UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

High frequency near-threshhold corrosion fatigue of AISI 316L stainless steel Fong, Clinton 1985

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-UBC_1985_A7 F65.pdf [ 16MB ]
Metadata
JSON: 831-1.0096297.json
JSON-LD: 831-1.0096297-ld.json
RDF/XML (Pretty): 831-1.0096297-rdf.xml
RDF/JSON: 831-1.0096297-rdf.json
Turtle: 831-1.0096297-turtle.txt
N-Triples: 831-1.0096297-rdf-ntriples.txt
Original Record: 831-1.0096297-source.json
Full Text
831-1.0096297-fulltext.txt
Citation
831-1.0096297.ris

Full Text

HIGH FREQUENCY NEAR-THRESHOLD CORROSION FATIGUE OF AISI 316L STAINLESS STEEL By CLINTON FONG B.A.Sc, The University of British Columbia, 1982 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF APPLIED SCIENCE in THE FACULTY OF GRADUATE STUDIES The Department of Metallurgical Engineering We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA May, 1985 © Clinton Fong, 1985 In presenting t h i s thesis i n p a r t i a l f u l f i l m e n t of the requirements for an advanced degree at the University of B r i t i s h Columbia, I agree that the Library s h a l l make i t f r e e l y available for reference and study. I further agree that permission for extensive copying of t h i s thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. I t i s understood that copying or publication of t h i s thesis for f i n a n c i a l gain s h a l l not be allowed without my written permission. Department of The University of B r i t i s h Columbia 1956 Main Mall Vancouver, Canada V6T 1Y3 Date DE-6 (3/81) ABSTRACT High frequency corrosion fatigue crack propagation behavior of AISI 316L stainless steel was studied in 1M NaCl and 1M NaCl + 0.01M Na2S2C»2, under various anodically and cathodically polarized potentials, and dessicated air at 22°C and 1 atmosphere pressure. Constant load amplitude fracture mechanics techniques employing single edge notch specimens were used to assess the fatigue crack growth rate in the various environments. Unique specimen preparation procedures were developed which allowed near-threshold behaviors to be studied under gradually rising crack t i p stress intensity conditions. Polarization studies showed that the presence of thiosulphate catalyzed the dissolution of stainless steel in low pH solutions(pH~1), due to reduction of thiosulphate species to H2S, but had no effect in the near neutral solutions. Fatigue tests conducted in the neutral NaCl + Na2S2C>2 solution at cathodic potentials showed that the presence of thiosulphate had an insignificant effect. This indicated that high frequency fatigue produces efficient exchange of bulk solution with the crack t i p environment, which prevented the lowering of pH in the crack by hydrolysis effects and prevented reduction of thiosulphate to H^S. Fatigue crack retardation phenomena were very pronounced in II the n e a r - t h r e s h o l d r e g i o n s i n most of the f a t i g u e t e s t s . The cause of t h i s r e t a r d a t i o n was a t t r i b u t e d mainly to the s u r f a c e -roughness- induced crack c l o s u r e e f f e c t , which reduced the e f f e c t i v e crack t i p c y c l i c s t r e s s i n t e n s i t y AKfc^ t o a lower l e v e l . T h i s c l o s u r e e f f e c t only predominated i n the near-t h r e s h o l d r e g i o n where s i g n i f i c a n t Mode II l o a d i n g was p r e s e n t . The i n f l u e n c e of v a r i o u s imposed anodic and c a t h o d i c p o t e n t i a l s was found to be c o n s i s t e n t with the surface-roughness-induced crack c l o s u r e e f f e c t s . The observed crack growth a c c e l e r a t i n g e f f e c t of hig h anodic p o t e n t i a l s was a t t r i b u t e d to the c o r r e s p o n d i n g high removal rate of s u r f a c e roughness i n the wake of the c r a c k , which kept the e f f e c t i v e c y c l i c s t r e s s i n t e n s i t y l e v e l near the a p p l i e d v a l u e s . Crack f r a c t o g r a p h y was s t u d i e d by scanning e l e c t r o n microscopy. I t showed that the fr a c t o g r a p h y g e n e r a l l y c o n s i s t e d of t h r e e r e g i o n s ; a c r y s t a l l o g r a p h i c c l e a v a g e - l i k e n e a r - t h r e s h o l d r e g i o n , a f e a t h e r y and f i b r o u s t r a n s i t i o n r e g i o n , and a s t r i a t e d r e g i o n . Using an e t c h p i t t i n g technique, i t was determined that the crac k plane and crack propagation d i r e c t i o n s i n the near-t h r e s h o l d r egion were mainly those of {111} <110>, {110} < T l 2 > , {110} <001>, and other higher indexed p l a n e s . These crack o r i e n t a t i o n s were e f f e c t e d by the a c t i v a t i o n of a s i n g l e s l i p system or the a l t e r n a t e a c t i v a t i o n of two i n t e r s e c t i n g s l i p systems. I l l TABLE OF CONTENTS Page Abstract II Table of Contents IV List of Tables VII List of Figures VIII List of Symbols and Abbreviations XIV Acknowledgement XVI Chapter 1. INTRODUCTION 1 1.1 Corrosion Fatigue Failure of Suction Press Rolls 2 1.2 Brief History of Roll Material 3 1.3 Cyclic Stresses in Suction Press Rolls 4 1.4 Chemistry of the Operating Environments in Paper Machines 5 1.5 Project Objective 6 2. GENERAL CORROSION FATIGUE CRACK GROWTH BEHAVIOR .. 8 2.1 Application of Linear Elastic Fracture Mechanics 8 2.2 Fatigue Crack Growth Development 9 2.2.1 Fatigue Crack Propagation 10 IV 2.3 Fatigue Crack Advance Mechanisms 14 2.4 Environmental Enhancement Effects 17 2.5 Solution Chemistry Within Corrosion Fatigue Crack 19 2.6 Effect of Cyclic Frequency 20 2.7 Effect of Applied Potentials 21 2.8 Effect of Thiosulphate Ions 23 2.9 Effect of Load Ratio 24 2.10 Crack Closure Effect 24 3. EXPERIMENTAL PROCEDURE 28 3.1 Material Specification 28 3.2 Metallography 29 3.3 Specimen Geometry 29 3.4 Specimen Preparation 30 3.5 Constant Load Magnitude Fatigue Crack Growth Testing 33 3.6 Cycle-To-Failure Testing 36 3.7 Potentiodynamic Testing 36 3.8 Optical and Electron Microscopy 39 3.9 Etch Pitting 40 4. EXPERIMENTAL RESULTS 42 4.1 Fatigue Crack Growth Curves •... 42 4.2 Fatigue Crack Propagation Rates 54 4.3 Near-Threshold Crack Propagation 64 V 4.4 Fractography 66 4.5 Cyclic Dissolution Current 84 4.6 Polarization Behavior 87 5. DISCUSSION 92 5.1 Corrosion Fatigue Crack Propagation Rates .... 92 5.2 Crack Closure Effect 93 5.3 Surface-Roughness-Induced Crack Closure 97 5.3.1 Crack Propagation Mode 97 5.4 Effect of the Applied Potential 100 5.5 Effect of Thiosulphate 101 5.6 Validity of the Measured Threshold AK in Corrosion Fatigue 103 5.7 Relevance to the Industrial Situation 104 5.8 Mechanisms of Fatigue Crack Propagation 105 5.8.1 Cleavage Fracture Mechanism 106 5.8.2 The Alternating Shear Mechanism 110 5.8.3 Transition Zone 117 5.8.4 Striated Region 119 5.9 Effect of Cyclic Stress Intensity on Fractography 119 5.10 Effect of the Cyclic Frequency 120 5.11 Stage I vs. Stage II Growth 123 6. CONCLUSION 124 BIBLIOGRAPHY 126 VI LIST OF TABLES Table Page I Testing Conditions for Fatigue Crack Propagation Studies 44 II Fatigue Crack Growth Properties of AISI 316L under Various Testing Conditions 65 III Fatigue Crack Orientation in the Near-Threshold Regions 114 IV Fractographic Correspondence with the Applied Cyclic Stress Intensities and Fatigue Crack Propagation Rates under Various Testing Conditions 121 VII LIST OF FIGURES Figure Page 1 Schematic plot of fatigue crack propagation 11 rate, depicting the commonly recognized three regions of crack propagation. 2 Schematic representation of fatigue crack propaga- 16 tion by the irreversible s l i p mechanism, (a) zero load; (b) forward s l i p during the opening load; (c) reverse s l i p on a different, coplanar s l i p plane. 3 Schematic representation of fatigue crack propaga- 16 tion by the plastic blunting mechanism, (a) zero load; (b) small tensile load; (c) maximum tensile load of the cycle; (d) small compressive load; (e) maximum compressive load of the cycle; (f) small tensile load in the succeeding cycle. Stress axis is vertical. 4 The Single Edge Notch specimen geometry and its 31 crack tip cyclic stress intensity calibration formula. 5 (a) Schematic of the environmental c e l l arrangement 34 for fatigue testing; (b) and (c) Photographs of fatigue testing in progress. 6 Schematic of the polarization test c e l l . 38 7 Scanning electron micrograph showing etch pits 41 formed on the indented polished AISI 316L stainless steel. The edges of the etch pits were parallel to s l i p traces induced by the indentation marks, indicating the facet faces are of {111} type. 8 Fatigue crack growth of AISI 316L in 1M NaCl at 45 free corrosion potential(-100 to -173 m V ( s c E ) ^ and 123 Hz. 9 Fatigue crack growth of AISI 316L in 1M NaCl at 46 +600 m v( S C E) and 115 Hz. VIII 10 Fatigue crack growth of AISI 316L in 1M NaCl at 47 +0 n>V^ S C Ej and 127 Hz. 11 Fatigue crack growth of AISI 316L in 1M NaCl at 48 -540 mV^ S C E) and 125 Hz. 12 Fatigue crack growth of AISI 316L in IM NaCl at 49 -540 m V ( S C E ) and 1 5 H z « 13 Fatigue crack growth of AISI 316L in 1M NaCl & 50 1M Na 2S 20 3 at -300 mV^SCE^ and 122 Hz. 14 Fatigue crack growth of AISI 316L in 1M NaCl & 51 1M Na^jO^ at free corrosion p o t e n t i a l (-280 to -340 m V ( S C E ) ) and 121 Hz. 15 Fatigue crack growth of AISI 316L in IM NaCl & 52 1M Na 2S 20 3 at -450 m V ( S C E ) a n d 122 Hz. 16 Fatigue crack growth of AISI 316L in dessicated 53 a i r and 124 Hz. 17 Fatigue crack propagation rate (da/dN vs. c y c l i c 55 stress intensity AK) of AISI 316L in 1M NaCl at free corrosion potential(-100 to -180 m V(scE)^ and 123 Hz. 18 Fatigue crack propagation rate (da/dN vs. c y c l i c 56 stress intensity AK) of AISI 316L in 1M NaCl at +600 mV( S C Ej and 115 Hz. 19 Fatigue crack propagation rate (da/dN vs. c y c l i c 57 stress intensity AK) of AISI 316L in 1M NaCl at 0 mV( S C E) and 127 Hz. 20 Fatigue crack propagation rate (da/dN vs. c y c l i c 58 stress intensity AK) of AISI 316L in 1M NaCl at -540 mV( S C j and 125 Hz. IX 21 22 23 24 25 26 Fatigue crack propagation rate (da/dN vs. c y c l i c 59 stress intensity AK) of AISI 316L in IM NaCl at -540 mV( S C E) and 15 Hz. Fatigue crack propagation rate (da/dN vs. c y c l i c 60 stress intensity AK) of AISI 316L in IM NaCl & 0.01M Na 2S 20 3 at -300 mV^ S C Ej and 122 Hz. Fatigue crack propagation rate (da/dN vs. c y c l i c 61 stress intensity AK) of AISI 316L in IM NaCl & 0.01M Na 2S 20 3 at free corrosion potential(-280 to -340 n>V ( S C E )) and 121 Hz. Fatigue crack propagation rate (da/dN vs. c y c l i c 62 stress intensity AK) of AISI 316L in IM NaCL & 0.01M Na 2S 20 3 at -450 mVj S C E) and 122 Hz. Fatigue crack propagation rate (da/dN vs. c y c l i c 63 stress intensity AK) of AISI 316L in dessicated a i r and 124 Hz. Fatigue crack surfaces of AISI 316L in 1M NaCl at 69 corrosion condition and 123 Hz. (a) AK=l8.64MPav/m, V=1.30E-09 m/cycle; (b) AK=20.93MPav/m, V=1.30E-09 m/cycle; (c) AK=26.15MPav/m, V=1.77E-09 m/cycle: (d) AK=32.16MPa»/m, V=7.50E-09 m/cycle; (e) AK=42.94MPat/m, V=7.20E-08 m/cycle; (f) AK=52.38MPay/m, V=1.05E-07 m/cycle. 27 Fatigue crack surfaces of AISI 316L in IM NaCl at +600 mV( S C Ej and 115 Hz. (a) AK=9.21MPav/m, V=1.00E-10 m/cycle; (b) AK=10.43MPa/m, V=1.00E-08 m/cycle; (c) AK=12.60MPa/m, V=1.lOE-07 m/cycle; (d) AK=14.90MPav/m, V=1.70E-07 m/cycle; (e) AK=l8.50MPa/m, V=2.37E-07 m/cycle; (f) AK=22.60MPai/m, V=2.85E-07 m/cycle; (g) AK=28.50MPav/m, V=3.50E-07 m/cycle; (h) AK=35.90MPa/m, V=4.00E-07 m/cycle. 71 X 28 Fatigue crack surfaces of AISI 316L in IM NaCl at 73 -540 mV( S C E) and 125 Hz. (a) AK=21 .35MPa»/m, V=1.27E-02 m/cycle; (b) AK=24.72MPa-/m, V=1.25E-09 m/cycle; (c) AK=29.60MPa»/m, V=9.00E-09 m/cycle; (d) AK=36.07MPa/m, V=3.10E-0B m/cycle; (e) AK=44.29MPav/m, V=8.20E-08 m/cycle; (f) AK=52.52MPav/m, V=2.l5E-07 m/cycle; (g) AK=84.5lMPai/m, V=3.60E-07 m/cycle. 29 Fatigue crack surfaces of AISI 316L in 1M NaCl & 75 0.01M Na 2S 20 3 at - 3 0 0 ( S C E ) and 122 Hz. (a) AK=20.53MPat/m, V=1.35E-09 m/cycle; (b) AK=25.33MPav/m, V=2.05E-09 m/cycle; (c) AK=30.1 9MPav/m, V=1.53E-07 m/cycle; (d) AK=35.73MPav/m, V=2.30E-07 m/cycle; (e) AK=44.28MPav/m, V=3.20E-07 m/cycle; (f) AK=58.30MPay/m, V=4.15E~07 m/cycle. 30 Fatigue crack surfaces of AISI 316L in 1M NaCl & 77 0.01M Na2S2C>2 at free corrosion potential and 121 Hz. (a) AK=11.84MPa»/m, V=5.00E-10 m/cycle (b) AK=12.24MPa»/m, V=1.20E-10 m/cycle (c) AK=1 9.62MPav/m, V=2 .00E-1 0 m/cycle (d) AK=25.34MPav/m, V=5.00E-10 m/cycle (e) AK=28.02MPai/m, V=1.00E-09 m/cycle (f) AK=30.86MPa»/m, V=1.30E-09 m/cycle (g) AK=38.35MPav/m, V=3.20E-08 m/cycle (h) AK=47.1 2MPav/m, V=1.00E-07 m/cycle. 31 Fatigue crack surfaces of AISI 316L in 1M NaCl & 79 0.01M Na 2 S 20- at -450 mV (SCE) ' and 122 Hz. (a) AK= 13 . 30MPaj/m, V= 4. 20E- 1 1 m/cycle; (b) AK= 15 .10MPa/m, V= 4. 00E- 10 m/cycle; (c) AK= 17 .90MPa»/m, V= 8. 76E- 09 m/cycle; (d) AK= 21 . 40MPav/m, V= 4. 30E- 08 m/cycle; (e) AK= 26 . 1 OMPai/m, V= 1 . 1 0E-07 m/cycle; (f) AK= 31 .60MPa/m, V= 2. 10E- 07 m/cycle; (g) AK= 40 .86MPav/m, V= 3. 90E- 07 m/cycle, 1000X, (h) AK= 40 .86MPa/m, V= 3. 90E- 07 m/cycle, 8000x. XI 32 Fatigue crack surfaces of AISI 316L in dessicated 81 air and 124 Hz. (a) AK=16.32MPa/m, V=2.10E-10 m/cycle; (b) AK=l8.86MPa/m, V=4.09E-09 m/cycle; (c) AK=22.60MPa/m, V=2.40E-08 m/cycle; (d) AK=27.40MPa/m, V=5.30E-08 m/cycle; (e) AK=35.60MPav/m, V=1.40E-07 m/cycle; (f) AK=45.l8MPa/m, V=2.48E-07 m/cycle; (g) AK=58.33MPav/m, V=3.80E-07 m/cycle. 33 Scanning electron micrograph of etch pits formed on 83 the corrosion fatigue fracture surface of AISI 316L 34 Matching features of two opposing crack surfaces 83 in the striated region of AISI 316L in 1M NaCl & 0.01M Na 2S 20 3 at -450 mV ( S C E ) and 122 Hz. Peaks match with peaks and valleys match with valleys. 35 Corrosion fatigue fracture surfaces of cylindrical 85 fatigue specimen of AISI 316L in 1M NaCl under free corrosion condition at 30 Hz. (a) Crystallographic region; (b) Transition zone; (c) striated region. 36 (a) Corrosion current during non-cracking stage and 86 (b) corrosion current during cracking stage of AISI 316L in 1M NaCl +0.01M Na2S2C>3 at -300 mV ( S C E ) and 122 Hz. 37 Acidification of the crack t i p solution during the 89 load cessation period as indicated by the increase of the anodic dissolution current of AISI 316L in 1M NaCl at 0 mV( S C E) and 127 Hz. 38 Anodic polarization curves of AISI 316L in 1M NaCl 90 and 1M NaCl + 0.01M Na 2S 20 3. 39 Anodic polarization curves of AISI 316L in simu- 91 lated crack tip solution(lM FeCl 2 adjusted to pH 1 with saturated CrCl 3); and 0.01M Na 2S 20 3 addition. XII 40 Fatigue crack path on polished surface. 96 (a) Crack branching effect is evident. (b) Both {111} and non {111} types of crack path are shown. 41 Orientation of fracture surfaces from etch p i t s . 111 (a) {111} crack plane with <110> crack direction at the near-threshold region. (b) Both {110} <112> and {100} <011> crack orientation are shown. 42 Schematics of the shapes of etch pits on {111} , 113 {110} , and {100} planes and their corresponding edge directions. 43 Schematic of the alternate s l i p model 116 44 Cleavage-like fracture surfaces in the near-threshold region, (a) Mag. 10000. (b) Mag. 8000. 118 X I I I LIST OF SYMBOLS AND ABBREVIATION Symbols a Crack l e n g t h W SEN specimen w i d t h B SEN specimen t h i c k n e s s K c C r i t i c a l s t r e s s i n t e n s i t y f a c t o r K m a x Maximum a p p l i e d c r a c k t i p s t r e s s i n t e n s i t y K m ^ n Minimum a p p l i e d c r a c k t i p s t r e s s i n t e n s i t y K c^ Crack c l o s u r e s t r e s s i n t e n s i t y AK C y c l i c s t r e s s i n t e n s i t y range AK ej£ E f f e c t i v e s t r e s s i n t e n s i t y range AK f c^ T h r e s h o l d s t r e s s i n t e n s i t y range P m Maximum a p p l i e d l o a d t o SEN specimen max c P m ^ n Minimum a p p l i e d l o a d t o SEN specimen R S t r e s s r a t i o o T e n s i l e s t r e s s T Shear s t r e s s N Number of f a t i g u e c y c l e s m Exponent i n P a r i s ' s power x law r e l a t i o n s h i p E Q E q u i l i b r i u m p o t e n t i a l ( s t a n d a r d hydrogen e l e c t r o d e ) X I V A b b r e v i a t i o n Y S 0 . 2 % UTS SEN S C E SEM 0.2 % o f f s e t y i e l d s t r e n g t h U l t i m a t e t e n s i l e s t r e n g t h S i n g l e edge n o t c h specimen S a t u r a t e d c a l o m e l e l e c t r o d e Scanning e l e c t r o n m i c r o s c o p e X V ACKNOWLEDGEMENT The author wishes to express his sincere gratitude and appreciation to Dr. Desmond Tromans for his invaluable advice and supervision throughout this project. He would also like to extend his thanks to Dr. Ivan Dickson of Ecole Polytechnique of Montreal, Quebec for a helpful discussion. The discussions and assistance of fellow graduate students, technicians, and faculty members are gratefully acknowledged by the author. In addition, special thanks should be given to Mr. Mitsuhiro Yasuda for his unfailing help on numerous occasions during the project. The financial support for this project was provided by the Science Council of British Columbia through the GREAT awards and is highly appreciated. XVI 1. INTRODUCTION Corrosion fatigue may be defined as the conjoint action of an alternating stress and environmental interaction, which leads to premature failure of metals by cracking. In the absence of either the cyclic stress or the environment, the performance of metals is generally more superior than when they are subjected to fatigue and corrosive action simultaneously. The majority of fatigue failures are caused by corrosion fatigue since only the absolute vacuum environment can be qualified as being inert. Environments as innocuous as dry air sometimes greatly enhance the fatigue crack growth rate as a result of metal interaction with air[1 ] . Many industrial corrosion fatigue failures have been recorded up to date. It has been reported that corrosion fatigue cracks form on steam turbine blades[2], ventilator impellers[3 ], and rotor blade bolts of helicopters[4] during service. Thermal stresses and gaseous pressures produced in the periodic heating cycles in light water reactor pressure vessels[5] and incessant tidal impacts on offshore o i l rig structures[6] can also provide conditions for the occurrence of corrosion fatigue, which may later cause grave consequences both in terms of human suffering * The number in the square brackets refers to the l i s t of references at the end of the thesis 1 and f i n a n c i a l l o s s . Occurrence of these f a i l u r e s p r e s e n t s a s p e c i a l c h a l l e n g e f o r s c i e n t i s t s and e n g i n e e r s to s e l e c t or develop b e t t e r m a t e r i a l s and to d e s i g n a s a f e r s t r u c t u r a l geometry so as to minimize the combined damage caused by f a t i g u e and environmental e f f e c t s . The goal w i l l be a c h i e v e d more completely through an improved understanding of the fundamental p r o c e s s e s governing c o r r o s i o n f a t i g u e . The major t h r u s t of t h i s t h e s i s was d i r e c t e d towards fundamental s t u d i e s r e l e v a n t to c o r r o s i o n f a t i g u e of s u c t i o n press r o l l s i n the p ulp and paper i n d u s t r y . 1.1 C o r r o s i o n F a t i g u e F a i l u r e of S u c t i o n P r e s s R o l l s : One of the s e r i o u s problems faced by paper m i l l s i s the c o r r o s i o n of s t a i n l e s s s t e e l s u c t i o n p r e s s r o l l s . The f u n c t i o n of s u c t i o n press r o l l s i s to compress the wet p u l p or paper f a b r i c i n t o a denser m a t e r i a l and , at the same time, remove as much water as p o s s i b l e to minimize the c o s t of d r y i n g the paper. The removal of water i s achieved by vacuum s u c t i o n of excess water through hole p e r f o r a t i o n s on the r o l l s u r f a c e . I t i s not s u r p r i s i n g t hat i n t h i s environmental c o n d i t i o n c o r r o s i o n f a t i g u e problems are s evere. I t has been r e p o r t e d t h a t about a t h i r d of s t a i n l e s s s u c t i o n r o l l f a i l u r e s were caused by c o r r o s i o n f a t i g u e [ 7 ] . The c u r r e n t replacement c o s t of s u c t i o n r o l l s c o u l d reach as h i g h as one m i l l i o n d o l l a r s [ 8 ] . In a d d i t i o n , the 2 installation cost and production loss can push the real cost to a much higher value. It is estimated that the production loss is in the range of $10,000 per hour[8]. A Quebec plant was shut down for two and half weeks as a result of a suction r o l l failure[9]. It is apparent that the financial damage can be enormously high. There is also the possibility of human casualties i f plant workers are in the v i c i n i t y when the catastrophic fracture of a r o l l occurs. In order to avoid costly replacement and fatal injuries, i t is imperative to understand how the suction press r o l l s f a i l in their service conditions. Preventive measures can then be taken to reduce the chance of i t s occurrence. 1.2 Brief History of Roll Material: Bronze was i n i t i a l l y selected as suction press r o l l material[10 ]. High castability, machinability, and corrosion resistance have made i t highly desirable as r o l l material. However, as the need for higher paper production rate arose, attempts have been made to design a faster and wider paper machine. For this type of machine, bronze failed utterly. Its lack of stiffness and low yield strength could easily lead to gross plastic deformation of r o l l s , rendering i t unsuitable for large paper machines. Apparently, alloys of higher strength and stiffness should be selected instead. The superior corrosion resistance and substantially high 3 strength of stainless steels have caused them to be selected as the alternative materials. Various types of stainless steels have entered service since the late 1950'S[10]. Limited success was met at the beginning. However, as the corrosive operating environments became more severe as the result of more stringent environmental regulation and accompanying process modifications from open water to closed water systems, failures of stainless steel r o l l s have occurred. In the worst cases, the service l i f e of certain r o l l s is less than one year[lO]. Evidently, there is great urgency for the development of improved durable suction press r o l l materials. 1.3 Cyclic Stresses in Suction Press Rolls; It is reported that most suction press r o l l shells f a i l as a result of growing circumferential cracks that in i t i a t e at suction holes and travel outwards from the interior diameter of the shell[7]. As the rolls compress wet pulp into a much more compact paper material, the point of contact on the exterior surface is experiencing a large compressive force, whereas the opposite interior surface is under a large tensile force. A circumferential bending stress f i e l d is thus developed in the suction press r o l l s , which reaches a maximum at the point of contact with pulp and drops off to a minimum at the opposite side of the r o l l s . The magnitudes of the operating cyclic loads on 4 suction r o l l s will depend mainly upon the designs of paper machines, types of pulp processed, and the process conditions. The larger the cyclic loads and more severe the environments, the more detrimental they w i l l be to the performance of r o l l s . Residual stresses are also considered to hasten the corrosion fatigue failures[11]. These stresses usually originate from uneven cooling after casting or heat treatments of the r o l l s during the manufacturing processes. Other types of residual stress can develop due to the subsequent machining and d r i l l i n g operations. Residual tensile stress is undesirable because i t is additive to the applied tensile cyclic stress. Its presence w i l l not only increase the maximum tensile stress, therefore reducing the c r i t i c a l crack length which is required for the occurrence of catastrophic failures, but also raise the minimum tensile stress to a level above the crack closure stress, thus accelerating cyclic crack propagation rates. These aspects of fatigue crack growth behavior will be further discussed in more detail in the latter parts of the thesis. 1.4 Chemistry of the Operating Environments in Paper Machines: The composition of the process water in paper machines is very complex and can vary markedly from process to process. Anything which is either brought into or produced in the pulp 5 making process can end up in the process water. However, the major constituents, depending on the process system, could be identified as sulphate ions(up to 600 ppm) including thiosulphate(up to 35 ppm), chloride ions (up to 300 ppm), dissolved organic solids (up to 1000 ppm), and other dissolved inorganic solids (up to 1600 ppm)[12]. The temperature of the process water ranges from 40°C to 50°C and i t s acidity runs roughly from pH 4 to 6. The corrosivity of this water increases with the quantity of dissolved solid content and water acidity, due to the increase of electrolytic conductivity of the water. Chloride ions are particularly harmful to the integrity of the protective surface films of stainless steels, because chloride ions reduce their pitting resistance[13 ]. Thiosulphate ions are also known to lower the pitting resistance of stainless steels[l4], especially in highly acidic environments. In this type of aggressive environment, i t is not surprising that the corrosion fatigue problem is further aggravated. 1.5. Project Objective; This project was directed towards the study of corrosion fatigue crack growth behavior of an austenitic stainless steel, AISI 316L, at various anodically and cathodically polarized potentials in a chloride solution under closely controlled laboratory conditions. Furthermore, efforts to c l a r i f y the 6 effect of thiosulphate ions in conjunction with chloride ions on the corrosion fatigue resistance of stainless steel were also made in this study. It is hoped that this undertaking w i l l contribute to the general understanding of corrosion fatigue as a whole and ,in addition, assist in solving specific corrosion fatigue problems encoutered by suction press r o l l s in paper mills. 7 2. GENERAL CORROSION FATIGUE CRACK GROWTH BEHAVIOR: 2.1 Application of Linear Elastic Fracture Mechanics: Historically, experimental determination of fatigue crack propagation rates has been conducted on a large number of materials, in order to derive a set of empirical data from which design c r i t e r i a and maintenance inspection guidelines could be clearly specified for load-carrying mechanical and structural components employed in service. Various types of experimental tests were proposedf15]. However, none of them became generally acceptable, because their application to real service conditions was very limited in scope. The experimental conditions have to closely resemble those of service conditions in stress amplitude, geometry of external forces, geometry of stressed components, and crack shape and length. A basic breakthrough in testing techniques resulted from the application of linear elastic fracture mechanics, which relates the macroscopic crack behavior to the local characteristics of the stress-strain response of the material at the crack t i p [ l 6 ] . Therefore, i f the local stress state at the crack t i p can be quantitatively described, the fatigue crack propagation rate can be accurately determined for any cracked body of an arbitrary shape under external forces of any arbitrary arrangement. Linear 8 e l a s t i c f r a c t u r e mechanics a l l o w s r e a s o n a b l e d e s c r i p t i o n of the s t r e s s f i e l d , termed s t r e s s i n t e n s i t y , i n t h e v i c i n i t y of t h e c r a c k t i p , p r o v i d e d the a p p l i e d s t r e s s , c r a c k l e n g t h , and geometry of the c r a c k e d body a r e known. A s i m p l e r e l a t i o n between th e s t r e s s i n t e n s i t y and c r a c k p r o p a g a t i o n r a t e can then be e s t a b l i s h e d . Most i m p o r t a n t l y , t h i s r e l a t i o n s h i p can be a p p l i e d t o any c o n d i t i o n as l o n g as the s t r e s s i n t e n s i t y a t t h e c r a c k t i p i s w e l l d e f i n e d . 2.2 F a t i g u e Crack Growth Development; The l i f e t i m e of m e t a l s when s u b j e c t e d t o c y c l i c l o a d i n g u s u a l l y c o n t a i n s f o u r s t a g e s [ l 7 ] : (1) I n i t i a l m i c r o s t r u c t u r a l damages i n t h e form of p e r s i s t e n t s l i p band or c e l l u l a r f o r m a t i o n . (2) F o r m a t i o n of s u r f a c e or i n t e r n a l m i c r o f l a w s . (3) P r o p a g a t i o n of e x i s t i n g m i c r o c r a c k t o a c r i t i c a l m a c r o s c o p i c l e n g t h . (4) F i n a l c a t a s t r o p h i c f a i l u r e . However, i n r e a l s e r v i c e c o n d i t i o n s , t h e f i r s t two s t e p s c o u l d be bypassed. I n i t i a l m i c r o c r a c k s c o u l d a l r e a d y be p r e s e n t as a r e s u l t of c a s t i n g and m a c h i n i n g d e f e c t s d u r i n g t h e m a n u f a c t u r i n g p r o c e s s e s , d e f e c t s due t o second phase p r e c i p i t a t e s 9 and inclusions, and crevice or pit formation in the service environments[18]. The presence of these preexisting defects is practically unavoidable and should be taken into consideration in designing engineering components. Consequently, since the existence of microflaws as i n i t i a l cracks is inevitable, the fatigue crack propagation stage takes on particular importance. 2.2.1 Fatigue Crack Propagation: The crack propagation region is again made of three substages as shown schematically in Figure 1. Stage I is the near-threshold crack growth stage, in which a typical crack - 9 growth rate is below 10 m/cycle. The word "threshold" mentioned above is theoretically defined as the cyclic stress intensity value at the crack tip, below which no crack growth is detectable. However, for the practical purpose of the experiment, i t s t r i c t l y depends upon the resolution of monitoring instruments and the patience of experimenters. 10 Figure 1. Schematic plot of fatigue crack propagation rate, depicting the commonly recognized three regions of crack propagation[19 ]. 11 The second stage(Stage II) is the intermediate crack growth region. In this region the crack growth obeys the Paris power law relationship[20], da/dN = C* AKn (1) where da/dN is the crack growth increment per cycle, C and n are experimentally determined scaling factors, and AK ,the alternating stress intensity, is equal to ^Kmax Kmin ^ " At higher AK values, an accelerating crack growth stage is observed(Stage III) where K m a x approaches the c r i t i c a l stress intensity for fracture Kc. Catastrophic failure occurs when K = K . max c Near-threshold fatigue crack growth is of particular interest, since a major portion of the fatigue l i f e of a cracked component may be spent in this slow crack growth region. Especially in the case of high cycle fatigue applications, such as steam turbines which have to withstand extreme high frequency and low amplitude loading for lifetimes in the range of 101^ to 1 o 10 cycles, i t is possible for any undetectable i n i t i a l cracks to grow to the c r i t i c a l failure size even at the low near-threshold crack propagation rate. In reality, failure can occur at a much shorter time since the crack propagation rate w i l l tend to increase as the fatigue crack length becomes longer. In addition, the determination of the threshold stress intensity value may assist in designing a fail-safe structure. As of now, 12 there is s t i l l a substantial lack of fatigue crack growth data concerning the near-threshold region. Any undertaking in the study of fatigue crack grow in the near-threshold region would contribute to better understanding of fatigue behavior as a whole. Before the possible fatigue crack growth mechanisms are discussed, the distinction between Stage I and Stage II fatigue behavior should be c l a r i f i e d . One attempt is to differentiate these two regions by the degree of plastic activity occurring ahead of the crack t i p . In Stage I, at low stress intensity, limited s l i p activity is attained at the crack tip. Only one primary s l i p system is believed to be activated during each cycle. It is observed that the local crack tip orientation generally follows the maximum shear plane, which is usually a s l i p system with the largest Schmid factor inclined with respect to the applied tensile direction. The transition from Stage I to Stage II crack growth occur at a higher stress intensity, where an increasing amount of plastic flow at the crack t i p results in extensive s l i p a c t i v i t i e s on non-coplanar s l i p planes. The difference between Stage I and Stage II can also be revealed in the sigmoid fatigue crack growth rate vs. cyclic crack t i p stress intensity(logv-log AK) plot. The substantial reduction of slope in Stage II as opposed to Stage I indicates the crack increment per cycle may be 13 achieved through different fatigue mechanisms, due to the profuse plastic deformation ahead of the crack t i p . 2.3 Fatigue Crack Advance Mechanisms: There are several popularly espoused fatigue mechanisms which can explain crack growth at various levels of stress intensity. (1) Irreversible s l i p action: This is mainly the activity of a single s l i p system. During the uploading portion of a fatigue cycle(i.e. crack opening), s l i p occurs in the forward direction and exposes an area of fresh metal surface at the crack t i p , on which adsorption of foreign species from the local environment or formation of a salt or oxide film occurs[2l]. In the unloading portion(i.e. crack closing), the exposed s l i p step w i l l try to return to i t s original position, but reverse s l i p on the original s l i p plane is prevented due to blockage by the surface film. Instead, a neighboring coplanar s l i p plane is activated in the reverse direction. Repeated Irreversible s l i p on parallel s l i p planes during repeated forward and reverse stress cycles can therefore lead to crack advancement as shown in Figure 2. The crack plane w i l l generally be that of the s l i p plane and take on a 14 crystallographic morphology with l i t t l e evidence of plastic deformation(i.e. no s l i p steps due to intersecting s l i p systems). (2) Alternate s l i p action: This mechanism is generally believed to be associated with the i n i t i a l period of Stage II fatigue crack growth. It involves the alternate activation of two intersecting s l i p planes[22]. One favorably oriented s l i p system is f i r s t activated during one fatigue cycle (or series of cycles) and the crack will extend in the direction of slip movement as described for irreversible s l i p . However, on the second cycle (or second series of cycles) strain hardening temporarily inhibits s l i p on the f i r s t system and s l i p now occurs on a second intersecting s l i p system of equally favorable orientation. As a result, the crack continues to grow on a different plane. In the next cycle (or series of cycles), the f i r s t s l i p system will again be activated since the crack tip has passed the strain-hardened zone. Crack growth by this mechanism w i l l retain an overall crystallographic fracture surface. The alternating movement of two roughly symmetrical s l i p systems with respect to the crack plane w i l l yield either {110} or {100} fracture planes in FCC metal, depending on the orientations of the operating s l i p systemsf 23]. (3) The Plastic Blunting Model: At higher stress intensity, a greater degree of crack tip plastic deformation results and a greater number of s l i p systems 15 Figure 2. Schematic representation of fatigue crack propagation by the irreversible s l i p mechanism, (a) zero load; (b) forward sl i p during the opening load; (c) reverse s l i p on a different, coplanar s l i p plane. Figure 3. Schematic representation of fatigue crack propagation by the plastic blunting mechanism, (a) zero load; (b) small tensile load; (c) maximum tensile load of the cycle; (d) small compressive load; (e) maximum compre-ssive load of the cycle; (f) small tensile load in the succeeding cycle. Stress axis is ver t i c a l . 1 6 are activated simultaneously[24]. As Figure 3 indicates, during the uploading portion of the fatigue cycle, the crack t i p is blunted due to the simultaneous sliding off of two s l i p systems. Upon the unloading portion, s l i p systems are reversed, the crack faces approach each other, and the newly created surface is folded by buckling of material at the crack t i p . A double notched crack t i p and a unit of crack increment are produced at the end of a fatigue cycle. A repeated occurrence of this blunting process w i l l give the fracture a long parallel striated appearance. Strong evidence of ductile tearing and frequent presence of secondary cracks are usually observed. 2.4 Environmental Enhancement Effects: Aside from the purely mechanical crack growth process by the cyclic stress, the environment can also exert considerable influence by either synergistically accelerating the propagation rate[25] or serving as an additional growth mechanism which is superimposed upon the mechanical process[26]. The environmental influences can normally be divided into three categories. (1) Film rupture-metal dissolution effect: This mechanism was originally applied to the crack growth process of the stress corrosion cracking phenomenon!27]. It can be similarly utilized in the fatigue case. The mechanism proposes that the protective film usually present at a metal surface is 17 broken by the cyclic strain produced by the alternating stress at the crack tip, where the exposed metal becomes a comparatively small and anodic region, whereas the.rest of the metal serves as cathodic site for electrochemical reactions. Intense dissolution w i l l take place at the crack t i p . The anodic dissolution can either extend the crack length by removing the metal lying directly ahead of the crack t i p . Or i t can contribute by dissolving the cyclic strain-hardened surface layer, thereby further increasing the sliding distance of the activated s l i p systems. (2) Hydrogen embrittlement effect: Through this mechanism, crack growth can be enhanced by the introduction of atomic hydrogen to the region immediately ahead of the crack t i p . The hydrogen gets into the metal by f i r s t adsorbing onto the fresh metal surface during reduction of H + and later diffusing to either l a t t i c e defects or inclusion/matrix interfaces. A high internal pressure of hydrogen gas developed through gradual recombination of atomic hydrogen w i l l make up part of the tensile stress which is required for crackingt28]. Or the presence of adsorbed hydrogen will simply reduce the interatomic bonding energy[29]. Consequently, the required fracturing stress is lowered. (3) Rebinder effect: Similar to hydrogen embrittlement, this mechanism suggested 18 that a specific species absorbs and interacts with metallic bonds at the crack tip[30]. This interaction causes a reduction in bond strength and leads to a lowering of the required applied tensile stress to produce b r i t t l e fracture. The degree of the adsorption of the detrimental species strongly depends on the electrochemical potential in a given environment. If a c r i t i c a l potential range is present, the Rebinder effect could become highly deleterious. 2.5 Solution Chemistry Within Corrosion Fatigue Crack: Localized acidification of the solution has been observed to occur in stress corrosion cracks in aluminum alloys[3l] and steels[32]. Hydrolysis reactions of dissolved alloy components and restricted exchange with the bulk solution are responsible for the formation of an highly acidic environment at the crack tip . For the case of corrosion fatigue, the repeated opening and closing of the crack due to the alternating stress promotes the interchange of crack solution with the bulk environments. If the mixing is thorough, no significant modification of solution chemistry would be expected within the crack. On the other hand, if the pumping action of fatigue cycling is not strong, some degree of acidification w i l l occur, as observed in several 19 experiments[33,34]. The strength of the pumping force and therefore the degree of mixing are found to be positively dependent on crack length, cyclic frequency, and cyclic stress amplitude[35]. If the above mentioned operating parameters are at high values, very l i t t l e modification of crack tip solution occurs. 2.6 Effect of Cyclic Frequency; It has been observed that the fatigue crack growth increment per cycle is dependent on the cycli c frequency in the presence of corrosive environments. In contrast, the cyclic crack growth rate remains relatively unchanged when fatigue tests are conducted in vacuum; Generally, the corrosion fatigue crack propagation rate(da/dN) increases as the frequency is reduced[36]. This increase in fatigue crack growth rate with decreasing frequency is attributed to the fact that more time is available for corrosive attack to take place during each cycle. However, when the cyclic fatigue crack growth rate(da/dN) is converted to crack increment per period of time(da/dt), the crack growth dependence on cyclic frequency suggests otherwise[37]. The fatigue crack increment per unit of time is found to be increased when cyclic frequency is increased. This is to be expected, because the maximum effect of environmental degradation 20 of metals through dissolution occurs immediately after new fresh metal surface is exposed due to the fatigue action. The dissolution currents w i l l gradually decay to a steady state, which is believed to be caused by the formation of surface film on the exposed metal[38]. As the cyclic interval becomes shorter, there is less time for the formation of surface film. Therefore, metal degradation due to dissolution at the crack t i p per unit of time will increase significantly as a result of higher frequency. The frequency may increase to such a level that no repair of surface film occurs at the crack t i p . Then the crack t i p w i l l essentially exist as a bare metal surface which is under constant corrosion attack. 2.6 Effect of the Applied Potentials It has long been recognized that stress corrosion cracking of metals is accelerated considerably within certain c r i t i c a l potential ranges. The strong dependence on potential is believed to be the direct result of complex interaction between the dissolution and repassivation rate of metals or alloys, strength and stability of the surface film at the crack t i p , and production and aggressive action of detrimental species which might embrittle the alloys[39]. The most susceptible potential ranges are pitting and active-passive transition regions. In 21 these potential ranges, the film tends to be unstable and not readily reformed when ruptured, thereby allowing metal dissolution at exposed areas. However, film repair processes are necessary to prevent excessive dissolution which may lead to crack t i p blunting. The crack t i p dissolution increment is mainly achieved by removal of metal by anodic processes. In other cases, specific potential ranges lead to the production of harmful species in the surrounding environment. For instance, i f the corrosion potential should be below the hydrogen evolution potential, hydrogen embrittlement may have a pronounced effect. Similar considerations may be applied to corrosion fatigue. Electrochemical potentials which promote unstable surface films, inadequate repassivation kinetics, and production of embrittling species are expected to accelerate fatigue crack growth. However, there is a major difference with corrosion fatigue because of the high cyclic strain rate at the crack t i p . Very l i t t l e time is allowed for film reformation(repassivation) to occur before the next cyclic rupture event. Therefore even the usual beneficial passivation potentials may turn out to be deleterious since the associated transient dissolution currents prior to passivation are much higher. The crack w i l l then advance more quickly due to the greater amount of metal dissolution. 22 2.8 Effect of Thiosulphate Ions: It was found that the presence of thiosulphate ions would greatly enhance the penetration rate of crevice corrosion of AISI 316L stainless steel in chloride solution[14]. The aggressiveness of the thiosulphate ions is attributed to the formation of dissolved H2S in acidic solution, which in turn acts as a catalyst in the anodic dissolution process according to the following electrochemical reactions[14], consistent with aqueous thermodynamics[40]. S 20 3 2" + 6H+ + 4e~ = 2S + 3H20 (2) S + 2H+ + 2e" = H 2 S ( a g ) (3) F e + H2 S(aq) = F e H S " (ads) + H + { 4 ) F e H S " (ads) = F e H S + + 2 e + ( 5 ) FeHS+ + H + = F e 2 + + H 2S ( a g ) (6) Overall, H2 S(aq) * s n o t c o n s u m e ^ and acts as a true catalyst. Therefore, only a small quantity of H2S is required to have significant accelerating effect on the anodic reaction. Other experiments conducted on the stress corrosion cracking of stainless steels have suggested that the dramatically deleterious effect of thiosulphate was mainly due to the production of elemental sulphur around the crack t i p by cathodic reduction of the thiosulphate[41]. The adsorbed sulphur was 23 reported to increase dissolution of the metal and hinder the repassivation process. In both cases, localization of acidification of solution, either in crevices or cracks, is a necessary prerequisite for thiosulphate to have a pronounced influence on the rate of the anodic dissolution. The reduction of thiosulphate to either elemental sulphur or dissolved by reactions (2) and (3) is thermodynamically more favorable in acid solutions. 2.9 Effect of Load Ratio: In fatigue studies, the load ratio (K . /K ) has been 3 mm max found to be one of the most important mechanical variables which affect the fatigue crack growth behavior of metals. As the load ratio is raised, the AK., (K - K . ) is markedly decreased and th max min J the crack propagation rate (da/dN) is increasedt42]. In this study, no attempt was made to - investigate the extent of influence of the load ratios. However, the marked effect of load ratio is believed to be caused by the extent of crack closure at the crack t i p . The lower the load ratio, the greater the crack closure effect will be present at crack t i p . 2.10 Crack Closure Effect: Crack closure, as i t s name implies, is the premature closure 24 of the c r a c k t i p ( i . e . c o n t a c t of o p p o s i n g c r a c k s u r f a c e s ) b e f o r e the a p p l i e d c y c l i c l o a d i s c o m p l e t e l y u n l o a d e d t o t h e minimum v a l u e d u r i n g each c y c l e . T h i s phenomenon w i l l t h e n e f f e c t i v e l y reduce t h e c y c l i c s t r e s s i n t e n s i t y r a n g e , AK, by m a i n t a i n i n g t h e a c t u a l minimum K e x p e r i e n c e d by t h e c r a c k t i p above t h e a p p l i e d minimum K as shown by t h e f o l l o w i n g e q u a t i o n s . ^ a p p = Kmax " K m i n ( 7 ) * K e f f = Kmax " K c l ' ( 8 ) where K c^ i s the s t r e s s i n t e n s i t y a t c r a c k c l o s u r e . S i n c e K ^ i s h i g h e r than K m ^ n i n t h e case of premature c r a c k c l o s u r e , AK t c i s l e s s than AK . T h e r e f o r e , t h e f a t i g u e e f r a P P c r a c k growth r a t e w i l l t e n d t o d e c r e a s e , even when t h e a p p a r e n t s t r e s s i n t e n s i t y range .becomes l a r g e , i f t h e r e i s a pronounced c r a c k c l o s u r e e f f e c t p r e s e n t . Crack c l o s u r e can be a t t r i b u t e d t o t h r e e main c a u s e s . (A) P l a s t i c i t y - i n d u c e d c r a c k c l o s u r e : The phenomenon of c r a c k c l o s u r e was f i r s t o b s e r v e d by E l b e r f o r c r a c k growth a t h i g h s t r e s s i n t e n s i t y [ 4 3 ] . I t was thought t o be caused by e x t e n s i v e p l a s t i c d e f o r m a t i o n w h i c h t a k e s p l a c e by s t r e s s r e l a x a t i o n i n t h e wake of a g r o w i n g c r a c k . The deformed c r a c k s u r f a c e s w i l l t h e n impinge upon each o t h e r b e f o r e c o m p l e t e u n l o a d i n g o c c u r s , r e s u l t i n g i n t h e premature c r a c k c l o s u r e . A nother s c h o o l o f thought b e l i e v e s t h e p l a s t i c i t y - i n d u c e d c r a c k c l o s u r e i s t h e r e s u l t of f o r m a t i o n of a c o m p r e s s i v e p l a s t i c 25 zone ahead of c r a c k t i p d u r i n g p a r t of t h e f a t i g u e c y c l e [ 4 4 ] . D u r i n g u p l o a d i n g p a r t of the f a t i g u e c y c l e , a t e n s i l e p l a s t i c zone w i l l be formed due t o t h e h i g h s t r e s s r a i s i n g n o t c h e f f e c t of the c r a c k t i p , whereas the s u r r o u n d i n g r e g i o n i s j u s t e l a s t i c a l l y s t r e t c h e d . Upon u n l o a d i n g , t h e e l a s t i c a l l y deformed a r e a w i l l t r y t o r e t u r n t o i t s o r i g i n a l u n s t r e s s e d p o s i t i o n and t h e r e f o r e impose a c o m p r e s s i v e s t r e s s on t h e t e n s i o n - i n d u c e d p l a s t i c r e g i o n . As a r e s u l t , a r e v e r s e d c o m p r e s s i v e p l a s t i c zone i s formed d u r i n g p a r t of the c y c l e . S i n c e t h e c r a c k w i l l not p r o p a g a t e through a c o m p r e s s i v e p l a s t i c zone, the c r a c k p r o p a g a t i o n r a t e i s reduced. (B) S u r f a c e r o u g h n e s s - i n d u c e d c l o s u r e [ 4 5 ] : In t h e n e a r - t h r e s h o l d s t r e s s i n t e n s i t y r a n g e , f a t i g u e c r a c k s g e n e r a l l y propagate d i s c r e t e d i s t a n c e s a l o n g one s l i p p l a n e and t h e n a l o n g a d i f f e r e n t l y o r i e n t e d s l i p p l a n e , r e s u l t i n g i n a s e r r a t e d , c r y s t a l l o g r a p h i c c r a c k p a t h . Because the c r a c k advances l a r g e l y by shear a c t i o n , l a t e r a l d i s p l a c e m e n t of one c r a c k s u r f a c e w i t h r e s p e c t t o the o p p o s i n g c r a c k s u r f a c e i s v e r y l i k e l y t o o c c u r t h r o u g h the shear d i s p l a c e m e n t mechanism. The p r o t u b e r a n c e of the d i s p l a c e d s u r f a c e s w i l l c o n t a c t upon one a n o t h e r b e f o r e the c y c l i c s t r e s s i s c o m p l e t e l y u n l o a d e d . The e f f e c t i v e c r a c k p r o p a g a t i o n d r i v i n g f o r c e AK i s t h e r e f o r e r e d u c e d . At h i g h s t r e s s i n t e n s i t y r a n g e s , where the c y c l i c p l a s t i c zone exceeds the g r a i n s i z e , m u l t i p l e s l i p systems a r e 26 a c t i v a t e d t o m a i n t a i n g r a i n boundary c o n t i n u i t y . W i t h a . g r e a t number of a c t i v a t e d s l i p systems, shear d i s p l a c e m e n t s a t t h e c r a c k t i p ( t y p e I I l o a d i n g ) d e c r e a s e and opening mode d i s p l a c e m e n t s ( t y p e I l o a d i n g ) p r e d o m i n a t e . T h i s reduces b o t h t h e f a c e t e d n a t u r e of the c r a c k s u r f a c e and roughness i n d u c e d c r a c k c l o s u r e e f f e c t . (C) C o r r o s i o n - p r o d u c t - i n d u c e d c l o s u r e : Crack c l o s u r e e f f e c t s can a l s o be enhanced by the c o r r o s i o n p r o d u c t wedging mechanism[46]. The f o r m a t i o n of a t h i c k o x i d e f i l m or p r e c i p i t a t i o n of s a l t p r o d u c t s p r e v e n t s c r a c k s from c o m p l e t e l y c l o s i n g . Thus, the e f f e c t i v e s t r e s s i n t e n s i t y range i s r e d u c e d . Other c r a c k r e t a r d a t i o n phenomena c o u l d be the d i r e c t r e s u l t s of t h e c r a c k b r a n c h i n g e f f e c t [ 4 7 ] o r c r a c k t o r t u o s i t y e f f e c t [ 4 8 ] which a l s o lower the e f f e c t i v e c y c l i c s t r e s s i n t e n s i t y range by d e c r e a s i n g the r e s o l v e d t e n s i l e s t r e s s component normal t o t h e c r a c k p l a n e . There a r e many o t h e r parameters which can s t r o n g l y i n f l u e n c e t h e c o r r o s i o n f a t i g u e c r a c k growth b e h a v i o r of m e t a l s [ 4 9 ] . However, t h e i r e f f e c t s a r e not i n v e s t i g a t e d i n t h i s p r o j e c t and t h e r e f o r e w i l l not be d i s c u s s e d i n t h i s r e p o r t . 2 7 3 EXPERIMENTAL PROCEDURE 3.1 M a t e r i a l S p e c i f i c a t i o n t F a t i g u e s t u d i e s were c o n d u c t e d on a co m m e r c i a l grade AIS I 316L s t a i n l e s s s t e e l (SS 316L), s u p p l i e d i n the form of hot r o l l e d 25.4 mm(l") t h i c k p l a t e by Uddeholm L t d . i n Richmond, B.C. The n o m i n a l c h e m i c a l c o m p o s i t i o n of the s t a i n l e s s s t e e l was as f o l l o w s : Element C Cr N i Mo S i Mn P S Wt. % <0.026 ~17 ~12 2.5 0.99 1 .99 0.056 <0.01 The m e t a l l o g r a p h y o f b o t h t h e a s - r e c e i v e d s t e e l and the s u b s e q u e n t l y a n n e a l e d s t e e l s r e v e a l e d a p e r i o d i c p a r a l l e l s t r e a k s t r u c t u r e , which c o n s i s t e d of bands of two d i f f e r e n t g r a i n s i z e s . The s m a l l g r a i n r e g i o n c o n t a i n e d a g r a i n s i z e of a p p r o x i m a t e l y 0.05 mm i n d i a m e t e r , whereas t h e l a r g e g r a i n r e g i o n e x h i b i t e d an aver a g e d i a m e t e r of 0.08 mm. The cause of t h i s g r a i n s i z e p e r i o d i c i t y might be due t o s o l u t e s e g r e g a t i o n w h i c h e i t h e r i n h i b i t e d or a c c e l e r a t e d t h e g r a i n r e c r y s t a l l i z a t i o n and growth p r o c e s s e s . The m i c r o s t r u c t u r e was p r e d o m i n a n t l y e q u i a x e d and d i s p l a y e d numerous a n n e a l i n g t w i n s , which i s a common c h a r a c t e r i s t i c of h o t - r o l l e d o r a n n e a l e d a u s t e n i t i c s t a i n l e s s 28 s t e e l s . No o b s e r v a b l e i n c l u s i o n p a r t i c l e s were v i s i b l e under the o p t i c a l m i c r o s c o p e . T h i s was e x p e c t e d because of the low c a r b o n c o n t e n t . However, subsequent h i g h - m a g n i f i c a t i o n s c a n n i n g e l e c t r o n m i c r o s c o p i c a l e x a m i n a t i o n of f r a c t u r e s u r f a c e s of f a t i g u e c r a c k e d samples showed t h a t a v e r y s m a l l number of t i n y i n c l u s i o n s were s t i l l p r e s e n t . 3 . 2 M e t a l l o g r a p h y : The e t c h i n g s o l u t i o n used t o show the g r a i n s t r u c t u r e of A I S I 316L was V i l e l l a ' s r e a g e n t ( 5 ml HC1, 1 ml p i c r i c a c i d , and 100 ml m e t h y l a l c o h o l ) . The specimen, f i n e l y p o l i s h e d down t o 1 m i c r o n w i t h diamond p a s t e , was immersed i n the s o l u t i o n f o r up t o 10 m i n u t e s t o r e v e a l i t s m i c r o s t r u c t u r e . 3.3 Specimen Geometry: S i n c e i t was n e c e s s a r y t o s t u d y the f a t i g u e c r a c k growth b e h a v i o r of AISI 316L s t a i n l e s s s t e e l i n aqueous e n v i r o n m e n t s under v a r i o u s p o l a r i z a t i o n p o t e n t i a l c o n d i t i o n s , a s i n g l e edge notch(SEN) specimen was c h o s e n . I t s l o n g l e n g t h p e r m i t t e d easy a t t a c h m e n t of an environment c e l l and the n e c e s s a r y e l e c t r o d e s around t h e specimen m i d s e c t i o n , t h e r e b y making c o n t r o l of a p p l i e d p o t e n t i a l p o s s i b l e . In a d d i t i o n , t h e r e was an abundance of 29 f r a c t u r e mechanics i n f o r m a t i o n a v a i l a b l e c o n c e r n i n g t h e c r a c k t i p s t r e s s i n t e n s i t y of the SEN specimen. T h e r e f o r e , the e f f e c t of s t r e s s i n t e n s i t y on f a t i g u e c r a c k p r o p a g a t i o n r a t e c o u l d be d e t e r m i n e d r e l a t i v e l y e a s i l y . 3.4 Specimen P r e p a r a t i o n : Specimen b i a n k s w i t h d i m e n s i o n s a p p r o x i m a t e l y 203.2 mmx50.8 mmx25.4 mm(8"x2"x1") were c u t w i t h the l e n g t h normal t o the r o l l i n g d i r e c t i o n . S u b s e q u e n t l y , two 19.05 mm(3/4") dia m e t e r g r i p h o l e s were d r i l l e d a t b o t h ends. A c e n t r a l edge s l o t was made by i n i t i a l l y d r i l l i n g a 6.75 mm(17/64") dia m e t e r h o l e t h r o u g h the m i d s e c t i o n and s u b s e q u e n t l y removing the r e m a i n i n g l i g a m e n t w i t h a bench saw. A c e n t e r c r a c k s t a r t i n g n o t c h was f i n a l l y c u t by a m i l l i n g c u t t e r so t h a t the ma c r o s c o p i c c r a c k p l a n e was p a r a l l e l t o t h e o r i g i n a l r o l l i n g d i r e c t i o n ( F i g u r e 4) and t h e c r a c k f r o n t normal t o t h e r o l l i n g d i r e c t i o n . The specimen was then mounted i n a — s e t — o f u n i v e r s a l g r i p s , thus m i n i m i z i n g m i s a l i g n m e n t , and c y c l i c a l l y p r e c r a c k e d i n an I n s t r o n f a t i g u e machine a t a mean l o a d of 39 KN w i t h an a l t e r n a t i n g c y c l i c l o a d 25 K N ( a p p r o x i m a t e l y c o r r e s p o n d i n g t o a c r a c k t i p s t r e s s i n t e n s i t y o f 25.8±16.5 MPa/m). A f t e r a l l o w i n g the c r a c k t o propagate ~2 mm by f a t i g u e ahead of t h e n o t c h , the f a t i g u e p r e c r a c k o p e r a t i o n was t e r m i n a t e d . At t h i s s t a g e , t h e o v e r a l l c r a c k l e n g t h was ~25.4 mm. 30 SINGLE EDGE NOTCHED SPECIMEN L: LENCTH= 203.2mm W:WIDTH - 50.8 mm B: TH1CKNESS= 25.4 mm H'.CRACK LENGTH AX = P >0-°~'V • (an)V2*f(°-) P f(a ) _ Oan(Q)/Q)'/2*(0.752+2.02a/W+0.37(1.0-sin(Q))i) cos(Q) 0 = (an/2W) F i g u r e 4. The S i n g l e Edge N o t c h specimen geometry and i t s c r a c k t i p c y c l i c s t r e s s i n t e n s i t y c a l i b r a t i o n f o r m u l a . 31 A half hour 400°C oxide-indueing heat treatment was applied to the precracked specimen in order to prevent the rewelding of cracked surfaces in the following annealing treatment. After the oxide forming operation, the specimen was water quenched, dried, sealed inside a stainless steel bag to prevent excessive oxidation, and returned to the furnace to be annealed at 1050°C for one hour. This annealing treatment was intended to remove the work hardened zone produced around the crack tip during the precracking operation. Upon the completion of annealing treatment, the specimen was again water quenched down to room temperature to avoid the formation of any b r i t t l e a phase. The faces of the specimen were carefully polished down to 600 grit so that accurate measurements of crack length could be taken. Finally the specimen was thoroughly cleaned with ethanol and wrapped in polytetrafluroethylene(PTFE) tape with only the region ahead of crack tip exposed to reduce applied current requirements during potentiostatic fatigue testing. A cylindrical fatigue specimen of 5.08 mm(0.2") gauge diameter and 17.8 mm(0.7") gauge length was also used to compare its fractographic morphology to those of SEN specimens. The cylindrical specimen underwent the same heat treatments and i t s gauge section was polished with fine emery paper on a lathe before subjected to the fatigue testing. The cylindrical specimen was also used to determined the mechanical properties of the heat treated steel. These are shown below: 32 YS Q 2 % UTS Elongation 282.0 MPa 608.5 MPa 54% 3.5 Constant Load Magnitude Fatigue Crack Growth Testing: The majority of the SEN specimens were tested on the Instron high frequency electromechanical resonant fatigue machine (Model 1603). The cyclic frequency of the sinusoidal wave form varied, depending on the stiffness of the tested specimen and the applied cyclic load. However, the fatigue frequencies encountered in the experiments generally hovered around 125 Hz with the selected geometry and material. The environment c e l l was constructed with two half c e l l s , made of transparent acrylic material(Plexiglas). It was fastened onto the specimen midsection by a set of stainless steel screws and nuts, and sealed with silicone rubber along the specimen/cell interfaces(Figure 5). The corrosion potentials were varied and maintained by an ECO potentiostat (Model 549). Potentials were monitored with respect to an external saturated calomel electrode(SCE) via a KC1 salt bridge and PTFE Luggin capillary placed ~1 mm from the crack t i p . Located at the backface of the fatigue specimen was a platinum mesh serving as the counter electrode. Potentials 33 A. WORKING ELECTRODE. B. COUNTER ELECTRODE. C. LUGGIN CAPILLARY. D TRAVELLING MICROSCOPE. E- ACRYLIC CELL a AQUEOUS ENVIRONMENT. (a) (b) (c) F i g u r e 5. (a) S c h e m a t i c o f the f a t i g u e t e s t i n g ; (b) t e s t i n g i n p r o g r e s s e n v i r o n m e n t a l c e l l a rrangement f o r and ( c ) P h o t o g r a p h s o f f a t i g u e 34 varying from the cathodic to anodic range were applied in different corrosion fatigue tests and the corresponding applied currents were continuously recorded on a Kipp and Zonen recorder (Model BD41) from a 1 ohm resistor output. The corrosion environments employed in the tests were of three types: 1 M NaCl, 1 M NaCl with 0.01M Na 2S 20 3 addition, and dessicated a i r . The aqueous solutions were simply poured into the c e l l to a level well above the exposed region and d i s t i l l e d water was regularly added to compensate for any evaporation losses. To achieve a dry air condition, the environment chamber was f i l l e d with s i l i c a gel and i t s open top was tightly closed with a plastic sheet. A l l tests were conducted at room temperature(~22°C). A custom built internally illuminated optical travelling microscope with a micrometer stage of 10 Aim resolution was used to monitor the progress of crack advance. The crack length measurements were always taken with respect to a fiducial reference line scribed on ^ he..specimen surface. Only single side crack length increases were recorded . Nevertheless, post-test examination of fracture surfaces indicated that crack length increments on both sides were generally equivalent because the crack front profiles remained relatively unchanged throughout the tests. A l l fatigue crack growth tests were i n i t i a l l y started with a low constant cyclic load at a constant R ratio 35 (min. load/max. load) of 0.22. If no detectable crack increment was observed within 2 to 3 million applied cycles, the cyclic load was then further increased. The process was repeated until crack growth was detected. Once the crack growth had initiated, the crack length and number of applied cycles were recorded at regular intervals. One SEN specimen was also tested on a MTS closed-loop electrohydraulic testing machine (Model 810) at a cyclic frequency of 15 Hz. Due to the lengthy time required in a low frequency test, only half a million cycles were allowed to elapse without detectable crack growth prior to the next load increase. 3.6 Cycle-To-Failure Testing: One cylindrical fatigue specimen was cycled to failure under a cyclic load of 312 MPa and a R-ratio of 0 while immersed in aerated 1 M NaCl. The frequency applied was 30 Hz and the fatigue -machine employed was the MTS electrohydraulic system. 3.7 Potentiodynamic Testing: The electrochemical behavior of a metal immersed in aqueous solution can be most conveniently illustrated by a polarization curve. By supplying current to the metal through an external 36 c i r c u i t , the electrochemical potential of a metal may be altered. The resulting relationship between current and potential characterizes the behavior of the metal under a range of electrochemical conditions. A cylindrical electrode, 10 mm length x 9.3 mm diameter, was sectioned from a stainless steel rod. A Ni wire was spot-welded to one end of the specimen for el e c t r i c a l connection purposes and enclosed in a glass tube. The specimen was mounted in self-setting acrylic resin to leave one diametrical face exposed and to seal the glass tube to the specimen. The resin surrounded sample was polished down to 600 grit and thoroughly cleaned with alcohol. It was then lowered into a multineck flask. The aqueous solution contained in the flask was purged with nitrogen gas at least one day before the start of each test and purging continued during a test. A platinum f o i l counter electrode and Luggin capillary were also fitted inside the flask(Figure 6). The Luggin capillary was connected to an external reference SCE via a saturated KCl salt bridge. The applied potential and corresponding applied currents were automatically controlled and monitored by an EG & G microprocessor-based corrosion measurement system (Model 350). Electrochemical behavior of AISI 316L in several aqueous solutions was investigated at 25°C. Each test simulated in a simplistic fashion a condition which the steel might experience 37 A, WORKING ELECTRODE. E . LUGGIN CAPILLARY. C. COUNTER ELECTRODE. D . NITROGEN PURGE. E . MULTINECK FLASK. F. RUBBER STOPPER. Schematic of the p o l a r i z a t i o n test c e l l 38 I in r e a l i t y . The test solutions were: 1. 1 M NaCl. 2. 1 M NaCl and 0.01 M Na 2S 20 3. 3. 1 M F e C l 2 adjusted to a pH value of 1 with saturated C r C l ^ s o l u t i o n . (This solution simulated conditions which may occur inside cracks and crevices.) 4. same as 3, but with 0.01 M N a ^ ^ ^ addition. After the test, each sample was examined under the o p t i c a l microscope for sign of gross crevice corrosion along the metal/resin interface, as t h i s would y i e l d misleading p o l a r i z a t i o n data. 3.8 O p t i c a l and Electron Micorscopy: The microstructure of the test material was mainly examined and photographed under the Zeiss Ultraphot l i g h t microscope. The specimens were usually f i n e l y polished and etched before examination to a s s i s t clear observation. The fratographic morphology was investigated on the ETEC Autoscan scanning electron microscope(SEM). Prior to examination, the specimens were a l l given a thorough ultrasonic cleaning in an i n h i b i t e d acid ( 3 ml concentrated HC1, 4 ml of a 35% aqueous solution of 2 butyne-1,4 d i o l , and 50 ml of H 20) to remove surface films, which could cause considerable reduction of 39 microscopic resolution!50]. A l l fracture surfaces were examined at 0 degree t i l t angle and 25 KV accelerating voltage. Stereographic photographs were taken at approximately 10 degree t i l t angle apart(i.e. ±5°tilt). 3.9 Etch pitting: An etch pitting technique was employed to determine the crystallographic orientations of fracture planes and directions. By electroetching in 1 N sulphuric acid and 1 mg/1 of ammonium thiocyanate at -0.15 V o l t ( s C E ) a n d 4 0 ° C ' well-delineated crystallographic pits were observed to formed on f l a t surfaces[51]. A highly polished specimen, which was subjected to hardness indentation to produce s l i p steps, was used to determine the orientation of the facets of the etch pits formed with this technique. Indentations were produced on a Vicker hardness machine. From Figure 7, i t was clear that the edges of the etch pits were parallel to the {111} s l i p traces induced by the indentation. Therefore, i t was demonstrated that the faceted surfaces of the pits were {111} planes. 40 Figure 7. Scanning electron micrograph showing etch pits formed on the indented polished AISI 316L stainless steel. The edges of the etch pits were parall e l to the s l i p traces induced by the indentation marks, indicating the facet faces are of { 111 } type. 41 4. EXPERIMENTAL RESULTS: 4.1 Fatigue Crack Growth Curves: The fatigue crack growth curves of AISI 316L in various environmental conditions were obtained by p l o t t i n g the experimental data of the fatigue crack length vs. the applied loading cycles as discrete points on l i n e a r graphs. The experimental crack length was f i r s t corrected for the lagging distance of the surface measurement behind the leading point on the curved crack front. The lagging distance was assessed a f t e r breaking open the specimen afte r t e s t i n g and was added to every experimentally measured crack length in the t e s t . This correct i o n was done in order to co r r e l a t e the crack propagation rate with the maximum c y c l i c stress i n t e n s i t y experienced at any point on the crack front. A smooth l i n e was then f i t t e d through a l l corrected data points using an incremental orthogonal polynomial method. This method prescribes s e l e c t i n g the f i r s t seven points from the data set and least square f i t t i n g these points by a fourth degree polynomial as the following: 4 a = L C n-N n (9) n=0 where a i s the fatigue crack length, C n i s a scal i n g c o e f f i c i e n t , and N i s the number of the applied c y c l e s . 42 New a r t i f i c i a l points were then generated with the selected polynomial for the intermediate range between point 4 and 5 to maintain the smoothness of the line. The f i r s t point was then removed and the next point added to the group for the same calculation. The pattern was repeated until the last datum point was processed. Along with the newly generated points, a continuously smooth line could be drawn out for each environment(Figure 8-16). The experimentally observed data points were also superimposed on the a r t i f i c i a l curve to show i t s degree of f i t t i n g . The test condition for each experiment is described in Table I. Some of the curves were plotted from the start of the fatigue test and some were plotted after millions of fatigue cycles had elapsed. The difference in crack growth i n i t i a t i n g cycles demonstrates the varying d i f f i c u l t y of continuously propagating fatigue cracks during the i n i t i a l stages. 43 TEST No. APP. LOAD FREQUENCY ENVIRONMENT I 17.5±11.2 KN 123 Hz IM NaCl, free corrosion II 10.6±6.8 KN 115 Hz IM NaCl, +600mV(SCE) III 23.3±14.9 KN 127 Hz IM NaCl, OmV(SCE) IV 13.6±8.7 KN 118 HZ IM NaCl, -540mV(SCE) 15.0±9.6 KN 120 Hz • • 16.4±10.4 KN 122 HZ • • 17.7±11.4 KN 123 HZ • 18.4±11.7 KN 124 HZ • 19.1±12.2 KN 125 HZ • V 17.2±11.0 KN 15 HZ IM NaCl, -540mV(SCE) VI 18.7±12.0 KN 122 HZ IM NaCl & 0.01M Na0S-0-,, -300mV(SCE) VII 13.3±8.5 KN 116 HZ IM NaCl & 0.01M Na 2S 20 3, • 14.6±9.3 KN 118 HZ free corrosion • 15.8±10.1 KN 119 HZ * • 17.2±11.0 KN 120 HZ * • 18.6±11.9 KN 121 HZ • VIII 14.7±9.4 KN 118 HZ IM NaCl & 0.01 Na,S00,, • 16.0±10.2 KN 120 HZ -450n\V(SCE) 17.2±11.0 KN 122 HZ • IX 16.3±10.4 KN 124 HZ Dessicated a i r . Table I. Test Conditions for Fatigue Crack Propagation Studies. 44 E E. 42 40 H 38H 36-1 O z Ld _l 34 (J < O O 26-f AISI 316L SS in 1M NaO at free corrosion potential Cyclic frequency, 123 HZ R ratio, 0.220 2 4 6 NUMBER OF THE APPLIED CYCLES (Millions) 8 Figure 8. Fatigue crack growth of AISI 316L in IM NaCl at free corrosion potential(-100 to -180 mV ( S C E )) and 123 Hz. 36 34-E E. O 32 z o < O 30 LU ID O 5 28-26 -f i r-rV AISI 316L SS in 1M NaCl at +600 mV(SCE) Cyclic frequency, 115 HZ R ratio, 0518 • A A A 0.5 1 1.5 2 NUMBER OF THE APPLIED CYCLES (Millions) Figure 9. Fatigue crack growth of AISI 31 and 115 Hz. 6L in IM NaCl at +600 mV 2.5 (SCE) 40 4^ —1 E E < a: U Ld O 38H 36H 34H 32 H 30H 28-\ 26 H AISI 316L SS in 1M NaCl at +0 mV(SCE) Cyclic frequency, 127 HZ R ratio, 0.220 24 H 99 Figure 10. 100 101 102 103 104 105 NUMBER OF THE APPLIED CYCLES (Millions) 106 Fatigue crack growth of AISI 316L in IM NaCl at +0 mV and 127 Hz. (SCE) 44 CO 42 A 40H 38H X o I . I 36 34H < o UJ 1 3 32 O 30 H AISI 316L SS in 1M NaCl at -540 mV(SCE) Cyclic frequency, 125 HZ R ratio, 0.223 Wvwg F i g u r e 11. I I I I I . 20 30 40 50 60 70 80 NUMBER OF THE APPLIED CYCLES (Millions) Fatigue crack growth of AISI 316L in IM NaCl at -540 mV, and 125 Hz. (SCE) 34 24 J | | | , T 0 1 2 3 4 5 NUMBER OF THE APPLIED CYCLES (Millions) Figure 12. Fatigue crack growth of AISI 316L in IM NaCl at -540 mV,^, and 15 Hz. (SCE) 50 o 45 H x O 40 o < O 35 Ld O 30-25-J AISI 316L SS in 1M NaCl & 0.01 M Na.S.0. Solution at -300 rrMSCE) Cyclic frequency, 122 HZ R ratio, 0218 13 Figure 13. -A «*r 14 15 16 17 18 NUMBER OF THE APPLIED CYCLES (Millions) 19 Fatigue crack growth of AISI 316L in IM NaCl & 0.01M Na 2S 20 3 at -300 mV (SCE) and 122 Hz. AISI 316L SS in IM NaCl A & 0.01 M Na,S,0, Solution under free corrosion condition Cyclic frequency, 121 MZ A R ratio, 0220 A A A I I I I I 0 10 20 30 40 50 60 70 NUMBER OF THE APPLIED CYCLES (Millions) e 14. Fatigue crack growth of AISI 316L i n IM NaCl & 0.01M Na_S~0, at free corrosion potential(-280 to -340 mV,___,.) and 121 Hz. [ S ^ E ) 40 to £ 38 3 6 -3 4 -3 2 -3 0 -O UJ _J < o r o Ld O L J 28-1 2 6 -24 + Figure 15 AISI 316L SS in 1M NaCl & 0.01 M Na.S.O. Solution at-450 mVtSCE) Cyclic frequency, 122 HZ R ratio, 0220 10 20 30 40 50 60 NUMBER OF THE APPLIED CYCLES (Millions) Fatigue crack growth of AISI 316L in at -450 m v ( S C E ) and 122 Hz. IM NaCl & 0.01M 70 Na 2S 20 3 e s 4.2 Fatigue propagation Rates; In the previous section, the fatigue crack growth curves were considered as a continuous series of differentiable functions of applied cycles in the form of equation (9). By differentiating the crack length with respect to the applied cycles, the cyclic crack propagation rate can be expressed in the following form: 4 da/dN = I n-C -N n _ 1 (10) n n=1 The fatigue crack growth rate is customarily plotted as a function of the applied alternating crack tip stress intensity AK. The alternating stress intensity of single edge notched specimen can be calculated from the load, P, by the following equat ions[52], AK = P m a x-(1 .0-R)/(W-B).(a7r) l / 2.f (a/W) (11) where R = P • /P m . (12) min max f(a/W) = (tan(Q)/Q)1/2-(0.752+2.02a/W+0.37-( 1.O-sin(Q)) 3)/cos(Q), (13) and Q = (TTR/2) . The log plots of the fatigue crack propagation rate vs. the alternating stress intensity curves , commonly known as a v- AK plot, are presented in Figures 17-25, for various testing conditions. 54 10 0) 3. CO Ed H OS o o OS 0.1 O.Old 0.001 0.0001 10 100 AISI 316L SS in 1M NaQ under free corrosion condition Cycle frequency, 123 HZ R ratio, 0219 b o . i 20 50 1:0.01 0.001 F i g u r e 17, 30 40 AK (MPa.m l / 8 ) Fatigue crack propagation r a t e (da/dN v s . c y c l i c s t r e s s i n t e n s i t y AK) of AISI 316L in 1M NaCl at f r e e c o r r o s i o n p o t e n t i a K - 1 0 0 to - 1 8 0 n » v ( S C E ) ) a n d 1 2 3 H z * 0.0001 100 CTi o o S 3 cd w O S o o SS o o •< O H 1 o . H 0.0H 0.00H 0.0001 0.0000H 1 10 I 100 J 1 4 AISI 316L SS in 1M NaCl at +600 mV(SCE) Cyclic frequency, 115 HZ R ratio, 02183 t o . i to.01 bo.001 5 10 20 AK (MPa.m l / 8 ) to.0001 30 J-0.00001 100 Figure 18 Fat i g u e crack propagation r a t e (da/dN vs. c y c l i c s t r e s s i n t e n s i t y AK ) of AISI 316L i n IM NaCl at +600 mV, and 115 Hz. (SCE) F i g u r e 19. F a t i g u e c r a c k p r o p a g a t i o n r a t e (da/dN v s . c y c l i c s t r e s s i n t e n s i t y AK) of A I S I 316L i n IM NaCl a t +0 mV,„ % and 127 Hz. <SCE) 10 100 Figure 20. Fatigue crack propagation rate (da/dN vs. c y c l i c stress intensity AK) of AISI 316L in IM NaCl at -540 mV, n„ n, and 125 Hz. ( S C E ' 10 0.1-100 —f 0.1 0.0H O . O O H AISI 316L SS in 1M NaCi at -540 mV(SCE) Cyclic frequency, 15 HZ R ratio, 0.220 o.oi o.ooi o.oooi-10 20 30 40 AK (MPa.m1/8) 50 •f o.oooi 100 Figure 21 Fatigue crack propagation rate (da/dN vs. cyclic stress intensity AK) of AISI 316L in IM NaCl at -540 mV, and 15 Hz. (SCE) O a g :d CO •d OS o a O H o 05 O H tt O H 10 1+ 0.1 O.Olt 0.001-0.0001-0.00001-10 100 • i 1000 ' I ' l l AISI 316L SS in 1M NaCl & 0.01 M NaLS O Solution at -300mV(SCE) Cyclic frequency, 122 HZ R ratio, 0.2182 ro.i rO.Ol 0.001 0.0001 20 30 40 50 100 Figure 22. AK (MPa*ml/a) I 0.00001 500 1000 Fatigue crack propagation rate (da/dN vs. cyclic stress intensity AK) of AISI 316L in IM NaCl & 0.01M Na„S„0. at -300 mV ( S C Ej and 122 Hz. l2°2^3 CTi CO o >. o ed w < OS o Ou o OS Ou tt o OS 10 1-0.1-0.01-0.001 0.0001-0.0000H 100 — t l AISI 316L SS in IM NaCl k 0.01 M Na,St0, Solution under free corrosion condition Cyclic frequency, 121 HZ R ratio, 0.220 rO.l rO.01 rO.001 M A X I M U M A P P L I E D L O A D A 21.8 K N X • 23.0 K N 28.fi K N B 28 .2 K N rO.0001 B 30.6 K N a 10 Figure 23 20 30 40 50 AK (MPa.ml/a) i—^-O.OOOOl 100 Fatigue crack propagation rate (da/dN vs. cyclic stress intensity AK) of AISI 316L in IM NaCl & 0.01M Na^O. at free corrosion potential(-280 '-- "*"n ~" and 121 Hz. to -340 n,V (" }? Figure 24. Fatigue crack propagation rate (da/dN vs. cyclic stress intensity AK) of AISI 316L in IM NaCl & 0.01M Na_So0o at -450 mV/e,,_. and 122 Hz. 2 2 3 Figure 25. Fatigue crack propagation rate (da/dN vs. cyclic stress intensity AK) of AISI 316L in dessicated air and 124 Hz. 4.3 Near-Threshold Crack Propagation; The power law dependence of crack growth rate on AK in Stage II, as described by the Paris law in equation(1), is believed to be an indication of the direct influence of either crack t i p plastic zone size or crack tip opening displacement when the exponent li e s between 2 to 4[53]. An attempt was made to obtain a similar Paris relationship between crack propagation rate and cyclic stress intensity in the near-threshold region. Using the linear regression method, the results are presented in Table II. The value of the exponential factor of the stage II crack growth rate relationship l i e s generally in the range of 2 to 4 in the absence of environmental effects for most materials. In the present study, i t ranged from 1.10 to 2.66. A large scatter was observed. It was due to the inability of the resonant fatigue machine to maintain the specified load for rapidly growing cracks, which led to a large change of specimen compliance at the latter stage of fatigue testing. 64 * AKth(MPa*/m) ** *** TEST No. m1 I 16.1 1 1 .47 1 .56 II 9.2 12.63 1 .42 III 17.0 23.76 2.66 IV 19.3 10.03 1.10 V 12.8 7.60 VI 17.1 15.21 1 .02 VII 22.3 9.97 1 .77 VIII 13.2 15.90 2.06 IX 15.9 12.81 2.48 * See Table I on page 44. * * ntj is the near-threshold fatigue crack growth exponential factor. m1 « Qog(da/dN)-logC)/log AK *** n»2 is the stage II fatigue crack growth exponential factor. m2 = (log(da/dN)-logC)/log AK Table II. Fatigue Crack Growth Properties of AISI 316L under Various Testing Conditions. 65 4.4 Fractography: Fratographic analysis of a l l fractured surfaces from various test conditions revealed fatigue failures occurred exclusively by transgranular cracking. On the transgranular crack surfaces, three distinct areas could be discerned. I n i t i a l l y , the fracture surfaces exhibited cleavage or quasi-cleavage characteristics. These consisted of crystallographic facets, together with parallel running river lines, as typically observed, in b r i t t l e cleavage failure of metals(Figure 26(a) and (c), 27(a), 28(a), 29(a), 30(a), (b), (c), and (d), 31(a), and 32(a)). The direction of the river lines generally pointed in the local crack propagation direction. At high magnification, the river lines appeared to be serrated in shape with faint intersecting s l i p traces oriented parallel to one of the serrated edges. From etch pitting studies of the cleavage-like surfaces, i t was determined that cracks could proceed along {111} , {110} , and {100} low index planes(Figure 33).Possibly cracks may also propagate on some other undetermined higher index 2 planes. From the SEM photo-analysis of 0.5 mm of the fractured area in the i n i t i a l stages, at least 15% of the crack surface arose from propagation along {111} planes. The remaining area was divided between {110} , {100} , and higher index planes. No clear distinction could be made between the fractographic features of differently oriented crack surfaces. 66 The second region was characterized by a rough and feathery types of surface(Figure 26(d) and (e), 27(e) and ( f ) , 28(c) and (d), 29(b), 30(e) and ( f ) , 31(c), (d), and (e), and 32(b), (c),and (d)). Slip traces and crack markings could also be faintly seen, though less well-defined. Among the rough regions, f l a t , almost perfectly reflective facets occasionally can be observed.(Figure 27(f), 31(e), and 32(e)) It was more d i f f i c u l t to form etch pits on the feathery regions, mainly due to the surface roughness. However, from the analysis of pits developed on the feathery surfaces, {110} and {100} planes were the predominant fracture planes, whereas reflective facets were of the {111} type. The second region was believed to be a transition region between the i n i t i a l ( f i r s t ) and subsequent(third) type of fracture surface region. The third type of fracture surface was the easily recognized striated region(Figure 26(f), 27(g) and (h), 28(g), 29(d), (e), and (f), 30(h), 31(f) and (g), and 32(f) and (g)). It consisted of fine continuous parallel lines, running perpendicular to the crack propagation direction(i.e. parallel to the crack front). Large surface plastic deformation and secondary cracks at the trough of the striation lines were easily observed(Figure 31(h)). By comparing the photographs of the same crack area on two opposing surfaces, i t was found that the striation was peak-to-peak and valley-to-valley type(Figure 34). Very few etch pits were found on the striated surfaces. Of those found, {110} 67 and {100} planes were the fracture planes. As far as the ef f e c t s of the p o t e n t i a l and environments were concerned, they did not exert s i g n i f i c a n t influence on the fractographic features of the cracked specimens. However, some fracture surfaces exhibited signs of anodic dissolution(Figure 27(a) and (b)) and corrosion product(Figure 26(b), 27(d), 30(d), and 31(a)). Others showed very clean surfaces(Figure 28(a) and (b), and 32(a)). Photographs selected from d i f f e r e n t locations at various crack lengths in each test are presented in the following pages. They are organized in the ascending order of the applied AK and crack growth rate(Figure 26-32). 6 8 (a) AK=l8.64MPav/m, (b) AK=20.93MPai/m, V=1.30E-09 m/c y c l e , Mag. 2000 V=1.30E-09 m / c y c l e , Mag. 2000 CRACK DIRECTION (c) AK=26.1 5MPav/ni, V=1.77E-09 m / c y c l e , Mag, 2000 (d) AK=32.1 6MPa»/m, V=7.50E-09 m / c y c l e , Mag. 2000 F i g u r e 26. F a t i g u e c r a c k s u r f a c e s of AISI 316L i n IM NaCl a t f r e e c o r r o s i o n c o n d i t i o n and 123 Hz. 69 Cont inue. (e) AR=42.94MPat/m, V=7.20E-08 m/cycle, Mag. 2000 CRACK DIRECTION (f) AK=52.38MPa/m, V=1.05E-07 m/cycle, Mag. 2000 70 (a) AK=9.21MPav/m, (b) AK= 1 0 . 43MPa/m, V=1.00E-10 m/cycle, Mag. 2000 V=1.00E-08 m/cycle, Mag. 2000 CRACK DIRECTION (c) AK=l2.60MPav/m, (d) AK= 1 4 .90MPav/m, V=1.lOE-07 m/cycle, Mag. 2000 V=1.70E-07 m/cycle, Mag. 2000 Figure 27. Fatigue crack surfaces of AISI 316L in 1M NaCl at +600 mVle.„„, and 115 Hz. 71 Cont inue (e) AK=l8.50MPa/m, (f) AK=22.60MPa/m, V=2.37E-07 m/cycle, Mag. 2000 V=2.85E-07 m/cycle, Mag. 2000 CRACK DIRECTION (g) AK=28.50MPai/m, V=3.50E-07 m/cycle, Mag. 2000 (h) AK=35.90MPav/m, V=4.00E-07 m/cycle, Mag. 2000 72 (a) AK=21 .35MPai/m,~ (b) AK=24.72MPav/m, V=1.27E-09 m / c y c l e , Mag. 2000 V=1.25E-09 m / c y c l e , Mag. 2000 CRACK DIRECTION (c) AK=29.60MPa/m, (d) AK=36. 07MPav/m, V=9.00E-09 m / c y c l e , Mag. 2000 V=3.1OE-08 m / c y c l e , Mag. 2000 F i g u r e 28. F a t i g u e c r a c k s u r f a c e s of AISI 316L i n 1M NaCl a t -540 mV/c.-,clN and 125 Hz. 7 3 Continue (e) AK=44.29MPav/m, V=8.20E-08 m/cycle, Mag. 2000 CRACK DIRECTION (f) M=52.52MPa»/m, V=2.15E-07 m/cycle, Mag. 2000 74 (a) AK=20.53MPaj/m, (b) AK=25.33MPa/m, V=1.35E-09 m/cycle, Mag. 1000 V=2.05E-09 m/cycle, Mag. 1000 CRACK DIRECTION (c) AK=30.19MPa/m, V=1.53E-07 m/cycle, Mag. 1000 (d) AK=35.73MPa^m, V=2.30E-07 m/cycle r Mag. 1000 Figure 29. Fatigue crack surfaces of AISI 316L in 1M NaCl & 0.01 M Na 2S 20 3 at -300 m v( S C E) and 122 Hz. 75 Cont inue (e) AK=44.28MPa»/m, V=3.20E-07 m/cycle, Mag. 2000 CRACK DIRECTION (f) AK=58.30MPai/m, V=4.!5E-07 m/cycle, Mag. 1000 76 (a) AK=1 1 .84MPav/m,~ (b) AK=12.24MPa/m, V=5.00E-10 m / c y c l e , Mag. 4000 V=1.20E-10 m / c y c l e , Mag. CRACK DIRECTION (c) AK=l9.62MPav/m, (d) AK=25. 34MPa/m, V=2.00E-10 m / c y c l e , Mag. 2000 V=5.00E-10 m / c y c l e , Mag. F i g u r e 30. F a t i g u e c r a c k s u r f a c e s of AISI 316L i n 1M NaCl 0.01M Na oS o0 - j a t f r e e c o r r o s i o n p o t e n t i a l and 121 77 Continue (e) AK=28.02MPat/m, ( f ) AK=30.86MPa»/m, V=1.00E-09 m/cycle, Mag. 1000 V=1.30E-09 m/cycle, Mag. 2000 CRACK DIRECTION (g) AK=38.35MPav/m, (h) AK=47.1 2MPa/m, V=3.20E-08 m/cycle, Mag. 2000 V=1.00E-07 m/cycle, Mag. 2000 78 (a) AK=1 3.30MPaj/m, V-4.20E-11 m/c y c l e , Mag. 1000 (b) AK=15.lOMPa/m, V=4.00E-10 m / c y c l e , Mag. 1 000 CRACK DIRECTION : . . _^  , (c) AK=17.90MPa/m, V=8.76E-09 m / c y c l e , Mag. 1000 (d) AK=21 .40MPav/m, V=4.30E-08 m / c y c l e , Mag. 1000 F i g u r e 31. F a t i g u e c r a c k s u r f a c e s of AISI 316 L i n 1M NaCl & 0.01M N a 2 S 2 0 3 a t -450 m V ( S C E ) a n d ^ R ^ 79 Cont inue (e) AK=26.1OMPa^m, ( f ) AK=3 1 . 60MPav/m, V-1.10E-07 m/cycle, Mag. 1000 V=2.lOE-07 m/cycle, Mag. 1000 CRACK DIRECTION (g) AK=40.86MPa/m, (h) AK=40.86MPav/m, V=3.90E-07 m/cycle, Mag. 1000 V=3.90E-07 m/cycle, Mag. 8000 8 0 CRACK DIRECTION (a) AK=l6.32MPai/m, V=2.10E-10 m/cycle, Mag. 2000 (c) M=22.60MPa/m, V=2.40E-08 m/cycle, Mag. 2000 F i g u r e 32. F a t i g u e crack su a i r . (b) AK=l8.86MPav/m, V=4.09E-09 m/cycle, Mag. 2000 (d) AK=27.40MPay/m, V=5.30E-08 m/cycle, Mag. 2000 of AISI 316L i n d e s s i c a t e d 81 Cont inue (e) AK=35.60MPa/m, V=1.40E-07 m/cycle, Mag. 2000 ( f ) AK=4 5. 1 8MPav/m, V=2.48E-07 m/cycle, Mag. 2000 CRACK DIRECTION (g) AK=58.33MPav/m, V=3.80E-07 m/cycle, Mag. 2000 82 Figure 33. Scanning electron micrograph of etch pits formed on corrosion fatigue fracture surface of AISI 316L. Figure 34. Matching features of two opposing crack surfaces in striated region of AISI 316L in IM NaCl & 0.01M Na 2S 20 3 at -450 mV(gc » and 122 Hz. Peaks match with peaks and valleys match with valley. Mag. 2000 83 Similar fractographic features were also observed on the fractured cylindrical fatigue specimen(Figure 35). In addition to the three distinct fatigue crack zones, a highly dimpled fatigue zone adjacent to the overload region was revealed, indicative of stage III fatigue crack growth which was attained in this case. The similarity in fractographic features demonstrated their universal occurrence, independent of any specific specimen geometry. 4.5 Cyclic Dissolution Current; As far as the applied corrosion current was concerned, there was hardly any substantial difference between crack propagating and non-propagating stages for a major portion of test time. During crack growth under an applied potential of -300 m V S C E in thiosulphate solution, the corrosion current time curve appeared a bit more uneven than the non-propagation current and fluctuated over a wider range of current(Figure 36). The difference seemed very minor. However i t was observed that corrosion products and gas bubbles were constantly pumped out of the crack t i p during propagation, especially in the highly anodic polarized test of +600 m V s c E * Towards the end of fatigue tests, where cracks advanced at much higher rates, large corrosion current fluctuations were recorded. Towards the end of fatigue testing, 84 (a) Crystallographic region, Mag. 400 (b) T r a n s i t i o n Zone Mag. 400 (c) S t r i a t e d region. Mag. 400 Figure 35. Corrosion fatigue fracture surfaces of c y l i n d r i c a l fatigue specimen of AISI 316L in IM NaCl under free corrosion condition at 30 Hz. 85 E 0.7 pi 0.0 D 0.7 0.0 20 ( a ) 72 ( b ) 80.5 TIME (hrs) Figure 36. (a) Corrosion current during non-crack stage and (b) corrosion current during cracking stage of AISI 316L in IM NaCl + 0.01M N a ^ O j at -300 mV , c ^ v and 122Hz. 86 large anodic currents were observed for anodic t e s t s , whereas in the cathodic tests cathodic current increases occurred instead. In addition, during the anodic t e s t s , when the test was temporarily stopped for the o p t i c a l measurement of fatigue crack length, the anodic current was observed to increase gradually with the time during the period of c y c l i c load cessation(Figure 37). As soon as the test was resumed, the corrosion current instantaneously decayed to the previous c y c l i n g l e v e l . In the ca t h o d i c a l l y polarized tests, t h i s phenomenon did not take place. Instead, the cathodic current dropped down step-wise to a lower l e v e l when c y c l i c " l o a d i n g was terminated and returned immediately back to the same position as soon as the c y c l i c load was reapplied. 4.6 P o l a r i z a t i o n Behavior: Anodic p o l a r i z a t i o n curves of AISI 316L s t a i n l e s s s t e e l in deaerated 1M NaCl solution and 1M NaCl + 0.01M Na 2S 20 3 are presented in Figure 38. The shape of the p o l a r i z a t i o n curve and magnitude of the d i s s o l u t i o n current density were not s i g n i f i c a n t l y a l t e r e d by the presence of thiosulphate ions. In contrast, the addition of sodium thiosulphate d r a s t i c a l l y changed the p o l a r i z a t i o n behavior of s t a i n l e s s s t e e l in a c i d i c solution(pH 1) as can be seen in Figure 39. The order of 8 7 magnitude of the dissolution current density was raised by a factor of approximately 3 to 4. This effect is especially accentuated in the low potential range just above the free corrosion potential. At around -100 mVgCE, the dissolution current density underwent a sharp decrease, which was not similarly observed in the test without thiosulphate. 88 E H Z. PC PS O 0.7 0.0 -TEST STOPS TEST CONTINUES TIME (hrs) Figure 37 A c i d i f i c a t i o n of the crack t i p solution during the load cessation period as indicated by the increase of the anodic d i s s o l u t i o n current of AISI 316L i n IM NaCl at 0 mV( j and 127 Hz. 89 LOG(CURRENT DENSITY) (rjA/cm8) Figure 38. Anodic polarization curves of AISI 316L in IM NaCl and IM NaCl + 0.01M Na0S90.. VO U U CO < W H O OH 1 0.8 0.6 0.4 0.2 0 -0.2 -0.4 -0.6 -0.8 H -1 AISI 316L in IM FeCl8 A without N« 2 S & 03 X wi th 0.01M N a t S t 0 , addition T — 1 — r 8 9 2> 0 1 2 3 4 5 6 7 L0G(CURRENT DENSITY) (rjA/cm2) Figure 39. Anodic polarization curves of AISI 316L in simulated crack tip solutionQM FeCl 2 adjusted to pH 1 with saturated CrCl 3); and 0.01M Na 2S 20 3. 10 5.DISCUSSION: 5.1 Corrosion Fatigue Propagation Rates: It i s apparent that the fatigue crack propagation curves, namely log-log v- AK plots, of AISI 316L generally attained the common sigmoidal shape. The Stage III portions were absent because the tests were terminated before v i o l a t i n g the general y i e l d i n g l i m i t o f _ l i n e a r e l a s t i c fracture mechanics and specimen symmetry requirement of the resonance fatigue machine. Crack retardation e f f e c t s were evident in the near-threshold region, as indicated by numerous "dips" and growth hesitation in almost a l l te s t s . Some tests exhibited large crack retardation over a large range of AK, as in Test I and VII(Figure 17 and 23). Others even showed crack arrest phenomena, in which cracks ceased to propagate(Figure 20, 23, and 24). The applied c y c l i c load had to be increased to r e i n i t i a t e crack growth as in the case of Test IV, VII, and VIII. The remaining tests did not show s i g n i f i c a n t crack growth retardation. In general, the crack retardation phenomenon was most prominent in the near-threshold and was normally absent in the higher AK region. 92 5.2 Crack Closure Effect: The various degrees of crack retardation phenomenon exhibited by each specimen can be explained by the crack closure effect. As mentioned in section 2.10, crack closure can be caused by several factors. It was clear that the most prominent crack closure effect observed in this project occurred in the near-threshold region, where limited crack t ip elastic deformation occurred. However, by examining the near-threshold fractographs, moderate(Figure 26(a)) to extensive(Figure 32(a)) surface plastic deformation can be observed after passage of the crack front. For example, in Figure 32(a), i t seems that the crystallographic step-like features were produced f i r s t during the cracking process and subsequently flattened by another mechanism. If this crack closure was purely a result of premature contact of surfaces by plasticity-induced closure due to the applied tensile loads[43], a l l the surface should be equally deformed. However, this was not the case since other parts of the region did not exhibit the same degree of deformation. In addition, i f this plasticity-induced crack closure was the mechanism which caused crack growth retardation as described by Elber[43], its effect should be expected to be greater as the applied AK became larger and the corresponding plastic zone size increased. On the contrary, crack retardation was virtually absent in the higher AK region. 93 Also, the presence of a compressive plastic zone ahead of the crack tip during a part of each fatigue cycle can not account for the slow-down of crack growth in the near-threshold region. F i r s t l y , the extent of the reversed plastic zone due to the residual compressive stress at the crack tip is a function of the applied cyclic stress intensity and R-ratio[44], It is expected that as the R-ratio decreases, the extent of reversed compressive plastic deformation ahead of the crack t i p would become greater. Since the R-ratio was equal for every test and maintained constant throughout the tests, the effect of crack closure should be equivalent for a l l tests at any AK. Therefore, no difference in apparent crack retardation should result on the basis of equivalent crack closure effect. Hence, the reversed compressive plastic zone is eliminated as a possible cause for the observed crack retardation. Crack retardation can also be attributed to a corrosion-product wedging effect. It is apparent from fractographs that corrosion product did form on the fractured surfaces, as in Figure 26(b), 27(d), 29(a), 30(d) and 31(a). It was very d i f f i c u l t to estimated the extent of coverage and thickness of corrosion products since the surfaces had to undergo inhibited acid cleaning for clear SEM examination. However, the oxide films are usually anticipated at anodic potentials in the passive region. In contrast, at the highly cathodic potentials, the 94 surface w i l l remain relatively clean of oxide film. In Test II, where +600 m V ( g r j E ) w a s applied, the lowest cyclic threshold value was obtained and no significant crack retardation phenomenon was shown. And when a cathodic potential of -540 m V(scE) w a S imposed, as in Test IV, not only crack retardation but also crack arrests were observed several times throughout the tests. Furthermore the dessicated air fatigue test of Test IX, where the environmental corrosiveness was considered as minimal, again exhibited signs of crack closure, as manifested by the relatively high threshold AK when compared to Test II. Nevertheless, the importance of corrosion-product-induced crack closure cannot be simply dismissed by the contradictory evidence mentioned above. Its failure to explain logically the crack closure phenomena can only indicate i t was not playing a dominant role in controlling the crack growth behavior of AISI 316L in the environments studied. Crack branching was occasionally observed in some of the fatigue tests.(Figure 40(a)) It was possible that the branching effect might be responsible to some degree for crack retardation by relieving the crack tip cyclic stress intensity via the creation of new crack surfaces. However, i t was highly unlikely that i t would play any dominant role in the near-threshold region, since i t s occurrence was much more frequent at higher stress intensities. The remaining possible cause was surface-roughness-induced 95 CRACK DIRECTION (b) F i g u r e 40. F a t i g u e crack path on p o l i s h e d s u r f a c e . (a) Crack Branching e f f e c t i s e v i d e n t . (b) Both {111} and non {111} types of c r a c k path are shown. 96 crack closure. Surface roughness was evident from the fractographs and Figure 40. This w i l l be discussed extensively in the following section. 5.3 Surface-Roughness-Induced Crack Closure: To understand f u l l y the extent of surface-roughness-induced crack closure in the fatigue crack propagation behavior of AISI 316L, the mode of crack growth at various AK should be considered f i r s t . 5.3.1 Crack Propagation Mode: A crack in a s o l i d can be stressed in three d i f f e r e n t modest 16], Normal stresses can lead to the "opening" mode, designated as Mode I. The displacement of the crack surface i s in the dire c t i o n of the p r i n c i p a l t e n s i l e stress and normal to the crack plane. In-plane shear gives r i s e to the displacement of the crack surfaces normal to the leading crack front. This " s l i d i n g " mode is known as Mode II . F i n a l l y the "tearing" mode or Mode III i s caused by out-of-plane shear. The displacement of crack surfaces i s in the plane of the crack and p a r a l l e l to the crack front. At high AK, where several s l i p systems are activated 97 simultaneously, macroscopic crack surfaces w i l l be planar in nature and generally perpendicular to the principal tensile direction. Mode I cracking predominates in this region. In the near-threshold AK, at which only the activation of one primary s l i p system is possible, the crack propagates along the maximum shear direction, which is inclined approximately 45 degree to the principal stresses. The crack surfaces produced w i l l tend to be more faceted and serrated. When these serrated surface features are coupled with crack tip mode II displacements, high crack closure load is expected because of the early impingement of the displaced surfaces. At higher AK, Type I loading predominates and there is insufficient Type II displacements to produce premature contact of roughened surfaces. Hence, the crack closure effect should diminish at higher AK. The fractographs suggest exactly what has been proposed above. At low AK, frequently the fracture surfaces displayed marks of surface impingement(i.e. closure). Faceted planes with serrated river lines were observed to have been crushed and flattened as shown in Figure 26(a), 27(a), 27(b), and 32(a). These flattened areas were i n i t i a l l y highly crystallographic facets. As the cyclic load decreased in the unloading part of the cycle, the displaced surfaces contacted each other prematurely before the minimum, cy c l i c load was reached. The local applied cyclic load was thus prevented from dropping lower by the 9 8 wedging action of mismatched surfaces. The e f f e c t i v e AK was reduced by the presence of higher ^ K c ^ a n d consequently the fatigue crack propagation rate decreased. At the points of contact, large compressive stresses were imposed on the surfaces. The sharp crystallographic features were then fl a t t e n e d out as the compressive stresses exceeded the y i e l d l i m i t . The degree of p l a s t i c deformation on the impinged areas depended on the roughness of the surfaces, the l a t e r a l surface displacement, and the l e v e l of crack closure load. Aside from the impinged areas, the crack surfaces appeared to undergo very l i m i t e d p l a s t i c deformation and retained t h e i r fine c r y s t a l l o g r a p h i c features. As the c y c l i c stress i n t e n s i t y increased, the surfaces became less faceted and more fibrous in nature.(Figure 26(d), 27(e) and ( f ) , 28(c) and (d), 29(b) and (c), 30(e) and ( f ) , and 31(c) and (d), and 32(b) and (c) . River l i n e s were s t i l l produced on the surfaces but were masked increasingly by tearing marks. Localized deformation occurred l e s s frequently, indicating lower roughness-induced crack closure e f f e c t s . In the region where fatigue s t r i a t i o n s were observed, stereogrphic photographs revealed that the crack surfaces were r e l a t i v e l y f l a t as compared to other regions. Eventhough the surfaces appeared to experience higher p l a s t i c deformation(i.e. s t r i a t i o n s ) , t h i s deformation was not l o c a l i z e d but uniform throughout the whole area. It can be concluded that the deformed surfaces were the result of the applied c y c l i c t e n s i l e stresses 99 and not the contact stresses. The absence of prominent localized impingement marks can be attributed to the flatter surface features, with a corresponding lower crack closure load, and Mode I opening. It is now clear, in the case of AISI 316L, that surface-roughness-induced closure had greater influence in the near-threshold region, where crack surfaces were rougher and the lateral crack surface displacement was greater. In the intermediate and striated regions, the combination of predominant Mode I crack displacement and fla t surface features rendered the crack closure effect relatively unimportant. 5.4 Effect of the Applied Potential; The imposition of high anodic potential has been found to have significant effect on the crack propagation rate. In the Test II, the threshold cyclic stress intensity of 9.2 MPa/m was the lowest of a l l tests. Only a few minor crack retardation events took place during the test. The low an<3 absence of crack retardation effect are attributed to the high dissolution rate of AISI 316L at +600 m V(gcE)* X t w a s n o t t h e h i 9 h r a t e o f removal of metal at the exposed crack t i p area or instability of surface film which led to the low AK^. B u t i t was the removal of the impinged surface asperities by anodic dissolution 100 processes, which resulted in the reduction of crack closure effects. As the opposing surfaces abraded each other, the integrity of the protective oxide film was destroyed. The metal at the exposed areas then underwent a rapid dissolution transient prior to repassivation. The surface film could reform i t s e l f quickly. However since the cyclic frequency was around 115 Hz, the film was again ruptured very quickly and the exposed metal dissolved. As this process repeated i t s e l f many times, the impinged asperities would then be removed. The effective AK was then roughly equivalent to the applied AK. As Figure 27(a) and 27(b) show, the dissolution marks can be clearly observed. The darker and deformed areas appear to suffer from extensive corrosion attack. In other areas, the surfaces remained relatively undisturbed. Also at higher AK, the fractographs do not show signs of environmental attack. In other tests, the imposition of lower anodic or cathodic potentials did not have the same effect on the crack propagation behavior as that of +600 m V ( s c E ) * I f c w a s m a * n l v because the rate of metal removal was too slow at the lower potentials to reduce the crack closure load to any substantial extent. 5.5 Effect of Thiosulphate: The addition of 0.01M of sodium thiosulphate did not affect 101 the fatigue crack growth behavior significantly. This was consistent with the similarity in anodic polarization behavior of AISI 316L in 1M NaCl and 1M NaCl + 0.01M Na 2S 20 3(Figure 38). The only difference arose when the test was conducted in highly acidic solution. In solution of low pH, the thiosulphate ion is easily reduced to H2S by electrochemical reactionst40]. The dissolved H2S will then serve as a catalyst for the dissolution of metal at low cathodic potentials, as demonstrated by the high dissolution current of Figure 39. The failure-of thiosulphate ions to influence the fatigue crack behavior in Tests VI, VII, and VIII can be simply explained. Applied cathodic potentials together with crack solution pH were not sufficiently low to reduce the thiosulphate to H2S in the testing condition. Therefore no catalytic effect on metal dissolution was produced and metal was not removed at a sufficiently high rate to reduce the crack closure load significantly. The absence of the catalytic effect was indicative of efficient-mixing of the solution inside the crack with the bulk environment, which prevented the localized ' acidification of solution near the crack t i p . Acidification was -2 necessary in order to reduce S 20 3 at the applied potential. From Figure 37, i t is very clear that the acidic solution, which was formed during a short cessation of cyclic loading, was quickly neutralized as soon as the cyclic load was reapplied. 102 The g r a d u a l i n c r e a s e of anodic d i s s o l u t i o n c u r r e n t when the t e s t was stopped was i n d i c a t i v e of l o c a l i z e d p a s s i v i t y breakdown due t o the crack behaving as a c r e v i c e wherein d i s s o l u t i o n and h y d r o l y s i s phenomena lowered the pH and promoted d i s s o l u t i o n of the m e t a l [ l 4 ] . During f a t i g u e , the high r a t e of s o l u t i o n exchange by the pumping a c t i o n of the high c y c l i c frequency f a t i g u e , the crack t i p environment should be c o n s i d e r e d to be s i m i l a r t o that of the bulk environment. At the bulk s o l u t i o n pH v a l u e s near 6, a p o t e n t i a l of -401 m V ( g c E ) * s r e q u i r e d f o r t h i o s u l p h a t e to be reduced thermodynamically[ 1 4]. The lowest t e s t p o t e n t i a l employed was -450 m V ^ S C E j ( T e s t V I I I ) t h i s was c o n s i d e r e d to be too c l o s e to the c a l c u l a t e d thermodynamic v a l u e so t h a t the k i n e t i c s of p r o d u c t i o n were too slow to y i e l d any s u b s t a n t i a l q u a n t i t y of J^S to c a t a l y s e metal d i s s o l u t i o n . T h e r e f o r e , a low metal d i s s o l u t i o n r a t e was expected and the removal of s u r f a c e roughness would not r e s u l t . 5.6 V a l i d i t y of the Measured T h r e s h o l d AK i n C o r r o s i o n F a t i g u e ; Because of the l a r g e s c a t t e r of t h r e s h o l d c y c l i c s t r e s s i n t e n s i t y f o r f a t i g u e behavior of AISI 316L i n the v a r i o u s environments, i t was almost i m p o s s i b l e to deduce the exact A R t h v a l u e s , below which no f a t i g u e c r a c k growth was p o s s i b l e . S i n c e the ^ K t h v a ^ - u e s a r e s t r o n g l y d i c t a t e d by the degree of c r a c k 103 c l o s u r e e f f e c t , the e f f e c t i v e AK d r i v i n g c r a c k p r o p a g a t i o n , AK cc( AK - AK , ) , can not be a c c u r a t e l y e s t i m a t e d . At t h i s e f f app c l p o i n t i n t i m e , i t can o n l y be s a i d t h a t the l o w e s t o b s e r v e d t h r e s h o l d AK f c h i s about 9MPa/m i n Te s t I I . I t was argued t h a t t h e removal of s u r f a c e roughness by m e t a l d i s s o l u t i o n was m a i n l y r e s p o n s i b l e f o r the low A K t ^ measured. The e f f e c t of t h e e n v i r o n m e n t s on the t r u e A K ^ w i l l remain unknown u n l e s s the e f f e c t i v e c y c l i c s t r e s s i n t e n s i t y can be d e t e r m i n e d by measuring the c r a c k c l o s u r e s t r e s s . W i t h o u t knowing A K e f f o n l y a s k e t c h y q u a l i t a t i v e i n f l u e n c e of c o r r o s i o n e n v i r o n m e n t s on ^ K t n c a n D e made w i t h r e g a r d t o the f a t i g u e c r a c k p r o p a g a t i o n b e h a v i o r . 5 . 7 R e l e v a n c e t o the I n d u s t r i a l S i t u a t i o n : The p r e s e n t f i n d i n g s have r e v e a l e d t h a t t h e i m p o s i t i o n of a n o d i c p o t e n t i a l s as a p r e v e n t i v e measure t o a l l e v i a t e t h e c o r r o s i o n problem may sometimes have an o p p o s i t e d e l e t e r i o u s e f f e c t i n a c c e l e r a t i n g f a t i g u e c r a c k growth by i n c r e a s i n g t h e AK g££ a t the c r a c k t i p . The problem may become much more s e r i o u s i n the case of s u c t i o n p r e s s r o l l s i n t h e p r e s e n c e of t h e i n d u s t r i a l e n v i r o n m e n t s due t o the b u i l d u p of v a r i o u s d e p o s i t s ( e g . p u l p f i b r e s ) . These d e p o s i t s , a c t i n g as p h y s i c a l b a r r i e r s , w i l l h i n d e r s o l u t i o n exchange between the b u l k e n v i ronment and produce o c c l u d e d c e l l s u n d e r n e a t h the d e p o s i t s . 104 Therefore the application of anodic protection method w i l l probably lead to higher rates of dissolution and higher crack growth rate i f cracks form in these isolated c e l l s . Furthermore the operating cyclic frequencies of suction press r o l l s normally are from 3 to 12 Hz[9] which w i l l ylied a less efficient solution mixing momentum, which again promotes localized acidification. The presence of highly acidic c e l l s w i l l present an additional problem because i t provides a favorable condition for the reduction of thiosulphate to HjS. However, the production of could be hindered by the imposition of high anodic potentials above the reduction potential of HjS. As a result, the advantage and disadvantage of the anodic protection method and the applied anodic potentials should be carefully considered and selected for the optimal result. 5.8 Mechanisms of Fatigue Crack Propagation: Fractography of the near-threshold crack surfaces of AISI 316L stainless steels revealed a fracture morphology which is commonly associated with b r i t t l e failures of metals. Flat cleavage facets with a parallel plateau and ledge features were present in a l l fracture specimens. The process of b r i t t l e cleavage involves transgranular fracture along specific crystal planes and is usually identified with low fracture energy 105 cracking in BCC and HCP metals[54]. Cleavage fracture is not commonly observed on FCC metals. They exhibit features characteristic of b r i t t l e failure only when they are subjected to severe environmental conditions[55]. 5.8.1 Cleavage Fracture mechanisms: Similar cleavage fracture features have been observed in corrosion fatigue experiments of other materials. It was observed on an aluminum alloy specimen when fatigued in ambient air[56]. The same crystallographic cleavage appearance was found on AISI 304 stainless steel when tested in sodium sulphate solution at 250°C[57]. In a nickel-base superalloy and an austenitic Fe-NI-Cr alloy, 2RK65, similar fatigue fracture features were again exhibited!58,59]. On the f i r s t glance, the cleavage type of fatigue propagation may be a typical cracking mode for most austenitic materials at low AK, independent of environmental influence, as long as the near-threshold condition is provided at the crack t i p . In the present study, the cleavage fracture was only evident in the near-threshold region. The fracture morphology gradually changed from cleavage to a fibrous and feathery feature at a higher AK, and later transformed into the easily recognized striation pattern as the applied AK was further increased. The 106 following discussion is an attempt to c l a r i f y the transformation of fracture morphology at different cyclic stress intensities in terms of fatigue crack growth mechanisms. There are many conflicting fatigue crack growth mechanisms in the near-threshold AK region. Fatigue crack advancement has been attributed by some researchers to the possibility of hydrogen embrittlement of metals[60]. At low AK, fatigue crack growth purely by the application of a cyclic load is extremely slow. There is enough time for hydrogen to diffuse through the metal and accumulate ahead of the crack t i p to a harmful concentration before the crack" is further mechanically extended due to local embrittlement. The crack growth rate is then determined by the single crack extension, which is the distance between the crack tip and the hydrogen accumulation zone, and the time required for the build-up of a c r i t i c a l level of hydrogen concentration. As the applied AK increases, fatigue crack growth s t r i c t l y by mechanical means becomes faster. The available accumulation time is then insufficient for hydrogen diffusion processes to produce a detrimental concentration of hydrogen in the metal l a t t i c e . Therefore a substantial reduction of cohesive bonding energy of the metal by hydrogen is not obtained. Consequently, the effect of hydrogen embrittlement diminishes at higher AK. It may thus seem reasonable to adopt the hydrogen embrittlement mechanism as the main driving force for fatigue crack growth of AISI 316L in the present environmental system on 107 the basis of fractographic examination. The generation of atomic hydrogens by electrochemical reduction reactions in aqueous solutions or chemical reactions in humid air conditions could lead to hydrogen embrittlement. However according to the Nernst equation of hydrogen evolution at 25°C and 1 atm[6l]: H 0 = 2H+ +2e~ E = 0.000 - 0.0591 pH (14) 2 O the highest reduction potential possible for the reduction of H of a highly acidic solution of a pH value of 1 is -59 m V ( s n E ) ( i . e . -301 mV^ S C Ej). Because of the absence of localized acidification of environment at the crack t i p and the application of high anodic potentials of +600 m V ( S C E ) a n <3 0 m V ( S C E ) * n Test II and III, which were above the hydrogen evolution potential, the production of atomic hydrogen was definitely not thermodynamically possible in these situations. Nevertheless, cleavage fracture was obtained. Therefore, while hydrogen embrittlement may play a secondary role in corrosion fatigue of AISI 316L in those cases where hydrogen reduction was attained, i t was not the essential component for near-threshold fatigue crack propagation. Others have proposed a dislocation pile-up model[62], which induces an additional tensile stress f i e l d on top of the applied stresses near the crack t i p , or a dislocations coalescence model[63], which involves coalescence of dislocations on 108 intesecting planes. The interaction of intesecting disloactions w i l l lead to incremental crack growth. However these mechanisms are only applicable to the BCC structure. The corresponding dislocation interactions in the FCC materials are not thermodynamically favorable and hence highly unlikely to occur. Furthermore, if these dislocation interaction models are operative, there is no reason why they cannot occur in tensile loading failures. B r i t t l e fracture has not been observed in austenitic materials. The possibility of b r i t t l e fracture by the normal stress developed from a dislocation pile-up or by coalescence of dislocations seems rather remote. It is possible to embrittle the metal by the presence of a high vacancy density produced by the interaction of dislocation during fatigue cycling[64]. However the exact mechanism of vacancy embrittlement is not clear at this point in time. In summary, therefore, cleavage in practice w i l l only occur when the shear stress, T , required to cause dislocation motion on the most favorably oriented s l i p plane exceeds the tensile stress needed to fracture the plane least resistant to cleavage. Hydrogen embrittlement or dislocation interaction have been ruled out as a possibility for causing the cleavage failure. Absorption of a detrimental species, which lowers the atomic bonding energy of a metal, can also be considered as a potential cleavage fracture mechanism. However, i t is d i f f i c u l t with the absorption argument to consistently interpret the similar 109 fractography obtained under different environmental conditions. 5.8.2 The Alternating Shear Model; The failure of the commonly espoused cleavage mechanism to interpret the near-threshold b r i t t l e fracture features c a l l s for a different approach to explain the fractography. The cracking process may be due solely to s l i p related processes. The crack propagation velocity is around 5 x 10 1 1 m/cycle and the smallest dislocation displacement for 13Crl2Ni austenitic stainless steel ,which is used as an estimate for AISI 316L stainless steel, is the atom diameter and corresponds to about 3.60 x 10 1^ m[65]. These two values are roughly equal in magnitude. The discrepancy can be attributed to nonuniform crack growth along the advancing crack front which then lowers the average crack velocity or the possibility that sl i p may not occur on every cycle but over a series of cycles. It is then possible to obtain the near-threshold crack velocity based solely on s l i p displacement by dislocation motion. Etch pits formed on the cleavage-like fracture surfaces indicated the possible cracking planes to be {111} , {110} , and {100} (Figure 41(a) and (b)). Other types of etch pits were also observed, but their shapes were quite irregular and possibly they formed on higher indexed planes of undetermined orientation. 110 (a) (b) F i g u r e 41. O r i e n t a t i o n of f r a c t u r e s u r f a c e s from e t c h p i t s . (a) {111} c r a c k p l a n e w i t h <110> c r a c k d i r e c t i o n a t the n e a r - t h r e s h o l d r e g i o n . (b) Both {110} < 1 l 2 > a n d {100} <011> c r a c k o r i e n t a t i o n s a r e shown. 11 1 The coexistence of the various crack surface orientations was further confirmed by the crack orientation profile on the polished side faces of the specimen. The {111} traces introduced through s l i p activity indicated the crack path could follow {111} planes and other crystallographic planes(Figure 40(b)). By analyzing the edges of well-defined etch pits(Figure 41 and 42), i t was concluded that the crack directions on {111} , {110} , and {001} are < T l O > , <Tl2>, and <TT0> respectively. The determination of crack direction was based on the fact that the direction of the river lines, which were essentially the tear ledges l e f t behind between crack front segments advancing on different but parallel planes, were parallel to the crack propagation direction. At least one side of the edges of most of the observed etch pits followed the same direction of the nearby river lines. The observed crack orientations can be obtained by either s l i p action of one single most favorably oriented s l i p system or alternate activation of two symmetrically arranged s l i p systems as indicated in the Table III. The maximum Schmid factor, which is an indicator for ease of yield for the s l i p system in question, is obtained for the single s l i p system when the s l i p plane and s l i p directions(eg. {111} ,<110>) are inclined 45 degree to the applied tensile stress. Under these circumstances the Schmid factor is 0.50. The 112 <»*> (111) l"0) (100) F i g u r e 42. Schematics of t h e shapes of e t c h p i t s on {111} , {110} , and {100} p l a n e s and t h e c o r r e s p o n d i n g edge d i r e c t i o n s . 113 S l i p system conjugate s l i p r e s u l t a n t crack' o r i e n t a t i o n Maximum Schimid f a c t o r occurrence {111} <TlO> {111} <Ti0> 0.50 observed {111} <01T> {1iT} <ToT> {110} <Tl2> 0.0408 observed { i l l } <OlT> { i i T } <oTT> {110} <ooT> 0.0408 none {111} <T01> {TTi} <oTT> {001} <TTo> 0.0408 observed {111} <To1> {TTi} <ToT> {001} <Too> 0.0408 none III. Fatigue Crack Orientation in the Near-Threshold Regions. corresponding Schmid factors for other possible s l i p systems are much less. In this case, the crack will continue to grow on the {111} plane. For the symmetrically oriented duplex s l i p systems with respect to a tensile axis normal to {110} or {001} crack planes, the Schmid factors are equal for each s l i p system at a value of 0.408. Therefore, i t is equally li k e l y that either s l i p system can be activated because the resolved shear stress is equivalent for both s l i p systems. The observed crack orientation( {110} or {001} ) can be produced by the alternate s l i p model as Figure 43 suggests. On the uploading cycle, one s l i p system is activated and advances the crack by a multiple of the unit Burger displacement. On the downloading cycle, the same s l i p system is reversed in the opposite direction but on a different parallel s l i p plane. A crack increment is thus produced. In the next cycle, the movement of an equally favorable s l i p system of the same Schmid factor is triggered by the uploading stress and again reversed in the downloading cycle. By the combination movement of the dual s l i p systems, the observed s l i p crack orientation results. In the Schmid factor analysis, i t is also found that another crack direction can occur with equal propensity. Hence, the {001} crack plane with the crack direction <100> and {110} crack plane with the crack direction <001> should also be observed in the fractographs. In examination of a l l the etch pitted fractographs, no such crack orientations were clearly 115 (a) F i g u r e 43. Schematic of t h e a l t e r n a t e s l i p model. 116 identified. However, these types of orientations have been observed on other metal systems. In aluminum alloys, (001)[100] faceted fatigue crack growth was found[56]. It is not known why these kinds of cracking systems were not observed in AISI 316L near-threshold fatigue fractography. The faint parallel s l i p traces(Figure 44), cutting across the river line marking at a fixed angle, can either be caused by the relaxation s l i p offset in the wake of crack growth[56] or the striation markings of the active duplex s l i p systems. A l l the fractogrphic features seem to agree with the alternate s l i p model. Thus, the"cleavage-like b r i t t l e morphology near AKfch * s in fact a result of crack growth via dislocation-activated s l i p processes. 5.7.3 Transition Zones; The fibrous fracture features which appeared at a higher AK were basically an extension of the faceted crack surfaces. They s t i l l maintained roughly some of the basic crystallographic characteristics of the near-threshold fatigue crack surfaces. Similar fracture surfaces were also found on austenitic stainless single crystals[66], AISI 304[67], and high-strength steel[68]. Some researchers have suggested that this fracture feature is developed by the operation of multiple s l i p planes, 117 (a) 118 some of which do not intersect along the crack front. These non-intersecting crack planes have to be accommodated by additional localized plastic flow at the crack tip[66]. An alternative explanation is now proposed whereby the feathery appearance is the result of a large degree of tearing along the river lines. Since some of the river lines are caused by the intersection of the crack front with screw dislocations[69], a higher AK w i l l increase the number of screw disloctions and increase the resulting frequency of river lines. When coupled with a higher degree of tearing, the fibrous, feathery fractography is produced. 5.7.5 Striated Region; The mechanism for the production of ductile striation lines in fatigue fracture is well established. It is the plastic crack blunting model, which requires simultaneous activation of at least two s l i p systems. The crack advances through the blunting and resharpening action of the crack tip[24]. The fractography of the striated region agrees with the proposed model with the peak-to-peak and trough-to-trough matching of striations on opposite fracture surfaces. 119 5.8 Effect of Cyclic Stress Intensity on Fractography: There is no general relation between the characteristics of particular fracture features and specific ranges of AK. This lack of correspondence is understandable because of the large surface-roughness-induced crack closure effect in the near-threshold region. To derive a sensible relationship between the crack t i p cyclic stress intensity, fractography and the crack propagation rate, the crack closure load has to be measured to obtain an accurate AR^. The importance of effective AK is indirectly suggested by the surprisingly good correlation between the crack growth rate and the fractography(Table IV) since i t is believed the crack propagation rate is a function of the crack ti p AK e f f. 5.9 Effect of The Cyclic Frequency: According to Figure 20 and 21, the difference between Test IV and Test V seems very large. At lower frequency, the threshold value was considerably lower due to the absence of large crack closure effect. However, i f the crack arrest in Test IV is ignored, the cyclic stress intensities from which cracks started to grow are very close for both cases at the value of around 12.0 MPa/m. The difference is then not very significant. The result also indicates a smaller crack closure effect at the 120 Test No. M (MPav'm) Growth rt.(m/cycle) Surface Morphology ' j 18.64 1 .30E-09 Faceted I 20.93 1 .30E-09 Faceted I 26.15 1.77E-09 Faceted I 32. 16 7.50E-09 Fibrous I 42.94 7 .20E-08 Fibrous I 52.38 1.05E-07 Striated n 9.21 1 .00E-10 Faceted 11 10.43 1.00E-08 Faceted-Fibrous 11 12.60 1.1OE-07 Faceted-Fibrous I j 14.90 1 .70E-07 Fibrous 11 18.50 2.37E-07 Fibrous II 22.60 2.85E-07* Fibrous II 28.50 3.50E-07* Striated II 35r90 4.00E-07 Striated IV 21 .35 1.27E-10 Faceted IV 24.72 1 .25E-09 Faceted IV 29.60 9.00E-09 Fibrous IV 36.07 3.10E-08 Fibrous IV 44.29 8.20E-08 Striated IV 52.52 2.15E-07 Striated IV 84.51 3.60E-07 Striated VI 20.53 1.35E-09 Faceted VI 25.33 2.05E-09 Fibrous VI 30. 19 1.53E-07 Fibrous-Striated VI 35.73 2.30E-07 Striated VI 44.28 3.20E-07 Striated VI 58.30 4. 15E-07 Striated VII 1 1 .84 5.00E-10 Faceted VII 12.24 1.20E-10 Faceted VII 19.62 2.00E-10 Faceted VII 25.34 5.00E-10 Faceted VII 28.02 1.00E-09 Faceted VII 30.86 1.30E-09 Fibrous VII 38.35 3.20E-08 Fibrous-Striated VII 47. 12 1.00E-07 Striated — — Table IV. Fractographic Correspondence with the Applied Cyclic Stress Intensities and Fatigue Crack Propagation Rates under Various Testing Conditions. 121 Cont i nue. VI I I 1 3. 30 4.20E-11 Faceted VII I 15.10 4.00E-10 Faceted VII I 17.90 8.76E-09 Fibrous VI11 21 .40 4.30E-08 Fibrous VII I 26. 10 1.10E-07 Fibrous-Str iated VI11 31 .60 2.10E-07 Striated VI I I 40.86 3.90E-07 Striated IX 16.32 2.10E-10 Faceted IX 18.86 4.00E-09 Fibrous IX 22.60 2.40E-08 Fibrous IX 27.40 5.30E-08 Fi brous IX 35.60 1.40E-07 Fibrous-Str iated IX 45.18 2.48E-07 Str iated IX 58.33 3.80E-07 Striated * extrapolated values. \ 1 2 2 lower frequency. It might be due to the higher degree of crack tip acidification as the result of inefficient mixing at the lower frequency. This could have produced more dissolution of the surface protuberances and reduced the crack closure effect. 5.10 Stage I vs. Stage II: The conventional definition of the stage I and stage II fatigue behavior based purely on the number of activated s l i p systems may seem extremely arbitrary and inapplicable in the AISI 316L system. In* fact, the number of s l i p planes capable of movement depends more on their crystallographic orientation with respect to the tensile axis than the applied cyclic stress intensity. The crystallographic cleavage-like fractography usually associated with stage I fracture can also be the composite result of the alternating s l i p movement. Furthermore, the typical crack propagation rate of the cleavage-like fracture rests well within the near-threshold region, away from the commonly recognized Stage II region, described by the Paris power law relationship. The exact definition of fatigue crack behavior may require some modification in light of this finding. From the present study, i t is more appropriate to define Stage I crack growth mechanism as the crack extension by the activation of one or more s l i p systems, of which only one is active at the crack ti p at any time on any given cycle. And the Stage II crack 123 growth mechanism is the simultaneous activation of multiple s l i p systems on each cycle. 124 CONCLUSION: Studies of the high frequency corrosion fatigue crack propagation behavior of AISI 316L stainless steel in 1M NaCl, 1M NaCl + 0.01M Na2S203r and dessicated air under various anodically and cathodically polarized conditions at room temperature and 1 atmosphere pressure have revealed the following conclusions: (1) The crack F«tardation phenomena at the near-threshold was mainly the result of surface-roughness-induced crack closure, which reduced the effective crack t i p cyclic stress intensity ^ K e f f a n c^ decreased the fatigue crack propagation rate. (2) The effect of the applied high anodic potentials on the fatigue crack growth behavior was to increase the effective crack t i p cyclic stress intensity AK eff by reducing the crack closure effect via the dissolution of surface protuberances(roughness) created in the wake of the advancing crack. (3) The effect of the environments on the high frequency corrosion crack propagation behavior, apart from their influence on the crack surface roughness, has not been conclusively obtained due to the lack of reliable estimates of the crack 125 closure stress intensity AK (4) A deleterious effect of the thiosulphate ions was not observed during high frequency corrosion fatigue behavior due to the efficient mixing between the bulk and crack t i p environments. The absence of a highly acidic crack tip solution provided a much less favorable condition for the production of catalytic H2S. (5) The fatigue~crack growth of AISI 316L in the near-threshold region was achieved by the activation of one s l i p system or the alternate activation of two intersecting s l i p systems. The predominant crack orientations observed were {111} <110>, {110} <Tl2>, and {001} <TTo> types. The near-threshold fractography mainly exhibited a cleavage-like crystallographic features. (6) The fatigue crack growth of AISI 316L in the Stage II region was effected by the plastic blunting mechanism, which resulted in a highly plastically deformed and striated fractography. 126 BIBLIOGRAPHY [I] . Enochs, J. S. and Devereux, F., Met. Trans. A, Vol. 6A, 1975, pp. 298-304. [2]. Spahn, H., 'Corrosion Fatigue: Chemistry, Mechanics and Microstructure', Ed. Devereux, 0. F., McEvily, A. J., and Staehle, R. W., NACE-2, 1972, pp. 40-51. [3]. Levy, M. and Morrossi, J. L., 'Corrosion Fatigue Technology', Ed. Craig, Jr. H. L., Crooker, T. W., and Hoeppner, D. W., ASTM STP 642, 1978, pp. 300-312. [4]. Scott, P. M. and Tomkins, B., 'Corrosion Fatigue', Proc. of the First USSR-UK Seminar on Corrosion Fatigue of Metals, Ed. Parkins, R. N. and Kolotyrkin, Ya. M., The Metals Society, 1983, pp. 156-164. [5]. Gooch, T. G. and Booth G. S., Met. Sci., Vol. 13, No. 7, 1979, pp. .396-401 [6]. Tomkins, B. as in reference [3], pp. 135-147. [7]. Nathan, C. C. and Piluso, A. J., 'Pulp and Paper Industry Corrosion Problems', Vol. II, NACE, 1977, pp. 126-132. [8]. Wensley, D. A., private communication. [9]. Garner, A., Paprican Corrosion Workshop, F a l l , 1984. [10]. Dahl, C. B. and Rainger, C. W., ibid, pp. 105-110. [II] . Moskal, M. D. and Eggeman, G. W., ibid, pp. 94-99. [12]. Thorpe, P. H., Third Internation Symposium on Corrosion in the Pulp and Paper Industry, 1980, pp. 26.1-26.20 [13]. Szklarska-Smialowska, S., 'Localized Corrosion*, NACE-3, 1974, pp. 312-341. [14]. Tromans, D. and Frederick, L., Corrosion, Vol. 40, No. 12, 1984, pp. 633-639. [15]. Pook, L. P., 'The Role of Crack Growth in Metal Fatigue', The Metal Society, London, 1983. [16]. Broek, D., 'Elemetary Engineering Fracture Mechanics', Third Rev. Ed., Martinus Nijhoff Publishers, 1983. 127 [17]. Grosskreutz, J. C , 'Metal Fatigue Damage - Mechanism, Dectection, Avoidance, and Repair', Ed. Manson, S. S., ASTM, STP 495, 1971, pp. 5-60. [18]. Laird, C. and Duquette, D., as in reference [2] , pp. 88-115. [19]. Fine, M. E. and Ritchie, R. 0 . , 'Fatigue and Microstruture', ASM, 1979, pp. 245-278. [20]. Paris, P. C. and Erdogan, F., Journal of Basic Engineering, Vol. 88, 1963, pp. 528-535. [21]. Gell, M. and Leverant, G. R., Trans, of the Met. of AIME, Vol. 242, 1968, pp. 1869-1879. [22]. Pelloux, R. M. N., 'Fracture*, Proc. 2nd Int. Conf., Ed. Pratt,. Chapman and Hall, 1969, pp. 731-744. [23]. Broek, D. t Int. Met. Rev., Vol. 19, 1974, pp. 135-182. [24]. Laird, C., 'Fatigue Crack Propagation', ASTM STP 415, 1967, pp. 131-168. [25]. Wei, R. P., Eng. Frac. Mech., Vol. 1, 1970, pp. 633-651. [26]. Wei, R. P. and Landes, J. D., Materials Research and Standards, Vol. 9, No. 7, 1969, pp. 25-46. [27]. Scully, J. C , Met. Sci., Vol. 12, No. 6, 1978, pp. 290-300. [28]. Tetelman, A. S. and Robertson, W. D., ACTA Met., Vol. 11, 1963, pp. 415-426. [29]. Oriani, R. A., and Josephic, P. H., ACTA Met., Vol. 22, 1974, pp. 1065-1074. [30]. Kruger, J., 'Stress Corrosion Cracking', Ed. Yahalom, J. and Aladjem, A., Freund Pulishing House, Tel-aviv, 1980, pp. 5-37. [31]. Sedriks, A. J., Green, J. A. S., and Novak, D. L., Corrosion, Vol. 27, No. 5, 1971, pp. 198-202. [32]. Brown, B. F., F u j i i , C. T., and Dahlberg, E. P., Journal of Electrochemical Society, Vol. 116, No. 2, 1969, pp. 218-219. 128 [ 3 3 ] . H a r t , W. H., Henke, T. E., and F l e u t , J . E., C o r r o s i o n , V o l . 33, No. 4, 1 9 7 7 , pp. 151-152. [34] Meyn, D. A., Met. T r a n s . , V o l . 2, 1971, pp. 853-865. [ 3 5 ] . H a r t , W. H., Tennant, J . S., and Hooper, W. S., as i n Ref. [ 3 ] , pp. 5-18. [ 3 6 ] . Pao, P. S., Wei, W., and Wei, R. P., 'Environment-S e n s i t i v e F r a c t u r e o f E n g i n e e r i n g M a t e r i a l s ' , Ed. F o r o u l i s , Z. A., AIME, New Y o r k , 1979, pp. 565-580. [ 3 7 ] . V a s i l e n k o , I . I . and K a p i n o s , V. I . , as i n R e f . [ 4 ] , pp. 74-84. [ 3 8 ] . P y l e , T. and R o l l i n s , V., P r o c . o f the 6th I n t . Congr. on M e t a l l i c C o r r o s i o n , A u s t r a l a s i a n C o r r o s i o n A s s o c i a t i o n , 198 , pp. 556-569. [ 3 9 ] . Lee, H. J3. and U h l i g , H. H., J . E l e c t r o c h e m . S o c , V o l . 117, No. 1, 1970, pp. 18-22. [ 4 0 ] . V a l e n s i , G., Van Mu u y l d e r , J . , P o u r b a i x , M., ' A t l a s o f E l e c t r o c h e m i c a l E q u i l i b r i a i n Aqueous S o l u t i o n ' , Ed. P o u r b a i x , M., NACE & CEBLCOR, 1974, pp. 545-553. [ 4 1 ] . Newman, R. C , S i e r a d z k i , K., and I s a a c s , H. S., Met. T r a n s . A, V o l . 13A, 1982, pp. 2015-2026. [ 4 2 ] . N a k a i , Y., Tanaka, K., and N a k a n i s h i , T., E n g i n e e r i n g F r a c t u r e M echanics, V o l . 15, No. 3-4, 1981, pp. 291-302. [ 4 3 ] . E l b e r , W., Eng. F r a c . Mech., V o l . 13, 1980, pp. 37-45. [ 4 4 ] . S c h i j v e , J . , Eng. F r a c . Mech., V o l . , 11, pp. 1979, pp. 167-221. [ 4 5 ] . Gray, G. T., W i l l i a m s , J . C , and Thompson, A. W., Met. T r a n s . A, V o l . 14A. 1983, pp. 421-433. [ 4 6 ] . van der V e l d e n , R., Ewalds, H. L., and S c h u l t z e , W. A., ' C o r r o s i o n F a t i g u e ' , Ed. C r o o k e r , T. W. and L e i s , B. N., ASTM STP 801, 1983, pp. 64-80. [ 4 7 ] . V i t e k , V., I n t e r n a t i o n a l J o u r n a l of F r a c t u r e , V o l . 13, No. 4, 1977, pp. 481-501. [ 4 8 ] . S u r e s h , S., Met. T r a n s . A. V o l . 14A, 1983, pp. 2375-2385. 1 2 9 [49]. Ritchie, R. 0., 'Environment-Sensitive Fracture of Engineering Materials', Ed. Foroulis, Z. A., AIME, 1979, pp. 538-564. [50]. Russell, A. J. and Tromans, D., Met. Trans. A, Vol. 12A, 1981, pp. 613-621. [51]. Mukai, Y., Watanabe M., and Murata, M. 'Fractography in Failure Analysis' Ed. Strauss, B. M. and Cullen, W. H. Jr., ASTM STP 645, 1978, pp. 164-175. [52]. Tada, Hiroshi, 'The Stress Analysis of Cracks Handbook', Del Research Corporation, Hellertown, Pennsylvania, 19 73 pp. 210-211. [53]. Tomkins, B., Met. Sci., Vol. 13, No. 7, 1979, pp 387-395. [54]. Petch, N. J., 'Fracture', Ed. Liebowitz, H., Vol. 1, Academic Press, New York, 1968, pp. 351-393. [55]. Liu, R., Narita, N., Altststter, C , Birnbaum, H., and Pugh, E. N., Met. Trans. A, Vol. 11A, 1980, pp. 1563-1574. [56]. Garrett, G. G. and Knott, J. F., ACTA Met., Vol. 23, 1975, pp. 841-848. [57]. Tsai, W. T., Moccari, A., Szklarska-Smialowska, Z., and Macdonald, D. D., Corrosion, Vol. 40, No. 11, 1984, pp. 573-583. [58]. Duquette, D. J. and Gell, M., Metallurgical Transactions, Vol. 2, 1971, pp. 1325-1331. [59]. Keys, L. H., Bursle, A. J., Thompson, K. R. L., Ward, I . A., and Flower, P. J., as in reference [49], pp. 614-627. [60], Smith, P. and Stewart, A. T., as in reference [53], pp. 429-435. [61]. Pourbaix, M. and de Zoubov, N., as in reference [40], pp. 112-121. [62], Stroh, A. H., 'Fracture', Ed. Averbach, B. L., Felbeck, D. K., Hahn, G. T., and Thomas, D. A., The Technology Press of MIT, 1959, pp. 117-122. 130 [ 63 ] . Cottrell, A. H . , Trans, of the Met. Soc. of AIME, 1958, pp. 192-203. [ 6 4 ] , Thompson, N. and Wadsworth, N. J., Advances in Physics, Vol. 7, 1958, pp. 71 -172 . [ 65 ] . 'Power Diffraction F i l e ' , Ed. Smith, J. V., ASTM, Philadelphia, Pa., 1963. [ 6 6 ] , Rieux, P. Driver, J., and Rieu, J., ACTA Met., Vol. 27 , pp. 145-153. [ 67 ] . Stanzl, S. and Tschegg, E., Met. Sci., Vol. 14, No. 4 , 1980, pp. 137-143. [ 68 ] . Ritchie, R. 0 . , Int. Met. Rev. Vol. 24, NO. 5 -6 , 1979, pp. 205-230. [ 69 ] . Low, J. R. Jr., as in reference [ 6 1 ] , pp.68-90. 131 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0096297/manifest

Comment

Related Items