack growth and tensilefracture specimens. This chamber will be evacuated by the ultra-
27
high vacuum roughing/oil diffusion pump system which iscurrently used in the corrosion fatigue experiments reported inProgram #1. We anticipate that this system will achieve chamberpressures of better that 4 x 10~7 torr with full specimeninstrumentation; including load, displacement, crack growth byelectrical potential and temperature. Initial experiments arelikely to be simple tensile or creep rupture types with a minimumof complex instrumentation. The system will be made operationalduring the next reporting period.
Secondly, Mr. Porr has collaborated in the design andconstruction of an improved system to purify gaseous helium forintroduction to the vacuum system employed in corrosion fatigueexperimentation, Program #1. This new system purifies He by aroom temperature molecular sieve followed by a commercial, hotmetal chip gettering furnace. The connections between thisultra-low oxygen and water vapor gas source are of minimal lengthand were made according to good high vacuum procedures tominimize recontamination of gettered helium. The construction ofthis purification system was completed during the this reportingperiod. Corrosion fatigue experiments are planned for the nexttwo months, and the apparatus will also be used in conjunctionwith the temperature environment chamber described in theprevious paragraph.
28
I n v i t e d paper fo r t he A S T M 20th N a t i o n a l S y m p o s i u m on F r a c t u r eMechanics , Lehigh U n i v e r s i t y , Beth lehem, PA, 23-25 June 1987
ENVIRONMENTALLY ASSISTED CRACK GROWTH IN STRUCTURAL ALLOYS:
PERSPECTIVES AND NEW DIRECTIONS
Robert P. Weil7 and Richard P. Gangloff-?/
ABSTRACT
Envi ronmenta l ly assisted crack growth (namely , stress corro-sion c r a c k i n g and co r ros ion f a t i g u e ) in al loys is one of theprincipal de t e rmin ing fac tors for durab i l i ty and re l iabi l i ty ofe n g i n e e r i n g s t r u c t u r e s . O v e r the past 20 years , a c t i v i t i e s inthis area have t r ans fo rmed f r o m pr inc ipal ly that of screening andof q u a l i t a t i v e c h a r a c t e r i z a t i o n s of the p h e n o m e n a , to that ofquant i ta t ive assessment and s c i en t i f i c unders tanding. This workhas enabled the recent development of l i fe predict ion procedures.
In th is p a p e r , the c o n t r i b u t i o n s o f f r a c t u r e m e c h a n i c s inth is t r a n s f o r m a t i o n a r e r e v i e w e d . C u r r e n t m e c h a n i s t i c u n d e r -s t a n d i n g o f e n v i r o n m e n t a l l y a s s i s t ed c r a c k g r o w t h by h y d r o g e ne m b r i t t l e m e n t is s u m m a r i z e d , and is placed in perspective. Ap-p l i c a t i ons to m i t i g a t e s t ress c o r r o s i o n and co r ro s ion f a t i g u ecracking in mar ine envi ronments are summar i zed . Some ou ts tandingissues and new direct ions for research are discussed.
— Professor , Depar tment of Mechanical Engineer ing and Mechan ics ,Lehigh Un ive r s i t y , Bethlehem, PA 18015
o / - .— Associate Professor, Department of Materials Science, Univer-
sity of Virginia, Charlottesville, VA 22901
ENVIRONMENTALLY ASSISTED CRACK GROWTH IN STRUCTURAL ALLOYS:
PERSPECTIVES AND NEW DIRECTIONS
Robert P. WeiV and Richard P. Gangloff^/
INTRODUCTION
Envi ronmenta l ly assisted cracking of s tructural alloys (in-
corporat ing the well known phenomena of stress corrosion cracking
and corrosion f a t i g u e ) is well recognized as an impor tan t cause
for the f a i l u r e or ea r ly r e t i r e m e n t o f e n g i n e e r i n g s t ruc tu re s .
Stress corrosion cracking (SCC) was f i r s t recognized as a. techno-
logical p r o b l e m in the last half of the 19th cen tu ry as "season
c r a c k i n g " i n cold d r a w n brass [1 ] , w i t h cor ros ion f a t i g u e ( C F )
b e i n g r e c o g n i z e d in t he ea r l y 1900 ' s [2 ,3] . B r o w n , t r a c i n g the
h i s to r i ca l b a c k g r o u n d t h r o u g h 1 9 7 2 , c o n c l u d e d tha t "SCC, once
though t c o n f i n e d to a few s y s t e m s ( c o m b i n a t i o n s of m e t a l s and
env i ronmen t s ) , must now be regarded as a general phenomenon which
any alloy f a m i l y may exper ience , given the wrong combina t ion of
heat t r ea tment and envi ronment" [1].
Bolstered by defense related interests and by safe ty issues
in the e n e r g y i n d u s t r y , t he re was a decade of u n u s u a l l y h igh
research ac t iv i ty f r o m the early 1960's to the early 1970's [1].
Because of these i n t e r e s t s and of the i m p a c t of the "ene rgy
c r i s i s " , w o r k c o n t i n u e d t h r o u g h the 1970 ' s in suppor t o f o f f -
shore o i l e x p l o r a t i o n and v a r i o u s a l t e r n a t i v e e n e r g y s y s t e m s ,
such as , coal g a s i f i c a t i o n and l i q u e f a c t i o n , and solar e n e r g y .
Coincidenta l to these ac t iv i t i e s , f r a c t u r e mechanics was under-
- 2 -
going considerable development and maturation, and became in-
creasingly accepted as an important tool in structural analysis
and materials research.
In this paper a heuristic summary of the key developments in
environmentally assisted ,crack growth in structural alloys is
given, and key issues and new directions for research are dis-
cussed. The intent of this summary is to highlight the signi-
ficant milestones and the contributions of fracture mechanics.
As such, it does not include a complete chronology of all of the
developments and contributions to this field. A complete view of
these developments over the past 20 years may be gleaned from
several monographs and from the proceedings of a large number of
international conferences and symposia [1,4-23].
Chronologically, it is convenient to think of three periods,
1966 to 1972, 1972 to 1978, and 1978 to the present. Much of the
groundwork in the United States was established during the first
of these periods under two major programs, one sponsored by the
Advanced Research Projects Agency of the Department of Defense
(ARPA Coupling Program on Stress Corrosion Cracking) and the
other by the Air Force Materials Laboratory [1,24]. It was
during this period that the fracture mechanics methodology was
first introduced.
Activities in the 1972 to 1978 period were devoted princi-
pally to phenomenological characterizations of cracking response,
and to the development of empirically based design and failure
analysis methods. Initial efforts toward quantitative mechanis-
- 3 -
t i c u n d e r s t a n d i n g w e r e l a u n c h e d , and sc ience based a p p r o a c h e s
began to be fo rmula ted .
The past decade, beginning w i t h 1978, is a period of transi-
tion. Sc ien t i f i c unders tanding of environmenta l ly assisted crack
g r o w t h and e n g i n e e r i n g a p p l i c a t i o n o f t h i s u n d e r s t a n d i n g a re
placed on a quant i ta t ive foot ing. Signif icant advances have been
m a d e , and are possible w i t h the development of more sophis t ica ted
ana ly t i ca l i n s t r u m e n t a t i o n a n d e x p e r i m e n t a l t e c h n i q u e s . M o r e
e m p h a s i s i s now n e e d e d to t r ans l a t e the i m p r o v e d u n d e r s t a n d i n g
in to m e t h o d s f o r q u a n t i t a t i v e des ign a n d t o i m p r o v e s t r u c t u r a l
. rel iabi l i ty.
INITIATION vs PROPAGATION (Dawn of an Era)
A key turning point in the s tudy of environmenta l ly assisted
c r a c k g r o w t h and in the a p p r o a c h to d e s i g n o c c u r r e d in 1965.
Brown and his coworkers [25,26] at the Naval Research -Laboratory
w e r e e n c o u r a g e d to i n v e s t i g a t e the stress c o r r o s i o n ' c r a c k i n g
suscept ib i l i ty of t i t an ium alloys by using specimens wh ich were
d e l i b e r a t e l y p r e c r a c k e d i n f a t i g u e . Due t o th i s i n i t i a l c r a c k ,
the a l loy was h i g h l y s u s c e p t i b l e and f r a c t u r e d in a m a t t e r of
m i n u t e s , even t hough i t a p p e a r e d i m m u n e t o s tress co r ros ion
c r a c k i n g w h e n s t ressed in the s m o o t h ( u n c r a c k e d ) s ta te in the
same electrolyte.
As a resu l t of these f i n d i n g s , a m a j o r s h i f t in e m p h a s i s was
made f r o m tes t ing of smooth and mi ld ly notched specimens to that
of cracked bodies. Fracture mechanics was in t roduced as a basis
for ana lyz ing env i ronmen ta l l y assisted c rack ing in a paper by H.
- 4 -
H. Johnson and P. C. Pa r i s [27] in the F i r s t N a t i o n a l S y m p o s i u m
on Fracture Mechanics in 1967. Exper imenta l support for the use
of the crack t ip stress i n t ens i t y f a c t o r (K) to desc r ibe the
mechanical d r iv ing force was provided by Smi th , Piper and D o w n e y
[28] for stress corrosion crack g rowth (Fig. 1) , and by Feeney ,
M c M i l l a n and We i [29 ] fo r co r ros ion f a t i g u e ( F i g . 2 ) . A f o r m a l
d i scuss ion of the use of c rack tip stress in tens i ty to d e s c r i b e
t h e m e c h a n i c a l d r i v i n g f o r c e f o r c rack g r o w t h w a s g i v e n b y W e i
[30].
Early users of this e m e r g i n g technology for env i ronmen ta l
c r a c k i n g i n v e s t i g a t i o n s inc lude S t e i g e r w a l d a n d T ro i ano [ 3 1 ] >
Hanna and Ste igerwald [32], Johnson and Wil lner [33], and Hancock
and Johnson [34] for stress corrosion cracking (or crack growth
under sustained loading) , and Bradshaw and Wheeler [35], Har tman
[36], and Li , Talda and Wei [37] for cor ros ion f a t i gue . Some
ea r ly w o r k p r e d a t e d the usage of the K concep t . The e f f o r t
associated wi th the A R P A Coupl ing Program made extensive use of
f r ac tu re mechanics , and cont r ibuted to the development of f rac-
ture mechanics based technology for materials evaluation and for
design [1 ].
PHENOMENOLOGICAL CHARACTERIZATION (1966 to 1972)
Two impor tan t approaches emerged f rom the early ac t iv i t i es
on stress corrosion cracking using precracked specimens [30,38];
n a m e l y the threshold and the k ine t ics approaches. The choice of
a p a r t i c u l a r a p p r o a c h was d e t e r m i n e d in par t by t r a d i t i o n and
des ign p h i l o s o p h y , and in pa r t by p r a c t i c a l c o n s i d e r a t i o n s of
- 5 -
exper imenta t ion and cost.
The s i m p l e r and less expens ive approach invo lves the m e a -
surement of t ime- to- fa i lu re for precracked specimens under dif-
f e ren t appl ied loads, and the de t e rmina t ion of a so-called thres-
hold stress in tensi ty factor (designated as K I s c c) , below which
(p resumab ly ) no fa i lu re would occur by stress corrosion cracking
[ 2 5 , 3 0 , 3 8 , 3 9 ] . The level of K I S C C in r e l a t i o n to K I c , the plane
s t r a i n f r a c t u r e toughness of a m a t e r i a l , p r o v i d e d a m e a s u r e of
stress corrosion cracking suscept ibi l i ty . The use of the thres-
hold a p p r o a c h was f a v o r e d fo r m a t e r i a l s e l ec t ion and fo r s a fe -
l i fe design.
The second a p p r o a c h was m o r e c o m p l e x , and i nvo lved the .
de terminat ion of crack growth kinetics [30,38]. It required the
m e a s u r e m e n t of c rack g rowth rate ( d a / d t ) under control led en-
vi ronmenta l condit ions and as a func t i on of the mechanical crack
d r i v i n g f o r c e , w h i c h i s c h a r a c t e r i z e d by the stress i n t e n s i t y
fac tor K. This approach required greater e f fo r t and more sophis-
t icated in s t rumen ta t i on , and was favored for mechanis t ic studies
and for f a i l - s a f e des ign . A s i m i l a r d i s t i n c t i o n in a p p r o a c h
exis ted for corrosion f a t i gue [40],
Both of these approaches were wide ly used. The period f r o m
1966 to 1972 was pr inc ipa l ly devoted to phenomenological charac-
t e r i z a t i o n s of m a t e r i a l s r e sponse and to the d e v e l o p m e n t of
e m p i r i c a l l y based d e s i g n a p p r o a c h e s . T h e f i n a l r e p o r t f o r t h e
A R P A Coupl ing P rog ram, pub l i shed as a monograph [1], ref lects the
typical e f fo r t s du r ing this period.
- 6 -
I t was d u r i n g th i s pe r iod that a n u m b e r of key s tud ies
[30,33-37,41-47] took place which helped to set the stage for the
d e v e l o p m e n t of q u a n t i t a t i v e u n d e r s t a n d i n g of e n v i r o n m e n t a l l y
ass i s ted c rack g r o w t h over the nex t decade . These s tud ie s also
s h o w e d the i m p o r t a n c e of the k i n e t i c s a p p r o a c h , and se rved to
es tab l i sh i ts use in s u b s e q u e n t i n v e s t i g a t i o n s . Some of the
important f i n d i n g s are as fo l lows:
o There was an increasing awareness of the impor tance of the
k ine t ics approach, and a recognit ion of the fact that stress
c o r r o s i o n c r ack g r o w t h p rog re s se s in t h r ee s tages ( F i g . 3 )
[31,41] . The par t icular s ign i f icance of the K-independent
Stage II in t e r m s of m e c h a n i s t i c u n d e r s t a n d i n g of c rack
growth response was also recognized. The approach was used
extensively by Speidel [42] to s tudy the inf luence of halide
ions on s t ress cor ros ion c r ack g r o w t h in h i g h s t r e n g t h
a l u m i n u m a l l o y s , - a n d b y W e i a n d h i s c o w o r k e r s [ 4 3 - 4 4 ] t o
examine hydrogen e m b r i t t l e m e n t of high strength steels,
o The e f fec t iveness of using crack g rowth rate as a means for
unders tanding the mechan i sms for environmental enhancement
of crack g rowth under sustained loading was demons t ra ted by
Johnson and coworkers [33,34] , and of f a t i g u e crack growth
by B r a d s h a w and W h e e l e r [35], H a r t m a n [36] , and Wei e t a l
[37]. Through s tudies of the in f luences of d i f f e r e n t envi-
r o n m e n t s and of the i n h i b i t i n g e f f e c t of t race a m o u n t s of
o x y g e n , these r e s e a r c h e r s d e m o n s t r a t e d the i m p o r t a n c e o f
s u r f a c e r e a c t i o n s as a par t of the e m b r i t t l e m e n t s equence
( F i g . 4 ) .
- 7 -
o Pressure and tempera ture were recognized as impor tan t probes
for i den t i fy ing the processes that control environmenta l ly
assisted crack g rowth under sustained loading [31 ,33 ,34 ,45 ,
46] and in fa t igue [31,35,37,47].
o Ear ly recognit ion of the relat ionship between stress corro-
sion and c o r r o s i o n f a t i g u e c rack g r o w t h , and m o d e l i n g of
corrosion f a t igue crack g rowth as a superpos i t ion of f a t i gue
and stress corrosion cracking processes [48].
As an i l lustrat ion of the phenomenological character iza t ion
e f f o r t , t yp i ca l S tage I I c r ack g r o w t h rate d a t a , fo r a h i g h
strength (AISI 4340) steel, stressed in various env i ronments , are
shown as a f unc t i on of t empera tu re in Fig. 4 [44 ,49 ] . These data
demonst ra ted that the crack growth response can vary w i d e l y w i t h
environmenta l condit ions, and can depend uniquely on t empera tu re .
F r a c t o g r a p h i c e v i d e n c e s u g g e s t e d tha t t h e m i c r o m e c h a n i s m f o r
c rack g r o w t h i n these d i v e r s e , h y d r o g e n p r o d u c i n g e n v i r o n m e n t s
was the same. Thus, the d i f f e r e n t response had to be a t t r ibu ted
to one of several chemical processes in the overall chain.
Based on these in i t i a l f i n d i n g s , i t was r e c o g n i z e d tha t
fu r the r progress in unders tanding envi ronmenta l ly assisted crack
growth was not possible wi thou t in tegra t ing , at least, the frac-
ture mechanics based k ine t i c measurements w i th independent mea-
surements of the k inet ics of pa r t i c ipa t ing chemical processes.
- 8 -
DEVELOPMENT OF A SCIENTIFIC APPROACH (1972-1978)
B e g i n n i n g in 1 9 7 2 , a s c i e n t i f i c approach to the s tudy of
e n v i r o n m e n t a l l y ass i s ted c rack g r o w t h in h igh s t r e n g t h al loys
began to be d e v e l o p e d , and e v o l v e d over the f o l l o w i n g years
[ 4 0 , 4 3 , 4 4 , 5 0 , 5 1 3 . The a p p r o a c h was g r o u n d e d in l i nea r f r a c t u r e
m e c h a n i c s and was p r e d i c a t e d on the r e c o g n i t i o n that e n v i r o n -
m e n t a l l y ass is ted c r a c k g r o w t h is the resul t of a s equence of
processes and is c o n t r o l l e d by the s lowes t process in the se-
quence [ 4 0 , 5 0 ] .
The processes that are i n v o l v e d in the e n v i r o n m e n t a l en-
hancement of crack growth in high-strength alloys by hydrogen and
hydrogenous gases (such as H 20 and H 2 S) and by aqueous e n v i r o n -
ments are as fo l lows , and are il lustrated schematical ly in Fig. 5
for the case of gaseous environments [40,50] :
1. Transport of the deleterious envi ronment to the crack tip.
2. React ions of the deleterious environment w i t h newly pro-duced crack surfaces to e f f ec t localized dissolution andto produce hydrogen.
3. Hydrogen entry (or absorp t ion) .
4. D i f f u s i o n of hydrogen to the f r ac tu re (or e m b r i t t l e m e n t )site.
5. P a r t i t i o n i n g or d i s t r i b u t i o n of h y d r o g e n a m o n g v a r i o u smicros t ructura l sites.
6. H y d r o g e n - m e t al i n t e r a c t i o n s l e a d i n g to embr i t t lement atthe micros t ruc tura l sites.
The actual processes depend on the mechan i sm of crack g rowth
e n h a n c e m e n t ; n a m e l y , a c t i v e p a t h d i s s o l u t i o n o r h y d r o g e n e m -
b r i t t l e m e n t . Fo r a d i s s o l u t i o n m e c h a n i s m (see ( 6 ) ) , on ly the
f i r s t two s teps . in the s e q u e n c e need to be c o n s i d e r e d , and the
- 9 -
anodic (dissolution) reactions in the second step are directly
responsible for crack growth enhancement. On the other hand, if
hydrogen embrittleraent is the responsible mechanism, then the
reaction step serves only as the source for hydrogen; the remain-
ing processes (3 through 6) must be considered. These steps are
identical for aqueous and gaseous environments.
The overall crack growth response is governed by the rate
controlling process in conjunction with the mechanical driving
force, which is characterized here by either the local stress or
the crack tip stress intensity factor, K [27,31,40].
Embrittlement, or the final step in the sequence is a func-
tion of microstructure. The extent of embrittlement, or the rate
of cracking along each microstructural path, depends on the local
hydrogen concentration, which depends in turn upon the external
conditions (i.e., pressure, pH, electrode potential, tempera-
ture). Cracking along the various microstructural paths can take
place concurrently (in parallel) with the overall crack growth
rate given as the weighted average of the individual rates.
Clearly, the understanding of crack growth required identi-
fication and quantification of the rate controlling process. The
evolving scientific approach, therefore, focused upon well de-
fined and coordinated chemical/electrochemical, mechanical and
metallurgical experiments.
It was recognized from the outset that the concept of rate
controlling processes applied to both stress corrosion (or sus-
tained-load) and corrosion fatigue crack growth, and that con-
siderations of both modes of crack growth would afford a syner-
- 10 -
gisra in exploring the fundamenta l issues. Because of the ready
l i n k a g e of a K - i n d e p e n d e n t r e g i m e (S tage I I ) of c r ack g r o w t h to
the under ly ing rate controlling process, early e f fo r t s were di-
rected at sustained-load crack growth. Detai led studies of the
k i n e t i c s and m e c h a n i s m s of r eac t ions of w a t e r w i t h i ron and
s teel , and of Stage II c rack g r o w t h w e r e c a r r i e d out as a f u n c -
tion of t empera tu re [51-55]. These e f fo r t s led to an unambiguous
iden t i f i c a t i on of the i ron-water oxidat ion reaction as the rate
controlling process for crack growth in high-strength steels in
1 9 7 7 , and a r e a l i z a t i o n of the ve ry l i m i t e d e x t e n t of these
reactions [6-10] (see Fig. 6).
W i t h a d d i t i o n a l w o r k on s u r f a c e r eac t ions and on c rack
g r o w t h , a b roade r based u n d e r s t a n d i n g of ra te con t ro l l i ng pro-
cesses emerged dur ing the remainder of the 1970's, and model ing
of sustained-load crack growth was begun. In 1977, this approach
was used to examine the role of water vapor in .enhancing fa t igue
crack growth in high-strength a luminum alloys [56], and the role
of hydrogen su l f ide in enhancing fa t igue crack g rowth in steels
[57]. These e f f o r t s led to the m o d e l i n g of co r ros ion f a t i g u e
c r a c k g r o w t h [58 ,59] , w h i c h l ed in t u r n to the r e c o g n i t i o n and
m o d e l i n g o f t r a n s p o r t - c o n t r o l l e d c rack g r o w t h unde r s u s t a i n e d
loading [60].
CHEMICAL AND MICROSTRUCTURAL MODELING
The mode l ing e f f o r t , wh ich began in the late 1970's , emerged
as an impor tan t ad junc t of the s c i e n t i f i c approach over the past
10 y e a r s . I t has p r o v i d e d g u i d a n c e and a f o r m a l i z e d f r a m e w o r k
- 11 -
for e x a m i n i n g the f u n d a m e n t a l i s sues , and has s e rved as a bas is
for the u t i l iza t ion of data in design. Chemical and micros t ruc-
tu ra l m o d e l i n g of s t ress c o r r o s i o n and cor ros ion f a t i g u e c r a c k
growth are b r i e f l y s u m m a r i z e d to provide perspect ive and to serve
as a basis for discussing direct ions for f u t u r e research.
M o d e l i n g of c r ack g r o w t h may be s u b - d i v i d e d in t e r m s of
l oad ing c o n d i t i o n s , and a d d i t i o n a l l y acco rd ing to the va r ious
processes that a f f e c t crack g r o w t h (see Fig. 5). Assuming hydro-
gen embr i t t l emen t as the responsible mechanism, model ing may then
be grouped in te rms of hydrogen supply ( t ranspor t , reactions and
d i f f u s i o n ) , h y d r o g e n d i s t r i b u t i o n w i t h i n t h e m a t e r i a l (pa r t i -
t i on ) , and e m b r i t t l e m e n t reactions (which determine the rates of
cracking along the various microst ructural paths) , and incorpo-
ra tes s u i t a b l e m e t h o d s fo r o b t a i n i n g an average o r m a c r o s c o p i c
c rack g r o w t h ra te . Mos t o f the e f f o r t d u r i n g th is p e r i o d has
concentrated on chemical mode l ing , and reflects developments in
u n d e r s t a n d i n g the c h e m i c a l and f r a c t u r e m e c h a n i c s aspec t s of
env i ronmenta l ly assisted crack growth . More recently, this ef-
f o r t has been e x t e n d e d to i n c l u d e the i n f l u e n c e of m i c r o s t r u c -
ture .
Sustained-Load Crack Growth
Ini t ia l model ing of sustained-load crack g r o w t h was largely
phenomenological and was l i m i t e d to the case of hydrogen-supply
controlled Stage II crack g r o w t h in the lower t e m p e r a t u r e region
( R e g i o n A in F i g . 6) . The p r i n c i p a l t h r u s t was d i r e c t e d a t
o b t a i n i n g r e l i a b l e c h e m i c a l r e a c t i o n a n d c rack g r o w t h d a t a t o
- 12 -
confirm the concept of rate controlled crack growth and to iden-
tify the controlling process. Obvious deviations of the crack
growth response curves (Fig. 6) from those of single rate con-
trolling processes and the observed changes in fracture paths (or
micromechanisms) led to the consideration of transfers of control
and of the role of microstructure.
Models Based On Hydrogen Supply. Models for Stage II crack
growth were proposed on the basis of extensive data on the kinet-
ics of surface reactions and crack growth for high-strength
steels in water/water vapor, hydrogen and hydrogen sulfide
[45,46,50,52-62]. When the reactions are slow (e.g., in hydrogen
and in water), Stage II crack growth rate is controlled by the
rate of surface reactions. On the other hand, for very rapid
reactions (e.g., in hydrogen sulfide), the growth rate1" is deter-
mined either by the rate of transport of the gases to the crack
tip, or by the rate of diffusion of hydrogen to the embri ttlement
site. The models, expressing the specific dependence on gas
pressure (po) and temperature (T), are as follows [44,63]:
Transport Control: (da/dt)jj = Ctpo/T1/2 (1)
(for Knudsen flow [64])
Surface ReactionControl: (da/dt)IZ = Csp™ exp(-Es/RT) (2)
Diffusion Control: (da/dt)ZI = CdpJ/2 exp(-Ed/2RT) (3)
The constants C^ contain chemical and physical quantities that
relate to gas transport, surface reaction, etc., and reflect the
susceptibility of specific alloys to embrittlement by specific
- 13 -
environments (viz., the embrittlement reaction term). ES and E^
are the activation energies for surface reaction and hydrogen
diffusion respectively. Good agreement with experimental obser-
vations in the low temperature region (Region A) is indicated in
Fig. 6.
In these models, a single process is assumed to be in con-
trol, and the terms C^ are assumed to be sensibly constant. It
is recognized that transfer of control from one process to anoth-
er may occur as the environmental conditions are changed. The
consequences of this transfer have been discussed [63]. In
formulating these phenomenological models, simple (single step)
reactions were implicitly assumed. Because the reactions tend to
be more complex, these models are viewed as starting points for
developing further understanding of environmentally assisted
crack growth.
Partitioning of Hydrogen and Surface Phase Transformation.
The observed decrease in sustained-load crack growth rate with
increasing temperature in the "high temperature" region (Region C
in Fig. 6) for carbon martensitic steels (such as, AISI 4340
steel) has been analyzed based on surface chemistry [46,65,66].
Similar analyses have been attempted to account for the much
steeper decrease in crack growth rate for 18N1 maraging steels
[67]. These analyses, however, have ignored the important role
of microstructure and micromechanism in hydrogen embrittlement,
or have made unrealistic assumptions regarding surface coverage
by hydrogen. The models proposed on the basis of these anal-
yses, therefore, cannot explain the observed changes in fracture
- 14 -
mode wi th tempera ture [49].
To q u a n t i t a t i v e l y account for the role of h y d r o g e n - m i c r o -
s t ructure interactions, a "hydrogen par t i t ioning" model was deve-
loped [49] . The m o d e l sugges t ed that the ra te of h y d r o g e n as-
sisted crack growth is de te rmined by two factors: (i) the rate of
hydrogen supply to the f r ac tu r e process zone, and ( i i ) the parti-
tioning of hydrogen amongst d i f fe ren t microstructural elements or
traps (pr incipal ly be tween pr ior-austeni te grain boundaries and
the m a r t e n s i t i c m a t r i x ) . The p a r t i t i o n i n g of h y d r o g e n is con-
trolled by hydrogen- t rap interact ions and de termines the contri-
bution by each element to the overall crack growth rate , which is
the w e i g h t e d ave rage of ra tes of c r a c k i n g along the d i f f e r e n t
microstructural paths. This model is i l lustrated schemat ical ly in
F ig . 7 for h y d r o g e n ass i s ted c r a c k g r o w t h in a h i g h s t r e n g t h
steel. Deta i led considerations and a der ivat ion of the model are
given in [49].
At low tempera tures (in Region A of Fig. 6), hydrogen would
res ide p r i m a r i l y a t gra in b o u n d a r i e s and s l ip p lanes . C r a c k
g r o w t h w o u l d t end t o b e p r e d o m i n a n t l y i n t e r g r a n u l a r ( I G ) , a n d
w o u l d c o n f o r m t o Eqns . ( 1 ) t o ( 3 ) . W i t h i n c r e a s i n g t e m p e r a t u r e
( into Reg ion C) , the hydrogen supply processes remain in control ,
but h y d r o g e n c o n c e n t r a t i o n a t the gra in b o u n d a r i e s and in the
s l ip planes dec reases and m o r e h y d r o g e n w o u l d r e s i d e in t he
m a r t e n s i t e l a t t i c e . This t e m p e r a t u r e i n d u c e d p a r t i t i o n i n g o f
h y d r o g e n leads to i n c r e a s i n g a m o u n t s o f m i c r o v o i d coa lescence
( M V C ) and t o s l o w e r c r a c k g r o w t h rates. The change i n c r a c k
- 15 -
g r o w t h ra te w i t h t e m p e r a t u r e r e f l ec t s a t r a n s f e r of c r a c k i n g
m e c h a n i s m s , r a the r than (or in a d d i t i o n to) a change in the
process of hydrogen supply. The predicted tempera ture and pres-
sure d e p e n d e n c e s fo r Stage I I c r ack g r o w t h in h i g h - s t r e n g t h
steels were in good agreement wi th crack growth data for an AISI
11340 steel in hydrogen and in hydrogen su l f ide (see, for example ,
Fig . 8) [49] .
For the l 8 N i maraging steels, a phase t rans i t ion model was
proposed to account for the abrupt decrease in crack growth rate
tha t had been o b s e r v e d a t the h igher t e m p e r a t u r e s [50 ,68] . The
model was based on the suggestion by Hart [69] that solute atoms
can u n d e r g o a phase t r a n s f o r m a t i o n at c e r t a i n t e m p e r a t u r e s and
pressures, and conformed w i th experimental observations.
C l e a r l y , a n u m b e r of f a c t o r s can i n f l u e n c e the k i n e t i c s of
environmenta l crack growth. The hydrogen par t i t ion ing and sur-
f a c e phase t r a n s i t i o n m o d e l s have p r o v i d e d s o m e i n s i g h t , and a
clear i n d i c a t i o n of the need for a b road ly based u n d e r s t a n d i n g ,
including that of the embr i t t l emen t mechanisms .
Fatigue Crack Growth
Based on the u n d e r s t a n d i n g d e v e l o p e d fo r s u s t a i n e d - l o a d
crack g rowth , models for sur face reaction and transport control-
l e d f a t i g u e c r a c k g r o w t h w e r e d e v e l o p e d [58 ,59] , a n d have been
used successful ly to explain the observed dependence of f a t i g u e
c r a c k g r o w t h ra tes on c y c l i c load f r e q u e n c y and p r e s s u r e in
gaseous envi ronments [56-59]. Insight obta ined f r o m this e f f o r t
- 16 -
has been a p p l i e d to the c o n s i d e r a t i o n of cor ros ion f a t i g u e in
aqueous envi ronments .
Superposi t ion Model. Model ing was based on the proposi t ion
tha t the m e c h a n i c a l and e n v i r o n m e n t a l c o n t r i b u t i o n s could be
d e c o u p l e d such that the r a t e of c rack g r o w t h in a de l e t e r ious
e n v i r o n m e n t , ( d a / d N ) e > may be w r i t t e n as the sum of th ree com-
p o n e n t s , Eqn . ( 4 ) [40 ,70,71] .
( d a / d N ) e = ( d a / d N ) r ( 1 - c j> ) + ( d a / d N ) c < j > + ( da /dN) s c c (4)
In th is e q u a t i o n , ( d a / d N ) r i s the m e c h a n i c a l f a t i g u e ra te ;
( d a / d N ) c is the "pure" co r ros ion f a t i g u e r a t e ; <J> as the f r a c -
t iona l area of c r a c k tha t i s u n d e r g o i n g pure co r ro s ion f a t i g u e ;
and ( d a / d N ) s c c is the c o n t r i b u t i o n of sus ta ined- load crack
growth. These rates may be composed of contr ibut ions f r o m sev-
eral concurrent mic romechan i sms . For s impl i c i ty , the sustained-
load g r o w t h t e r m i s no t i n c l u d e d in the f o l l o w i n g d i scus s ions .
Eqn. (4) is r ewr i t t en as fol lows [66]:
( d a / d N ) e = ( d a / d N ) r (1 - < f r ) + ( d a / d N ) c < j >
= ( d a / d N ) r + [ ( da /dN) c - ( d a / d N ) r ] 4 » (5)
or,
( d a / d N ) c f = [ ( d a / d N ) c - (da /dN) r ]< |> . (6)
where (da /dN) c f - denotes the incremental increase in growth rate
above the re fe rence level result ing f r o m the e m b r i t t l i n g environ-\
ment .
- 17 -
In the l i m i t for <|> = 0, or for a test in an ine r t e n v i r o n -
m e n t , ( d a / d N ) e = ( d a / d N ) r and co r responds to pu re f a t i g u e . For
<j> = 1 , c o r r e s p o n d i n g to c h e m i c a l r e ac t i on s a t u r a t i o n [ 5 3 , 5 M ] ,
( d a / d N ) e = ( d a / d N ) e s = ( d a / d N ) c , and m e a s u r e d g r o w t h rates
c o r r e s p o n d to pure c o r r o s i o n f a t i g u e rates . In essence , the
parameter <|> represents mater ia l response to changes in environ-
m e n t a l c o n d i t i o n s . I t i s d i r e c t l y re la ted to i t s c o u n t e r p a r t ,
the f rac t ional surface coverage ( 9 ) , in chemical model ing; i.e.,
<J> = 9 [58,59] . The m a x i m u m in co r ro s ion f a t i g u e c r a c k g r o w t h
r a t e , t h e r e f o r e , co r responds to the m a x i m u m ex t en t o f c h e m i c a l
r e a c t i o n (9 = 1 ).
Chemical Modeling
Important understanding of corrosion fatigue crack growth
response in gaseous environments was developed through chemical
modeling (58,59) and through experimental verification of the
role of gas transport and surface reactions on (da/dN)cf-
[56,57,72]. Similar understanding is being developed for aqueous
environments [17,23-26].
Assuming that the environmental enhancement of fatigue crack
growth results from embrittlement by hydrogen produced by the
reactions of hydrogenous gases with freshly produced crack sur-
faces, models for transport and surface reaction were developed
[58,59]. An analogous model for electrochemical reaction con-
trolled crack growth was proposed for steels in aqueous environ-
ments, where the kinetics of reaction are assumed to be slow
[73-75]. In these models, the environmental contribution is
- 18 -
assumed to be proportional to the extent of chemical or electro-
c h e m i c a l r eac t ion per cyc le , w h i c h is g iven by the f r a c t i o n a l
surface coverage e, and the crack g rowth rate ( d a / d N ) c f > is given
as by Eqn . (6) w i t h <j> = e.
Models for d i f f u s i o n controlled growth (76) and strain in-
duced h y d r i d e f o r m a t i o n [77 -79] have also been sugges ted . The
latter model relates to metal lurgical changes and the consequent
e f fec t on crack growth rates, and is considered later. D i f f u s i o n
controlled crack growth occurs when the preceding transport and
s u r f a c e r e a c t i o n processes a re r a p i d , and m u s t be c o n s i d e r e d
outside of the context of l im i t ed surface coverage per cycle.
Transport controlled growth. For highly reactive gas-metal
s y s t e m s , c r a c k g r o w t h is c o n t r o l l e d by the r a t e of t r a n s p o r t of
gases to the c r ack t ip [58,59] . The s u r f a c e cove rage (e) i s
l inearly proportional to pressure (pQ) and inversely proportional
t o f r e q u e n c y ( f ) . The e n v i r o n m e n t a l c o n t r i b u t i o n t o f a t i g u e
crack growth is given by the fo l lowing relationships:
( d a / d N ) G f = C ( d a / d N ) c - ( d a / d N ) r H(po /f ) / (p 0 / f ) s] (7a)
for ( P 0 / f ) < ( P 0 / f ) s
X d a / d N ) c f = [ ( d a / d N ) c - ( d a / d N ) r ] = constant (7b )
for ( p Q / f ) £ ( P 0 / f ) a
The te rm [ ( d a / d N ) c - ( d a / d N ) r ] is the m a x i m u m enhancement in
the rate of cycle-dependent corrosion fa t igue crack g r o w t h , and
is a c o n s e q u e n c e of the f i n i t e e x t e n t of s u r f a c e r e a c t i o n ( i .e . ,
6 + 1 ) [56 ,57 ] . T h e s a t u r a t i o n e x p o s u r e , ( p o / f ) s » i s a f u n c t i o n
- 19 -
of gas p res su re , t e m p e r a t u r e , and m o l e c u l a r w e i g h t of the gas ,
and of stress i n t e n s i t y leve l , load r a t io , y ie ld s t r eng th and
elastic constants; through their influences on crack geometry and
gas transport [58,59,72].
Surface and electrochemical reaction controlled growth.
W i t h less r e a c t i v e s y s t e m s , c r a c k g r o w t h i s cont ro l led by the
ra te of su r f ace or e l e c t r o c h e m i c a l r eac t ions at the c r a c k t ip .
For s i m p l e f i r s t - o r d e r r e a c t i o n s , the c rack g r o w t h ra te in
gaseous e n v i r o n m e n t s is g i v e n by Eqn . (8) in t e r m s of p r e s su re ,
f r equency and the reaction rate constant kc [58,59].
(da/dN) c f = [ ( d a / d N ) c - (da /dN) r >[1 - exp(-k c p o / f ) ] (8)
A more general in terpre ta t ion of surface coverage can be made to
accommoda te mul t i - s t ep react ions, w i t h the actual response re-
f l e c t i n g the n a t u r e and k i n e t i c s o f the i n d i v i d u a l r e a c t i o n
steps.
For aqueous e n v i r o n m e n t s , ( d a / d N ) c f may be e x p r e s s e d as an
ana logue to Eqn . (8) [75] :
( d a / d N ) c f = C ( d a / d N ) c - ( d a / d N ) r ] - C q / q g ] (9)
where q is the amount of electrochemical charge t ransferred per
cycle; qs is the "saturation" amount .or that requi red to complete
the reactions; and the ra t io q/q s is i d e n t i f i e d w i t h 6.
D i f f u s i o n controlled g r o w t h . When transport and surface
react ion processes are s u f f i c i e n t l y r ap id , the crack g r o w t h rate
is de t e rmined by the rate of d i f f u s i o n of hydrogen f r o m the crack
- 20 -
tip to the "fracture process zone". According to Kim (76),
(da/dN)cf is given by Eqn. (10):
(da/dN)cf = A0exp(-HB/RT)-(p0D/f)1/2AK2 (10)
where AQ is an empirical constant, HB is the binding enthalpy of
hydrogen to dislocations, and R is the universal gas constant, D
is the hydrogven diffusivity, and f is the frequency.
In the transport and reaction controlled models, a growthp
rate dependence upon AK is implicitly assumed to reflect the
expected proportionality between, the sizes of the "hydrogen
damaged" zone and the crack-tip plastic zone [59,72]. This
dependence is explicitly incorporated in the diffusion controlled
model [76]. The temperature dependence is reflected through its
influence on the reaction rates, the fatigue process, the mecha-
nical properties and on gas transport [58,59,73,74]. If the
reaction mechanisms remain unchanged, the maximum enhancement in
rate (or (da/dN)c) is expected to remain constant. The tempera-
ture dependence for corrosion fatigue would be reflected princi-
pally through its frequency dependence, with the maximum remain-
ing constant.
Experimental support. The transport and surface (and elec-
trochemical) reaction controlled models have been examined by
coordinated studies of the kinetics and mechanisms of gas-metal
reactions, and of corrosion fatigue crack growth response as
functions of pressure, temperature, and loading time (or frequen-
cy). Good agreement between these models and the experimental
data on crack growth and surface and electrochemical reactions
- 21 -
has been o b t a i n e d (see Figs . 9 to 11, for e x a m p l e ) . The t rans-
port con t ro l l ed case is r ep re sen t ed by a l u m i n u m and t i t a n i u m
alloys in w a t e r vapor (F ig . 9 ) , and steels in h y d r o g e n s u l f i d e
(F ig . 10) at low pressures; the reaction controlled case, by high
strength steels in aqueous electrolytes (see Fig. 11, for example
[80]).
I t i s r e c o g n i z e d that the f o r m of the c r a c k g r o w t h ra te
re sponse d e p e n d s on the k i n e t i c s and on the s p e c i f i c n a t u r e and
m e c h a n i s m ( s ) of the react ions, and may reflect both t ransport and
reac t i on control . For the case of r eac t ion cont ro l led c r ack
g r o w t h , the r e sponse may r e f l e c t the f a c t that the r eac t ions do
not fo l low s imple f i r s t -order kinet ics , and the presence of more
than one r e a c t i o n step. For e x a m p l e , for the case of 7075-T651
a l u m i n u m alloy ( F i g . 1 0 ) , the add i t i ona l enhancemen t a t the
higher pressures is surface reaction controlled and is a t t r ibu ted
to a slow s tep in the reactions of water w i th segregated magne-
s i u m [81]. S i m i l a r l y , the inc rease in ra te o b s e r v e d on the 2-
1 / i l C r - 1 M o steel in h y d r o g e n s u l f i d e a t the h ighe r p ressures is
s u r f a c e r e a c t i o n c o n t r o l l e d , a n d i s i d e n t i f i e d w i t h t h e s lower
second step in the reactions of H2S w i t h iron [57,60]. A s imilar
s i t u a t i o n ex i s t s fo r c r a c k g r o w t h o f h i g h - s t r e n g t h s teels in
w a t e r vapor and in aqueous so lu t ions . The s i t u a t i o n in w a t e r
vapor may be fu r the r compl ica ted by capil lary condensation at the
crack t ip [53 ,73 ,74] .
Evidence for d i f f u s i o n - c o n t r o l l e d crack g rowth is provided
by da ta on t i t a n i u m al loys (see F ig . 12 , fo r e x a m p l e ) . A t t he
higher f r e q u e n c i e s , ( d a / d N ) c f was f o u n d to be inverse ly proper-
- 22 -
t ional to the square root of f r e q u e n c y [82]. This d e p e n d e n c e ,
coupled w i t h the known reac t iv i ty of t i t a n i u m , is consistent w i t h
d i f f u s i o n cont ro l . The a b r u p t decrease in g r o w t h ra tes a t the
l o w e r f r e q u e n c i e s i s a t t r i b u t e d to a h y d r i d e m e c h a n i s m that
d e p e n d s on both s t r a in and s t r a in ra te [29-31 ,36] . T h e r e i s ,
however , no quant i ta t ive model for hydr ide induced crack growth.
Micros t ructura l Model ing
The impor tan t role of d i f f e r e n t mic romechan i sms was discus-
sed by G e r b e r i c h and P e t e r s o n (83) . The role of m i c r o m e c h a n i s m
(or of m i c r o s t r u c t u r e ) i s e x p l i c i t l y i n c o r p o r a t e d in Eqns . (4)
and ( 6 ) . The i m p l i c a t i o n s of the m o d e l are as f o l l o w s : ( i ) the
p a r t i t i o n i n g of h y d r o g e n to the va r ious m i c r o s t r u c t u r a l s i tes
need not be u n i f o r m , and ( i i ) the f r a c t i o n a l area of f r a c t u r e
s u r f a c e ( < £ ) p r o d u c e d by pure corros ion f a t i g u e is equal to the
f r a c t i o n a l s u r f a c e coverage ( 9 ) f o r c h e m i c a l r e ac t i ons . T h e
relat ionship between the micros t ructura l and environmenta l param-
eters ( < j > and 9) was e x a m i n e d by Ress le r ( 8 4 ) and by Gao et a l
( 8 5 ) .
For an A I S I 4 3 4 0 steel in w a t e r vapor (585 Pa) a t room
tempera ture , Ressler (84) de te rmined the corrosion fa t igue crack
growth rate as a f u n c t i o n of f r equency (Fig. 13). Fractographic
data show a change in f r a c t u r e surface morphology w i t h decreasing
f r e q u e n c y f r o m a p r e d o m i n a n t l y t ransgranular mode (relat ive to
the pr ior -aus ten i te gra ins) to one that is p r edominan t ly inter-
granular , Fig. 14. By i d e n t i f y i n g the in tergranular f a i l u r e mode
w i t h p u r e c o r r o s i o n f a t i g u e a n d t h e t r a n s g r a n u l a r m o d e w i t h
- 23 -
mechanical fatigue, the fraction of pure corrosion fatigue ((j>),
was estimated from the microfractographs. A comparison was made
between <j> and 9, based on independent surface reaction measure-
ments [51] and an adjustment of exposure to account for capillary
condensation in the fatigue crack (Fig. 15). Agreement is excel-
lent. A similar good correlation was reported by Gao et al [85]
for a 7075-T651 aluminum alloy.
These results indicate the important role of microstructure,
and of the interactions between the environmental and microstruc-
tural variables.
More work is needed to broaden the scope of this under-
standing, and to provide statistically reliable support. Never-
theless, the framework for understanding has been set in place.
IMPLEMENTATION OF THE FRACTURE MECHANICS APPROACH (1980s)
Significant complexities must be overcome in implementing
the fracture mechanics approach for the quantitative prediction
of component life to control environmentally assisted crack
growth. During the 1980s, two critical issues emerged. Firstly,
the principle of fracture mechanics similitude (that is, equal
subcritical crack growth rates are produced by equal applied
stress intensities) may be violated because of the effects of
crack closure and variations in environmental chemistry within
the crack. Secondly, the large number of relevant variables and
their time-dependent interactions greatly complicate life predic-
tions, particularly for lower strength materials which were not
extensively characterized during the 1970s and where linear frac-
- 24 -
ture mechanics may not s u f f i c e .
M e c h a n i s t i c u n d e r s t a n d i n g and s c i e n t i f i c model ing provide
the means for character iz ing subcri t ical cracking in terras of a
scalable c r ack t i p d r i v i n g f o r c e , f o r e x t r a p o l a t i n g shor t t e r m
data to p r e d i c t long t e r m c o m p o n e n t b e h a v i o r , and fo r d e f i n i n g
the e f f ec t s of mechanical , chemical and metallurgical variables.
Fracture Mechanics S imi l i tude
The use of f r ac tu re mechanics s imi l i tude to scale environ-
menta l ly assisted crack growth rates, for d i f f e r e n t crack sizes
and loadings, is pe rmi t t ed only if the fo l lowing two conditions
a r e m e t : ( 1 ) t h e a p p l i e d d r i v i n g f o r c e p a r a m e t e r (e .g. , s t ress
i n t e n s i t y f a c t o r ) u n i q u e l y d e f i n e s t he s t resses , s t r a ins and
s t r a i n rates n e a r the c rack t ip ; and (2) the c o m p o s i t i o n and
c o n d i t i o n s of the e n v i r o n m e n t at the c r a c k t ip are cons t an t for a
given appl ied dr iv ing force pa ramete r , i r respect ive of crack size
and o p e n i n g shape . I n v e s t i g a t i o n s o f e n v i r o n m e n t a l l y ass i s ted
crack g rowth at low stress intensi t ies , in low strength or aniso-
t r o p i c m a t e r i a l s , f o r smal l c rack g e o m e t r i e s and i n c o m p l e x
embr i t t l i ng env i ronments demonst ra te that the val id i ty of s imil i -
tude cannnot be always assumed [16-18,86-88]. The applied stress
i n t e n s i t y bas is s h o w n in F igs . 1 and 2 may need to be m o d i f i e d to
r e f l e c t the ac tua l c r a c k d r i v i n g f o r c e , a s d e s c r i b e d in the
fo l l owing subsections.
Crack Closure P r o b l e m . P r e m a t u r e contact of f a t i gue crack
surfaces dur ing unloading or crack closure reduces the e f f e c t i v e
c r a c k d r i v i n g f o r c e r e l a t i v e t o t h e a p p l i e d va lue (e .g . , A K ) .
- 25 -
For co r ros ion f a t i g u e , l i k e l y c rack c losure m e c h a n i s m s i n c l u d e
env i ronmenta l ly assisted crack def lec t ion , intergranular cracking
and enhanced surface roughness, enhanced plast ici ty, f lu id pres-
sure, and corrosion product wedging [89,90].
The e f fec t of corrosion product wedg ing is shown in Fig. 1 6.
Here , the reduct ion in fa t igue g rowth rates at low mean stress is
a t t r i b u t e d to w e d g i n g by a t h i n s u r f a c e o x i d e p r o d u c e d by f r e t -
t ing of the f r ac tu r e surfaces in moist or oxygenated envi ronments
[91] . Th i s c losure m e c h a n i s m i s r e l evan t w h e n the c rack t ip
opening d isplacements are small , and may become more pronounced
because o f e n h a n c e d ( c r e v i c e ) co r ros ion w i t h i n f a t i g u e cracks .
The e f f e c t was d o c u m e n t e d fo r c a thod ica l l y p o l a r i z e d s teels in
seawater [92,93] .
For l a b o r a t o r y s p e c i m e n s , c r a c k c losure e f f e c t s a re ac -
c o u n t e d fo r t h r o u g h m e a s u r e m e n t s o f s p e c i m e n c o m p l i a n c e [89].
The i m p l i c a t i o n s o f e n v i r o n m e n t i n d u c e d c losure fo r c o m p o n e n t
b e h a v i o r , h o w e v e r , a re unc lea r and r e q u i r e a t t en t ion . Even
though e lementa ry micromechan ica l models have been proposed for
spec i f i c closure mechan i sms , the envi ronment sensit ive processes
are p o o r l y u n d e r s t o o d [90] . The p r o b l e m is e x a c e r b a t e d by the
t ime dependent nature of corrosion processes.
Small Crack Problem. Because of the impor tance of early
crack f o r m a t i o n and growth to component f a t i gue l i f e , consider-
able emphas is has been placed on the f r a c tu r e mechanics of small
c r a c k s t h r o u g h the 1980s [8?]. G e n e r a l l y , s t ress c o r r o s i o n and
c o r r o s i o n f a t i g u e c r a c k s s i z e d b e l o w 1 to 5 mm have been f o u n d to
grow unexpectedly rapidly re la t ive to long cracks at the same K
- 26 -
level , and to grow be low the threshold K level for long cracks
[87, 90, 91!, 95]. This behav io r m u s t be u n d e r s t o o d in o rde r to
apply the f r ac tu re mechanics approach to predict environmental
cracking l i fe in components .
The obse rved c rack s ize e f f e c t may be caused by inappro-
p r i a t e f o r m u l a t i o n o f the m e c h a n i c a l d r i v i n g f o r c e and by d i f -
ferences in crack tip env i ronment . A var ie ty of mechanisms and
bounding crack sizes were ident i f ied [96]. For f a t i gue in benign
e n v i r o n m e n t s , i nc reased c rack t i p p l a s t i c i t y , u n d e r d e v e l o p e d
crack w a k e closure, three dimensional crack shape and large scale
yielding can increase growth rates of small cracks. Interactions
w i t h grain boundaries , on the other hand, can arrest growth. The
s a m e m e c h a n i s m s m a y a f f e c t e n v i r o n m e n t a l l y a ss i s t ed c r a c k i n g
through changes in crack- t ip env i ronment , envi ronmenta l m o d i f i c a -
t ion of crack closure, or s t ra in enhanced f i l m rup tu re [90,95,97-
99] .
. The o c c l u d e d e n v i r o n m e n t s w i t h i n short c racks can d i f f e r
f r o m the b u l k and f r o m those w i t h i n long c racks . Fo r the cases
examined to date, the short crack env i ronments appear to be more
deleterious. Data for a high strength steel in aqueous chloride
solution (Fig. 17) show that stress in tens i ty s i m i l i t u d e was not
o b e y e d , w i t h K l s c c d e c r e a s i n g w i t h c r ack size b e l o w about 1 mm
[100] . Va lues fo r K j s c c a t t he s m a l l e s t c r a c k s ize w e r e abou t
1 /3 those o b t a i n e d w i t h s p e c i m e n s c o n t a i n i n g 15 to 30 mm long
c r a c k s . T h i s c r a c k s ize e f f e c t i s c h e m i c a l i n o r i g i n , and has
been predic ted successful ly through calculations of solut ion pH,
electrode potential and total rate of hydrogen p roduc t ion w i t h i n
- 27 -
cracks of v a r y i n g l eng th , and the use of an e m p i r i c a l re la t ion-
ship between K I s c c and adsorbed hydrogen concentration [100-102].
Gangloff showed that small corrosion fa t igue cracks (below
about 3 mm) g r e w a t rates up to 10^ t i m e s f a s t e r than long c racks
a t e q u i v a l e n t a p p l i e d AK [ 1 0 3 , 1 0 4 ] . This e f f e c t i s i l l u s t r a t ed
in F ig . 18 by c o m p a r i n g da ta for 25 to 40 mm long c racks w i t h
those of 0.1 to 3 mm long elliptical surface and through-thick-
ness edge c r a c k s for a h igh s t r e n g t h steel in aqueous 3% N a C l
so lu t ions . The c o n c o m i t a n t absence of a c rack s ize e f f e c t in
vacuum and moist air fu r ther demonstrates the chemical origin of
the breakdown in s imi l i tude. The chemical crack size e f f ec t has
been conf i rmed by l i terature data, and by experiments w i th lower
strength steels. The e f f e c t , however , decreased f r o m a factor of
10^ to 2 as the yield strength of the steel decreased [95].
The e f f e c t of c r ack g e o m e t r y shown in F i g . 18 is qua l i t a -
t ively understood in te rms of a complex interact ion between mass
transport by d i f f u s i o n , ion migra t ion and convect ion, and elec-
trochemical reactions [94 ,104 ,105] . Initial e f for t at model ing
this e f f ec t has been m a d e , but needs to be suppor ted by cri t ical
exper iments [104-106] . Poss ib l e v a r i a t i o n s in c r a c k c h e m i s t r y
and their e f f ec t s on the growth of longer cracks must be considered.
In essence i t is n e c e s s a r y to e x a m i n e the coup led r e a c t i o n
and t r a n s p o r t p r o b l e m as a w h o l e in d e a l i n g w i t h the a p p a r e n t
v i o l a t i o n of s i m i l i t u d e . T u r n b u l l has a t t e m p t e d to do th i s by
m o d e l i n g s t e a d y s t a t e r e a c t i o n s a n d m a s s t r a n s p o r t i n s i m p l e
t rapezoidal ly shaped cracks [107,108]. This w o r k must be modi-
f i e d t o i n c l u d e t h e i m p o r t a n t t r a n s i e n t r e a c t i o n k i n e t i c s dis-
- 28 -
cussed in a previous section, and must be related to the raicrome-
chan ics of c rack advance . Such an approach w i l l p r o v i d e the
f o u n d a t i o n f o r d e v e l o p i n g q u a n t i t a t i v e p r e d i c t i v e m e t h o d s f o r
long te rm component service.
L i f e P red ic t ion for Environmenta l Cracking
The t i m e dependence of environmental crack g rowth , the many
r e l e v a n t va r i ab l e s a n d t h e c o m p l e x i t i e s a f f e c t i n g s i m i l i t u d e
h i n d e r f r a c t u r e m e c h a n i c s l i f e p r e d i c t i o n . None - the - l e s s , t h e
phenomenological and sc ient i f ic foundat ions have been developed
d u r i n g the past 15 years . The cha l lenge is to r e f i n e and inte-
gra te this u n d e r s t a n d i n g to p roduce m e t h o d s w h i c h o v e r c o m e the
w e a k n e s s e s o f s m o o t h s p e c i m e n , t i m e - i n d i f f e r e n t des ign rules.
Signi f icant progress has been reported in this regard for stress
corrosion and corrosion f a t i gue in nuclear sys tems, based on f i l m
rup tu re mechanis t ic modeling [109].
Recent advances and f u t u r e directions for l i fe predict ion of
steels cracking by hydrogen embr i t t l emen t are rev iewed for static
and cyclic loading.
Stress C o r r o s i o n C r a c k i n g . Stress c o r r o s i o n c r a c k i n g i s
s ens ib ly p r e d i c t e d based on the th resho ld s t ress i n t e n s i t y
( K I s c c ) c o n c e p t , w i t h i n t h e b o u n d s s u m m a r i z e d e l s e w h e r e [110] .
Two variables cr i t ical ly a f f ec t Kj s c c : steel yield s trength and
hydrogen uptake .
Lower bound re la t ionships be tween Kj s c c and yield s t rength
are presented in Fig. 19, based on over 500 measu remen t s reported
d u r i n g the past 20 years [111]. The bene f i c i a l e f f e c t of de-
- 29 -
creasing strength is shown for f i ve speci f ic environments . For
constant strength, K i s c c > correlates wi th the steady state con-
centration of dissolved hydrogen, produced by gaseous chemical or
aqueous electrochemical reactions on the input surface of a steel
foi l and measured by a permeation expe r imen t , as il lustrated in
F ig . 2 0 [ 1 1 1 , 1 1 2 ] , Such d a t a , w h e n e m p l o y e d w i t h p e r m e a t i o n
based sensors of h y d r o g e n u p t a k e in p lan t c o m p o n e n t s [ 1 1 3 ] »
enable conservative l i f e predict ions for environmental cracking
u n d e r m o n o t o n i c load ing . This a p p r o a c h may be c o m p r o m i s e d by
cyclic loading, unique micros t ructures and new envi ronment chem-
istries [ 1 1 0 , 1 1 1 ] .
Corrosion Fatigue. S igni f icant progress was achieved in the
1980s for f rac ture mechanics predictions of long term corrosion
fa t i gue in o f f shore structural applicat ions, par t icular ly large
w e l d e d j o i n t s b e t w e e n l ow s t r e n g t h ca rbon steel t u b u l a r s
[ 1 0 9 , 1 1 ^ ] - The elements of this approach are rev iewed.
(1) Data base. Corrosion fa t igue cracking in the low alloy
s t e e l / a q u e o u s e n v i r o n m e n t s y s t e m occurs a t s tress i n t e n s i t i e s
well below Kj s c c and wi th in the regime relevant to tubular joint
p e r f o r m a n c e , as i l l u s t r a t ed by the da t a and a r r o w in F ig . 21
[93,115-119]. Crack growth is enhanced by reduced loading fre-
quency and increased cathodic polarization; such effects mus t beQ
u n d e r s t o o d to p r e d i c t the long t e r m (10 cycles a t 0 .1 Hz) l i f e
of cathodical ly protected tubulars .
(2) 'Hydrogen e m b r i t t l e m e n t models. The aim of mechan i s t i c
mode l ing is to predict corrosion f a t i g u e crack growth rate as a
- 30 -
funct ion of stress intensi ty, stress ratio, environment composi-
t ion, micros t ruc ture , electrode potential and loading f requency.
No s ing le , q u a n t i t a t i v e m o d e l exis ts . The f o u n d a t i o n has been
establ ished, however , by research on crack tip reactions and the
f r e q u e n c y dependence of corrosion f a t i g u e [ 4 4 , 5 9 , 7 5 ] , on crack
c l o s u r e [ 8 9 ] , o n s t e a d y s t a t e c r a c k c h e m i s t r y m o d e l i n g
[100 ,102 ,105-108] , and on hydrogen embr i t t lement [115,120] . Me-
chanisms for near- threshold corrosion fa t igue have been proposed
also [ 9 0 , 9 1 , 9 3 , 9 4 , 1 1 5 , 1 2 1 ] , h o w e v e r , q u a n t i t a t i v e f o r m u l a t i o n s
and experimental evaluations are lacking.
(3) Stress i n t e n s i t y and l i f e p r e d i c t i o n s . To da t e , only
a p p r o x i m a t e K so lu t ions for c o m p l e x t u b u l a r j o in t s have been
developed [122]. Such equations were integrated w i th laboratory
crack g rowth rate data (Fig. 21) to yield the predic t ion shown in
Fig. 22 [122] .
Full scale componen t testing. Sophisticated capabili-
ties exist in several countries to conduct f a t igue and corrosion
f a t i g u e e x p e r i m e n t s w i t h large w e l d e d t ubu la r j o i n t s , a n d t o
moni tor crack growth continuously by electrical potential tech-
niques [123-125]. As shown in Fig. 22, f a t igue l i fe data on ful l -
scale joints are in excellent agreement w i th f r ac tu re mechanics
predict ions for cycle dependent c racking in mo i s t air. Measure-
ments of surface crack shape development are also consistent w i t h
model predic t ions .
L i m i t e d corrosion f a t i g u e expe r imen t s on full-scale jo ints
demons t ra te the deleter ious e f f e c t of the seawater e n v i r o n m e n t ,
- 31 -
potential problems due to cathodic protection and the inability
of simple cycle based fatigue design rules to adequately describe
tubular life [109]. Modeling of these effects requires improved
crack growth rate data, mechanistic models and understanding of
similitude, particularly for near threshold corrosion fatigue.
OUTSTANDING ISSUES AND NEW DIRECTIONS
The foregoing sections provide a perspective summary of the
modeling effort to quantitatively connect chemical and physical
processes with environmental crack growth response. From an ini-
tial effort that was narrowly focused on the rate controlling
processes for crack growth in steels exposed to gaseous environ-
ments, ensuing studies broadened the understanding to cracking in
high-strength aluminum and titanium alloys, and extended the
approach to the complex problem of crack growth in aqueous envi-
ronments. The simplified models serve as a framework for re-
search and design, as illustrated for offshore structures. They
serve also as a basis for furthering the understanding of en-
vironmentally assisted crack growth. Results from these studies
will aid in the development of alloys and of methods to minimize
sensitivity to environmentally assisted cracking, and of pro-
cedures for making reliable predictions of long-term service.
Fracture mechanics has played a major role in the development of
this understanding. Significant issues remain to be resolved.
(i) Detailed understanding of the kinetics and mechanisms of
surface/electrochemical reactions with clean metal surfaces is
needed, over a broad range of environmental conditions: (a) to
- 32 -
establish the fo rm and quanti ty of hydrogen that is produced, and
the f r a c t i o n that enters the m a t e r i a l to e f f e c t e m b r i t t l e m e n t ,
(b) to exp lore and c o n f i r m the i n d i c a t e d t r a n s f e r of ra te con-
trolling processes w i t h changes in environmental condi t ions, and
(c) to de t e rmine the relationship between the kinet ics of surface
reac t ions and c rack g r o w t h response to i m p r o v e the p r e d i c t i v e
capabil i ty of crack growth models.
( i i ) Fo r e n v i r o n m e n t a l l y ass i s ted c rack g r o w t h in aqueous
e n v i r o n m e n t s , the t r a d i t i o n a l e l e c t r o c h e m i c a l m e a s u r e m e n t o f
p o l a r i z a t i o n re sponse is i n a d e q u a t e . A new t e c h n i q u e for mea-
suring the kinetics of equil ibrat ion reactions at the crack tip
has been developed. These measuremen t s , however , mus t be coupled
w i t h a de t a i l ed u n d e r s t a n d i n g of the r e a c t i o n m e c h a n i s m s and
model ing of the crack tip chemis t ry .
( i i i ) Greater e f f o r t is needed to unders tand the factors and
processes tha t control or i n h i b i t e l e c t r o c h e m i c a l r eac t i ons of
bare, s t ra in ing metal surfaces w i th electrolytes.
( i v ) Q u a n t i t a t i v e u n d e r s t a n d i n g o f the p h y s i c a l - c h e m i c a l
interact ions be tween hydrogen and metal (i.e., the embr i t t lement
m e c h a n i s m s ) is needed to e s t ab l i sh the roles of m i c r o s t r u c t u r e
and of other metal lurgical variables in de te rmin ing the rate of
crack g rowth , or the degree of suscept ibi l i ty .
(v) Bet ter unders tanding of the inf luences of al loying and
i m p u r i t y e l emen t s , and of mic ros t ruc tu re is needed. It is essen-
tial to d e t e r m i n e whether such inf luences on env i ronmen ta l crack-
i n g r e s u l t f r o m a l t e r a t i o n s o f t h e r e a c t i o n k i n e t i c s ( c h e m i c a l
e f f e c t s ) , f r o m t h e i r i n f l u e n c e o n t h e m e c h a n i c a l p r o p e r t i e s o f
- 33 -
alloys (physical e f f e c t s ) , or both (physical-chemical e f fec ts ) .
( v i ) Be t t e r u n d e r s t a n d i n g of the processes that cont ro l
environmental crack ini t ia t ion and early growth (viz . , threshold
and Stage I crack growth) is needed.
( v i i ) Mechanis t ic descriptions of crack chemis t ry , transient
react ions and micromechanica l embr i t t l emen t must be integrated to
produce a p r e d i c t i v e mode l of e n v i r o n m e n t a l l y assisted crack
growth rate. The model must quan t i t a t ive ly predict both specimen
and c o m p o n e n t c r a c k i n g p e r f o r m a n c e , and mus t be amenab le to
exper imenta l conf i rma t ions and mechanis t ic r e f inemen t s .
(v i i i ) Exper imenta l research and modeling mus t be exploited
to d e v e l o p i n - s i t u m o n i t o r s of e n v i r o n m e n t a l c rack g r o w t h in
c o m p l e x c o m p o n e n t s . Sensors f o r h y d r o g e n u p t a k e f r o m se rv ice
envi ronments mus t be fu r ther developed and imp lemen ted .
( i x ) New exper imental methods must be developed for direct
measurement of environmental damage processes at the crack tip.
(x) Research on envi ronmenta l ly assisted crack ing must be
extended to include novel monol i th ic and compos i te materials .
Summary
E x p e r i m e n t a l and a n a l y t i c a l w o r k in the past decade has
contr ibuted s ign i f i can t ly to both the phenomenological and mecha-
n i s t i c u n d e r s t a n d i n g of e n v i r o n m e n t a l l y assisted crack growth.
Crack growth response ref lects the complex interplay among chemi-
cal , m e c h a n i c a l and m e t a l l u r g i c a l f a c t o r s , and i s d e p e n d e n t on
the ra te c o n t r o l l i n g p rocesses , and on the m i c r o m e c h a n i s m s fo r
c r ack g r o w t h and the m e c h a n i s m s o f the r e l evan t c h e m i c a l reac-
- 34 -
tions. On the basis of this understanding, modeling of environ-
mentally assisted crack growth, under sustained-load and in fa-
tigue, has been made. This modeling effort has placed the study
of this technologically important problem on a sound footing, and
provides a framework for new understanding and for the develop-
ment and utilization of data in design. To make significant
further advances in understanding, continued emphasis on multi-
disciplinary approaches which incorporate chemistry, physics,
materials science and fracture mechanics, and long-term support
are essential.
ACKNOWLEDGEMENT
This work is supported by the Office of Naval Research under
Contract N00014-83-K-0107 NR 036-097 and by the National Aeronau-
tics and Space Administration, Langley Research Center, Research
Grant NAG-1-745.
REFERENCES
1. Brown, B. F., in Stress-Corrosion Cracking in High StrengthSteels and in Titanium and Aluminum Alloys, e d., B. F.Brown, U. S. Naval Research Laboratory, Washington, DC(1972), pp. 2-16. (Library of Congress Catalog Card No. 72-600053)
2. Haigh, B. P., J. Inst. Metals, Vol. 18, 1917, p. 55.
3. Gough, H. J. and Sopwith, D. G.,. J. Inst. Metals, Vol. 49,' 1932, p. 93.
4. The Theory of Stresss Corrosion Cracking in Alloys, ed. ,J. C. Scully, NATO, Brussels, Belfium, 1971.
5. Fundamental Aspects of Stress Corrosion Cracking, NACE-1,eds., R. W. Staehle, A. J. Forty, and D. van Rooyen, Natl.Assoc. Corro. Engrs., Houston, TX, 1969.
- 35 -
6. Corrosion Fatigue: Chemis t ry , Mechanics and Mic ros t ruc tu re ,N A C E - 2 , eds . , 0 . F . D e v e r e u x , A. J . M c E v i l y , and R. W.Staehle , Nat l . Assoc. Corro. Engrs . , Hous ton , TX, 1972.
7. Stress Corrosion Cracking and Hydrogen E m b r i t t l e m e n t of IronBase A l loys , N A C E - 5 , eds . , R . W. S taeh le , J . H o c h m a n n , R .D. M c C r i g h t and J . E . S la te r , N a t l . Assoc . C o r r o . E n g r s . ,Hous ton , TX, 1978.
8 . L ' H y d r o g e n e Dans Les M e t a u x , eds. , J . P . F i d e l l e , and M.R a p i n , C o m m i s s a r i a t Energie A t o m i q u e , France, 1967.
9 • H yjLL2IL® H AH M-lJi.3. A^.» e d s - » I • M . B e r n s t e i n and A. W.Thompson , Am. Soc. Metals , Metal Park , OH, 1974.
10. E f f e c t of H y d r o g e n on B e h a v i o r of M a t e r i a l s , eds . , A. W.T h o m p s o n and I . M. B e r n s t e i n , Am. Soc. M e t a l s , M e t a lP a r k , OH, 1 976.
11. H y d r o g e n in M e t a l s , Proc . of Second Japan Inst . of M e t a l sIn t e r . S y m p o s i u m ( J I M S - 2 ) , Suppl . t o Trans . J apan Inst .M e t a l s , Vol . 21 , 1980 .
12. Env i ronmen t -Sens i t i ve Fracture of Engineer ing Mater ia ls ,ed . , Z . A . F o r o u l i s , T h e M e t a l l . Soc. A I M E , N e w Y o r k , 1 9 7 9 .
13- Hydrogen E m b r i t t l e m e n t and Stress Corrosion Crack ing , aF e s t s h r i f t , e d s . , R . G i b a l a and R . F . H o c h m a n n , Am. Soc.Meta l s , Meta ls P a r k , OH, 1984.
1 4 . W e i , R . P . , i n E n v i r o n m e n t a l D e g r a d a t i o n o f E n g i n e e r i n gMate r i a l s in Aggress ive E n v i r o n m e n t s , eds., M. R. Lou t rhan ,Jr . , R . P . M c N i t t and R. D. Sisson, Jr., V i r g i n i a Po ly-t e c h n i c I n s t i t u t e a n d S ta te U n i v e r s i t y , B l a c k s b u r g , V A ,1 981 , pp. 73-81 , Vol . 2.
1 5. Env i ronment -Sens i t ive Fracture; Evalua t ion and Compar i son ofTest M e t h o d s , A S T M STP 821 , eds . , S . W. D e a n , E . N. P u g h ,a n d G . M . U g i a n s k y , A m e r i c a n S o c i e t y f o r T e s t i n g a n d M a t e -rials, 1916 Race St., Phi ladelphia , PA, 19103, 1984.
16. E m b r i t t l e m e n t by the Local ized Crack E n v i r o n m e n t , ed., R. P.Ganglof f , The Metall. Soc. A I M E , New York , 1984.
17. M o d e l i n g E n v i r o n m e n t a l E f f e c t s on Crack In i t i a t ion and Prop a g a t i o n , eds . , R . H . Jones and W. W. G e r b e r i c h , The M e t a l l .Soc. A I M E , W a r r e n d a l e , PA, 1986.
18 . C o r r o s i o n F a t i g u e T e c h n o l o g y , A S T M STP 6 4 2 , eds . , H . L .C r a i g , J r . , T . W . C r o c k e r a n d D . W . H o e p p n e r , A S T M ,Phi lade lph ia , P A ( 1 9 7 8 ) .
- 36 -
19. Hydrogen E f f ec t s in Meta ls , eds., I. M. Bernstein and A. W.Thompson, T M S - A I M E , War renda le , PA, p . 1005 ( 1 9 8 1 ) .
20. Bas i c Q u e s t i o n s in F a t i g u e , Vols . I and II , A S T M S T P , eds. ,W. P. W e i , J. T. Fong , R. J. F i e ld s , and R. P. G a n g l o f f ,ASTM, Phi ladelphia , PA, in press ( 1 9 8 7 ) .
21 . Env i ronmen ta l Degradat ion of Engineer ing Mater ia ls-I l l ,e d s . , M . R . L o u t h a n , R . P . M c N i t t a n d R . D . S i s s o n ,Pennsylvania State U n i v e r s i t y , 1987.
22. M e c h a n i s m s of E n v i r o n m e m t Sensit ive Crack ing of Mater ia l s ,eds. , P. R. S w a n n , F. P. Ford and A. R. C. W e s t w o o d , TheMetals Society, London, 1977.
23. C o r r o s i o n F a t i g u e , A S T M STP 801, eds . , T. W. Crocke r and B.N . L e i s , A m . Soc. T e s t i n g M a t l s . , P h i l a d e l p h i a , P A , 1983.
24 . S t aeh l e , R . W. , F ina l R e p o r t on "Stress C o r r o s i o n C r a c k i n gof H i g h S t r e n g t h A l l o y s " , A F M L - T R - , 1 9 7 0 ( to be com-pleted ) .
25. B r o w n , B. F . and B e a c h e m , C. D. , Cor ro . Sci . , Vo l . 5 , 1 9 6 5 ,pp. 745-750.
26. B r o w n , B. F . , M a t e r . Res . S tand . , Vo l . 6 , No. 3 , 1966 ,p. 129 .
27. J o h n s o n , H. H. and P a r i s , P. C. , Eng . F rac t . M e c h . , Vol . 1 ,1968, pp. 3-^5.
28. S m i t h , H. R. , P i p e r , D. E . and D o w n e y , F . K. , Eng. Frac t .M e c h . , Vol . 1 , 1 9 6 8 , pp . 123-128.
29 . F e e n e y , J . A . , M c M i l l a n , J . C . and W e i , R . P . , Me ta l l .T rans . . Vol . 1 , 1 9 7 0 , pp . 1 7 4 1 - 1 7 5 7 .
30 . W e i , R . P . , in F u n d a m e n t a l A s p e c t s o f Stress C o r r o s i o nC r a c k i n g , N A C E - 1 , eds . , R. W. S t aeh le , A. J . Fo r ty and D.v a n R o o y e n , N a t l . Assoc . C o r r o . E n g r s . , H o u s t o n , T X , 1969 ,pp . 104 -112 .
31. S t e i g e r w a l d , E. A., "Delayed f a i lu re of h igh-s t rength steeli n l i q u i d e n v i r o n m e n t s " , P r o c . A . S . T . M . 6 0 , 1 9 6 0 , p . 750.
32 . H a n n a , G . L . , T r o i a n o , A . R . , and S t e i g e r w a l d , E . A . , ASMTrans . Q u a r t e r l y , Vol . 57, 1 9 6 4 , p . 658.
33 . J o h n s o n , H . H . and W i l l n e r , A . M. , A p p l i e d M a t e r . R e s . ,Vol. 1 , 1 9 6 5 , p . 3^.
34 . H a n c o c k , G . G . and J o h n s o n , H . H . , T rans . M e t a l l . Soc. A I M S ,Vol . 236 , 1 966 , p. 513.
- 37 -
35. B r a d s h a w , F . J . and W h e e l e r , C. , A p p l i e d M a t e r . R e s . ,Vol. 5, 1 9 6 6 , p. 112.
36. H a r t m a n , A., Inter. J. Fract . M e c h . , Vol. 1, 1965, p. 167.
37 . W e i , R . P . , Ta lda , P . M. and L i , C h e - Y u , i n F a t i g u e C r a c kP r o p a g a t i o n , A S T M S T P 4 1 5 , A m . Soc. T e s t i n g M a t e r . , Ph i la -d e l p h i a , P A , 1 9 6 7 , p p . 4 6 0 - 4 8 5 .
38. W e i , R. P . , N o v a k , S . R. and W i l l i a m s , D. P . , M a t e r . Res .S t and . , Vol . 12 , 1 9 7 2 , p . 25.
39. N o v a k , S. R. and R o l f e , S. T. , C o r r o s i o n , Vol . 26, No. 4,1970, pp. 1 21-130 .
40 . W e i , R . P . , in F a t i g u e M e c h a n i s m s , A S T M STP 675 , ed. , J . T .F o n g , A m . Soc. T e s t i n g M a t e r . , P h i l a d e l p h i a , P A , 1 9 7 9 ,pp. 816-840.
41 . W e i , R . P . , i n "Steels", A R P A C o u p l i n g P r o g r a m on StressCorrosion Cracking: (Seventh Quar ter ly R e p o r t ) , NRL Memoran-dum Repor t 1 9 4 1 , Naval Research Laboratory, Wash ing ton , DC,O c t o b e r 1 9 6 8 , p . 49 .
42 . S p e i d e l , M. 0 . , " C u r r e n t U n d e r s t a n d i n g o f Stress C o r r o s i o nCrack Growth in A l u m i n u m Alloys", in The Theory of StressCorros ion Crack ing in Alloys, ed., J. C. Scully, NATO Scien-t i f i c A f f a i r s Div i s ion , Brussels, 1971 , pp . 289-344.
4 3 . W e i , R . P . , K l i e r , K . , S i m m o n s , G . W . a n d C h o u , Y . T . , i nHydrogen E m b r i t t l e m e n t and Stress Corrosion Crack ing , eds.,R . G i b a l a and R . F . H e h e m a n n , A m e r i c a n S o c i e t y fo r M e t a l s ,Metals Park , OH, 1984, pp. 103-133-
44. R. P. Wei and M i n g Gao: in Hydrogen Degradat ion of FerrousA l l o y s , eds . , R . A. O r i a n i , J . P . H i r t h and M. S m i a l o w s k i ,pp . 5 7 9 - 6 0 7 ; N o y e s P u b l i c a t i o n s , P a r k R i d g e , N . J . , 1985.
45 . W i l l i a m s , D. P . and N e l s o n , H. G. , Me ta l l . Trans . , Vol . 1 ,1.970, p. 63.
46. N e l s o n , H. G. , W i l l i a m s , D. P. and T e t e l m a n , A. S. , M e t a l l .T r a n s . , Vol . 2 , 1 9 7 1 , p . 953.
47 . W e i , R . P . , In te r . J . F rac t . M e c h . , Vol . 4 , 1968, pp . 159-170.
48 . W e i , R . P . and L a n d e s , J . D . , M a t e r . Res . S t a n d . , Vo l . 9 ,No. 7, 1 9 6 9 , pp. 25-28 .
4 9 . G a o , M . a n d W e i , R . P . , M e t . T rans . A , Vol . 1 6 A , 1985,pp. 2039-2050.
- 38 -
50. G a n g l o f f , R. P . and W e i , R. P . , M e t . Trans . A, Vol . 8A,1977, pp. 1043-1053-
51. S i m m o n s , G. W., Pao , P. S. and W e i , R. P. : M e t . Trans A, 9_A(1 978 ) p. 1 147 .
52 . S i m m o n s , G . W. and D w y e r , D . J . : S u r f . Sci. , 48 ( 1 9 7 5 )p.373.
53. D w y e r , D. J . , S i m m o n s , G. W. and W e i , R. P . : S u r f . Sci., 64( 1 9 7 7 ) p . 617 .
54. D w y e r , D. J. : Ph. D. Disser ta t ion , Lehigh Un ive r s i t y , 1977.
55. W e i , R. P. and S i m m o n s , G. W.: in Stress Corrosion Crackingand Hydrogen E m b r i t t l e m e n t of Iron Base Alloys, eds., R. W.S t aeh l e , J . H o c h m a n n , R. D. M c C r i g h t and J . E. S la te r , pp.7 5 1 - 7 6 5 ; N A C E - 5 , N a t . Assoc . Cor ro . Engr s . , H o u s t o n , T X ,1979.
5 6 . W e i , R . P . , P a o , P . S . , H a r t , R . G . , W e i r , T . W . a n dS i m m o n s , G. W.: Met. Trans A, 11A (1980) pp. 151-158.
57 . B r a z i l l , R . S i m m o n s , G . W. and W e i , R . P . , J . Engr . M a t l s .& Tech . , Trans . A S M E , 1 01 ( J u l y 1 9 7 9 ) pp. 1 9 9 - 2 0 4 .
58 . W e i r , T . W. , S i m m o n s , G . W. , H a r t , R . G . and W e i , R . P . :Scripta Met . , 1 4 (1980) pp. 357-364.
59 . W e i , R . P . and S i m m o n s , G . W. : i n F A T I G U E ; E n v i r o n m e n t andT e m p e r a t u r e E f f e c t s , eds. , John J . B u r k e and V o l k e r W e i s s ;Sagamore A r m y Mater ia l s Research Conference Proceedings , 27( 1 9 8 3 ) PP. 59-70.
60 . Lu , M. , Pao , P . S . , W e i r , T . W. , S i m m o n s , G . W. and W e i , R .P. , M e t . Trans . A, 1 2A ( 1 9 8 1 ) pp. 805-811.
61 . Lu , M. , Pao , P . S . , C h a n , N . H. , K l i e r , K . and W e i , R . P . ,in Hydrogen in Meta ls , Suppl. to Trans. Japan Inst. Metals ,2J_ (1 980 ) p . 449 .
62 . C h a n , N . H . , K l i e r , K . and W e i , R . P . , i n H y d r o g e n in M e t -als, Suppl. to Trans. Japan Inst. Metals, 21 (1980) p. 305.
63. W e i , R. P., in Hydrogen E f f e c t s in Metals , eds., I. M. Bern-s t e in a n d A . W . T h o m p s o n , p . 677 ; T h e M e t a l l . S o c . - A I M E ,W a r r e n d a l e , PA, 1981.
64. D u s h m a n , S. , "Molecular F low" , in S c i e n t i f i c Foundat ions ofV a c u u m Technique , 2nd ed. , ed., J . M. L a f f e r t y , W i l e y , 1 962,p p . 8 7 - 1 0 4 .
- 39 -
65. Pasco, R. W. and F ico lo ra , P. J. , Sc r ip t a M e t . , 1 5 ( 1 9 8 0 ) p.1019.
66. P a s c o , R . W. , S i e r a d z k i , K. and F i c a l o r a , J . P . , Sc r ip t aM e t . , 1_6, 1982 , p. 881.
6 7 . C h a n , N . H . , K l i e r , K . a n d W e i , R . P . , S c r i p t a M e t . ,Vol . 12, 1978, pp. 1043-1046.
68. Pao , P. S. and W e i , R. P. , Sc r i p t a M e t . , 11 ( 1 9 7 7 ) p. 515-520.
69. Har t , E. W., in The Nature and Behavior of Grain Boundaries,H s u n H u , ed. , P l e n u m P u b l . C o r p . , N e w Y o r k , 1 9 7 0 , p . 155.
70. W e i , R. P. and L a n d e s , J . D.: M a t e r . Res . S tand. , 9^ (7)( 1 9 6 9 ) pp . 25-28.
71. W e i , R. P . and Gao , M i n g , Sc r ip t a M e t . , 17 ( 1 9 8 3 ) pp . 959-962.
72. Sh in , T r u e - H w a and W e i , R. P . , Engr . Frac. M e c h . , 1 8 (4)( 1 9 8 3 ) PP- 827-837.
73. W e i , R. P . and S h i m , G. , in C o r r o s i o n F a t i g u e , A S T M STP 801,eds . , T. W. C r o o k e r and B. N. L e i s , Am. Soc. T e s t i n g &Mater . , Phi ladelphia , PA 19103, pp. 5-25.
74. S h i m , G u n c h o o , and W e i , R. P . , Ma t s . Sci. & E n g r . , Vol. 86,1 9 8 6 , pp. 1 21-135 .
75. W e i , R. P., "Corrosion Fatigue Crack Growth" , Proceedings ofI n t e r n a t i o n a l S y m p o s i u m on M i c r o s t r u c t u r e and M e c h a n i c a lB e h a v i o u r o f M a t e r i a l s , 2 1 - 2 4 Oc t . , 1985, X i ' a n , Peop le ' sR e p u b l i c o f C h i n a , E . M . A . S . , W a r l e y , E n g l a n d ( t o b ep u b l i s h e d ) .
76 . K i m , Y. H. , S p e a k e r , S . M. , G o r d o n , D. E . , M a n n i n g , S . D . ,a n d W e i , R . P . , N a v a l A i r D e v e l o p m e n t C e n t e r ( 6 0 4 ) , W a r -m i n s t e r , PA, Report No. NADC-83126-60 , Vol. I , March , 1983.
77. Pao , P. S. , and W e i , R. P. , in T i t a n i u m : Sc ience and Tech-nology, eds., G . L u t j e r i n g , U . Z w i c k e r and W. B u n k , FRG:
' Deutsche Gesel lschaft Fur Meta l l -kunde e.V., 1985, p. 2503.
78 . B i r n b a u m , H. K. , "On the m e c h a n i s m s of h y d r o g e n r e l a t edf r a c t u r e i n me t a l s " , P r o c . o f ONR W o r k s h o p on E n v i r o n m e n t -S e n s i t i v e F r a c t u r e o f M e t a l s a n d A l l o y s , W a s h i n g t o n , D C ,J u n e 3 , 1985 ( to be p u b l i s h e d ) .
7 9 . P e t e r s o n , K . P . , S c h w a n e b e c k , J . C . , a n d G e r b e r i c h , W . W . ,M e t . Trans . A , Vo l . 9 A , 1 9 7 8 , p . 1 1 6 9 .
- 40 -
80. Nakai, Y., Alavi, A., and Wei, R. P., "Effects of Frequencyand Temperature on Short Fatigue Crack Growth in AqueousEnvironments", Submitted for publication in Met. Trans. A.
81. Wei, R. P., Gao, M. and Pao, P. S., Scripta Met., Vol. 18,1984, pp. 1195-1198.
82. Chiou, S., and Wei, R. P., "Corrosion Fatigue CrackingResponse of Beta Annealed Ti-6Al-4V Alloy in 3.5% NaClSolution", Report No. NADC-83126-60 (Vol. V), U. S. NavalAir Development Center, Warminster, PA, 30 June 1984.
83. Gerberich, W. W., and Peterson, K. A., in Micro and MacroMechanics of Crack Growth, eds., K. Sadanada, B. B. Rath andD. J. Michel, The Metall. Soc. of AIME, Warrendale, PA,1 982, pp. 1-1 7.
84. Ressler, D., "An Examination of Fatigue Crack Growth In AISI4340 Steel in Respect to Two Corrosion Fatigue Models", M.S.Thesis, Dept. of Mech. Engg. & Mechanics, Lehigh University,Bethlehem, PA, 1984.
85. Gao, M., Pao, P. S., and Wei, R. P., in FRACTURE; Interac-tions of Microstructure, Mechanisms and Mechanics, eds., JTM. Wells and J.D. Landes, The Metall. Soc. of AIME, Warren-dale, PA, 1985, pp. 303-319.
86. Fatigue Crack Growth Threshold Concepts, eds., David David-son and Subra Suresh, TMS-AIME, Warrendale, PA, 1984.
87. Small Fatigue Cracks, eds., R.O. Ritchie and J. Lankford,TMS-AIME, Warrendale, PA, 1986.
88. Aluminum-Lithium Alloys II. eds., T.H. Sanders, Jr. and E.A.Starke, Jr., TMS-AIME, Warrendale, PA, 1984.
89. Suresh, S. and Ritchie, R. 0., in Fatigue Crack GrowthThreshold Concepts, eds., David Davidson and Subra Suresh,TMS-AIME, Warrendale, PA, 1984, pp. 227-261.
90. Gangloff, R. P. and Ritchie, R. 0., in Fundamentals ofDeformation and Fracture, eds., B.A. Bilby, K.J. Miller andJ.R. Willis, Cambridge University Press, Cambridge, UK,
' 1985, pp. 529-558.
91. Suresh, S., Zamiski, G. F. and Ritchie, R. 0., Metall.Trans. A., Vol. 12A, 1981, pp. 1435-1443.
92. Hartt, W. H., Culberson, C. H. and Smith, S. W., Corrosion,Vol. 40., 1 984, pp. 609-61 8.
93. Thorpe, T. W., Scott, P. M., Ranee, A. and Silvester, D.,Intl. J. Fatigue, Vol. 5., 1983, pp. 123-133.
94. G a n g l o f f , R. P . and D u q u e t t e , D. J . , in C h e m i s t r y and Phy ic sof F r a c t u r e , eds. , R . M . L a t a n i s i o n and R .H . Jones , M a r t i n u sN i j h o f f Publishers BV, Netherlands, in press, 1987.
95. G a n g l o f f , R. P . and W e i , R. P . , in Smal l F a t i g u e Cracks ,eds. , R .O. R i t c h i e and J . L a n k f o r d , T M S - A I M E , W a r r e n d a l e ,P A , 1 9 8 6 , p p . 239-264 .
96. R i t c h i e , R. 0. and L a n k f o r d , J . , M a t l s . Sci. E n g r . , Vol . 84,1986.
97 . T a n a k a , K. and W e i , R . P . , Engr . Frac t . M e c h . , Vol. 21 ,1985 , p . 293-305.
98 . Z e g h l o u l , A. , and P e t i t , J . , in Smal l F a t i g u e C r a c k s , eds . ,R.O. R i t c h i e a n d J . L a n k f o r d , T M S - A I M E , W a r r e n d a l e , P A ,1986, pp . 225-235.
99. Fo rd , F. P. and H u d a k , Jr. , S. J. , in Smal l F a t i g u e C r a c k s ,eds. , R .O . R i t c h i e a n d J . L a n k f o r d , T M S - A I M E , W a r r e n d a l e ,PA, 1986, pp. 289-308.
100. G a n g l o f f , R. P. and Turnbul l , A., in Model ing Envi ronmenta lE f f e c t s on C r a c k I n i t i a t i o n and P r o p a g a t i o n , eds. , R . H .Jones and W . W . Gerber ich , T M S - A I M E , Warrendale , Pa., 1986,pp. 55-81.
101. C l a r k , Jr . , W. G . , i n C r a c k s and F r a c t u r e , A S T M STP 601 ,ASTM, Philadelphia, PA, 1976, pp. 138-153.
102. Brown, B.F., in Stress Corrosion Cracking and HydrogenE m b r i t t l e m e n t of I ron Based Al loys , eds., J. Hoch raann , J.S la te r , R . D . M c C r i g h t a n d R . W . S taeh le , N A C E , H o u s t o n , T X ,1 9 7 6 , pp . 747 -751 .
103. G a n g l o f f , R. P . , Res. Mech . Let . , Vol . 1 , 1981 , pp. 299-306.
104 . G a n g l o f f , R . P . , M e t a l l . Trans . A._, Vol . 1 6 A . , 1985,.pp. 953-969.
105. Gangloff, R. P., in Embrittlement by the Localized CrackEnvironment, ed., R.P. Gangloff, TMS-AIME, Warrendale, PA,1984, pp. 265-290.
106. Gangloff, R. P., in Critical Issues in Reducing the Corrosion of Steel, eds., H. Leidheiser, Jr. and S. Haruyama,NSF/JSPS, Tokyo, Japan, 1985, pp. 28-50.
107. Turnbull, A. and Ferriss, D. H., in Proc. Conf. CorrosionChemistry within Pits, Crevices and Cracks, ed., A.Turnbull, NPL, Teddington, UK, in press,1987.
- 42 -
108. Turnbul l , A., in Embr i t t l emen t by the Localized Crack Env i r o n m e n t , ed . , R .P . G a n g l o f f , T M S - A I M E , W a r r e n d a l e , P A ,1 9 8 4 , pp. 3-48.
109. A n d r e s e n , P. L. , G a n g l o f f , R. P. , C o f f i n , L. F. and F o r d ,F. P. , in F a t i g u e 87. eds. , R.O. R i t c h i e and E.A. S t a r k e ,Jr., EMAS, Wes t Mid l ands , England, in press, 1987.
110. "Character izat ion of Environmenta l ly Assisted Crack ing forD e s i g n " , N a t i o n a l M a t e r i a l s A d v i s o r y Board , R e p o r t N M A B -386, 1982.
1 1 1 . G a n g l o f f , R. P., in Corrosion Prevent ion and Control, Proc.33rd Sagamore Army Materials Research Conference, , ed., S.Isserow, U.S. A r m y Mater ia ls Technology Laboratory , Water-town, MA, in press, 1987.
112. Y a m a k a w a , K. , T s u b a k i n o , H. and Y o s h i z a w a , S . , in C r i t i c a lIssues in R e d u c i n g the Cor ros ion of Steel , eds. , H. Le id -h e i s e r , J r . and S . H a r u y a m a , N S F / J S P S , T o k y o , J a p a n , 1985,pp. 348-358.
113. Y a m a k a w a , K., Harushige , T. and Yosh izawa , S., in CorrosionMoni to r ing in Industr ial Plants Using Nondest ruct ive Testing and E l e c t r o c h e m i c a l M e t h o d s , A S T M STP 908, eds., G.C.Moran and P. Labine, ASTM, Philadelphia, PA, 1986, pp. 221-236.
114 . D o v e r , W. D. and D h a r m a v a s a n , S . in F a t i g u e 84 , ed . , C.J .Beevers, E M A S , West Midlands , England, 1984 , pp . 1417-1434 .
115. A u s t e n , I . M . and W a l k e r , E . F . , in F a t i g u e 84 , ed. , C.J.Beevers, E M A S , West Mid lands , England, 1984, pp. 1457-1469.
116. C i a i o n e , H. J . and H o l b r o o k , H. J . , Me ta l l . Trans . A. , Vol.1 6 A , 1985 , p p . 115 -122 .
117 . V o s i k o v s k y , 0. and C o o k e , R. J . , In t . J . Pres . Ves . & P i p -_ing_, Vol . 6 . , 1978 , pp. 1 1 3 - 1 2 9 .
118. . V o s i k o v s k y , 0. J . Test . Eva l . , 1978, p p . 1 7 5 - 1 8 2 .
119. V o s i k o v s k y , 0. J. Test . Eva l . , Vol . 8., 1 9 8 0 , pp. 68-73.
120. Sco t t , P . M. , T h o r p e , T . W. and S i l v e s t e r , D . R . V . , Corros ion Sc i ence , Vol . 2 3 - , No. 6 , 1 9 8 3 , pp . 559-575.
121. D u q u e t t e , D . J . and U h l i g , H . H . , Trans . A S M , V o l . 61 . ,1 968 , pp. 4 4 9 - 4 5 6 .
122. H u d a k , S. J . , B u r n s i d e , 0. H. and C h a n , K. S. , J . E n e r g yR e s o u r c e s Tech . , A S M E T r a n s . , Vol . 107 , 1985 , p p . 2 1 2 - 2 1 9 .
123- Fatigue in Offshore Structural Steels, Thomas Telford,Ltd., London, 1 981 .
124. Proceedings of the International Conference on Steel inMarine Structures, Comptoir des produits Siderurgiques,Paris,1961.
125. Proceedings Institute of Mechanical Engineers Conference onFatigue and Crack Growth in Offshore Structures, Inst.Mech. Engr., London, 1986.
3O
28r
< 25
I2'H
5400
COcou
5200-^ooZo
5000^I
H
Z
46OO
io'4 icr3 icr2
CRACK GROWTH RAID, d2a/dt, in/sec.
Fig . 1 - Stress in tens i ty fac to r and net-sect ion stress versusc r a c k g r o w t h r a t e fo r a w e d g e - f o r c e - l o a d e d s p e c i m e n[28].
i<r*
c
^z
o
UJ
(E
H
0CC
Xo
IT
10'
Pool 1*
AWEOGE FORCE1 . ff
^REMOTE J /^
• WEDGE FORCE »5 »
oRCMOTE *• °
A
3.5%NaCI jt
«
A
f#
Co
a
I
=1 "
\v-(_ flcn.o!e f
1 _i_J L
Pon.l• WEDOE FORCS1•REMOTE J "• WEOOE FORCE *t•REMOTE «5
c
iDry Air (IO%R.K) *•>
&
1*
***
^*
J^O *
3rf
4* •>
7 $^
t ;»rtgexce *
^) 681" 2O 2 4 6 8 IO
Fig. 2 - C o m p a t i b i l i t y da ta f o r a l u m i n u m alloy 7075-T6 u p o nintensity fac tor .
IE
Fig. 3 - Schematic of monotonic load environment enhanced crackgrowth response.
.24
.20
.16
.2 .12
' .08
.04
Woter
^Humidified Argon Plus Oxygen(Equal Volumes)
midified ArgonPlus Hydrogdn
Pure Hydrogen
Humidified Argon Plus Oxygen(Equal Volumes)
I I
10 15 2O 25Time(Minutes)
30
Fig. S u b c r i t i c a l c r ack g r o w t h i n d i f f e r e n t w a t e r , w a t e rvapor , hydrogen , and oxygen envi ronments , H-11 steel,230 ksi yield strength [34].
Local Stress
Crack Tip RegionFracture
Zone
1. Gas Phase Transport2. Physical Adsorption
3. Dissociative Chemical Adsorption4. Hydrogen Entry5. Diffusion
EmbrittlemenfReaction
Fig. 5 - Schematic illustration of sequential processes inenvironmental enhancement of crack growth by hydro-genous gases. Embrittlement by hydrogen is assumedand is schematically depicted by the metal-hydrogen-metal bond.
TEMPERATURE CO140 120 00 80 60 40 20 0 -20 -40
! I 'AISI 434O STEEL
10 |— 'I 90pet ond 95pct confidence intervals
i 4.6:3.6kJ/mol«»90pct)
oa:o
2.5 3.0 3.5
I03/T IK"1)
4.0
RATE CONTROLLINGPROCESS
'E. (a) Diffusion
Z l10" g (b 1 Cos Phose Transport
IO"3 S 'c' Surface Reaction2 (H,-Metal)
IO'4(d) Surface Reaction
(H20-Metal)
Fig. 6 - Temperature dependence for Stage II crack growth andthe corresponding rate controlling processes at lowtemperatures for AISI 4340 steel in various hydro-genous environments: (a) H2S at 2.66 kPa, (b) H2S at0.13 kPa, (c) H2 at 133 kPa, and (d) H?0 (liquid)[51,60,61].
HYDROGEN SUPPLY
Surface Reaction ControlTransport ControlDiffusion Control
PRIOR AUSTENITEGRAIN BOUNDARIES'
{IIO}a'{ll2}a'Planes
Martensite Lath Boundariesor Patch Boundaries
MARTENSITE LATTICE
IG(T,P)
• Q C ( T , P )
MVC(T.P)
(da/dt) II
Fig. 7 - Schemat ic i l lus t ra t ion of the p a r t i t i o n i n g of hydrogenand i t s r e l a t i o n s h i p to h y d r o g e n s u p p l y and the re-su l t ing Stage II crack g rowth rate [49].
TEMPERATURE (°C)
ISO I 2 5 I O O 7 5 50 25 O -25
ID'
O<r
oH
10'
1
AISI 4340 STEEL IN HYDROGEN
J$ Measurements, 95% ConfidenceInterval
-50
10"
UJ
s
O<£O
cco1=1Ul
2.5 3.0 3.51000/T (K-1)
4.010"
Fig . 8 - C o m p a r i s o n b e t w e e n mode l p r e d i c t i o n s sus ta ined loadand c rack g r o w t h data on AISI 4 3 4 0 steel tes ted inh y d r o g e n a t d i f f e r e n t pressures and t e m p e r a t u r e s[ 4 9 , 6 1 ] .
Fig. 9 -
icf
icr1
-s-t_-t
-I^Q
torr
I0'2 10'' 5x10'' ,i I 1 I i i I | I i i i i i i i | i i ' | '
7O75-T65I Aluminum AlloyRoom Temperature ^j-
FrequencySHz ^-r' " "viwerLoad Ratio O.I AK= 15.5 MPa Vm" I3kp°
* • ' 1
invwc___ -̂ "̂ ^" vvwer
in°* " AKMLSMPo^m" '-3kPB
I. ... « 1
•̂ ^in Ar ^^ "̂̂in v°c_.^^^^'^|
inOi
I i t i i i 1 t i i i i i i i 1 t t i i i i i ii t
o-
01
KT5!.g
pn-»
WATER VAPOR PRESSURE, Pa
The i n f l u e n c e of w a t e r vapor p r e s su re and of o there n v i r o n m e n t s on f a t i g u e c rack g r o w t h in 7075-T651( A l M g Z n ) a l loy a t room t e m p e r a t u r e . The so l id l inesr e p r e s e n t p r e d i c t i o n s of a m o d e l for t r a n s p o r t - c o n -trolled crack g r o w t h . The dashed lines ind ica te sur-f a c e - r e a c t i o n - c o n t r o l l e d g r o w t h and r e f l e c t t he in -f luence of segregated m a g n e s i u m [85].
IOH
PRESSURE/FREQUENCY (Pa-s)
I 10 I02
io-
io-
10'
x
110''
I03
HjS, 5Hz
<IO'6 Pa, 5 Hz
AK.300
•
MPa40a
•
-m'"50
*T
1I 10 I02 IO3
HYDROGEN SULFIDE PRESSURE (Pa)
10"*
I0"
<JV.c
oT3
Fig. 10 -
Fig. 11 -
The inf luence of hydrogen sulf ide pressure on f a t iguecrack growth in a 2 - l A C r - 1 M o (A542 Class 2) steel atroom tempera ture . The dashed lines represent predic-t ions of a mode l for t r a n s p o r t - c o n t r o l l e d c rackg r o w t h . The solid l ines i n d i c a t e s u r f a c e - r e a c t i o n -control led growth and ref lect the second step of hy-drogen sul f ide- i ron surface reactions.
'10-6 10'FREQUENCY f (Hi)
i l O-7
CC
n:
acco
o
o
i-B
i—i—i " 1 1 1 1 1 1 1 rSHORT CRACK
10°1 1 i i 10-1 10-2
0.3M Na2S04 F.C.
CHARGE CRACK TEMP.TRANSFER GROWTH
RATEo 295 K* 348 K
10-5
UJ
10-6
10-' 10° io1EFFECTIVE TIME 1/f (sec)
acco
cco
The influence ofcrack growth for[80].
frequency and temperature on fatigueNiCrMoV steel in 0.3N Na2S04 solution
FREO.(Hr)10
o<ro
io-
I
Ti-6AI-4V</3)3.5 pet NoCI Solutionot Room Temperoture(R'O.OS)
X
N
5 AK'59MPoi/m
>44 MPoVin
»~^ AK'SZMPo/m
-—s-
AK-IBMPO/S;
to-s
10 I0I/FREO.(S)
Fig. 12 - The inf luence of f requency on fa t igue crack growth ina T i - 6 A l - 4 V al loy exposed to 0 .6M N a C l so lu t ion atroom tempera tu re and R = 0.1 [833-
0) O AK-22.2 MPa-m1/2
A AK-38.5 MPa-s!1/2
0. 1 0. 2 0. 5 1
Cycle PQriod:
2 5
1/f Cs]
Fig . 13 - I n f l u e n c e of f r e q u e n c y on f a t i g u e c r a c k g r o w t h r a t efor an A I S I 4 3 4 0 steel in w a t e r vapor at 585 Pa atroom t e m p e r a t u r e [84].
ORIGINAL PAGE ISOE POOR QUALITY
- 50 -
10 pm
8 Hz 1 Hz
10 pm
0.5 Hz 0.1 Hz
Fig. 14 - Microfractographs of AISI 4340 steel stressed in watervapor, showing changes in fracture surface morphologywith frequency [84].
- 51 -
i.o01
^ o. a
ou.LD
0. 6
COLcnL01
0. 4
0. 2
0.0
O AK-22.2 MPa-m1/z
A AK-38. 5 MPa-m1/2
0.0 0.2 0. 4 0.6 0.8 1.0
SurfacQ Reaction Coverage
Fig. 15 - Correlation between surface coverage and fraction ofintergranular failure (or 4>) for AISI 43^0 steel.
5 6 7 8 9 10 30 10 50i i I i i r
21/4 Cf-I Mo STEEL
SA 54Z Clais 3Frequency * 5O Hz , Ambient Temperature
B.
« O.O5 «et nyanxjen 038 kPo)• OO5 Or, hetum (IJ8 kPo)
o Threshold AKn
J I* It l ti l
I latticeipacjno -per cycle
1 1
10
3 * 5 6 7 8 9 1 0ALTERNATING STRESS INTENSITY. AK(MPo^m)
10 50 60
Fig. 16 - Oxide induced closure and hydrogen embrittlement ef-fects on corrosion fatigue crack propagation in a lowstrength bainitic steel exposed to moist, oxidizingand dry gaseous environments, (after Ritchie et al.[91])
- 52 -
sao.Sc1"EEin1O
XCM
"Jo
O0
£>.*•*'tocd>1=toto<u
30
25
20
15
10
5
n
_ 4130 SteelLiterature
3% NaCI-600 mV (SCE)
~~ K = 1.1 MPaVm/hroys=1330MPa ^ 0
— o ^
/
_ •• _L*• o°
'
_ • Theory
Ii *
i 1 . I0.1 1.0 10
Crack Depth Ahead of Notch (mm)
Fig. 17 - The measured and predicted e f fec t of precrack depth onthe t h r e s h o l d s t ress i n t e n s i t y fo r s t ress co r ros ionc r a c k i n g of high s t rength steel exposed to 3% NaCI.( a f t e r Ganglof f and Turnbull [100])
- 53 -
10" E
=4130 STEEL
10
I 10-
k-20)
O>.O
UJ
10
OCdO
O<ct:O
-4
10'
10.-6
10 I I I I I
AERATED 3?iNACLR=0.1 0.1 HZ
A SURFACE CRACK(.1-.9 MM)
• EDGE CRACK(.1-3. MM)
+ COMPACT TENSION(25-40 MM)
i i i
5 10 20 50 100
STRESS INTENSITY RANGE (MPa*m1/2)
Fig . 18 - The e f f e c t of shor t c r a c k s ize on c o r r o s i o n f a t i g u ep r o p a g a t i o n ra tes in h igh s t r e n g t h steel in 3% Nad.A p p l i e d s t ress r a n g e s a r e s h o w n f o r s u r f a c e c r acks ,( a f t e r Ganglof f [104 ] )
If
in
63
tn<now
CO
oeno
100
Lower Bound Threshold Stress Intensities for_\ • Hydrogen Embrit t lement of Alloy and C-Mn
\ Steels in 23°C Environments
40
20H2S +
Water Vapor
400 600 800 1000 1200 1400
Fig. 19 -'
Yield Strength (MPa)
L o w e r b o u n d s on K i s c c ~ ° y s f o r f e r r i te -pear l i te ,bain i t ic and t empered mar tens i t ic steels stressed inf i v e hydrogen producing envi ronments . Over 500 meas-urements are represented, (a f te r Gangloff [111] )
E(0Q.5
U)c
inin
co2o
C£.
2
1.9
1.8
1.7
1.6
1.5
1.4
1.3
1.2
1.1
1
0.9
0.8 r-
3 O
-K,1CQuenched and Tempered Steel
Oys = 1400 MPaAqueous ElectrolytesGaseous Hydrogen
300K
0.7-0.6
f t1 psi H2 3% NaCI 750° Psi H2~Mi I i I i I i I i I i i-0.2 0.2 0.6 1 1.4 1.8 2.2
Log Hydrogen Solubility (ppm wt)
2.6
Fig . 20 - Corre la t ion between K I S QQ and hydrogen uptake for highs t r e n g t h s teels f r a c t u r e d in gaseous h y d r o g e n andelectrolytes. ( a f t e r Gangloff [ 1 1 1 ] )
- 55 -
o
o
EN-^
UJ
XI-£oQCO
*o<DCO
10-6
10-7
10-8
10-9
i t i TTTT-T
8S4360 :50DStee l
5-10°C
O.IHz
-1 .10V(Ag/AgCI )
10
AK (MPa^/m)
100
Fig. 21 - Corrosion fatigue crack growth rate versus AK for lowstrength BS4360:50D C-Mn steel in seawater after Scottet al. [931; s e a w a t e r +
1 1 " L . [117] , a n d highH9S [115], oil
pressure - [116] .H2S
CD0.3
UJOz
COCOUJor\-coKOO.COI
H-OI
- 56 -
1000
100
10
I ' r n M T ' ' 'I ' 'ENVIRONMENT: AIRCHORD: 457mmXiemm
LOW R.BENDINQLOW R.AXIAL
. . .I
EXPERIMENTALA NEL. R = -1a TNO. R = 00 TNO. R=-1• Wl. R=0O DNV, R=0
• I
10' 10' 10' 10' 10'
CYCLIC LIFE
Fig. 22 - Fracture mechanics life prediction for fatigue crackpropagation in welded tubular joints exposed to moistair and compared to full scale component measurements[121].
AQUEOUS ENVIRONMENT EFFECTS ON INTRINSIC CORROSION FATIGUE
CRACK PROPAGATION IN AN Al-Li-Cu ALLOY
ROBERT S. PIASCIK AND RICHARD P. GANGLOFF
Department of Materials Science
University of Virginia
Charlottesville, VA 22901 USA
INTRODUCTION
The objective of this research is to characterize and
understand intrinsic fatigue crack propagation in advanced
aluminum-lithium based alloys, with emphasis on the damage
mechanisms for environmentally assisted transgranular cracking.
While the importance of complex extrinsic (namely crack
closure, delamination and crack deflection) contributions to
fatigue crack propagation (FCP) in Al-Li alloys is established
(1), gaseous and aqueous environmental effects have not been
examined. In this study, fracture mechanics experiments;
developed to characterize intrinsic FCP for benign environment
exposure and based on short crack, programmed stress intensity
methods (2); are extended to investigate the effect of aqueous
sodium chloride. Emphasis is placed on electrochemical control
of the environment and on high resolution measurements of fatigue
crack length changes in response to chemical variations.
EXPERIMENTAL PROCEDURE
A 3.8 cm thick plate of an advanced Al-Li-Cu-Zr alloy (AA
2090) was studied in the rolled, unrecrystallized, peakaged
Submitted for publication in: "Proceedings, World Congress onEnvironmental Cracking of Metals", M. B. Ives and R. P. Gangloff,eds., NACE, Houston, TX (1989).
condition. Alloy composition, heat treatment, mechanical
properties and microstructure, defined by optical and
transmission electron metallography, are summarized in Table 1.
Corrosion behavior was characterized by potentiodynamic
polarization experiments at a scan rate of 9.6 mV/min in helium
deaerated, aqueous NaCl at pH 8 and 297 K (3). Corrosion fatigue
experiments were conducted with specimens which were fully
immersed in flowing (30 ml/min), helium deaerated, 1% NaCl (pH 8)
in a sealed plexiglass chamber. Electrode potential was
maintained constant by a Wenking potentiostat and platinum
counter electrodes isolated from the solution through asbestos
frits.
Crack propagation experiments were conducted with single
edge notched specimens (10.16 mm wide, 2.54 mm thick, 0.25 or
0.89 mm notch depth) machined in the LT orientation at the one-
third position in the plate. Pinned, freely rotating grips were
employed, consistent with the stress intensity solution (4). The
growth of short edge cracks, sized between 0.25 and 5 mm, was
monitored continuously by a direct current electrical potential
method including 8 to 12 amps current, 0.1 uV potential
measurement resolution and an analytical calibration relation
(5). This method has been successfully applied to short cracks
in aluminum alloys, with better than 3 urn crack growth resolution
and long term (10 day) stability for near threshold growth rate
measurements (2).
Intrinsic and transient or steady-state corrosion fatigue
crack propagation is studied for both the moderate stress
intensity ( AK) Paris regime and near threshold growth rates.
Edge cracked specimens are stressed cyclically (sine waveform;
0.07 Hz < frequency < 20 Hz; 0.05 < stress ratio < 0.95) in a
closed loop servohydraulic machine operated in load control.
Employing computer measurement and feedback control, AK is
maintained constant over segments of growing crack length (about
0.5 to 1 mm), with step reductions in AK at constant K a ax,
producing increasing stress ratio (R). Crack growth rates
(da/dN) are calculated by linear regression analysis of crack
length versus load cycles data. For a single specimen, the load
sequence initially probes high AK, low R cracking; followed by
near threshold AK, high R fatigue for deeper crack depths.
Short crack length for the former and high stress ratio for the
latter conditions minimize complicating extrinsic effects.
Continuous reductions in A K at constant Kaiz provide an
effective characterization of intrinsic fatigue crack growth (6).
The procedure based on decreasing AK/step increasing R more
effectively characterizes transient to steady state crack
propagation associated with A^ changes, electrochemical
variables and microstructure.
RESULTS AND DISCUSSION
Corrosion Characteristics
Potentiodynamic polarization data are summarized in Fig. la
for peakaged alloy 2090 exposed to deaerated and oxygen
containing 1% NaCl solutions. In oxygen containing environments,
the open circuit potential of alloy 2090 is near the breakaway
potential, resulting in extensive surface pitting. Deaeration of
the NaCl solution limits the oxygen reduction reaction and shifts
the open circuit potential in the cathodic direction forming a
passive region extending from -1.0 V to -0.7 V (SCE). At passive
region potentials, general corrosion and pitting are greatly
reduced; constituent particles are none-the-less primary sites
for sub-breakaway pitting. Long term (8 day) constant potential
experiments performed at -0.760 V SCE in deaerated 1% NaCl reveal
that Ti precipitates, concentrated at subgrain boundaries, and
constituent particles are sites of localized attack, Fig. Ib.
Corrosion fatigue experiments were performed in deaerated 1%
NaCl to limit fracture surface corrosion and pitting, to reduce
crevice corrosion at the electrical potential probes and to
minimize electrolyte IR-type errors in measured potential. Since
the open circuit potential of alloy 2090 varies by 100 to 400 mV
during long term exposure, fatigue experiments were performed at
a constant anodic potential of -0.760 V (SCE) and a cathodic
potential of either -1.160 or -1.120 V (SCE).
Intrinsic Fatigue Crack Growth (Moist Air)
The intrinsic fatigue crack growth characteristics of alloy
2090 in moist air are shown in Fig. 2a. The data points ( O )
labeled with A K in MPa-m1/2 and R were calculated from the (a)
versus (N) response shown in Fig. 2b, and which shows the seven
constant A K levels of linear crack growth obtained by step
increases in R at a constant Knax of 17 MPa-m1/2. Specifically
note that the linear crack growth for the low AK region
(AK=1•68/R=0.90) reveals no evidence of delay retardation,
demonstrating that load shedding at constant K B ax does not induce
overload effects presumably because the maximum or forward
loading plastic zone size is constant.
Also plotted in Fig. 2a are the results of a continuously
(C=-177 m-1) decreasing AK experiment at constant K«ai. Above
the near threshold regime, good agreement is observed between the
constant AK-step increased R and continuously decreasing AK
constant K - ax techniques. The constant AK results exhibit crack
growth below the threshold stress intensity indicated by the
continuous load shedding experiment. The data contained in Fig.
2a are in excellent agreement with literature results for alloy
2090 (1,2,6) and represent the intrinsic FCG behavior of this
alloy.
Aqueous Corrosion Fatigue
Intrinsic CF Characteristics. Shown in Fig. 3 is a comparison of
the intrinsic corrosion fatigue crack growth characteristics of
alloy 2090 in moist air, deaerated 1% NaCl (-0.760 V and -1.160 V
SCE) and deaerated 1% NaCl+0.05M LiaCO3 (-0.760 V SCE). All
experiments were performed at a cyclic frequency of 5 Hz, using
the constant AK-step increased R/constant K>ax method and a AK-
step sequence similar to that shown in Fig. 2b. These results
show that alloy 2090 exhibits a moderate environmental effect at
high levels ofAK where differences in crack growth rate are less
than three-fold. Near threshold, considerable differences in
fatigue crack growth are observed, with a factor of ten increase
in da/dN revealed by comparing 1% NaCl with anodic polarization
(-0.760 V SCE) and moist air. Anodic polarization with NaCl
produced the largest corrosion fatigue effect. In contrast, at
moderate to high levels of AK, cathodic potential (-1.160 V SCE)
reduced fatigue crack growth to a level slightly greater than
that observed for moist air. Cathodic potential at low AK
produced crack arrest above the moist air threshold. The
addition of Li2CO3 to 1% NaCl (-0.760 V SCE) retarded crack
growth; the reduction is not as dramatic as that produced by mild
cathodic potential.
Corrosion fatigue experiments performed in aqueous 1% NaCl
at constant anodic potential (-0.760 V SCE) show that peakaged
alloy 2090 exhibits increased resistance to transgranular
corrosion fatigue crack growth, by a factor of three, when
compared to alloy 7075-T651 (2).
Frequency Effects. The classical frequency effect observed for
high strength aluminum alloys at higher AK and characteristic of
hydrogen environment embrittlement (7,8), namely increased
corrosion fatigue crack growth rate with decreased frequency, is
not observed for alloy 2090. In deaerated 1% NaCl solution (-
0.760 V SCE), alloy 2090 exhibits mildly increased corrosion
fatigue crack growth rates with increased frequency at high
cyclic stress intensity and little frequency effect at low AK,
as shown in Figs. 4, and 5.
Fig. 4 reveals the fatigue crack growth rate characteristics
of alloy 2090 in 1% NaCl (-0.760 V SCE), at a frequency of 5 and
0.1 Hz. Plotted are the results of a constant AK-step
increasing R/constant K m a i test performed at a frequency of 5 Hz
and three experiments performed at constant AK and R, at 0.1 and
5 Hz. Arrows indicate the increase in crack growth rate, at
constant AK, as the frequency was changed from 0.1 to 5 Hz. Best
fit lines are drawn for the 5 and 0.1 Hz data, revealing a factor
of 2 increase in fatigue da/dN at high AK and no change in
crack growth at low levels of AK.
The changes in slope of the crack length versus cycles plot,
Fig. 5a, highlight the changes in da/dN for three frequency. At
high constant AK (9.9 MPa-m1/2), a factor of two decrease in
fatigue crack growth rate is observed when decreasing the cyclic
frequency from 5 to 0.1 Hz. Similar results were reported by
Yoder and coworkers for alloy 2090 in aerated 3.5% NaCl (9).
At low cyclic stress intensities, distinct changes in
fatigue crack growth as a function of frequency are not observed.
Fig. 5b is a comparison of the average crack growth rate versus
frequency, and includes the variation in crack growth rate
(calculated with a point secant method) and the sequence in which
the experiments were conducted for two specimens. At low AK, no
difference in fatigue crack growth rate is observed for cyclic
frequencies of 0.2, 1, 5, and 20 Hz. As much as a factor of 1.6
increase in da/dn is suggested by the a-N data for the 0.07 Hz
condition and a 35% reduction in rate is recorded for the last
portion of the constant AK experiment. The origin of the low
frequency result is unclear. Fractographic analysis revealed
evidence of increased dissolution for this long term experiment
(0.25 mm of crack extension in 8 days), possibly related to an
increased driving force for fatigue; however; long term
electrical potential system variations may have contributed to
part or all of the increased growth rate. The reduction in da/dN
for segment 6 is attributed to a chemical or mechanical history
dependence associated with crack length and shape (10). These
factors complicate characterizations of moderate frequency
effects in corrosion fatigue. The important point to be made in
Fig. 4 and 5 is that corrosion fatigue in AA 2090 is not
exacerbated by decreasing frequency; to the contrary, the
opposite trend is likely, particularly at higher AK.
Polarization Effects. The results of constant AK experiments
show that corrosion fatigue crack growth in 1% Nad is mitigated
by cathodic polarization. Fig. 6a reveals that, at constant AK
(2.2 MPa-mi/ 2, R=0.8) and anodic polarization (-0.760 V SCE), a
constant crack growth rate is maintained to a crack length of
2.05 mm. Here, a cathodic potential of -1.160 V (SCE) was
applied, resulting in crack arrest. At 1.0 million load cycles a
potential of -0.760 V (SCE) was again applied. 76 hours or 2.2
million cycles elapsed before crack growth resumed, achieving a
steady state rate similar to the initial anodic condition. A
second cathodic polarization of -1.160 V (SCE) reproduced crack
arrest.
Cathodic polarization also reduced rates of corrosion
fatigue at high cyclic stress intensity (9.6 MPa-m1/2, R=0.1), as
shown in Fig. 6b. After initial cracking at -0.760 V (SCE), a
mild reduction in da/dN was produced by polarization to -1.160 V
(SCE), and a significant, time dependent retardation in crack
growth was observed at a cathodic potential of -1.260 V (SCE).
At a crack length of 3.12 mm, the applied stress intensity range
was decreased to a AK of 2.1 MPa-m»/ 2 at R=0.9 with K m a i
maintained constant. A slow recovery to a constant growth rate
was observed at an anodic potential of -0.760 V (SCE). Crack
growth arrest occurred due to cathodic polarization at -1.160 V
(SCE), reproducing the results shown in Fig. 6a. The results of
constant AK and R experimentation, Fig. 6, are consistant with
the single specimen constant AK-step increased R results
presented in Fig. 3.
The reduction in corrosion fatigue crack growth of alloy
2090 by cathodic polarization is consistent with the findings of
other researchers. Cathodic polarization of alloy 7075 in aqueous
3.5% NaCl reduced corrosion fatigue crack propagation, only at
AK=11 MPa-m1/2, (11), and the enhanced stress corrosion cracking
resistance of conventional 2000 and 7000 series alloys due to
cathodic polarization is documented (12). Consistent with the
above results, although not conclusive, the smooth specimen
fatigue life of an Al-4.2Mg-2.1Li alloy was increased by mild
cathodic potential (13).
Ion Addition Effects. The experimental results shown in Fig. 7
demonstrate that a lithium carbonate (LijCOs) addition to 1% NaCl
reduces the rate of transgranular corrosion fatigue crack growth
at constantAK, pH and potential. At a cyclic stress intensity
of 2.3 MPa-m1/2 (R=0.87) and an anodic potential of -0.760 V
(SCE), a constant crack growth rate was initially established in
deaerated 1% NaCl (pH=8.1). No change in crack growth is
observed as the solution pH is adjusted from 8.1 to 10.4. At a
crack length of 1.90 mm, the cell environment was changed by
continuously pumping deaerated 1% NaCl-0.05M Li2CO3 (pH=10.4)
solution from a separate reservoir. After purging the cell with
flowing 1% NaCl-0.05M Li2C03, the experiment was resumed,
resulting in a reduced fatigue crack growth rate. This
beneficial effect was reproduced at two higher AK levels, Fig. 3.
Fractographic Analysis. A comparison of the fracture surface
morphologies for moist air and deaerated 1% Nad (-0.760 V SCE)
environments is shown in Fig. 8. Both fracture surfaces
represent crack growth at low cyclic stress intensity. In moist
air, Fig. 8a, the transgranular fracture surface is
crystallographic with intermittent regions of flat and deflected
crack growth exhibiting a ductile appearance. The transgranular
fracture morphology for 1% NaCl, Fig. 8b, is characterized by a
flat "cleavage type" fracture surface exhibiting little evidence
of ductility (11). Detailed examination of the fatigue crack-
final overload fracture interface region revealed little evidence
of corrosion. Small and widely dispersed regions of sub grain
boundary corrosion, similar to that shown in Fig. Ib, were
observed. No evidence of corrosion was observed along high angle
grain boundaries.
The fracture surface morphology for aqueous NaCl depends on
cyclic stress intensity range. At high AK (15 MPa-m1/2) severe
planar crack surface deflections are dominant. Here fatigue crack
growth is possibly due to decohesion along intense slip bands.
As stress intensity range decreases, the amount of crack tip
deflection is reduced and the fracture surface is characterized
by increased areas exhibiting a flat crystallographic appearance.
MECHANISTIC IMPLICATIONS
Many mechanisms have been proposed for environmental
fracture of aluminum alloys including: hydrogen embrittlement,
adsorbate decohesion, film rupture/dissolution, film induced slip
irreversibility, enhanced localized plasticity, and crack tip
blunting (7,8,11,12,14,15). Paris regime corrosion fatigue has
been emphasized; the effects of aqueous environments on near
threshold cracking have not been considered. For the former, a
cleavage fracture morphology was circumstantially linked to both
hydrogen embrittlement and surface film effects. Hydrogen from
cathodic reduction in electrolytes and chemical reactions in
water vapor is held to embrittle aluminum with contributions from
dislocation and grain boundary transport processes. Alternately,
surface films suppress plastic deformation by blocking the crack
tip emission of dislocations; the local normal stress is elevated
causing cleavage.
Based on the experimental observations of FCP response to
AK, frequency, cathodic polarization and LizCOs addition, we
speculate that surface films play an important role in crack tip
damage during transgranular corrosion fatigue of Al-Li-Cu alloys.
Faster crack growth is correlated with those environmental
conditions which are likely to favor less protective crack tip
surface films. While hydrogen embrittlement may contribute to
crack propagation, the frequency and cathodic polarization
dependencies observed for AA 2090 are opposite to the classical
relationships which are diagnostic of this failure mechanism, as
discussed for 7000 series alloys (7,8,12).
Increased crack growth rates with increased frequency
suggest that time dependent chemical reactions and mass transport
are rapid and do not limit crack tip environmental damage.
Considering mechanical rate effects, crack tip cyclic strain rate
is approximately proportional to loading frequency and stress
intensity range raised to a power on the order of 5 (16). At
high AK, mechanical film disruption provides a plausible
explanation for the mild increase in crack growth with increasing
frequency. Here, crack tip strain rate increases with increasing
frequency, the film is locally breached more frequently favoring
transient dissolution or hydrogen production reactions, and crack
extension per unit time and cycle increases. Near threshold,
crack tip deformation may further localize, however, the average
crack tip strain rate decreases strongly. The diminished effect
of frequency suggests that the latter effect dominates for AA
2090 in aqueous chloride. The environmental effect is strong
near threshold, suggesting that chemical factors control surface
film formation and environmental crack advance.
The beneficial effect of a surface film is likely to be
chemical in origin. Film or corrosion product induced crack
closure is not operative because the environmental influences are
observed at stress ratios above 0.85 and for short crack wakes.
Secondly, there is no evidence that surface films promote
irreversible dislocation damage (14). Rather, we speculate that
films interfere with crack tip environmental reactions (hydrogen
production or dissolution), resulting in a significant reduction
in fatigue crack growth rate. Once the film is damaged by
chemical or mechanical means, such as localized anodic pitting
dissolution or slip step formation, the film no longer acts as a
barrier; the alloy is rendered susceptible to brittle crack
growth.
Although experimental evidence strongly suggests that
surface films play a primary role in crack tip damage during
transgranular fatigue of alloy 2090 in aqueous environments,
further work must be performed to understand the roles of
hydrogen and transient localized anodic dissolution. Gaseous
environment corrosion fatigue experiments and high resolution
fractographic analyses are being conducted in this regard (2).
CONCLUSIONS
The high resolution potential difference technique, short
crack specimen geometry and constant AK-step increased R/constant
K»ax approach successfully characterizes intrinsic fatigue crack
growth in complex aluminum-lithium based alloys exposed to
aggressive aqueous environments.
Al-Li-Cu alloy 2090, peak aged, is susceptible to corrosion
fatigue crack propagation in aqueous 1% NaCl under anodic
polarization. At low AK/high R, near threshold crack growth
rates are significantly increased, with growth occurring by a
cleavage process. High AK/low R growth rates are increased by
the environment, with highly deflected slip band cracking the
dominate microscopic mode. Environmental effects on AA 2090 are
less severe than those reported for high strength AA 7075.
Crack tip films play an important role in corrosion fatigue
damage, as evidenced by: (1) increased da/dN with increased
cyclic frequency, particularly at high AK, (2) retarded crack
growth due to cathodic polarization in NaCl and (3) reduced da/dN
due to Li2CO3 additions.
ACKNOWLEDGEMENT
The research support provided by the NASA-Langley Research
Center, grant NAG-1-745 with D.L. Dicus as monitor, is gratefully
acknowledged. Alloy 2090 was provided by E.L. Colvin, Alcoa
Technical Center.
REFERENCES
1. K.T. Venkateswara Rao, W. Yu and R.O. Ritchie, Metall.Trans.. Vol. 19A, pp. 549-561 and pp. 563-569 (1988).
2. R.S. Piascik and R.P. Gangloff, "Environmental AssistedDegradation Mechanisms in Al-Li Alloys", Report No.UVA/528266/MS88/101, University of Virginia (1988).
3. ASTM 05-82,"Standard Reference Method for MakingPotentiostatic and Potentiodynamic Anodic PolarizationMeasurements", 1985 Annual Book of ASTM Standards-Methodsand Analytical Procedure. ASTM, Philadelphia, PA, Vol. 3,pp. 123-133 (1985).
4. H. Tada, P. Paris and G.I. Irwin, The Stress Analysis ofCracks Handbook. Del Research Corp., St. Louis, MO, pp.2.10-2.11 (1985).
5. R.P. Gangloff, Advances in Crack Length Measurements, C.J.Beevers, ed., EMAS, UK, pp. 175-229 (1982).
6. W.A. Herman, R.W. Hertzberg, and R. Jaccard,"A SimplifiedLaboratory Approach for the Prediction of Short CrackBehavior in Engineering Structures", The Journal ofFatigue and Fracture of Engineering Materials andStructures. in press (1988).
7. H.J.H. Holroyd and D. Hardie, Corr. Sci.. Vol.23, No.6, pp.527-546 (1983).
8. M.Gao, P.S. Pao and R.P. Wei, "Chemical and MetallurgicalAspects of Environmentally Assisted Fatigue Crack Growth In7075-T651 Aluminum Alloy, Metall. Trans. A. in press (1987).
9. G. R. Yoder, P.S. Pao, M.A. Imam and L.A. Cooley,Intl. SAMPE Metals Conf. Ser.. Vol.1, pp. 25-36 (1987).
10. R.P. Gangloff and D.J Duquette, Chemistry and Physicsof Fracture, R.M. Latanision and R.H. Jones, eds., NATOSeries E, No.130, pp. 612-645 (1987).
11. R.E. Stotz and R.M. Pelloux, Met. Trans. Vol. 3, pp. 2433-2441 (1972).
12. N.J.H. Holroyd and G.M. Scammans, Environment-SensitiveFracture. ASTM 821, S.W. Dean, E.N, Pugh and G.W. Ugiansky,eds., American Society for Testing and Materials, pp. 202-241 (1984).
13. R.E. Ricker and D.J. Duquette, "The Use of ElectrochemicalPotential to Control and Monitor Fatigue of Aluminum Alloys,Corrosion 85, 1985, NACE, Publications Dept., Houston, TX,Paper # 354,
14. K.V. Jata and E.A. Starke, Jr., Metall. Trans. A. Vol. 17A,pp. 1011-1026 (1986).
15. D.L. Davidson and J. Lankford, Fat. Engr. Matls. and Struc..Vol. 6, pp. 241-256 (1983).
16. S.J. Hudak, Jr., D.L. Davidson and R.A. Page, Embrittlementby the Localized Crack Environment, R.P. Gangloff, ed., TMS-AIME, Warrendale, PA, pp. 173-198 (1984).
17. J.G. Craig, R.C. Newman, M.R. Jarrett and N.J. H. Holroyd,Journal De Physique. Vol. C3, No. 9/48, pp. 825-833 (1987).
18. J. Gui and T.M. Devine, Scripta Metall., Vol. 21, pp. 853-(1987).
Figure Captions
Fig.l (a) Potentiodynamic polarization of alloy 2090 inaerated and deaerated 1% NaCl. (b) A photomicrographof alloy 2090 exposed to deaerated 1% NaCl for 8 daysat - 0.760 V (SCE). Note subgrain boundary, Tiprecipitate, corrosion (arrows) and constituentparticle pitting.
Fig.2 (a) A comparison of constant AK-step increasedR/constant K»a* results with decreasing AK-constantKmax data. (b) Crack length versus loading cycles plotfor the decreasing AK-increasing R/constant K m airesults in Fig. 2a.
Fig.3 The corrosion fatigue crack growth characteristics ofalloy 2090 in aqueous 1% NaCl at constant anodic (-0.760 V SCE) and cathodic (-1.160 V SCE) potentials, 1%NaCl + 0.05M LiaCO3 (-0.760 V SCE), and moist airenvironments.
Fig.4 The effect of cyclic loading frequency on the intrinsiccorrosion fatigue crack growth of alloy 2090 exposed to1% NaCl at constant anodic potential.
Fig.5 (a) The effect of frequency on corrosion fatigue crackpropagation in alloy 2090 at high AK. (b) Thedependence of corrosion fatigue crack growth rate oncyclic frequency at high and low AK.
Fig.6 The effect of cathodic polarization on intrinsiccorrosion fatigue crack growth in alloy 2090 atconstant AK level of (a) 2.2 MPa-m1/2 for high R and(b) 9.6 and 2.1 MPa-m1/2 for low and high R conditions,respectively, in 1% NaCl.
Fig.7 The effect of Li2CO3 addition and pH change on theintrinsic fatigue crack growth of alloy 2090 in IX NaClat constant anodic potential and constant cyclic stressintensity.
Fig.8 A comparison of the fracture morphology due tocorrosion fatigue at low AK in (a) moist air at 2.5MPa-mi/2 and (b) deaerated 1% NaCl (-0,760 V SCE) at2.2 MPa.m1/2. Crack growth is from right is from rightto left, with the fatigue crack-final overload fractureinterface shown.
TABLE 1 Material Properties
Chemical Composition:Li Cu Zr Fe Si Mn2.14 2.45 0.09 0.05 0.04 0.00Mg Cr Ni Ti Na Zn0.00 0.00 0.00 0.01 0.001 0.01Material Condition:Solutionized quenched and stretched.Peakaged 190«C - 4hrsMechanical Properties:(Long-Transverse)
Yield Ultimate496 MPa 517 MPa
Microatructure:Grain size - 3.3 mm (transverse)
- 0.11 mm (short transverse)Subgrain size - 15 urn (transverse)
- 5 um (short transverse)Precipitate morphology&' - Al3Li, 0^- Al2Li, Ti - Al2CuLi, Tt - Al6CuLi3,Matrix phases - $', Ti, Q>Sub grain boundary - TiHigh angle grain boundary - Tz
ORIGINAE PAGE ISOF POOR QUALITY
UJow
E-~r
UJ
-500
-600
-700
-800
-900
-1000
-1 100
-1200
-1300
-1400
At 2090Peakaged
1.0% NaCI.16MV/S6C
Aerated
Deaerated -
10~2 10"1 10° 101 102 10S 104
I (/iA/cm2)
(a) (b)
Figure 1
10 "3-3
ALLOY 2090 Packaged, L-TMoist Air, 5 Hza—Constant A.K. step increased R*-Decreasing A K. constant Kmax=17 MPaVrn, C=-177 m1
0a
-R-0.05/Kmox-17
-0.05/11
10 -9^O
oJXooO
0)
a"o
T£
.c-4— 'CJ>c
_J
oD
^
10 -VE
-
10--1
4.00 -.
3.50 ^
3.00 ^
2.50 \
2.00 ^
1.50 -,=
0-NLL____ (a)
j?.. — 0.84/17*.<a'
^ "~~ 0.87/17
•
~~~":— -0.90/17
1 I I 1 1 1 1 I | | | 1 1 1 1 I 1 |
10 10 *Stress Intensity Range (MPaVrn)
ALLOY 2090 tPeakaged. L-T fMoist Air, 5 Hz /Constant AK, step increased R I e.se/o.so
^T \. 3.29/0.80
^ff^^i ^^\
, * * * * * * * * * * * * * * * * * , N '̂VO-84f*** ** 2.20/0.87\
i ^^ 1.68/0.90 I
| 15.98/0.05i
O1.00
(b)
AK-10.43/R-0.05
0.50 -,
0.00500000 1000000 1500000 2000000 2500000 3000000
Cycle Count
Figure 2
10 -3
JDO>\O
110-S
JI-t-i
O
O
OOl_O
en
10-8-
10 -7.
ALLOY 2090 Packaged, L-T, f=5 Hz*—1* NaCI, -0.760 V (SCE)a —1* NaCI + 0.05M Li2C03 -0.760 V (SCE)x —1* NACI, -1.16 V (SCE), A—Moist Air
10Stress Intensity Range (MPaVrfT)
10'
Figure 3
10 "*qALLOY 2090, Peakaged, L-T. 1* NaCI (-0.760 mV SCE)Hi-Constant AK, step increased R
Constant AK, *-R=0.05 (16 MPaVm), A-R=0.1 (9.9and 0-R=0.8 (2.2
o^o
-*_
O
O
ooO
0)
10-*-
0.1 Hz
10-*_ t I
10Stress Intensity Range
10
Figure 4
EE
1.60 -q
1.40 •=
1.20 -_
1.00 i
CT>c 0.80 3
0.60 q
ALLOY 2090Peakaged, L-TAK=9.9 MPaVm. R=0.11.0* NaCI, deaerated, -0.760 V SCE
1 Hz
1.11E-04 mcn/cy
0.1 Hz
6.88E-05 mm/cy
(a)
0.40 -
0.20 -
O nn ".UU
S 5 Hz \
/ X
1.37E-04 mm/cy
(D
I 1 1 1 1 1 1 1 1 | 1 1 1 1 1 1 1 1 1 J 1 1 1 1 1 1 1 1 1 | 1 1 1 1 1 1 1 1 1 J 1 I J 1 1 1 1 1 1 | 1 1 1 1 1 1 1 1 1 | 1 I 1 1 1 1 1 1 1 |
0 2000 4000 6000 8000 10000 12000 14000Cycle Count
ALLOY 2090, Peakaged, L-T1.0* NaCI, deaerated, -760 V SCEA-AK=9.9 MPaVm", R=0.1+ -AK=2.2 MPaVm; R=0.8
0>O
0\E
£, 1 0 "4-
-t->
p ]o
o
2io- s ,
en'-*->o
u_
10--10
T 1
-r2--A^~
- r 3 ^ _ — -̂ ~^^
^--^— ^ "̂ -i.
(b)
T
1 4 T j J_ j
ir3 •+• 1 1 .f.j. ^~2 T -1-5
i i t i i i 1 1 j i i i i i i i 1 1 i i i i i i i t ) i i i i i i 1 1 1
~J 10"' 1 10 10Cyclic Frequency (f)
Figure 5
3.00 -i
0.00 Hn-rr
4.00
o
1 00
/i -1.16 V
* * , * * * * * * * * * * *
76 Mrs. -0.76 V-1.16 V
(a)
-0.76 V(SCE)ALLOY 2090Peakaged, L-T1.058 NaCI (Deaerated)R=0.8,AK=2.2 MPaVm, f=0.07-20 Hz
1000000 2000000 3000000 4000000Cycle Count
* * * * * * ***"
w- -1.18 V
FrequencyChange
20 to 2 Hi
-0.76 V
(b)
ALLOY 2090Peakaged, L—T1.0* NaCI (Deaerated)f=5,20,2 Hz
AK=9.6 MPaVm. R=0.1for a<3.12 mmA K=2.1 MPaTrn, R=0.9for a>3.12 mm
i i ' ' ' ' ' i ' ' i i''200000 400000
i i i i i i i i i600000
i i i i iii800000 1000000
Cycle Count
Figure 6
2.60 -q
2.40 -j-—x
EE 2.20 -j
cnc
2.0Q •=
1.80 •=
(J0 1.60 ̂
O
1.40 •=
ALLOY 2090 Peakaged, L-T1* NaCI and 0.05M Li2C03; -760 mV (SCE)f=5 Hz. R= 0.87, AK= 2.3 MPa^/mDeaerated
NaCI pH=10.4
NoCI pH=8.1
Cycle Count
NaCI-U2C03
pH=10.4
300000 400000 500000 600000 700000 800000
Figure 7
Extended Abstract for ICF-7; Houston, Texas (1989)"Fatigue and Fracture of Advanced Aerospace Materials"
(Session Organizer: T. W. Crocker)
INTRINSIC FATIGUE CRACK PROPAGATION IN ALUMINUM-LITHIUM ALLOYS:THE EFFECT OF GASEOUS ENVIRONMENTS
ROBERT S. PIASCIK AND RICHARD P. GANGLOFF
Department of Materials ScienceUniversity of Virginia
Charlottesville, VA 22901
Advanced aluminum-lithium based alloys exhibit outstandingfatigue crack propagation resistance, but attributable to theextrinsic effects of crack surface 'closure contact and crack tipdeflection. Extrinsic crack growth resistance is likely to begeometry, orientation and loading history dependent; thuscomplicating defect tolerant predictions. It is critical tounderstand the intrinsic fatigue response of such materials, andparticularly environmental and microstructural influences oncrack tip damage mechanisms.
Intrinsic fatigue crack propagation is effectively studiedby in situ, direct current, electrical potential monitoring ofsingle, short, but through-thickness, edge cracks sized between0.3 and 5 mm. Computer control enables constant stress intensityrange and constant Kmax load shedding at high mean stress levelsnear threshold. Results obtained to date complement studies ofmicrostructurally small cracks.
The intrinsic growth kinetics of such short cracks in peakaged Alloy 2090, exposed to moist air, are rapid at stressintensity ranges between 1 and 2 MPa-m1/2. The near thresholdbehavior of 2 to 5 mm edge cracks in freely rotating specimensagrees with data for standardized compact tension specimenssubjected to high mean stress loading programs; the apparentthreshold is near 2 MPa-m1/2. If rotation is restricted byeither short crack length or by mechanical constraint, crackgrowth continues to a AK level of about 1 MPa-m 1 / 2; similar tothe behavior of microstructurally small cracks. These resultsfor benign environments are interpreted based on crack closureand bending related, Mode II displacements.
Equal intrinsic fatigue behavior is observed for twoorientations of short cracks in anisotropic, unrecrystallizedAlloy 2090, peak aged. This result is significant because suchcracks sample either single or multiple grains along thepropagating front. The later case is given by an LT orientationwhere the 2.5 mm wide edge cracks grow normal to and intersect
about 50 thin, elongated grains. In the LS orientation, theshort cracks grow entirely within one or two grains and oftenencounter high angle boundaries. Electrical potential monitoringat constant stress intensity range gives no indication of grainby grain changes in growth rate for the LS orientation, despitesignificant crack deflections. No evidence is obtained toconfirm reports of uniquely high growth rates of small crackswithin single grains in Alloy 2090.
Gaseous environmental effects on intrinsic fatigue crackgrowth are significant for Alloy 2090. For high AK-low Rloading, crack growth rates decrease according to the environmentorder: purified water vapor, moist air, helium and oxygen. Forlow AK-high R, the fastest intrinsic crack growth rates areobserved for helium, with slower rates measured for water vapor,moist air and oxygen. These data are unique in that crack growthin oxygen is substantially reduced and growth in helium isunexpectedly rapid. The environmental effects are mostpronounced near threshold and are not obviously closuredominated. Similar results were obtained for Alloy 7075-T651.While these environments were established in an ultra-high vacuumchamber, the complicating effects of trace impurities cannot, atpresent, be eliminated; and the contributions of oxide debris,hydrogen and surface film-dislocation interactions are not yetunderstood. Experiments are in progress to probe these issues.
This research is supported by the NASA-Langley Research Center,grant NAG-1-745, with D.L. Dicus as monitor.
APPENDIX II. TRAVEL AND CONFERENCE PARTICIPATIONNOVEMBER, 1987 TO JUNE 1988
Travel
G.E. Stoner, R.P. Gangloff, R.S. Piascik and R.G. Buchheit, Jr.:Alcoa Technical Center; Pittsburgh, PA, February, 1988;Student presentations on environmental effects research.
R.P. Gangloff: NASA meeting on National Aerospace Plane-HydrogenEnvironment Effects; Phoenix, AZ, June, 1988; H.G. Nelson,organizer; Presented talk on "Fracture Mechanics andEnvironmental Fracture of Advanced Materials".
Loca_L
R.P. Gangloff, "Advanced Metallic Materials for Light AerospaceStructures" Virginia Academy of Science, Charlottesvilie,VA, May, 1988.
R.G. Buchheit, Jr. and G.E. Stoner, Abstracts follow.
EXTENDED ABSTRACTS
1/15/88 Presentation at The Alcoa Technical Center, Pittsburg,PA.
4/27/88 Presentation to the Industrial Associates of the LightMetals Center, Charlottesville, VA.
"Measurements and Mechanisms of Localized AqueousCorrosion in Aluminum-Lithium Alloys"
Based on a review of literature, microreference Ag|AgCl
electrodes have been constructed and tested. Microreference
electrode tips have been made as small as 5 micrometers (inner
bore diameter). Electrodes constructed show excellent reference
potential stability and have been used to measure potential
variations as small as 0.6 mV on an actively corroding metal
surface. Microreference electrodes have been used to measure the
variation in potential in the vicinity of corrosion pits on peak
aged 2090 (Al-2.7Cu-2.2Li-0.12Zr) during potentiodynamic
polarization scans.
Measurements of the change in pH versus time have been
performed for artificial crevices and immersed machine shavings
in solutionized and peak aged 2090. Preliminary results suggest
that pH vs. time behavior is dependent on alloy heat treatment.
Results obtained for 2090 are compared with results obtained in
similar tests performed on alloys 2024 and 7075.
Scanning electron microscopy, energy dispersive
spectroscopy and X-ray mapping have been used to examine pit
morphology and variations in the concentrations of metallic
elements like copper, iron and magnesium in pitted regions.
4/15/88 Electrochemical Seminar Series at the University ofVirginia.
"An Ongoing Investigation of the Stress Corrosion Cracking andLocalized Corrosion Behavior on an Al-Li-Cu Alloy"
(presentation with Jim Moran).
The localized aqueous corrosion and stress corrosion
cracking behavior of alloy 2090 (Al-Li-Cu) are presently being
studied. Progress to date includes a microstructural
characterization of the alloy, anodic polarization experiments,
and constant immersion and constant extension rate (CERT) SCC
tests. In addition, localized pH and localized potential
experiments have been performed using both conventional and
micro-electrodes. A brief review of pertinent microstructural
features of Al-Li alloys (2090, in particular), along with a
summary of the corrosion work to date will also be presented.
The polarization and SCC experiments were performed in
various NaCl based environments. Additions to the environment
included carbonate and sulfate anions, with Li, Na and K cations.
In addition to the quantitative information obtained from these
experiments, the SCC fracture surfaces and the polarization
corrosion surfaces were examined. The results to date suggest
that SCC and polarization behavior can be significantly changed
by the addition of these various species.
the pH and potential of the solution contained within
artificial crevices formed in 2090 and 2024 alloys has been
measured in situ. The effect of alloy heat treatment, alloy
composition, degree of solution agitation and aeration on the
crevice solution behavior has been monitored as a function of
time. The effect of coupling an external cathode to the crevice
has also been studied.
Potential variations in the solution near actively corroding
pits on aluminum alloy surfaces have been measured with specially
constructed microreference electrodes. Potential variations as
small as 0.6 mV have been detected. The pitting of polished
alloy specimens has been observed in situ using an optical
microscope equipped with long working distance objective lenses.
5/26/88 Meeting of the Virginia Academy of Science,Charlottesville, VA
"Characterization of Occluded Aqueous Environmentsin Al-Li-Cu Alloy 2090"
Changes in potential and pH occurring in simulated crevices
machined in alloy 2090 have been monitored as a function of bulk
environment condition and time. In this talk, the methods used
for simulating crevices and making measurements of potential and
pH in occluded environments will be described. The chemistry
developed in crevices formed in 2090 will be compared to that
developed in conventional aluminum alloys like 2024 and 7075.
The corrosion mechanisms thought to cause the observed potential
and pH changes will also be discussed.
DISTRIBUTION LIST
Copy No.
1 - 3
6
7 - 1 8
19 - 20
21 - 22
23
24 - 25--
Mr. Dennis DicusM/S 188A, Metallic Materials BranchNational Aeronautics and Space AdministrationLangley Research CenterHampton, VA 23665
Mr. W. Barry LisagorM/S 188A, Metallic Materials BranchNational Aeronautics and Space AdministrationLangley Research CenterHampton, VA 23665
National Aeronautics and Space AdministrationLangley Research CenterHampton, VA 23665
Attention: Mr. J. F. Royal1, Jr.Grants Officer, MS 126
Thomas H. Courtney, Chairman, MS
Richard P. Gangloff, MS
Glenn E. Stoner, MS
Robert E. Swanson, Assistant ProfessorDepartment of Materials EngineeringVirginia Polytechnic and State UniversityBlacksburg, VA 24061
SEAS Publications Files *
NASA Scientific and Technical InformationFacility
P.O. Box 8757Baltimore/Washington International AirportBaltimore, MD 21240
•"Reproducible copy
1656:ald:DW212R
UNIVERSITY OF VIRGINIASchool of Engineering and Applied Science
The University of Virginia's School of Engineering and Applied Science has an undergraduateenrollment of approximately 1,500 students with a graduate enrollment of approximately 560. Thereare 150 faculty members, a majority of whom conduct research in addition to teaching.
Research is a vital part of the educational program and interests parallel academic specialties.These range from the classical engineering disciplines of Chemical, Civil, Electrical, and Mechanicaland Aerospace to newer, more specialized fields of Biomedical Engineering, Systems Engineering,Materials Science, Nuclear Engineering and Engineering Physics, Applied Mathematics and ComputerScience. Within these disciplines there are well equipped laboratories for conducting highly specializedresearch. All departments offer the doctorate; Biomedical and Materials Science grant only graduatedegrees. In addition, courses in the humanities are offered within the School.
The University of Virginia (which includes approximately 2,000 faculty and a total of full-timestudent enrollment of about 16,400), also offers professional degrees under the schools of Architecture,Law, Medicine, Nursing, Commerce, Business Administration, and Education. In addition, the Collegeof Arts and Sciences houses departments of Mathematics, Physics, Chemistry and others relevantto the engineering research program. The School of Engineering and Applied Science is an integralpart of this University community which provides opportunities for interdisciplinary work in pursuitof the basic goals of education, research, and public service.