https://ntrs.nasa.gov/search.jsp?R=19960008465 2019-03-26T17:11:58+00:00Z 7 XL NASA Contractor Report 198230 t Time-Dependent Corrosion Fatigue Crack Propagation in 7000 Series Aluminum Alloys Mark E. Mason University of Virginia, Charlottesville, Virginia m </> 0s rH (C O i — m <o o oo o c o Z 13 O •t <M V, fO o Grant NAG1-745 * , o w H4 • K i- z: Z < Q. UoJ O< «>/) m October 1995 S o S ^ a. a _i ai a. < o ^ i *: z: • , uj u 3 > z: < z — >-i QC •-« C National Aeronautics and *~ u § ~) Space Administration /^ 3 < .2 Langley Research Center oa c Hampton, Virginia 23681-0001 S K 1C'^ 00 < •-« k- o> u. oc — rH UJ > I Z CO s-» oc o o i-t o I 1/5 O Ifl < o o — (/> oc f- in < OC 0> z o z x: TIME-DEPENDENT CORROSION FATIGUE CRACK PROPAGATION IN 7000 SERIES ALUMINUM ALLOYS A Thesis Presented to the faculty of the School of Engineering and Applied Science University of Virginia In Partial Fulfillment of the requirements for the Degree Master of Science (Materials Science and Engineering) by Mark E. Mason December, 1994 "PAGE MISSING FROM AVAILABLE VERSION" Ill TIME-DEPENDENT CORROSION FATIGUE CRACK PROPAGATION IN 7000 SERIES ALUMINUM ALLOYS ABSTRACT The goal of this research is to characterize environmentally assisted subcritical crack (EAC) growth for the susceptible S-L orientation of aluminum alloy 7075-T651, immersed in acidified and inhibited NaCI solution. This work is necessary in order to provide a basis for incorporating environmental effects into fatigue crack propagation life prediction codes such as NASA-FLAGRO. This effort concentrates on determining relevant inputs to a superposition model in order to more accurately model environmental fatigue crack propagation (EFCP). Time-based environmental crack growth rates from quasi-static load experiments in the 7075/NaCI system are an order of magnitude too small for accurate linear superposition prediction of da/dN for loading frequencies above 0.001 Hz. A combination of primarily intersubgranular fracture and overload fracture with few high angle boundary cracking features is observed for quasi-static EAC. Anodic dissolution to supply hydrogen to the crack tip process zone and the resulting hydrogen embrittlement (HE) are likely operative for EAC of 7075-T651 in chloride environments. Alternate methods for establishing da/dt, based on enhanced crack tip strain rate (de /dt) to disturb the passive film at the crack tip CT and enhance hydrogen production and uptake, do not provide a means of enabling linear superposition modeling of EFCP da/dN for frequencies above 0.001 Hz. Specifically, EAC da/dt is not enhanced for estimated crack tip strain rates iv from 2 x 10~5 to 2 sec"1, imposed by monotonic, ripple, and trapezoidal loading conditions. These observations indicate that passive film rupture may not be the rate limiting crack tip damage mechanism governing EAC of 7075-T651 in chloride. Fatigue crack propagation rates in the 7075/NaCI system are enhanced five to ten-fold over da/dN in moist air and are over an order of magnitude faster compared to cracking in helium. EFCP da/dN show a weak time-cycle frequency dependence above 0.001 Hz (viz, da/dN a f, where & is of order 0.1). The NaCI environmental effect on time-cycle-dependent EFCP, shown to be identical for acidified chromate inhibited and neutral uninhibited environments by comparison with literature results, is identical for the S-L and L-T orientations of 7075-T651, suggesting a minimal involvement of anisotropic high angle grain boundaries. Corrosion fatigue is predominantly intersubgranular for low frequencies and AK levels (below 0.1 Hz for AK of 9 MPa/m), while higher frequencies and AK levels produce primarily a transgranular fracture morphology. Compliance estimates of ^CLOSURE, and comparison with electrical potential based da/dN results, indicate a negligible effect of time-dependent crack closure on the frequency and rise time dependencies of da/dN for moderate AK. Trapezoidal loading EFCP experiments indicate a very small effect of hold-time (T ) at K . EFCP da/dN increases mildly H MAX with increasing loading rise time (t ) from K to K for a wide range of loading R MIN MAX waveforms, indicating that t is a more general parameter than frequency for R analyzing time-dependent EFCP da/dN. EFCP rates transition from time-cycle-independent to time-cycle-dependent V at a critical crack growth rate (da/dN ) and frequency (f ). Above f shorter CRrr CWT CRIT cycle periods provide less time for diffusion and accumulation of embrittling hydrogen in the process zone, and da/dN becomes time-cycle dependent. Derived from a model of crack tip hydrogen diffusion, a proportionality constant, Yam. relates da/dN to (fcwr)"05, and includes the effect of diffusivity of hydrogen CRIT in aluminum (D ), the critical concentration of hydrogen in the process zone for H embrittlement (C ), and the concentration of hydrogen at the crack tip surface CRrr (C ). The complex dependence of D , C and C on metallurgical, compositional, s H CRIT1 s and environmental parameters may be qualitatively assessed, however, these parameters cannot be precisely determined from the results of this study. A combination of many possible rate-limiting processes involved in EFCP of 7000- T651(S-L) series aluminum alloys in chloride evidently controls time-cycle- dependent EFCP da/dN above f . Results indicate the governing rate-limiting CRrr reaction changes with AK. Fractographic comparisons indicate that the fracture mode undergoes a transition at identical da/dN and f for both 7017-T651 (S-L) and 7075-T651 (S- CRrT CRIT L) in chloride. Fracture modes for the S-L orientation of 7475-T651 and 7075-T651 differ, despite nearly equivalent time-cycle dependent EFCP da/dN and identical environments. EFCP of 7000-T651(S-L) series aluminum alloys in NaCI is governed by competitive fracture modes. The crack path that offers the least resistance to cracking, and is favored by process zone mechanics and environment conditions, is operative. vi Linear superposition may be used to predict EFCP da/dN in the 7075- T651 (S-L)/NaCI system for corrosion fatigue waveforms having K levels in excess of K for more than 1000 seconds. A conservative approach using da/dN for ISCC CRITl frequencies above where da/dN becomes negligible, coupled with linear scc superposition below this frequency, is warranted for damage-tolerant life prediction for 7000 series aluminum alloys in aggressive chloride environments. VII TABLE OF CONTENTS Abstract iii List of Tables x List of Figures xi List of Symbols xvi CHAPTER 1. BACKGROUND 1 1.1 "PROBLEM 1 1.2 INTRODUCTION 1 1.2.1 Stress Corrosion Cracking (SCC) of High Strength Aluminum Alloys 2 1.2.2 EFCP of High Strength Aluminum Alloys 4 1.2.3 Modeling EFCP for 7000 Series Aluminum Alloys 9 1.3 OBJECTIVE 12 1.4 APPROACH 12 CHAPTER 2. EXPERIMENTAL PROCEDURE 14 2.1 MATERIAL 14 2.2 EAC EXPERIMENTS 16 2.2.1 Stress Intensity Calculation . 16 2.2.2 Crack Length Measurement 19 2.2.3 Monotonic Loading 21 2.2.31 Constant Load Line Displacement Rate 21 2.2.32 Constant Load Line Displacement 23 2.2.33 Constant Stress Intensity 23 2.2.4 Ripple Load 24 2.2.5 Fatigue 24 2.2.51 Fatigue Precracking 25 2.2.52 AK Decreasing 25 2.2.53 Constant A K: Variable Frequency and Waveform . 26 2.2.6 Environment 27 2.3 FRACTOGRAPHY 28 VIII CHAPTER 3. RESULTS 29 3.1 MONOTONIC LOAD EAC 29 3.1.1 EAC Under Quasi-Static Load 29 3.1.11 Background 29 3.1.12 Crack Growth Rate Results 29 3.1.2 EAC Under Constant Load Line Displacement Rate 35 3.1.21 Background 35 3.1.22 Crack Growth Results 36 3.1.3 EAC Under Ripple Load 38 3.1.31 Background 38 3.1.32 Crack Growth Rate Results 39 3.2 ENVIRONMENT ENHANCED FATIGUE CRACK GROWTH 41 3.2.1 Effect of Environment on da/dN versus AK 41 3.2.11 Background 41 3.2.12 Fatigue Crack Growth Rate Results 42 3.2.2 Effect of Sinusoidal Frequency on EFCP da/dN 46 3.2.21 Background 46 3.2.22 EFCP Results 46 3.2.23 Fatigue Crack Closure 50 3.2.3 Effect of Hold Time at K on EFCP da/dN 55 MAX 3.2.31 Background 55 3.2.32 EFCP Results 55 3.2.4 Effect of Rise Time to K on EFCP da/dN 58 MAX 3.2.41 Background 58 3.2.42 EFCP Results 58 3.2.43 Fatigue Crack Closure 61 3.2.5 Effect of Orientation on EFCP da/dN 62 3.2.6 Summary of EFCP da/dN 64 3.3 SCANNING ELECTRON MICROSCOPY OF ENVIRONMENT ENHANCED FRACTURE 66 3.3.1 SCC Fractography 66 3.3.2 Fractography of Ripple Loading Fracture 70 3.3.3 Fractography of Variable Frequency Constant AK Fracture 77 3.4 CONCLUSIONS 85 CHAPTER 4. DISCUSSION 88 4.1 LINEAR SUPERPOSITION MODELING 88 4.1.1 Background 88 4.1.2 Model Application to 7075-T651 (S-L) in NaCI 88 IX 4.2 TIME-DEPENDENT EFCP 91 4.2.1 Hydrogen Embrittlement Mechanism for EAC in 7075 ... 92 4.2.2 Crack Tip Strain Rate Effects on EAC 93 4.2.21 Background 93 4.2.22 Crack Tip Surface Strain Rate Estimations ...... 95 4.2.23 Effect of Strain Rate on Environmental Cracking . 100 4.2.3 Time Dependent EFCP da/dN 110 4.2.31 Mechanism-Based Linear Superposition Modeling 111 4.2.32 Mechanistic Interpretation of Frequency Dependent EFCP da/dN 115 4.2.4 Rise Time Dependence 135 4.3 IMPLICATIONS FOR LIFE PREDICTION MODELING 139 4.4 CONCLUSIONS 141 CHAPTER 5. CONCLUSIONS 144 CHAPTER 6. REFERENCES 150