Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Corrosion fatigue and pitting behaviour of duplex stainless steels in chloride solutions Sriram, Rajagopal 1989

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

Item Metadata

Download

Media
831-UBC_1989_A1 S68_5.pdf [ 23MB ]
Metadata
JSON: 831-1.0078527.json
JSON-LD: 831-1.0078527-ld.json
RDF/XML (Pretty): 831-1.0078527-rdf.xml
RDF/JSON: 831-1.0078527-rdf.json
Turtle: 831-1.0078527-turtle.txt
N-Triples: 831-1.0078527-rdf-ntriples.txt
Original Record: 831-1.0078527-source.json
Full Text
831-1.0078527-fulltext.txt
Citation
831-1.0078527.ris

Full Text

CORROSION FATIGUE AND PITTING BEHAVIOUR OF DUPLEX STAINLESS STEELS IN CHLORIDE SOLUTIONS by RAJAGOPAL SRIRAM M.A.Sc. (Metallurgical Engineering) The University of British Columbia, 1984 B.E. (Metallurgical Engineering) Indian Institute of Science, 1980 B.Sc Bangalore University, 1977. A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY i n THE FACULTY OF GRADUATE STUDIES DEPARTMENT OF METALS AND MATERIALS ENGINEERING We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA July 1989 (c) Rajagopal Sriram, 1989 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of P E T A L S £ ^TSI^MS ^Gi^^S^I C The University of British Columbia Vancouver, Canada Date AM ZO ' WSJ DE-6 (2/88) i i Abstract The pitting and corrosion fatigue (CF) behaviour of two commercial duplex stainless steels (SS), one cast and other wrought, were studied in chloride solutions. One solution was a simple chloride (1M NaCl) and the other was a synthetic white water containing a lower concentration of chloride, together with oxidized sulphur species (thiosulphate). Differences in composition between the ferrite and austenite phases were determined by micro-analytical techniques. The pitting studies showed that the pitting potentials and the preferential pitting of the ferrite or austenite phases were dependent upon partitioning of the elements Cr,Mo and N between the two phases. Alloying considerations leading to improved pitting resistance were discussed and it was concluded that the beneficial effects of alloyed nitrogen were due to surface enrichment of N atoms. Potentiostatically controlled CF tests were combined with studies of repassivation kinetics to determine the mechanism of CF crack propagation. The crack tip chemistry was maintained under well characterized conditions by using high frequency fatigue testing to produce good mixing between the crack solution and bulk solution. Supplementary experiments confirmed that such mixing was achieved. Near-threshold CF propagation rates were studied with compact tension specimens as a function of cyclic stress intensity, AK, and electrochemical potential. The propagation of cracks was measured by optical microscopy and a back face strain gauge. Repassivation kinetics were studied as a function of potential by using potentiostatically controlled rapid scratch tests and monitoring anodic current transients with a fast response oscilloscope. The experiments showed that near-threshold fatigue crack propagation (FCP) rates in 1M NaCl were influenced by the applied potential. The FCP rates were faster at very cathodic potentials (-1.2Vsce) and very anodic potentials (+0.3 Vsce). At intermediate potentials (-0.4 Vsce) crack propagation rates were slower. However, there was little effect of potential on FCP in synthetic white water. The rapid scratch tests showed that the anodic potential where the fastest FCP rates were observed coincided with the potential at which the peak transient current density was highest and the repassivation rate was most rapid. Fractographic observations showed that at potentials where hydrogen evolution was not possible, the fracture surface features were independent of potential. At cathodic potentials where hydrogen evolution was possible, more interfacial fracture regions were seen. Cracking was completely transgranular at anodic potentials and opposing fracture surfaces contained fine scale (-2000A) interlocking ridges. It was concluded that FCP at anodic potentials was consistent with a restricted slip reversibility (RSR) model of cracking, where potential affects the rate of oxidation of freshly exposed surface which, in turn, controls the degree of slip reversibility at the crack tip. At cathodic potentials, where hydrogen is evolved during fatigue, it was concluded that hydrogen transport and embrittlement processes can increase the rate of fatigue crack propagation. i v TABLE OF CONTENTS ABSTRACT . ii TABLE OF CONTENTS i v LIST OF TABLES viii LIST OF FIGURES ix LIST OF SYMBOLS AND ABBREVIATIONS xiv ACKNOWLEDGEMENTS xvi 1.INTRODUCTION 1 1.1 PAPER MACHINE CORROSION 1 1.2 SUCTION PRESS ROLLS 3 1.3 PITTING OF DUPLEX SS 4 1.4 CORROSION FATIGUE OF SUCTION ROLLS 6 1.5 OBJECTIVES OF THE PRESENT STUDY 8 2.LITERATURE SURVEY 10 2.1 PIT INITIATION AND GROWTH BEHAVIOUR 10 2.1.1 Pitting corrosion of S S 10 2.1.2 Models and mechanisms of pit initiation 11 2.1.3 Mechanisms of pit propagation 13 2.1.4 Factors affecting pitting corrosion of SS 13 2.1.4.1 Composition of the alloy 13 2.1.4.2 Effect of microstructure 15 2.1.4.3 Effect of environment and solution composition 16 2.2 GENERAL CORROSION FATIGUE CRACK INITIATION AND GROWTH 16 2.2.1 Corrosion fatigue 16 2.2.2 Corrosion fatigue crack initiation 19 2.2.3 Corrosion fatigue crack propagation 20 2.2.4 Near-threshold crack growth behaviour 20 V 2.2.5 Models and mechanisms of corrosion fatigue propagation 21 2.2.5.1 Oxidation-related mechanisms 23 2.2.5.2 Hydrogen embrittlement mechanism 24 2.2.6 Solution chemistry at the crack tip 26 2.2.7 Effect of applied potential 27 2.2.8 Potential drop down the crack 27 2.2.9 Effect of load ratio 28 2.2.10 Effect of frequency 29 2.2.11 Two phase alloys 29 3.SCOPE OF THE PROJECT 33 4.EXPERIMENTAL STUDIES ON PITTING OF DUPLEX SS 35 4.1..EXPERIMENTAL PROCEDURE 35 4.1.1. Specimens and Solutions 35 4.1.2 Pitting Tests 39 4.1.3 Microscopical Techniques 40 4.1.4 Micro-Chemical Analysis and Phase Identification 40 4.2 RESULTS 41 4.2.1 PPS Tests 41 4.2.2 Transmission Electron Microscopy 43 4.2.3 Long term exposure tests 47 4.2.4 Nitrogen Distribution 47 4.3.DISCUSSION 51 4.3.1 Alloying Effects 51 4.3.2 Role of Nitrogen 52 4.3.3 Effect of Copper 53 4.3.4 Industrial Considerations 54 4.3.5 Summary 54 v i 5 EXPERIMENTAL STUDIES ON CORROSION FATIGUE AND REPASSIVATION KINETICS OF DUPLEX SS 5 5 5.1 EXPERIMENTAL WORK 5 5 5.1.1 Fatigue Specimen Preparation 55 5.1.2 Fatigue Testing 58 5.1.3 Solution Mixing Test 61 5.1.4 Rapid Scratching Test 61 5.1.5 Fractography 62 5.2 RESULTS 63 5.2.1 Fatigue Crack Propagation 63 5.2.2.Solution Mixing Test 78 5.2.3 Rapid Scratching Tests 81 5.2.4 Fractography 86 5.2.4.1 Near-Threshold Fractography (Stage I) 86 5.2.4.2 Stage II Fatigue Fractography 93 5.3 DISCUSSION 104 5.3.1 General Behaviour 104 5.3.2 Consequences of Solution Mixing 104 5.3.3 Scratch Tests - Repassivation Kinetics 105 5.3.4 Dissolution Mechanism 110 5.3.5 Film-Related models of Corrosion Fatigue 111 5.3.5.1 Film-Induced cleavage (FTC) 111 5.3.5.2 Restricted Slip Reversibility(RSR) 114 5.3.6 Correlation between scratch tests and FCP rates 117 5.3.7 Hydrogen embrittlement 118 5.3.8 Effect of specimen orientation on crack propagation 120 5.3.9 Summary 121 v i i 6. CONCLUSIONS 122 7. SUGGESTIONS FOR FUTURE WORK 124 REFERENCES APPENDIX-1 APPENDLX-2 1 3 5 APPENDLX-3 1 3 6 v i i i LIST OF TABLES Table 1 Corrosion problems in a paper machine 2 Table 2 Effect of fatigue variables on mixing 30 Table 3 Composition of steels 35 Table 4 Composition of synthetic white water 36 Table 5 SEM-EDS analysis of a and y phases in wrought steel 45 Table 6 TEM-EDS analysis of a and y phases in cast and wrought steels 45 Table 7 Constants C and D for oxide growth rates 109 Table 8 Predicted crack velocities by dissolution mechanism 112 i x LIST OF FIGURES Figure 1 Schematic diagram of the wet end of a Fourdrinier paper machine. Figure 2a Cross-section of a suction roll Figure 2b Schematic pseudo-binary diagram of 70% Fe... Figure 3 A typical pitting scan Figure 4 Mechanism of pit propagation Figure 5a Basic modes of loading involving different.... Figure 5b Distributiorfof stresses in the vicinity of crack.. Figure 6a A typical log-log plot of corrosion fatigue crack growth Figure 6b Schematic figure showing possible forms of corrosion fatigue crack growth Figure 7 Mechanisms of crack closure Figure 8 Light micrograph showing microstructures (a) Cast duplex steel (b) Wrought duplex steel Figure 9 Light micrograph showing coarse grain structure of cast duplex steel Figure 10 Single cycle potentiodynamic pitting scans in in 1M NaCl (a) Cast duplex steel (b) Wrought duplex steel Figure 11 Appearance of pitted surfaces (a) Preferential pitting of austenite in cast duplex steel (b) Preferential pitting of ferrite in wrought duplex steel Figure 12 TEM micrograph of cast steel showing oc-oc interface Figure 13 TEM micrograph of cast steel showing ot-y interface Figure 14 SEM micrograph showing preferential pitting of austenite in cast steel in 1M NaCl solution... Figure 15 Effect of 24Hr potentiostatic tests on pitting of cast steel in 1M NaCl solution... (a) -200 mVsce (b) 0 mVsce (c) +200mVsce X Figure 16.Compact tension specimen Figure 17.Protective coatings Figure 18.Three wire connection for temperature compensation Figure 19.Schematic of test set-up Figure 20.Fatigue crack propagation rates for cast duplex steel in 1M NaCl at -0.4 AND 0.3 Vsce Figure 21.Fatigue crack propagation rates for cast duplex steel in 1M NaCl at -0.4 AND -1.2Vsce Figure 22.Fatigue crack propagation rates for cast duplex steel in syn.white water at -0.4 AND 0.3 Vsce Figure 23.Fatigue crack propagation rates for cast duplex steel in syn.white water at -0.4 AND -1.2Vsce Figure 24. Crack deviation due to presence of austenite (A) Optical micrograph (B) SEM micrograph Figure 25.Fatigue crack propagation for wrought duplex SS dessiccated air Figure 26.Fatigue crack propagation rates for wrought duplex steel in 1M NaCl at -0.4 and +0.3 Vsce (T-L orientation) Figure 27.Fatigue crack propagation rates for wrought duplex steel in 1M NaCl at -0.4 and -1.2 Vsce (T-L orientation) Figure 28.Fatigue crack propagation rates for wrought duplex steel in 1M NaCl at -0.4 and +0.3 Vsce (L-T orientation) Figure 29.Fatigue crack propagation rates for wrought duplex steel in 1M NaCl at -0.4 and -1.2 Vsce (L-T orientation) Figure 30.Fatigue crack propagation rates for wrought duplex steel in 1M NaCl in L-T and T-L orientations at +0.3 Vsce Figure 31.Fatigue crack propagation rates for wrought duplex steel in 1M NaCl in L-T and T-L orientations at -0.4 Vsce Figure 32.Fatigue crack propagation rates for wrought duplex steel in 1M NaCl in L-T and T-L orientations at-1.2Vsce x i Figure 33.Fracture surface from the "mixing test" sample (A) Optical micrograph (B) SEM micrograph Figure 34.Preferential pitting of fatigue sample Figure 35.Current transients for cast duplex in 1M NaCl and syn. white water Figure 36.Current transients for wrought duplex in 1M NaCl and syn. white water Figure 37.Cumulative charge passed in cast duplex tested in 1M NaCl Figure 38.Cumulative charge passed in cast duplex tested in syn. white water Figure 39.Cumulative charge passed in wrought duplex tested in 1M NaCl Figure 40.Cumulative charge passed in wrought duplex tested in syn.white water Figure 41.Typical appearance of fracture surface at potentials -0.4 & +0.3 Vsce. (A) +0.3 Vsce da/dN = 7.62 E-H m/cycle AK = 4.29 MPayflT (B) -0.4 Vsce da/dN = 4.54 E-11 m/cycle AK = 6.08 MPajrn Figure 42.Transgranular cracking of cast duplex in 1M Nacl Faint ridges on opposing faces (mirror image) (a) Cast duplex 1M NaCl +0.3 Vsce daMN = 7.62E-11 m/cycle AK = 4.29 MPaJffi (b) Cast duplex 1M NaCl +0.3 Vsce da\dN-7.62E-H m/cycle AK = 4.29 MPa/m Figure 43.Transgranular cracking of cast duplex in 1M Nacl Faint ridges on opposing faces (mirror image) x i i (a)Cast duplex lMNaCl +0 3 Vsce daViN = 7.62E-11 m/cycle AK = 4.29 MPaffS Figure 42.Transgranular cracking of cast duplex in 1M Nacl Faint ridges on opposing faces (mirror image) (High magnification) (a)Cast duplex 1M NaCl +0.3 Vsce da\dN = 7.62E-ll m/cycle AK = 4.29 MPa/m Figure 45(A,B.C).Examples of interfacial fracture in cast duplex at -1.2 Vsce Cast duplex 1M NaCl -1.2 Vsce daMN = 9.14E-10 m/cycle AK = 6.6 MPaJm Figure 46(A,B,C).Transgranular cracking with faint ridges in wrought duplex in 1M NaCl at +0.3 Vsce. Wrought duplex 1M NaCl +0.3 Vsce daNdN = 4.06E-11 m/cycle AK = 3.575 MPa/fTi Figure 47 Transgranular cracking with faint ridges in wrought duplex in 1M NaCl at -0.4 Vsce. Wrought duplex 1M NaCl -0.4 Vsce daMN = 6.77E-H m/cycle AK = 5.80 MPaJm Figure 48.(A,B)-Examples of interfacial fracture in wrought duplex at -1.2 Vsce Wrought duplex 1M NaCl -1.2 Vsce daW = 1.17E-09 m/cycle AK = 4.796 MPaJffi Figure 49(A,B)-Oassical fatigue striations in matching surfaces ;cast duplex SS; 1M NaCl; -0.6 Vsce. x i i i Cast duplex 1M NaCl -0.6 Vsce daUN = 2.11E-07 m/cycle AK = 26.37 MPaM Figure 50.Figure showing logarithmic film growth rates (A) Cast duplex /syn.white water /+0.3Vsce B) Wrought duplex /syn.white /OVsce Figure 51. Schematic diagram of stage I fatigue crack propagation by restricted slip reversibility mechanism Figure 52.Crack propagation through a and c-y interface Figure 53 Schematic of the procedure used for measuring crack closure stress intensity x i v LIST OF SYMBOLS AND ABBREVIATIONS a Crack length a Ferrite phase AES Scanning auger electron spectroscopy b Width of the compact tension specimen P Crack opening angle BFSG Back face strain gauge CF Corrosion fatigue 8 Thickness of the film da/dN Fatigue crack propagation per cycle Ec Corrosion potential EDS Energy dispersive x-ray spectroscopy Epit Pitting potential Eprot Protection potential Esce,EsCE Potential with respect to Saturated calomel electrode Eshe Potential with respect to Standard hydrogen electrode F Faraday FCP Fatigue crack propagation FIC Film induced cleavage y Austenite phase h Over-voltage ia Bare surface dissolution current density K Stress intensity Kclosure Closure stress intensity AK,(AK>rh) Alternating stress intensity .(threshold stress intensity) AKgff Effective stress intensity 1 Boundary free energy/unit length XV M Atomic weight m s Milli-seconds v Surface free energy PPS Potentiodynamic pitting scan PSB Persistent slip bands PTFE Poly-tetra-fluoro-ethylene APjk Alternating near-threshold load P.Pmax.Pmin Load, maximum load, minimum load p Density R Radius of the oxide nucleus RSR Restricted slip reversibility mechanism S Area covered by 1 mole of oxide in a mono-atomic layer Oapp Applied stress a v Yield stress SCF Stress corrosion fatigue SEM Scanning electron microscope t Thickness of the compact tension specimen TCP True corrosion fatigue TEM Transmission electron microscope W Activation energy for oxide growth Z Charge of the metal cation x v i ACKNOWLEDGEMENTS I would like to extend my sincere thanks to Prof.Desmond Tromans for his advice and assistance throughout this project. I would also like to thank Ken Wong and Kurt Bose of the Pulp and Paper centre for their help in securing supplies and machining. Prof E.B.Hawbolt (Metals and Materials engg.) and Prof.D.P.Romilly (Mech.Engg.) loaned their equipment for this project and I would like to thank them for their help. I am also grateful for the understanding and support my wife has given me throughout this project. At the end of the day when nothing seems to work, my daughter has consistendy removed the gloom and made me appreciate that there is lot more to life than failed experiments. Thanks are also extended to the fellow graduate students and the staff at the Department of Metals and Materials Engineering for their help in this project. Finally, I must gratefully acknowledge the financial support of the Pulp and Paper Research Institute of Canada, via the award of a Paprican scholarship. 1 1.INTR0DUCTI0N 1.1 PAPER MACHINE CORROSION Paper machine corrosion is recognized as one of the significant problems contributing to equipment failure in the pulp and paper industry 1. Figure 1 shows a schematic of the wet end of a paper machine. The stock piping brings in the water containing the suspended pulp fibres (white water) from the bleach plant to the head box. The function of the head box is to create intense rnicro-turbulence so as to disperse the pulp fibres and also to distribute the dilute pulp suspension (which has a concentration of ~1 wt%) onto a wire mesh (called the "wire"). The distribution and the amount of pulp suspension flowing onto the wire is controlled by the width of the slice(opening). The wire is usually made of synthetic material (polyester mono-filament). During its travel on the wire the concentration of the pulp suspension is increased to about 15-20 wt%, mostly due to suction of water by the vacuum boxes placed below the wire. At this point the resulting wet paper web is held together by surface tension and passes on to the suction press rolls. As the name suggests, these rolls both compress and suck the moisture from the paper web. At the end of suction roll operation the paper web has about 40% solids content and passes onto drying and calendering stages. Paper machine components which have high failure rates include stainless steel suction press rolls, process piping and paper machine roll journals 1. The chemistry of the white water in paper machines is very complex and varies from mill to mill. The major constituents are chloride ions (upto 300 ppm), thiosulphate (upto 35 ppm), sulphate (upto 600 ppm), dissolved organic solids (upto 1000 ppm) and other dissolved inorganic solids (upto 1600 ppm)2. The temperature of the process water ranges from 40 to 50 deg.C and its acidity ranges from a pH of 4 to 6. Table 1 lists the types of corrosion problems encountered by the various components in a paper machine. These components are exposed to white water by splashing and immersion. The chlorides and thiosulphate species cause rapid corrosion of 2 Figure 1 Schematic diagram of the wet end of a Fourdrinier paper machine1. | COMPONENT MATERIAL CORROSION PROBLEMS Stock piping 304L 316L Rarely a problem Occasional pitting and microbiological problem Head box 316L 317L Alloy 276 Pitting, Crevice and Microbiological corrosion Maintenance free Slice Lip 317L 254 SMO Pining can Occur No problems Wire Polyester Mono filament No problems Suction press Rolls Bronzes and SS Pining, Crevice Microbiological and Corrosion fatigue Table 1 Corrosion problems in a paper machine. • 3 plain carbon steels and localized corrosion of stainless steels. The corrosion-related failures of suction roll shells represent the most serious materials and corrosion problem in a modern paper machine 1. 1.2 SUCTION PRESS ROLLS Suction press rolls are used to remove the water from the wet paper web by passing it through a roll nip, one roll of which is the suction roll. These rolls, which can measure 2m in diameter by 12m in length and upto 100 mm thick, rotate at speeds ranging from 100 to 600 R P M . The rolls are either cast (centrifugal casting) or fabricated from wrought material (longitudinal welding). Figure-2a shows the cross section of a suction roll. The water is sucked through perforations in the roll by means of a vacuum (low pressure) applied to the inside of the roll. A typical paper machine may use 3 or more suction rolls (couch roll, wringer roll, pickup or press roll). The traditional material for rolls were cast bronze because of cost, castability, machinability and corrosion resistance. However, due to their low stiffness and strength, bronze was not used for the latest generation of paper machines which were designed for faster machine speeds, higher nip pressure and wider rolls. Other complications have arisen from the trend of the pulp and paper industry towards recycling of process effluents (mill closure), together with new procedures for brightening paper. These have caused paper machine components to be subjected to an increasingly corrosive environment (white water). Chloride levels have increased and oxidized sulphur species (eg., thiosulphates) are frequently present These, together with the higher loading stresses in modern machines, have increased the tendency to encounter corrosion fatigue (CF) cracking of suction press rolls. Efforts to minimize the fatigue problem have accelerated the move from bronze roll material towards stainless steel (SS). These include conventional austenitic and martensitic SS and more recently, the newer duplex SS. Austenitic SS (Type 304 and 316) have good corrosion resistance, but have 4 limited fatigue strength due to their low yield strength (Typically around 240-270 MPa). Ferritic stainless steels (Type 430 and 446) have low toughness and ductility. Their low toughness causes high fatigue crack propagation rates. However, they are more resistant to chloride stress corrosion cracking. Martensitic SS (Type 410) has higher yield strength (800-1000 MPa) and good toughness in the tempered condition but it has lower corrosion resistance. Austeno-ferritic duplex steels have been developed to get the best properties of both austenitic and ferritic SS; i.e., they have the corrosion resistance, ductility and toughness of austenitic SS and a high resistance to chloride stress corrosion cracking like ferritic stainless steels. The strength levels are intermediate between the two grades. However, localized corrosion (eg., pitting and crevice corrosion) is a common problem with all stainless steels in white water systems and is aggravated by the changes that have occurred in the white water chemistry. 1.3 PITTING OF DUPLEX SS Duplex steels are composed of austenite (y) and ferrite (a) phases and their pitting behaviour may differ from single phase alloys due to solute partitioning effects and the presence of a/y interfaces. Austeno-ferritic duplex SS can soUdify with an austenitic matrix or ferritic matrix, depending on the composition of the melt (particularly the chromium equivalent and nickel equivalent) as shown in Figure 2b. There will be preferential partitioning of Cr.Mo and Ni between the phases; i.e., Cr and Mo preferentially partition to ferrite while Ni partitions to austenite. Nitrogen content is known to affect the corrosion resistance of duplex steels3,4. The lower solubility of N in the a-phase suggests that N additions should have a preferential effect on pitting resistance of the y-phase-\ In an aqueous environment like white water (which contains chlorides and thiosulphate), these partitioning effects could lead to preferential pitting of one phase. If that phase happens to be the matrix (ie.,the continuous phase), fatigue crack initiation may occur more easily because of the stress raising characteristics associated Figure 2a Cross-section of a suction roil. — PRIMARY AUSTENITE—f PRIMARY DELTA FERRITE F e - C r - N i PSEUDO-BINARY CD I CD CHROMIUM - NICKEL Figure 2b Schematic pseudo-binary diagram for 70% Fe showing structures developed in Fe-Cr-Ni alloysl07. • ^ 6 with pits. It is becoming more widely recognized that localized corrosion is a major cause of fatigue crack initiation and premature failure of SS suction rolls, because the locally corroded sites act as stress raisers. Pits caused by aggressive white waters have been associated with fatigue cracking by Nathan6, Bowers^  and Yeske8. It has been recognized that between suction press roll alloys having equal residual stress, localized corrosion resistance (eg., pitting) could become the detennining factor in CF crack initiation and this could influence the alloy selection^ Consequently, pitting behaviour may be as important a consideration as mechanical properties when selecting optimum SS alloys for suction rolls. This leads to the necessity for more complete information on the pitting behaviour of duplex steels, because their suitability for suction press roll application is now being made primarily on mechanical property considerations. 1.4 CORROSION FATIGUE OF SUCTION ROLLS In a suction press roll, the line of contact at which the wet web is compressed to remove moisture experiences a large compressive force while the interior experiences a large tensile force. This induces a bending stress field which reaches a maximum at the point of contact with the paper web and a minimum at the opposite side of the rolls. Therefore, every point in the roll experiences cyclic loads. The largest bending moments and highest cyclic stresses are experienced by the middle two-thirds of the rolls. These cyclic stresses and the presence of a corrosive environment (white water) are responsible for CF cracking of the rolls. Corrosion fatigue failures of suction press rolls constructed from SS, including Duplex stainless steels, are not unknownlO. Bowers^ , has reviewed the literature on CF failure of suction roll alloys. Yeskel 1 has reported that in duplex SS, crack initiation and growth often occurred without any visible evidence of corrosion. Circumferential cracking usually occurs in these rolls. These cracks mostly initiate at the inside surface and almost all cracking is confined to the middle two thirds of the shell. Most industrial components that are subjected to fatigue, including suction 7 press rolls, spend the majority of their lifetime in the near-threshold region where the crack velocity is typically < 1X10-9 m/cycle. This is also the region where the environment plays an important role in fatigue cracking. Therefore, a mechanistic understanding of CF cracking in duplex SS alloys which are used for suction press rolls would contribute to the efforts in solving the problems of roll failure. Very few mechanistic studies on near-threshold CF cracking have been conducted on austeno-ferritic duplex SS, whereas considerable work has been conducted on austenitic and ferritic types of SS12-18. The proposed mechanisms by which CF occurs can be broadly classified into those which are related to oxidation events at the crack tip and those which are related to the generation of hydrogen and subsequent embrittlement of the region ahead of the crack tip. Oxidation mechanisms include models describing alternate cycles of film rupture, electrochemical dissolution and repassivation, plus a more recent model of restricted slip reversibility at the crack tip 12. The identification of either hydrogen or oxidation mechanisms of FCP is not always easy, because the fatigue crack may act as an occluded cell (eg.crevice) causing marked deviations in local solution from that of the bulk environment. One of the consequences of occluded cell behaviour in SS is a decrease in the local pHl9. Hence, attempts to use potentiostatic experiments to control the electrochemical environment and hydrogen generation in the crack are frustrated by the imprecise knowledge of the crack solution chemistry. Theoretical considerations20 and experiments^  suggest that good mixing may be obtained between the crack solution and bulk solution at high loading frequencies, due to the rapid pumping action of the opposing crack surfaces. Under these circumstances it should be possible to use potentiostatically controlled CF tests to either prevent or allow hydrogen generation, based on thermodynamic considerations of solution pH and electrochemical potential. Thus, the problem of distinguishing between hydrogen-related and oxidation mechanisms of CF becomes more tractable. The effect of aqueous environments on FCP behaviour of duplex SS is 8 expected to be complicated by the two phase structure. For example, ferrite is prone to hydrogen embrittlement and may provide an easier crack propagation path if the CF mechanism involved hydrogen. On the other hand, Cr and Mo alloying elements preferentially partition in the ferrite phase21 and lower the resistance of the austenite phase to breakdown of passivity and pitting corrosion. Consequently, crack propagation may be easier in the austenite phase if the CF mechanism involves passivity breakdown and dissolution processes. Therefore, in order to study the mechanistic aspects of FCP in duplex SS, it is advantageous to be able to conduct experiments under conditions where the presence or absence of hydrogen evolution at the crack tip is unambiguous. Repassivation kinetics, including nucleation and growth of oxide films, are important in all oxidation-related models of CF. Knowledge of these processes is necessary for a more complete understanding of anodic oxidation effects on FCP. Potentiostatically controlled rapid scratch tests22 provide a simple means of studying nucleation and film growth kinetics as a function of electrochemical potential. Furthermore, provided the response time of the technique is made sufficiently sensitive, the method should be capable of providing information on film nucleation. In this manner, the charge required to repassivate the scratched surface may be correlated to the rate at which the film grows at different potentials. These data may then be correlated with FCP behaviour. 1.5 OBJECTIVES OF THE PRESENT STUDY The present study was concerned with an evaluation of the pitting and CF behaviour of two duplex steels; one cast and other wrought having different a/y phase proportions and different alloy chemistry. The environments of interest were a neutral chloride solution (1M NaCl) and a solution containing chlorides and oxidized sulfur species (synthetic white water). A major objective of the pitting work was to provide a better understanding of the factors, particularly phase composition, controlling pitting behaviour of duplex 9 steels. From this, the relationship between alloy chemistry and service performance of duplex steels in white water may be understood more fully. Fatigue crack propagation studies were done in both cast and wrought duplex steels to determine the presence and mechanism of anodic oxidation enhanced FCP and hydrogen assisted FCP in the near-threshold region. In the wrought alloy, testing was done parallel and normal to the rolling direction. Testing conditions were chosen that gave good solution mixing and either allowed or prevented hydrogen evolution at the crack tip. The effects of applied cyclic stress intensity (AK) and electrochemical potential on FCP rates were studied at high cyclic frequency using potentiostatically controlled fracture mechanics testing techniques. Crack retardation effects due to premature contact of the opposing fracture surfaces on the unloading cycle (crack closure) were monitored and crack closure stresses were measured. Tests were conducted to assess the effectiveness of high frequency loading on the mixing of bulk and crack solutions. Rapid scratch tests were conducted in order to determine the kinetics of film formation in relation to the time spent by the crack tip in the opening and closing mode of the load cycle. The effects of potential on FCP, crack fractography and film formation were all considered in the determination of a mechanistic model of cracking. 1 0 2.LITERATURE SURVEY 2.1 PIT INITIATION AND GROWTH BEHAVIOUR 2.1.1 Pitting corrosion of SS Stainless steels rely upon the presence of a protective (passive) oxide film for corrosion resistance. Any localized breakdown of this film leads to rapid localized corrosion. Pitting is a form of localized corrosive attack that is destructive to engineering structures by causing perforation of equipment. Chloride ions are well known to aggravate pitting and are present in many environments. The successful use of stainless steel equipment in chloride-containing solutions depends on their ability to resist localized attack. Pit initiation occurs at or above a critical potential called Epit (the breakdown potential). There also exists a lower critical protection potential, Eprot* below which growing pits cease to propagate. Between Eprot and Epit, new pits cannot initiate but existing pits continue to grow. Pitting characteristics are often studied by electrochemical techniques. Both Epit and Eprot for an alloy-environment combination can be determined by a single cycle potential pitting scan, as shown in Figure 3, where the arrows show the direction of the scan. Whether or not pitting occurs during service depends on the position of Ec (corrosion potential) relative to Epit. If Ec is less than, but close to Epit then the addition of even a small amount of oxidizers (ferric or cupric ions) can produce pitting by raising Ec above Epit. If Ec is significantly active to Epit then the tendency for pit formation is decreased. Many other test techniques to assess the pitting resistance of alloys by electrochemical techniques have been proposed and reviewed23. Chemical methods are also used for assessing pitting corrosion behaviour of metals and alloys. The solution consists of an activator (Cl- ions) and an oxidizing agent in known concentrations. Although this method cannot be used to predict pitting 1 1 behaviour in actual service environments, it can be used to rank the alloys. The best known chemical method is the ferric chloride test (ASTM standard G48) in which specimens are immersed in 6% FeCl3 test for a definite interval of time. Brigham24 has proposed a modification of the ferric chloride test where temperature is used as the ranking parameter. The minimum temperature at which pitting occurrs on a sample exposed for 24 hours in 10% FeCl3 solution is defined as the critical pitting temperature(CPT). However, the correlation between results obtained from the ferric chloride test and pitting resistance in practical situations is not considered good23. Natural exposure of a metal or alloy to a given environment is obviously the most reliable way to determine the pitting susceptibility. However, the main drawback is the length of time the test takes and the difficulty in interpreting the results. For example, the induction time for start of pitting is difficult to detect because it requires a visual observation of pits. 2.1.2 Models and mechanisms of pit initiation The mechanism by which pit initiation occurs is a matter of controversy. Some of the main theories that have been used to explain pit initiation are summarised below. Adsorption Mechanism: Some of the proponents of adsorption mechanism25,26, consider passivity to be an adsorption of an oxygen layer on the metal surface. Pit initiation is caused by reversible competitive adsorption between chloride ions and oxygen for sites on the SS surface. The pitting potential is considered to be the minimum electrode potential at which the aggressive chloride ions are able to reversibly replace the passivating oxygen on the metal surface. The anodic overvoltage for the dissolution of alloy ions is appreciably reduced whenever the metal is in contact with CI- compared to metal in contact with adsorbed O and hence metal ions rapidly enter solution resulting in pit initiation. 1 2 Ion Penetration Mechanism: The passive film in these models is considered to be an oxide layer of finite thickness and not an adsorbed oxygen film. Rosenfield and Danilov27 suggested that the breakdown of the oxide film was caused by the exchange adsorption of 02" by CI- ions at sites where metal oxygen bonding is weakest. These CI- ions then penetrate into the passive film and concentrate at the metal-film interface causing pit initiation. Hoar28 has proposed a mechanism of anion penetration without oxygen exchange. He suggested that under the influence of electrostatic field across the oxide/solution interface, CI- adsorbs on the oxide film surface. Subsequently, due to its small diameter, CI- ions penetrate through protective oxide film. Breakdown occurs when these anions reach the bare metal surface. The most important consequence of penetration mechanisms is the incorporation of CI- in the passive film. While some authors have reported the presence of CI- ions29,30, others have reported the absence of Cl-31. Hoar32 has also suggested a "mechanical" model for anion penetration. The adsorbing anions were postulated to replace water in the passive film. Due to mutually repulsive forces between anions the interfacial tension or interfacial free energy of the oxide/solute interface is reduced. The adsorbed anions push one another and the oxide to which they are strongly attached comes apart. Any resulting crack or split facilitates further entry of CI- which marks the process progressive. A different mechanical model for passivity breakdown has been suggested by Sato33. The initially formed thin anodic oxide films are stabilized by the surface tension. But with increasing thickness, surface tension decreases. Thus a "critical thickness" has been postulated above which the film breaks down. Anion (C1-) adsorption lowers both the surface tension and "critical thickness" for the film to break. Therefore increasing anion concentration leads to much easier breakdown of passive film. 13 2.1.3 Mechanisms of pit propagation Pit propagation is an autocatalytic process. The mechanism of pit propagation 19 is shown schematically in Figure 4 for stainless steel pitting in neutral aerated chloride solutions. Metal dissolution occurs in the pit, M=M+ + e-, (1) and is balanced by the cathodic reduction of dissolved oxygen on the adjacent exterior surface, 02 + 4H20 + 4e = 40H- (2) The high concentration of cations leads to hydrolysis and lowering of pH as shown in equation 3. Mn+ + xH20 - M(0H)x(n-x)+ + xH+....(3) The imbalance in charge distribution leads to anions (principally CI") to migrate into the pit. Both chloride and hydrogen ions stimulate the dissolution of metals and alloys and makes repassivation difficult. Therefore, the entire process accelerates with time. Various test techniques are used to assess the pitting tendency of an alloy 23,34. Crevice corrosion, which is similar to pitting in terms of possible mechanisms for film break down (and differs from pitting in terms of geometries involved) has been mathematically modelled35,36. 2.1.4 Factors affecting pitting corrosion of SS 2.1.4.1 Composition of the alloy The effect of alloying elements on pitting of SS has been studied by investigators and reviewed34,37-43. Various alloying elements like Cr, Mo, Ni and N are known to be beneficial in both experimental and commercial alloys. Chromium, nickel and molybdenum are the three alloying variants that are used to prepare many proprietary alloys. In chloride solutions, increasing the chromium content of iron-chromium alloys increases the pitting potential, which thereby expands the passive potential range41. Although Ni is known to move the pitting potential in the noble 14 Figure 4 Mechanism of pit propagation. 15 direction, in the range of Ni contents found in stainless steels, its beneficial effect on the pitting potential is not very pronounced41. Brigham24 has confirmed the beneficial effect of Mo on pitting using the ferric chloride immersion test. The alloy containing the highest amount of molybdenum showed the highest critical pitting temperature. Boron, Carbon and copper are described as having variable effect4^ depending on whether they are present in solid solutions or precipitated as intermetallic compound or carbides. Nitrogen is considered beneficial in improving the pitting resistance of stainless steel34,43. Generally, the main effect of alloy composition is to control the stability of protective passive film. 2.1.4.2 Effect of microstructure Phases such as sulfides, delta ferrite, sigma and alpha prime, and sensitized grain boundaries and the heat affected zone near the weld, can have a deleterious effect on the pitting resistance of SS. Soluble MnS inclusions are known to act as pit initiation sites45. it has been pointed out by Henthome465 that the chromium content of MnS inclusions, or lack of it, determines their ability to behave as pit initiation sites. Low Mn steels contain CrS (chromium sulfide) inclusions which are thermodynamically stable and less prone to act as initiation sites for pitting. In steels containing higher manganese, an iron manganese spinel is formed (FeMn)Cr2S4 which act as an effective nucleation site for pit initiation. Delta ferrite in austenitic stainless steels is considered detrimental to pitting resistance47. Dundas et al47, measured the pitting resistance of austenitic stainless steel in three conditions, (i) free of ferrite, (ii) delta ferrite formed after heat treating at 1345 deg.C and (iii) delta ferrite that formed at 1345 deg.C and had been removed by annealing at 1120 deg.C. They found that the corrosion rate of steel containing delta ferrite was very much higher compared to ferrite-free conditions. Sensitized grain boundaries act as sites for preferential nucleation of pits in both austenitic SS48 and ferritic stainless steel49. Cast stainless steel CF8M (wrought 16 equivalent is 316SS) and CD4MCu (wrought equivalent is 329SS), both of which contain a second phase, undergo selective attack on the austenite phase when exposed to 10% FeCl3 test50. Preferential dissolution of austenite is seen in a two phase, low nitrogen austeno-ferritic duplex SS when exposed to sulphuric, phosphoric, hydrochloric and oxidizing chloride environments. However, in high nitrogen duplex SS, austenite preferentially dissolves in sulfuric and phosphoric acid, whereas ferrite dissolves in hydrochloric and oxidizing chlorides^ . 2.1.4.3 Effect of environment and solution composition Pitting in pulp and paper environments is aggravated by the presence of chloride and hypochlorites (which is found in bleaches) 19. Increasing the chloride concentration increases the tendency for pitting. Oxidizing cations like cupric, ferric and mercuric in chloride solutions are also aggressive. Anions that inhibit pitting by raising the pitting potential in the noble direction are nitrates, sulphates, hydroxides, carbonates and chromates45. The inhibiting action depends on their concentration and concentration of chlorides in solutions26. Temperature, pH and velocity of environment also affect the pitting behaviour of alloy. The effect of thiosulphate on the pitting of stainless steels (Type 304 and 316) in white water system has been studied by Newman et al.51. This study proved the existence of a critical anionic concentration ratio (Ratio of sulphate to thiosulphate) for different ionic strengths, for pitting to occur. It was also proved that chloride was not necessary for pitting of 304SS. Bisulfite (HSO3-) was found to inhibit pitting when present in the same concentration as sulphate (S042-). 2.2 GENERAL CORROSION FATIGUE CRACK INITIATION AND GROWTH BEHAVIOUR 2.2.1 Corrosion fatigue Corrosion fatigue is the time-dependent propagation of sub-critical cracks under the conjoint action of cyclic loads and a corrosive environment. Corrosion fatigue failure is known to occur in many industries including chemical, aerospace, 17 petrochemical and pulp and paper 1. Many reviews have been published on the subject52,53,54 Generally, air fatigue data is used as a basis of comparison although laboratory air is certainly not inert and is known to reduce the fatigue strength compared to data in vacuo55,56,57. Before the use of fracture mechanics, most of the data on corrosion fatigue was reported in terms of the effect of the corrosive environment on the form of the S-N curve (stress Vs. no. of cycles to failure). With the application of fracture mechanics to cracked components58,59, more meaningful comparisons could be made between fatigue crack propagation rates in corrosive environment (including air) and in vacuo. Briefly, the application of fracture mechanics involves the characterization of the stress field ahead of a sharp crack by a parameter called the stress intensity factor K. For the opening mode crack situation, as shown in Figure 5a, where the crack plane is normal to the applied load, the stress intensity is designated as Ki and the stresses at the crack tip are given by the following equations58, 9 39 cos^-l 1 +sin-r$in-=-Vlitr 2 I 2 2 K 9\ lltT 2 L - = r c o 4 f I -«n4sin3? .(4) K . 9 9 39 sin J COS J COS -y Vlvr where the stresses, angles (9) and displacement (r) are defined in Figure 5b. Clearly, the stresses at the crack tip are determined by Ki, which is seen to have the units (stress)(length)l/2. Note that Ki is independent of specimen geometry. There are many different formulae that have been developed by theoretical and experimental measurement to relate Ki to the applied load P and the specimen geometry. In all cases Ki is proportional to P. For example, the stress intensity for a central through thickness crack of length 2a in a infinite plate of width W and thickness t is given by Ki = (P/Wt) K a1N2. ..(4a) As the crack approaches the edge of the plate, this is modified.eg., Ki = (P/Wt) K a1/2 f(a/W)...(4b) 18 M o d a l Mode II Mode I I I Figure 5a Basic modes of loading involving different crack surface displacement59. Figure 5b Distribution of stresses in the vicinity of crack tip59. 19 More complex crack/loading geometries lead to more complex formulae for KT11**. The critical stress intensity at which mode I crack propagates in an unstable manner (catastrophic crack growth) is designated as KiC- In all cyclic fatigue tests, the maximum stress intensity,( K m a x ) , minimum stress intensity, (Kmin) and AK (Km^-K^n) are important variables. Additionally, mean stress, which is related to the R-Ratio (R=Kmin/Kmax=Prrnn/Pmax) is also important. Environmental variables like pH, temperature, composition and electrochemical potential play a crucial role in determining the corrosion fatigue behaviour in aqueous environments. Corrosion fatigue data is usually plotted logarithmically in the form of crack growth/cycle (da/dN) versus applied alternating stress intensity, AK, as shown in Figure 6a. 2.2.2 Corrosion fatigue crack initiation In smooth specimens free of stress raisers, fatigue cracks usually nucleate at intrusions and extrusions in persistent slip bands59. The relationship between nucleation sites and the presence of corrosive environment has been studied in order to understand the mechanism of CF nucleation. Two possible mechanims that have been postulated are (i) Slip dissolution reverse slip model60,61 and (ii) Enhanced slip model61,63. in the former, the passive oxide film on the surface is ruptured and fresh metal exposed due to the emerging slip step. This fresh surface, being highly anodic, dissolves rapidly. When the slip reversal occurs due to the reverse fatigue loading cycle, a notch is formed. In the second model, localized selective electrochemical dissolution of work hardened region which is produced due to dislocation interactions, effectively soften specific surface regions leading to accelerated crack initiation. Thus, this model consists of metal removal from specific highly deformed regions in the metal surface. Both the mechanisms depend on characteristics like adhesion, strength, structure and composition of the passive film. Congleton et al.53, quoted from the work of McAdam that pitting was likely to be associated with early initiation of corrosion fatigue cracks. Congleton also found 20 that a major reduction in fatigue strength occurred with a few days of pre-corrosion. Both aluminum and BS 4360-50D steel showed a dependence of crack initiation on pit formation in 3.5% NaCl solution. However, the pH of the solution was found to be a critical factor53. Conversely, many workers61,64 have concluded that crack nucleation is not enhanced by the presence of corrosion pits. It was also concluded that for steels, pits observed after failure were not the cause of corrosion fatigue but rather the result of it61. The contradictory interpretation has been explained53 on the basis that several mechanisms are possible and the one with the fastest kinetics will prevail. For example,Moskovitz et al.65, have invoked the enhanced slip model to account for rapid initiation in duplex SS (Alloy 63). 2.2.3 Corrosion fatigue crack propagation The fatigue crack propagation is usually divided into 3 stages as shown in Figure 6a. Briefly, detectable air fatigue crack growth occurs only above a threshold cyclic stress intensity factor (AKTh)- With increase in AK above AKj^, the crack grows quite rapidly (Stage I). At still higher AK, log da/dN is linearly dependent upon log AK (Stage II). In this stage fatigue crack growth is related to AK by a simple power law, da/dN = CAK n (5)66 where C and n are material constants. With further increase in AK crack growth again increases more rapidly as K J C is approached. Cracking is crystallographic in stage I. In stage II and ni both air and corrosion fatigue crack growth occurs by ductile striation mechanism. 2.2.4 Near-threshold crack growth behaviour Studies of ultra-low crack growth rates (< 10-9 m/cycle), where the alternating stress intensity,(AK), approaches the threshold stress intensity,(AKTh), are important for two reasons. Firstly, the majority of components including suction press rolls, spend most of their lifetime in this regime. Designers of such components need to have reliable engineering data to choose material that could withstand low amplitude loadings for 1011 to 101 2 cycles. Secondly, the mechanistic aspects of crack growth in this stage need to be elucidated. Ritchie67 has critically reviewed the literature on near-threshold crack propagation in steels. Near-threshold crack propagation is affected by numerous factors. These include mechanical considerations such as mean stress (load ratio), frequency, waveshape and material strength; Microstructural factors like grain size; and environmental factors that include temperature, pressure, and environmental species. The effect of some of these variables will be dealt with in subsequent sections. Very little work has been conducted on the near-threshold behaviour of duplex stainless steel. To the author's knowledge, only one such study has been reported in the open literature by Bassidi et.all^l. However, the electrochemical conditions for cracking were neither controlled nor reported. 2.2.5 Models and mechanisms of corrosion fatigue propagation There are three simplified general forms of log da/dN vs log A K corrosion fatigue curves as shown in Figure-6b(i,ii,iii). The form of the curve depends on whether or not there is a synergistic effect between the corrosion environment and the alternating stress intensity which is independent of any stress corrosion cracking (SCC) mechanism and independent of the value of KisCC i n relation to A K ^ . Figure 6b(i) is typical of a system in which stress corrosion cracking does not occur under static loads and is called true corrosion fatigue. Figures 6b(ii) and 6b(iii) are typical of systems which shows stress corrosion cracking under static loads. In Figure 6b, environment enhancement of crack growth is only seen when the alternating stress intensity range, A K , is such that the maximum cyclic stress intensity exceeds K J S C C -This is called stress corrosion fatigue. Figure 6b(iii), shows a system in which environmental effects are seen even when K m ax is less than KisCC- To explain stress corrosion fatigue, Wei and Landes68 proposed the superposition model in which the total corrosion fatigue propagation rate da/dN (CF) is the sum of the contribution from SCC and mechanical fatigue. 22 Z -•o 73 -o a> u >> CJ o o J£ o co O Stage Ul / Stage n / Stage I j AK-rh Aoolied Alternatina Stress Intensity Figure 6a A typical log-log plot of corrosion fatigue crack growth. Log (AK) Log (AK) Log (AK) Figure 6b Schematic figure showing possible forms of corrosion fatigue crack growth; TCF: True corrosion fatigue; SCF: Stress corrosion fatigue 23 (da/c^cF = (da/dN) s c c + ( d a / d N ) f a t . g u e ( 6 ) Obviously, this model cannot account for true corrosion fatigue behaviour (i.e., for material environment systems which do not exhibit stress corrosion cracking). However, Austin & Walker69 have shown that for certain metal/environment systems changing the frequency or R-Ratio changes the behaviour from Figure 6b(i), (True Corrosion Fatigue) to Figure 6b(iii),(Stress corrosion Fatigue); i.e., by increasing frequency or decreasing R-Ratio the contribution of stress corrosion can be decreased and eventually nullified. This led them to propose the process competition model; i.e., the processes of stress corrosion and corrosion fatigue are mutually competitive. The crack will propagate by the fastest available mechanism at the pertinent combination of stress intensity, frequency and R-Ratio. In studying alloys which are highly susceptible to SCC (Aluminum alloys in 3.5% NaCl solution) a process interaction model has been proposed^ O. in this model, the effects of predominantly mechanical and predominantly electrochemical crack propagation processes are added, but each is modified to account for the presence of the other. It can be logically argued that the phenomenon of true corrosion fatigue is due to synergistic action of the environment and cyclic loading at the crack tip. The main mechanisms by which the environment may affect the crack growth are by either oxidation related mechanisms or by hydrogen embrittlement. In the former, metal is oxidized to form a soluble ionic species (dissolution) or to form an oxide film and in the latter cathodic generation of hydrogen adsorbs and later absorbs, embrittling the metal. 2.2.5.1 Oxidation-related mechanisms The two main oxidation related mechanisms are (i) alternate cycles of film rupture, dissolution and repassivation mechanism and (ii) restricted slip reversibility mechanism. FILM RUPTURE AND DISSOLUTION MECHANISM In this mechanism the protective film at the crack tip is ruptured by the cyclic 24 strain produced by the alternating stress at the crack tip. The bare surface thus produced acts as an anode while the rest of the metal serves as cathode. Because of this, highly localized anodic current densities are maintained. Ford71,has related the maximum theoretical crack propagation rate V, to bare surface dissolution current density, i a , under activation control, V = M ia/ZFp (6) where M is the atomic weight of the dissolving metal of density p, Z = charge of the metal cation, F = Faraday's constant, 96500 coulombs/gm.equivalent. The anodic dissolution can also help by dissolving the cyclically strain hardened surface at the crack tip and thereby increasing the gliding distance of the activated slip system. RESTRICTED SUP REVERSIBILITY MECHANISM (RSR) In this mechanism it has been proposed that the bare metal that forms due to slip during the forward cycle in a fatigue test is covered with oxide patches 12. Therefore, in addition to work hardening effects on the slip plane, these oxide patches restrict slip reversibility along the same plane during the reverse cycle. The average increment by which slip is restricted constitutes the average crack growth per cycle. The model of RSR has been used to account for decreased crack growth rates in alloys with low stacking fault energy (SFE)72,73,74. in alloys with low SFE, (eg.Cu-13.5 at% Al) planar slip is seen. Slip planarity is thought to promote slip reversibility by preventing cross slip, which in turn reduces the per cycle crack tip growth increment resulting in lower crack growth rates. In alloys with high SFE (eg.Cu), due to ease of cross-slip on planes at the crack tip (wavy slip), slip reversibility is low which results in higher crack growth rates. 2.2.5.2 Hydrogen embrittlement mechanism The effect of hydrogen on deformation and fracture in austenitic, ferritic and martensitic SS has been recently reviewed75. The solubility and the movement of 25 hydrogen in stainless steels are controlled primarily by the crystal structure (FCC or BCC). The severity of hydrogen damage correlates well with the crystal structure, with BCC structures showing higher damage. Austenitic SS is affected less by hydrogen than are ferritic or most precipitation-hardenable stainless steels. This has been related to (i)the generally lower yield strength of austenitic steels, (ii)their higher hydrogen solubility and (iii)lesser tendency for hydrogen segregation75. Various mechanisms have been proposed to explain hydrogen embrittlement in steels. They are: (1) Pressure Theory-J^ According to this model cracking is due to diffusion of atomic hydrogen into the metal and its accumulation as molecular hydrogen in voids or at internal surfaces in the alloy. With the increase of concentration of hydrogen at these places, a high internal pressure is generated which enhances void growth or initiates cracking. (2) Surface Adsorption Theory-JI It has been proposed that cathodic generation of hydrogen can embrittle metal by direct adsorption at the crack tip to lower the surface free energy(v). This leads to a lowering of fracture stress (of) as shown by equation (7) af^Ev/b) 1/ 2 7 where E = Young's modulus and b = interatomic spacing. (3) Decohesion: 78 The maximum cohesive forces between metal atoms located in the region just ahead of the crack tip are reduced due to accumulation of high concentrations of atomic hydrogen which diffuse along the hydrostatic strain gradient. Therefore, local maximum tensile stress perpendicular to the plane of the crack becomes greater than the lattice cohesive force resulting in fracture. (4) Enhanced Plastic Flow: 79 This model proposes that the reduction in cohesive forces by adsorbed and 26 absorbed atomic hydrogen enhances dislocation motion, generally screw dislocations, and the creation of dislocation at surfaces and or crack tips leading to softening of the metal. This model has been based on fractographic observations using high resolution microscopy where crack tip plasticity has been noted on what appears to be brittle cleavage. In order to find the particular mechanism which is responsible for corrosion fatigue crack growth (either anodic or cathodic) it is necessary to define the electrochemical conditions at the crack tip. These conditions include crack tip pH and the electrochemical potential at the crack tip. 2.2.6 Solution chemistry at the crack tip A crack is a shielded site and resembles an occluded cell. Modification of solution chemistry at occluded cells is well known and is relevant to pit growth and stress corrosion cracking. Therefore corrosion fatigue processes inside CF cracks should be based on local crack tip chemistry and pH rather than bulk solution chemistry and pH. However, in corrosion fatigue, pumping of electrolyte into and out of the crack tip occurs due to alternate opening and closing of crack faces and solution mixing is possible. Hartt et al.20 have analyzed theoretically the effect of fatigue variables (frequency, mean stress, crack opening angle, crack length etc.) on mixing between crack and bulk solutions. Table 2, gives the effect of various fatigue variables on mixing. Notwithstanding the theoretical analysis, it is necessary to experimentally prove the occurrence of crack solution modification. Turnbull et al.80 have acknowledged that experimental determination of solution composition and electrode potential is inherently difficult due to restricted geometry of the crack. They add that mathematical models can be usedS 1 which, even though mathematically complex, represent a simplified picture of real events occurring at the crack tip. However, if mixing occurs, but is incomplete, the crack solution composition remains uncertain. The merits of experimental techniques (such as) freezing technique, extraction of solution from the crack tip, use of artificial 21 crack that have been developed to determine the solution composition and electrode potential at the crack tip have been reviewed82. A technique of inserting electrodes at holes drilled to intersect the growing crack has been developed80 to measure crack tip pH and potential. Nevertheless, the only effective way to deal with uncertainities in the crack solution composition is to ensure that the solution chemistry at the crack tip is made equivalent to the bulk chemistry by testing under conditions where efficient mixing is possible. One of the easier methods to effect good solution mixing is by increasing the frequency of the corrosion fatigue test. This is because, at higher frequencies, due to the rapid movement of the opposing crack faces, liquid is expelled from the crack during the decreasing load (closing) portion of the cycle and fresh liquid is drawn in during increasing load (opening) portion of the cycle. Hartt et al.20 have predicted that the mixing increases linearly with frequency. 2.2.7 Effect of applied potential The influence of applied electrochemical potential on fatigue crack propagation rates of materials in aqueous environments has not been investigated in detail. Almost all corrosion fatigue crack growth behaviour in different aqueous environments has been studied under freely corroding conditions. For example, although CF crack propagation in duplex SS in synthetic white water has been studied using fracture mechanics techniques! 1,105,106, none of the work was done under potential control. The effect of different applied potentials on CF experiments may shed some light on the underlying mechanism if combined with fractography. A few studies on the effect of cathodic potential on fatigue of low alloy steels have been reported. For small specimens made from low strength structural steels, cathodic protection has been shown to improve corrosion fatigue performance. However, for high strength steels, application of cathodic over protection can prove deleterious due to enhanced crack initiation83. According to Congleton et al.53 the effect of cathodic 28 potentials in reducing corrosion fatigue crack growth rate is minimal, while over-protection actually enhances crack propagation rate due to cathodic generation of hydrogen and subsequent hydrogen embrittlement. 2.2.8 Potential drop down the crack Several workers84,85,86 have shown that potential drops of at least 100 mV could occur for cracks of 1 cm depth in controlled potential stress corrosion tests. Vermilyea et al.84 have proposed a theory for concentration and potential variation in a crevice which can be used to estimate concentrations in crevices in real corrosion situations. A consequence of the theory, supported by experiments, is that a small potential difference in a crevice (or crack) can cause a large concentration change and, correspondingly, a large increase in corrosion current at a given applied potential. They have analyzed the situation assuming an active crack tip with inert sides. Ateya et al.85, have studied the situation under cathodic charging. However, no such potential drop was found by ParkinsSV, who found that for C-Mn steel both smooth and precracked specimens crack at the same potential. The contradictory results have been resolved53, based on the net current (anodic or cathodic) flowing in the potentiostat circuit The net current flowing depends on whether or not the crack sides are passivated. If the crack sides are not passivated larger currents are drawn. The higher the net current flowing in the circuit, the larger will be the potential drop. Hence it would be possible to predict that applied potential which would cause maximum potential drop by simple inspection of appropriate polarization diagram for the metal-environment system. 2.2.9 Effect of load ratio Near-threshold corrosion fatigue behaviour is much more affected by mean stress, temperature and environment than stage II crack propagation67,88-96. Increase in mean stress, which is related to the R-Ratio (0"min/o"max)» generally decreases AX-ph and increases crack propagation rates. However, there is a lack of R dependence for fatigue tests on low alloy steels conducted under inert conditions96 and for 316 SS in 29 vacuo and in helium97. These results suggest that the effect of R-Ratio on near threshold fatigue crack propagation under other conditions can be due to environmental interactions. The effect of load ratio has also been explained in terms of the phenomenon of crack closure98. The effect of crack closure mechanisms on fatigue crack growth rates below 10-9 m/cycle has been reviewed by Suresh et.al99. Briefly, an elastic-plastic medium that undergoes limited plastic deformation at the crack tip during loading, produces a plastic enclave in an elastically strained matrix. It is then possible for the crack faces to impinge during unloading before the cyclic tensile load reaches its minimum value. Therefore, the driving force for crack growth is reduced; i.e., the normally applied value of AK at the crack tip is reduced to AKgff (i.e.,AKeff=Kmax-Kclosure)- Crack closure at the crack tip can occur due to (a) plasticity (b) formation of oxide or corrosion products on the cracked surface (c) fracture surface roughness induced closure during Mode II loading (d) viscous fluid within the crack (e) phase transformation. The mechanisms are shown schematically in Figure 7. 2.2.10 Effect of frequency It is known that corrosion fatigue crack propagation (da/dN, fatigue crack increment per cycle) is enhanced at low frequency 100.101. This implies that at lower frequencies there is enough time for corrosive environment to make a contribution to crack growth. Congleton53 has shown that if da/dN (crack growth per cycle) is plotted as da/dt vs. log AK, the rate of growth due to environments effects actually increases with frequency. However, since the time is short and growth due to mechanical action being large the additional contribution of the environments seems negligible. 2.2.11 Two phase alloys It is known that second phase particles of different modulus can affect crack growth; lower modulus particles attract the crack while higher modulus particles deflect the crack. Fatigue crack growth behaviour in two phase steels (dual phase steel) 102, and a high strength steel having different tempered structuresl03, has been examined. It was 30 found in the dual phase steel that the continuous martensitic matrix exhibited higher crack closure levels compared to a continuous ferrite microstructure. This was explained on the basis that martensite constrained the plastic deformation in ferrite and bears a larger portion of the applied load, thereby shielding the crack tip and effectively reducing the crack tip stress intensity. In high strength steel, higher strength and higher strain hardening microsrructures shows lower threshold stress intensity while lower strength and lower strain hardening microstructure shows higher threshold. In the higher strain hardening microstructure (martensitic structure), localized plastic deformation can be prohibited either due to high strain hardening coefficient or operation of restricted number of slip systems. Hence the crack tip is relaxed by fracture in transgranular quasi cleavage mode. In the latter, crack growth occurred by slip along crystallographic planes resulting in transgranular facets. Both fatigue crack initiation65,104 and growth65,105,129-130 have been studied in duplex SS. Fukaura et al.!04? found that the persistent slip bands were found in both ferrite and austenite during fatigue. However, fatigue cracks leading to failure were found only in the ferrite phase. It was reasoned that the fine grain diameter of the austenite phase compared to ferrite, and the higher strength of austenite containing significant amounts of nitrogen, made it more difficult for slip bands to develop into fatigue cracks. However, in Alloy 63, a duplex SS, crack initiation in air occurred at inclusions or at phase boundaries at low stress levels and at persistent slip bands(PSB) in austenite at higher stress levels65. In duplex SS VKA 171 and VKA 271, crack initiation occurred along phase boundaries in air and at PSB in aqueous solutions (white water). Duplex SS IN744 shows crack initiation at phase boundaries regardless of environment. All the above results were rationalized on the basis that fatigue crack initiation is a function of stress level, environment, alloy composition and microstructure. At both low and high mean stress, crack propagation has been reported to occur by cleavage-like fracture in the ferrite and ductile failure in the austenite. This has TABLE -2 FATIGUE VARIABLE INFLUENCE UPON MIXING Crack angle opening ranged Mean stress Frequency Crack length,a Temperature Stress Wave form Applied current density Increases linearly with j3 Increases with decreasing mean stress Increases linearly with frequency Projected to increase with the cube of a Increases with temperature Increases with more rapid opening and closing Increases with current density Figure 7 Mechanisms of crack closure 32 been ascribed to the keying effect of austenite in retarding crack propagation 106. Moskovitz et al.106 quote from the work of Liljas that a similar effect is seen in studies of wrought duplex SS 3RE60. Bassidi et al.129, believe that hydrogen embrittlement processes occur based on fractographic observations. Therefore, in summary, the tendency of austenite to enhance crack initiation while retarding crack propagation and that of ferrite to act vice versa depends on the ductility of each individual phase. Adding to this is the contribution of the electrochemical action of the environment. Both anodic dissolution and cathodic generation of hydrogen affect the phases differently. It should be pointed out that none of the above work on two-phase alloys has been done under controlled potential conditions. 33 3.SC0PE O F T H E P R O J E C T It has been pointed out in the literature that corrosion pits on the surface could act as crack initiation sites for fatigue cracking to occur. These fatigue cracks can eventually propagate causing failure of suction roll shells. Although some work has been done to understand fatigue crack propagation of duplex steels in synthetic white water environment, crack tip variables like pH and chemistry were either unknown or uncontrolled. For example, Bassidi et al. 105,129-30 have published several studies on the fatigue of duplex stainless steels but the electrochemical conditions at the crack tip (eg., pH and electrochemical potential) were neither controlled nor reported.Further, they did not consider the possible changes that could occur in the localized solution chemistry within the crack. This leads to ambiguities in ascertaining the mechanism responsible for cracking. The main goals of this study are listed below: (i) To understand the effect of phase composition on pitting behaviour of two duplex steels, one cast and other wrought, in chloride solutions. (ii) To generate near-threshold fatigue crack propagation data in chloride solutions by performing corrosion fatigue tests under well defined crack tip conditions (crack tip pH and potential). Therefore, the potentials at which hydrogen evolution is possible can be defined unambiguously. In addition, high R-ratio will be used to eliminate crack tip closure effects. This will minimise crack retardation effects. At potentials where hydrogen evolution is impossible and oxidation mechanisms are possible, the kinetics of oxide growth rate on bare metals surface will be studied (scratch tests). The results of the scratch tests and fractography of the failed sample will be used to ascertain the mechanism of cracking. At potentials where hydrogen evolution is possible, no attempt will be made to ascertain the mechanism of hydrogen embrittlement 3 4 In accordance with Section 1.5 the main objectives of the present work are summarized below: 1. To find the effect of phase composition and partitioning effects on the pitting behaviour of duplex SS in chloride environments. 2. To determine region I (near-threshold) crack propagation rates of cast and wrought duplex SS in simple chloride solutions and synthetic white water as a function of applied stress intensity, controlled electrochemical potentials (anodic and cathodic) and rolling direction (wrought steel). 3. To measure crack closure stresses during the unloading part of the fatigue cycle. 4. To determine the repassivation kinetics of the steels in the environments of interest, including nucleation and growth of oxide films, and relate it to possible oxidation related models of crack growth at anodic potentials. 5. To determine whether good mixing occurs between the bulk and crack tip solutions at high frequency. 6. To determine the mechanistic model of cracking based on the effect of potential on FCP, crack fractography and repassivation kinetics. 35 4.EXPERIMENTAL STUDIES ON PITTING OF DUPLEX SS 4.1.. EXPERIMENTAL PROCEDURE 4.1.1. Specimens and Solutions Studies were conducted on both wrought (AVESTA 2205) and cast (VKA 378) duplex steels. The wrought material was received as 12.5mm thick plate and the cast alloy was obtained in 25mm thick sections that were machined from a large casting. The chemical compositions and etched microstructures are shown in Table 3 and Figure 8 respectively. The etchant was Kallings reagent, an acid chloride solution (1.5 gms of CuCl2, 33ml of HC1, 33ml of alcohol and 33ml of distilled water) which etches ferrite dark and austenite light. The cast alloy contained 65-70% ferrite with islands of Widmanstatten austenite along grain boundaries and within grains, as shown in Figure 9. The wrought steel was composed of approximately equal proportions (50%) of ferrite and austenite that were elongated along the rolling direction, as shown in Figure 8b. The average width of the elongated grain in the wrought steel was <50 urn. The cast steel contained very coarse grains with an average grain size of ~2 mm, as seen in Figure 9. Square specimens of 100 mm2 working area were cut from both steels and a high purity nickel wire spot welded to the back face for electrical connection. The nickel wire was insulated with polytetrafluoroethylene (PTFE) tubing and the end of the tube sealed to the specimen. The specimens were mounted in a cold mount epoxy resin and the working surface was metallographically polished to a lum alumina finish. The exposed specimen/epoxy interface was covered with lacquer to prevent crevice corrosion. Tests were run in 1M NaCl solution (pH=6) and a synthetic white water (pH=4.7) whose composition is given in Table 4. All solutions were prepared from reagent grade chemicals. Supplementary tests were conducted on unmounted and unlacquered specimens. All faces of the specimen were polished, a Ni wire spot welded to one edge, and the specimen partially submerged in the test solution. This completely eliminated Table 3 Composition of Steels (Wt%) STEEL C Cr Ni Mo Si Mn P S Cu N* Fe Cast Duplex 0.06 20 5.0 2 0.8 0.6 0.03 0.01 4.5 0.02 Bal Wrought Duplex 0.03 22.4 5.8 2.6 0.37 1.7 # # - 0.14 Bal * Determined from Fusion analysis using LECO analyzer. # Not determined. Table 4 Composition of Synthetic White Water (pH=4.7) Chemical Cone, of Master Proportions in PPM* in Solution Test Solutions Test Solutions (mL) NaCl 3.29 g/1 10 32.9 HC1 0.1 N 3.16 11.5 Na2S04 1.479 g/1 100 147.9 Na 2S 20 3 1.412 g/1 100 141.2 H 20 1000 " Parts per million by weight 37 (a) Cast duplex SS (b) Wrought duplex SS Figure 8 Light micrograph showing rnicrostructurcs. Figure 9 Light micrograph showing coarse grain structure of cast duplex SS. 39 interfaces between the metal and a coating material and prevented any possibility of crevice corrosion interfering with the experiment. This procedure was used primarily on the cast steel which was particularly prone to crevice corrosion. 4.1.2 Pitting Tests Single cycle potentiodynamic pitting scan (PPS) tests were conducted in a single compartment cell, using a platinum counter electrode and a microprocessor controlled potentiostat (Eg&G PARC Model 350A) operating at a scan rate of 0.5mV/sec. The potential was controlled with respect to an external saturated calomel electrode (SCE) that was connected to the test solution via a KC1 salt bridge and a PTFE Luggin capillary that terminated 1mm from the specimen. A porous zirconia plug was fitted in the end of the Luggin capillary. The PPS tests were not commenced until a stable free corrosion potential was obtained. The potential was then scanned in the noble direction. The scan direction was reversed once a preset threshold current density of 5X10-3 A/cmZ was reached. In most PPS tests, nitrogen purging of the test solution was conducted throughout the experiment. The solutions were not purged during PPS tests on partially submerged specimens, in order to prevent turbulence. However, a nitrogen atmosphere was maintained above the solution. Long term pitting tests were conducted on mounted and polished specimens that were immersed in the test solutions for periods of 100 to 200 Hours. Potentiostatic control was exercised throughout each test, using either an ECO Model 549 or Hokuto Model HA-211A potentiostat. At the end of the exposure period, specimens were removed, washed successively with distilled water and alcohol and stored in a desiccator for later examination. Some of the specimens were etched before the long term immersion tests in order to locate the position of phase boundaries prior to pit initiation. 40 4.1.3 Microscopical Techniques The general appearance of the microstructure and pitting morphology was studied by optical microscopy and conventional scanning electron microscopy (SEM) using 20 KeV excitation. The detailed microstructure of the steels in terms of second phase particles and non-metallic inclusions was determined by transmission electron microscopy(TEM). Thin specimens (25X25X3)mm were cut from the two steels and mechanically polished to 0.1mm thickness with a surface finish of 600 grit using SiC papers. Discs of 3mm dia. were cut from the thin sheets using a spark erosion machine. The discs were subsequently electropolished and thinned using a jet polisher (Tenupol-2) until a perforation appeared. The electrolyte was 5% perchloric acid and 95% acetic acid. The thinned areas around the perforations were observed by TEM at 200 KeV (Hitachi model H-800). 4.1.4 Micro-Chemical Analysis and Phase Identification The composition of phases and partitioning of elements in both steels was studied by quantitative energy dispersive spectroscopy (EDS) using EDS spectrometers interfaced to both the scanning and transmission electron microscopes. In the case of TEM-EDS studies, a scanning transmission beam control system produced a very fine diameter (~8nm) electron beam for extremely small area microanalysis. A beam excitation of 200 KeV was used in the TEM-EDS analysis to minimize beam broadening interactions in the specimen. Identification of phases during SEM observations of pitted surfaces was assisted by EDS analysis of local areas. It is well knownl07 that chromium and molybdenum preferentially partition to ferrite, leaving behind an austenite containing reduced chromium and molybdenum compared to the average alloy composition. Similarly, nickel preferentially partitions to austenite 107. Scanning Auger electron spectroscopy (AES) was used to determine the distribution (partitioning) of nitrogen in the two phases of the high nitrogen wrought 41 steel. The specimens were initially polished and etched to locate the position of the phases and phase boundaries. Micro-hardness indentations were then placed on the phases. The samples were subsequently repolished to remove the etched layer. AES analyses were performed on the regions around the indentations using a PHI-SAM 595 system.(Courtesy of Simon Fraser University). 4.2 RESULTS 4.2.1 PPS Tests Typical single cycle PPS diagrams in 1M NaCl are shown for the cast and wrought steels in Figures 10a and 10b respectively. The potential at which pits initiate (Epit), was determined by the potential value at which the anodic current underwent a rapid increase on the forward scan. Pitting of the cast steel occurred relatively easily. There was an initial rapid increase in current at +200mVsCE and a continuing increase thereafter. Thus, the pitting potential was identified as +200mVsCE and confirmed later by long term potentiostatic tests. The Epit value of the wrought steel was much higher, but its precise value was masked by anodic currents corresponding to oxygen evolution. For example, a polarization curve for Pt in 1M NaCl is superimposed on Figure 10b (dashed line), and shows that the oxygen evolution current closely coincides with the increased anodic current on the PPS diagram of the wrought steel. However, the wrought steel exhibited pits after the PPS test. Therefore, it is concluded that Epit of the wrought steel must be greater than or equal to +900 mVsCE-The potential at which pits repassivated (Eprot) during the PPS test was determined from the potential at which closure of the anodic hysteresis loop occurred during the reverse scan. Thus, based on Figures 10a and 10b, Eprot was near -140mVsCE for the cast steel and no lower than +900mVsCE for the wrought steel. The small hysteresis effect on the wrought steel, relative to the cast steel, was a reproducible 42 •0.400 io l io J 101 id 1 10S io* i o 7 IO1 10* i o l * Current Density, uA/m2 (a) Cast Duplex SS 1.200 0.800 LU o CO > 10.400 c 0.000 -0.400 io io' io2 io3 io4 io5 io6 to7 iog io9 10 Current Density, uA/m2 (b) Wrought Steel Duplex SS. Dashed line indicates oxygen evolution current on Pt elec-trode. Figure 10 Single cycle potentiodynamic pitting scans in 1M NaCl. 43 phenomenon. It is believed that the true nature of the hysteresis currents due to pitting are masked by oxygen evolution currents. Nevertheless, the results clearly show that Epit and Eprot of the wrought steel are more noble than the corresponding potentials for the cast steel, demonstrating that the wrought steel has superior pitting resistance. At the end of the cyclic pitting scan in 1M NaCl the microstructure showed preferential dissolution of austenite in the case of cast steel, and ferrite in the case of wrought steel, as shown in Figures 11a and lib. In the cast steel, the austenite phase was discontinuous, because of the solidification process, and readily identified as islands. No SEM-EDS analysis was required for phase analysis. However, in the wrought steel due to phase constituents being present in almost equal amounts with similar morphology, a SEM-EDS analysis of Cr,Ni and Mo was performed to confirm the identification of phases on the pitted surface. The results of the analyses are given in Table 5. Comparison of the phase analyses with the pitted regions confirmed that the ferrite dissolves preferentially. 4.2.2 Transmission Electron Microscopy The cast steel showed numerous small precipitates which were either spherical in shape (diameter of <10um) or ellipsoidal. The precipitates were present in the a phase, at a - a phase boundaries and at a-y phase boundaries, as shown in Figures 12 and 13. The y-phase regions were completely devoid of precipitates, as shown in Figure 12. The phases were identified by TEM-EDS analysis of their composition. These compositions are listed in Table 6 for both steels. It should be noted that the composition analysis was done on thin films rather than the bulk alloy. This could lead to errors in the accuracy of the analysis. Therefore, composition difference between the phases should be considered rather than the absolute values.(see Appendix-3). Table 6 shows that the small precipitates in the cast steel were Cu-rich. However, considering the difficulty of analyzing very small precipitates without obtaining contributions to the EDS spectrum from the surrounding matrix, it is concluded 44 (b) Preferential pitting of ferrite phase (dark areas) in wrought duplex S S . Figure 11 Appearance of pitted surfaces. Pits appear as black regions. (Light micrograph). 45 Table 5 SEM-EDS Analysis of a and y Phases in Wrought Steel* Cr Ni Mo (wt%) (wt%) (wt%) Ferrite (a) 25.53 4.43 3.86 Austenite (7) 21.98 6.94 2.33 Table 6 TEM-EDS Analysis of a and y Phases in Cast and Wrought Steels* Cast (wt%) Cr Ni Mo Cu Wrought (wt%) Cr Ni Mo Ferrite (a) Austenite (7) Precipitate 22.37 2.76 2.25 1.01 17.55 4.54 1.61 4.27 14.37 4.14 0.64 34.91 26.1 3.16 2.73 22.41 5.97 1.94 See Appendix-3 for accuracy of analysis 46 Fisure 12 TEM rnicrograph of cast steel showing a-a interface. Note Cu precipitates in both regions and at interface. Figure 13 TEM micrograph of cast steel showing a-y interface. Note Cu precipitates in a phase and at interface. 47 that the analyses indicate that the precipitates are composed principally of copper. This is consistent with the analyses of others 108. Note that in Figures 12 and 13, some of the precipitates have fallen out of the specimen due to preferential dissolution of the adjoining matrix during the thinning process. The wrought steel showed no precipitates in either of the phases or at the phase boundaries, consistent with the absence of copper in the alloy (see Table 3). Also, no nitrides were detected, despite the high nitrogen content in this alloy. 4.2.3 Long term exposure tests A cast steel specimen exposed to 1M NaCl solution at the freely corroding potential showed pitting at the oc-y boundary and propagation into the austenite, as shown in Figure 14. During the time period the corrosion potential rose from -200 mVsCE to -40 mVsCE. which was 100 mV more noble than the protection potential (Eprot=-140 mVsCE)- Figure 15 shows a series of micrographs of cast duplex SS exposed to 1M NaCl for 24 hours at different controlled potentials. The sequence a,b,c in Figure 15 shows the initiation of pits at the a-y boundary, propagation into the austenite and eventual dissolution of austenite. Similar pitting behaviour was obtained when the cast steel was exposed to synthetic white water. However, the times to pit initiation were 5 to 6 times longer (120 to 140 Hrs.). Wrought steel showed no pitting, even at potentials of +900 mV SCE in 1M NaCl solution and synthetic white water. This observation shows that the Epit of wrought SS must be greater than 900 mV SCE, consistent with the PPS tests. 4.2.4 Nitrogen Distribution The AES analyses of the wrought duplex SS (high N) showed that the atom ratio of Fe:N was ~20 in the y-phase and ~60 in the a-phase. This corresponds to a nitrogen enrichment of three times in the y-phase and is consistent with the reported low solubility of N in the a-phase^ .^ Note that based on the composition of the wrought steel in Table 3, the 48 Figure 14 SEM micrograph showing preferential pitting of austenite in cast steel in 1M NaCl solution at free corrosion potential. Pits initiate at a-y interface and propagate into y region. (a) -200 mVsCE; Pit initiation at ct-y interface. (c).+200 mVsCE. Dissolution of austenite. Figure 15 Effect of 24 hour potentiostatic tests on pitting of cast steel in 1M NaCl. (Light Micrographs). Fits appear as black areas. 51 average Fe:N ratio should be ~124:1. If all the N enters the y-phase only, the Fe:N ratio in this phase should be 62:1. This is higher than the observed value of 20:1. The discrepancy is attributed to (i)difficulties in obtaining absolute atom ratios by the AES technique when the solute concentration falls below 1 at%, which is the case for N (ie., 0.14 wt% = 0.55 at% N) and (ii) the peak to background ratio for N was low indicating that the analysis was only qualitative. Therefore, attention must be directed to relative differences in atom ratio between the phases, rather than absolute ratios, in order to confirm partitioning effects. No AES analyses were conducted on the cast steel because the residual N content was too low. 4.3.DISCUSSION 4.3.1 Alloying Effects The results clearly show that the y-phase is most susceptible to pitting in the cast duplex SS whereas a-phase is more susceptible in the wrought duplex SS. Also, the long term tests show that the wrought material is more resistant to pitting than the cast alloy, and PPS tests show that Epit and Eprot of the wrought alloy are more noble than that of the cast material. All of these observations can be rationalized in terms of phase composition, particularly the distribution of the elements Cr, Mo and N. It is well established that the pitting resistance of SS is improved by raising the Cr and Mo content 110. The analyses clearly show that the ferrite phase in both steels is enriched in Cr and Mo. This accounts for the higher resistance of the a-phase to pitting in the cast material. It also accounts for the high Epit and Eprot values of the a-phase in the wrought material. However, another factor must be important in the wrought steel because the ferrite pitted preferentially despite the observed partition of Cr and Mo. The wrought steel contained a high nitrogen content relative to the cast steel. The results show that alloyed nitrogen in the wrought steel is partitioned preferentially in the y-phase. It is known that N has a beneficial effect on the pitting behaviour of single 52 phase austenitic SS^.Hl. Therefore, it may be concluded that the beneficial effect of N in the duplex SS is conferred primarily on the v-phase. This raised Epit and Eprot of y-phase to values more noble than the a-phase, so that when pitting is observed in chloride-containing solutions it occurs preferentially in the a-phase. This clearly indicates that good pitting resistance of duplex steels requires both high N, which benefits the y-phase only, and high Cr and Mo to benefit a-phase by preferential partitioning of Cr and Mo. Consistent with the present study, other workers have reported that N raises the Epit of duplex SS to more noble valuesH2,and that pitting occurs preferentially in the a-phase of high-N duplex SS and in the y-phase of low-N duplex SS-*'113. 4.3.2 Role of Nitrogen The mechanism by which N confers additional corrosion resistance to duplex stainless SS is not clear. Truman et.allH have studied the effect of N additions on pitting corrosion resistance of a series of austenitic alloys. They found that with sufficient Cr and Mo contents a large increase in pitting resistance is seen with small additions of N, due to synergistic effects. They speculated that the increased availability of adsorbed N in the atomic form at the surface favored metal nitride formation which then increased the corrosion resistance. Osozowa et al.,H4 have proposed that when alloyed nitrogen dissolves it consumes protons in the pit to form ammonium ions. This prevents a local lowering of pH and helps to repassivate the pit before it propagates. Jargelius and Wallin also believe that the beneficial role of N is due to the formation of NH4+ ions 115. Thermodynamic considerations! 16 show that it is possible for molecular nitrogen to be reduced to ammonia. However, Pourbaixll6 has pointed out that the reaction is extremely irreversible and that N is practically non-reducible in solution. Similar considerations could well apply to N which exists in solid solution in the steel; i.e., kinetic 53 considerations severely retard formation of ammonium ions from alloyed nitrogen. Obviously, when pitting does occur in the N-containing alloys, the nitrogen atoms must dissolve (react) to produce a species consistent with the thermodynamics of aqueous equilibria. Hence, it is not surprising that NH4+ ions are detected in the aqueous environment after pitting has occurredll4,115. Consequently, it is now proposed that it is the rate at which N reacts with the environment and not the reduction of N, per se, that accounts for the beneficial effect of alloyed nitrogen on pitting resistance. Based on the preceding considerations, it is proposed that in the initial stages of pit growth on N-alloyed SS, the metal ions will preferentially dissolve, due to the irreversibility of N reduction, thereby enriching the surface with N atoms. This enrichment progressively decreases the sites at which metal dissolution occurs. Surface diffusion of N atoms to kink and ledge sites will further retard dissolution, (i.e., pit growth is inhibited). Some indirect evidence in support of a N enrichment model was obtained in the present work. It was found during AES analysis that wrought duplex SS surfaces etched in Kallings reagent (i.e., acid chloride solution) gave higher N signals than unetched surfaces. It should be noted that a surface enrichment effect has been proposed by Newman 117 to account for the beneficial effects of Mo on pitting resistance ofSS. 4.3.3 Effect of Copper The preferential nucleation of pits at the cc/y interfaces (eg.Fig 14) of the cast steel could be associated with Cu precipitates at these sites. Whenever an interfacial precipitate encounters the free surface, the local integrity of the passive film will be affected leading to easier film breakdown and pit initiation. This explains the somewhat surprising occurrence of pitting in the cast steel in Figure 14, where the free corrosion potential rose to a value >Eprot but <Epit, as defined by the pitting scan experiments. The Cu-rich precipitate may not weaken the film sufficiently for pits to initiate at potentials <Epit in the pitting scan test, but may weaken the film sufficiently for pits to 54 initiate at these potentials in long term exposure tests. The pits then grow into the y-phase because of the lower pitting resistance of this phase arising from its alloy chemistry. It should be noted that Cu is often added to stainless steels to raise the strength level by precipitation hardening and to improve corrosion behaviour under freely corroding conditions in non-oxidizing media. It has been observed that both pitting current density and corrosion rate are decreased 1 1 2 . Also, cavitation erosion resistance in sea water contaminated with H 2 S is improved by Cu addition 1 1 2 . However, in view of the possible role of Cu in pit initiation in chloride solutions it would appear that its presence may be detrimental in white water environments. 4.3.4 Industrial Considerations Cast duplex SS of similar composition to that used in the present study is being used as material for suction press rolls in the pulp and paper industry. It is now clear that such material has poor pitting (and crevice) corrosion resistance. Pitted sites will lead to early and easy fatigue crack initiation leading to reduced service lifetimes. Therefore, it is recommended that the alloy composition of such steels should be modified by raising the Mo content and adding N to obtain a composition similar to the pit resistant wrought alloy. The alloy modifications will probably require precautionary measures to prevent sigma-phase formation and nitride precipitation during the cooling of casting. 4.3.5 Summary Pitting studies of both cast and wrought duplex SS showed that the composition of the individual phase plays an important role in the general pitting behaviour of the alloy. Preferential pitting of the ferrite or austenite phase was dependent upon partitioning of the elements Cr, Mo, and N. 55 5 EXPERIMENTAL STUDIES ON CORROSION FATIGUE AND REPASSrVATION KINETICS OF DUPLEX SS 5.1 EXPERIMENTAL WORK 5.1.1 Fatigue Specimen Preparation Corrosion fatigue and repassivation studies were conducted on specimens obtained from the same as-received wrought and cast duplex steels that were used in the pitting experiments. The details of these steels have already been presented, including microstructure (Figures 8 and 9), average composition (Table 3), and compositions of the individual phases (Tables 5 and 6). Fatigue tests were conducted on pin loaded compact tension specimens that were machined from the starting material. A schematic of the specimen is shown in Figure 16, together with the stress intensity (KjJ calibration formula appropriate to the specimen geometry 118. The specimen thickness (t) was 12.7 mm, and the length (b) of the specimen from the loading line was 44 mm. An initial crack (a) of length 12 mm was introduced by machining. The specimen thickness was sufficient to maintain nominal plane strain conditions at the crack tip based on the conditionll9. t>= 2.5 (Ki/CTy)2 (8), where Oy is the tensile yield stress. The yield stresses were 480 MPa and 450 MPa for the cast and wrought steel respectively. The wrought specimens were machined in both T-L and L-T orientations58 and all specimens were ground and polished to a lum finish. The machined starter notch was sharpened and extended 2mm by fatigue pre-cracking, using an electromagnetic resonance fatigue machine (Instron model 1603). The cyclic stress intensity applied was approximately 10 + 5 MPa/m or 20 + 10 MPa/m. At the end of pre-cracking, a strain gauge of 3mm length and 120 ohm resistance (Micro-measurements CEA-06-125UN-120) was installed at the center of the back face. The gauge was chosen such that the self temperature compensation (STC) number of the gauge closely matched the thermal expansion coefficient of the stainless steel specimen. This made the gauge insensitive to temperature under the conditions employed. The gauge was protected from test environment by several coatings20) as o T T a t o Kj = 2P(2b+a)^ a") f(2) / t(b-a) Figure 16.COMPACT TENSION SPECIMEN 57 Air d r y i n g P r o p r i e t a r y c o a t i n g p o l y u r o t h a n o c o a t i n g Figure 17.PROTECTTVE COATINGS Figure 18.THREE-WIRE CONNECTION FOR TEMPERATURE COMPENSATION 58 shown in Figure 17. To eliminate the effect of temperature on lead wires connecting the gauge into the bridge, a three wire quarter bridge circuit was used as shown in Figure 18. 5.1.2 Fatigue Testing High frequency fatigue testing was conducted on an electromagnetic resonance machine (Instron Model 1603) using a sinusoidal wave form. The loading frequency was ~85Hz. Tests were conducted at room temperature (22 deg.C) using an R-Ratio of 0.33 (R = Pmin/Pmax = Kmin/Kmax)- The principal test environments were 1M NaCl (pH = 6) and synthetic white water of pH=4.7 (see Table 4). They were contained in a transparent acrylic (Plexiglass) cell that was mounted around the mid-section of the specimen. Some supplementary tests were conducted in desiccated air by placing silica gel in the cell and covering the top of the cell with a plastic sheet. In the aqueous environments, the specimen potential was controlled by means of a Hokuto Model HA-211A potentiostat and measured with respect to a saturated calomel reference electrode (SCE). The SCE was connected to the test solution via a KC1 salt bridge and a PTFE Luggin capillary that terminated 1mm from the specimen. A porous zirconia plug was fitted in the end of the Luggin capillary. A platinum wire was used for the counter electrode. A schematic of the test set-up is shown in Figure 19. The surface crack lengths were measured to within ±10 um using a custom designed travelling microscope with internal illumination. Only single side crack length measurements were recorded. The average crack length was monitored using the back face strain gauge (BFSG) and a signal conditioner and amplifier (Bridge Amplifier and Meter Model BAM-1). The output from BAM-1 was connected to an analog to digital (A/D) converter. The digitized signal was fed into a personal computer (Zenith model 150). The digitized signal from the A/D converter was acquired every 5 minutes at a rate of 8.5 KHz. At a time about 100 points were collected. During the time interval between the two data acquisitions, a computer program was used to analyze the collected data to 5 9 Luggin capillary PLCounter electrode Corrosion cell Strain Gauge Drilled hole (mixing test) Working electrode (Specimen) Figure 19.SCHEMATIC OF THE TEST SET-UP 60 identify the peak voltages (both minimum and maximum voltages). From a previous calibration of voltage versus crack length, (see Appendix 1) the measured voltage was read in terms of crack length. There was good agreement between the crack lengths measured using the travelling microscope and the BFSG. The DC amplifier was switched on atleast an hour before the start of the test and remained in continuous operation throughout the test to reduce drift. Corrosion fatigue experiments were conducted in two steps. The first involved a decreasing AK test to determine the cyclic threshold stress intensity for cracking (AKr/h)- The second step involved a rising AK test to determine the relationship between AK and crack propagation rate. A recommended load shedding procedure*^  w a s used to achieve decreasing AK conditions and in the present work decrements of 8% of the previous load were used. After each reduction, crack growth rates were monitored for a period of time. The threshold, AR^h' was defined as the condition where the crack growth rate (da/dN) was <= 1X10-10 m/cycle. The small incremental load reductions minimized problems with premature crack arrest due to plasticity-induced closure (ie., effects associated with residual plastic deformation at the crack tip. Near the threshold, crack closure stress intensities were determined by measuring the residual strain with the BFSG after reducing the load to zero. After reaching AK-j^, the cyclic loads were kept constant at values near AP^ /h and AK allowed to rise as the crack propagated (see Kj-calibration formula in Figure 16). Throughout all fatigue tests, the crack length was recorded periodically, by means of the travelling microscope and the BFSG, as a function of the number of elapsed load cycles. The incremental crack length recorded (da) was divided by the number of elapsed load cycles (dN) to determine da/dN at the stress intensity corresponding to the total crack length at the instant of measurement. Testing was continued until the increasing crack length caused AK to enter stage II fatigue behaviour. The test was then terminated, the specimen removed and washed successively with distilled water and 61 alcohol. Also, some experiments were terminated during the load shedding period near AK t h . 5.1.3 Solution Mixing Test In order to determine whether complete mixing occurs between the bulk solution and the crack tip solution, a unique test was devised. A small hole (dia.2.4mm) was drilled ahead of crack tip at the mid-thickness of a compact tension specimen so that it intersected the crack plane perpendicularly, as shown in Figure 19. The hole was filled with an aqueous vegetable red dye solution and sealed. The specimen was then subjected to corrosion fatigue in the test cell in 1M NaCl. The test was stopped as soon as a change in the colour of the bulk environment was detected, due to mixing of the dye with the test solution. The specimen was removed and broken open to check whether (1) crack front had just intersected the drilled hole or (2) advanced beyond the hole. The first condition implied that the mixing is good; i.e., as soon as the fatigue crack intersects the hole the dye comes out and is pumped into the bulk solution by the movement of the opposing crack faces. The second condition suggests that even after the dye had seeped into the crack it had taken some time to be pumped out, showing that the mixing between the crack solution and bulk solution is not good. Note that the top of the drilled hole was sealed with silicone rubber to prevent the dye running out under gravity when the advancing crack intersects the hole. It should be recognized that because of this sealing the dye must first diffuse into the crack tip solution before being pumped out. This would involve a finite amount of time during which the crack could go past the edge of the drilled hole. 5.1.4 Rapid Scratching Test Scratch tests were conducted on small 0.56 mm thick sheets of duplex steel that were sectioned from the as-received material. The sheet was mounted in cold setting epoxy resin with an insulated electrical connection attached. The thin edge of the sheet was the exposed working face and was metallographically polished to a 600 grit finish, 62 followed by cleaning in ethanol. The specimen potential was controlled by a potentiostat, (ECO MODEL 549), and measured with respect to an external saturated calomel electrode that was connected to the test solution via a KC1 salt bridge and a PTFE Luggin capillary. The counter electrode was a platinum wire. Scratch tests were done in 1M NaCl and synthetic white water at different potentials ranging from 0.3Vsce to -0.6Vsce. Scratching was done using a diamond tip pencil subjected to a constant load. Just before the test, the tip was placed alongside the thin edge of the specimen and then moved horizontally across the short specimen dimension (0.56mm). Because of the constant load, the depth of the scratch was uniform in all the tests. The resulting current transients were recorded using a digital storage oscilloscope (Tektronix model 2230) with a rise time of 1.6 ms. The projected area of the scratch was measured using an optical microscope and found to be 0.0616 mm2 in the case of the cast steel and 0.0682 mm2 in the case of the wrought steel. The digitized current transients were recorded and stored in a personal computer (IBM PC) for subsequent processing. The total contact time of the diamond tip on the exposed face was determined by mounting two thin slices of steel pieces parallel to each other at a known distance apart and scratching across both the pieces. The time lapse between the two resulting peak current transients was recorded and hence the velocity of the scratching was determined. Knowing this velocity, and the thickness of the sample, the contact time was determined to be 2.3 ms. 5.1.5 Fractography At the end of the fatigue tests, the uncracked ligament ahead of the crack was removed near to the crack tip and the specimen was broken open. The fracture surfaces were cleaned with an inhibited acid solution (4mls. of 35% 2butyne-l,4-diol + 3 mis. of HC1 + 50 mis. of distilled water) to remove corrosion films and improve resolution of fractographic features. The fatigue surfaces were examined by conventional scanning electron microscopy (SEM), using secondary electron imaging and excitation energies of 63 20 keV and 5keV. It was found that the low 5keV excitation, together with a pointed tungsten filament, improved the contrast between very small surface protuberances, due to the decreased volume of surface region that generated secondary electrons. This allowed opposing fracture surfaces to be examined at high magnification in order to determine whether fine topographic features matched peak to peak (mirror image) or peak to valley (interlocking). 5.2 RESULTS 5.2.1 Fatigue Crack Propagation Fatigue tests were conducted on the cast steel in 1M NaCl and synthetic white water over the potential range +0.3 Vsce to -1.2 Vsce- Under all conditions, the crack growth rate da/dN, increased with increasing AK. The essential features of the crack propagation behaviour are fully characterized by confining attention to three potentials, +0.3, -0.4 and -1.2 Vsce- The near threshold behaviour at these potentials is shown in Figures 20 to 23, where the data are presented in conventional fashion on a log-log plot. The data at +0.3 VSce and -1.2 VSce are each compared with -0.4 VS Ce and show that crack propagation rates at +0.3 Vsce and -1.2 Vsce tend to be higher than those at -0.4 VSce in 1M NaCl. No discernible effect of potential was detected in white water. Tests on the cast steel showed some data scatter that is evident in Figures 20 to 23. This behaviour suggests that crack growth retardation effects were present. However, analyses of the shape of the cyclic waveform signal from the BFSG showed that the retardation phenomena were not due to crack closure (ie., premature contact of opposing crack surfaces on the decreasing load segment of the load cycle). For example, if there existed some crack closure, then instead of a sinusoidal strain output, one would see a flattening of the strain signal when the crack closes prematurely. However, when the load and the strain signals were monitored simultaneously using an oscilloscope, no such CRACK PROPAGATION RATE, da/dN (m/cycle) © 1 1 1 1 , 1 O, III 1 I I 111 £ £ 2 2 > < J O ? H 8 g H Q O * oe " C S H C/3 CP - o o I 111111 o. I I 1111 i I I I , II i I I I I I I I I CO © oe O, v9 CAST D U P L E X SS 1M NaCL FREQ = 85 HZ -0.4 Vsce a -1.2 Vsce o ' ' I 10 3 0 STRESS INTENSITY, MPaJM gUlN1 N a C l ^ T ^ ^ T E S FOR CAST DUPLEX ,N -6 1 0 1 0 JO | . o 7 W H < O H < < Cu O OC Ou u u 1 0 -9 -8 -1 0 10 1 0 J I I L_ C A S T DUPLEX SS SYN.WHITE W A T E R F R E Q = 85 H Z - 0 . 4 Vsce Q +0.3 Vsce o • n a • D O CK • • J a o o, iff a® Ojbo D O 3 o8° o o 8 o 30 .-6 1 0 1 0 1 0 1 0 1 0 10 as as 1 0 -11 — 1 1 1 1 1 1 I — I — I I 10 30 STRESS INTENSITY, MPa/M Figure 22.FATIGUE CRACK PROPAGATION RATES FOR CAST DUPLEX IN SYNTHETIC WHITE WATER AT -0.4 AND 0.3 Vsce 1 0 11 I I I I 10 J_L_ 30 -6 J - 1 0 CAST DUPLEX SS SYN.WHITE WATER FREQ = 85 HZ -0.4 Vsce o -1.2 Vsce o • • • O o o D ^ a o • • o o o o OP 6b o , o o o o o o 1 1 1 1 1—i—i—r— 10 STRESS INTENSITY, M P a M 1 0 1 0 -8 1 0 1 0 -10 1 0 II 30 Figure 23.FATIGUE CRACK PROPAGATION RATES FOR CAST DUPLEX IN SYNTHETIC WHITE WATER AT -0.4 AND -1.2 Vsce 68 (A) OPTICAL MICROGRAPH 0 9 E 6 1 9 £8KV X i ! 0 0 K ' ' " 3 0 u m (B) SEM MICROGRAPH Figure 24.CRACK DEVIATION DUE TO PRESENCE OF AUSTENITE PHASE CAST DUPLEX IN 1M NaCl AT-0.4 Vsce 69 behaviour was observed. This was consistent with the anticipated absence of closure effects at an R-ratio of 0.33. Furthermore, an examination of the crack profile and topography of the fracture surface suggested that crack deflection could have caused crack growth retardation. It was observed that the propagating crack deviated locally away from and back towards the macroscopic fracture plane. This deviation caused the effective crack tip stress intensity to be less than the applied stress intensity because of a decrease in the local tensile components at the crack tip. Crack deflection retardation is not unexpected in large grained specimens during near-threshold stage I behaviour, because cracking is crystallographic in nature. Figure 24 shows optical and SEM micrographs of crack deviation in the large grained cast steel, where it is seen that crack deviation occurs frequently and tends to be influenced by the islands of austenite. Fatigue tests on the wrought steel were conducted in desiccated air (Figure 25) and in 1M NaCl at potentials of -1.2 V s c e , -0.4 VS Ce and 0.3 VSce- The general behaviour was similar to the cast steel, where crack propagation rates increased with increasing AK. However, there was considerably less data scatter, consistent with the finer grain size of the wrought steel and a decreased probability of crack deflection retardation. Details of .the near-threshold crack propagation behaviour in the wrought steel are shown for the T-L orientation in Figures 26 and 27. The behaviour at +0.3 VS Ce is compared with -0.4 VS Ce in Figure 26 and the behaviour at -1.2 Vsce is compared with -0.4 Vsce in Figure 27. These figures clearly show that crack propagation rates were higher at +0.3 VSce and -1.2 VSce, relative to -0.4 VSce, at AK values <= 10 MPavffn. Also, a comparison of these figures with Figure 25 shows that the fatigue behaviour in desiccated air was similar to that in -0.4 Vsce in 1M NaCl. Above 10 MPa>fm fatigue behaviour was similar under all testing conditions, indicating that crack propagation behaviour was dominated by mechanical effects at the crack tip and not by environmental 1 0 1 0 1 0 -8 1 0 1 0 10 1 0 II 10 J I 1 I I I L W R O U G H T DUPLEX DESSICATED AIR F R E Q = 85 H Z T - L ORIENTATION o o o 8, o o 9> o I I I I I I I -I 10 STRESS INTENSITY, MPa/M 30 1 0 1 0 1 0 -8 1 0 1 0 10 1 0 II 30 Figure 25.FATIGUE CRACK PROPAGATION FOR WROUGHT DUPLEX SS IN DESSICATED AIR o 1 10 I I I I I L_L W R O U G H T DUPLEX SS 1M NaCl T - L ORIENTATION -0.4 Vsce o +0.3 Vsce • • Q D • • o 0 o i i 1 i 1 r—i—i | 10 STRESS INTENSITY, MPa/M 30 1 0 1 0 1 0 -8 1 0 -9 1 0 10 30 1 0 M Figure 26.FATIGUE CRACK PROPAGATION RATES FOR WROUGHT DUPLEX IN 1M NaCl AT -0.4 AND 0.3 Vsce (T-L ORIENTATION) o "a MJ H < O H < < a, O a, (J 10 10 -7 10 10 1 WROUGHT DUPLEX SS 1M NaCl T-L ORIENTATION -0.4 Vsce o -1.2 Vsce o -9 -10 -10 J L a on o 10 J i i o oo 8 30 10 10 -7 10 - 8 10 10 10 - j 10 -11 1 1 1 1 1 r — i — | — 10 STRESS INTENSITY, MPa/M 30 10 II Figure 27.FATIGUE CRACK PROPAGATION RATES FOR WROUGHT DUPLEX IN 1M NaCl AT -0.4 AND -1.2 Vsce (T-L ORIENTATION) o >» W H < 2; O >—i H < O < DM O Of a, U U 1 1 0 J L 10 J — I I I 1 0 -7 W R O U G H T DUPLEX SS 1 M NaCl L - T ORIENTATION -0 .4 Vsce • +0.3 Vsce o 1 0 1 0 -9 4 1 0 10 1 0 i i a i i i—i—i— 10 STRESS INTENSITY. MPa/M 30 f 1 0 6 1 0 10 1 0 1 0 -10 30 1 0 II R gTJM F^? Un PROPAGATION RATES FOR WROUGHT DUPLEX IN 1M NaCl AT -0.4 AND 0.3 Vsce (L-T ORIENTATION) J O 30 J L WROUGHT DUPLEX SS 1M NaCl L-T ORIENTATION -0.4 Vsce o -1.2 Vsce o O O • o ° °uTJ O °° £ o o Jl o o B a u D D ° a 10 30 STRESS INTENSITY, MPa/M Figure 29.FATIGUE CRACK PROPAGATION RATES FOR WROUGHT DUPLEX IN 1M NaCl AT -0.4 AND -1.2 Vsce (T-L ORIENTATION) 6 10 11 J I L 10 J I I W R O U G H T DUPLEX 1M NaCl F R E Q = 85 H Z +0.3 Vsce T - L ORIENTATION a L - T ORIENTATION Q • a* CO rP o o (5) - I — I — I — r — 10 STRESS INTENSITY, MPa/M 30 , 1 0 -7 1 0 10 9 1 0 10 1 0 II 30 Figure 30.FATIGUE CRACK PROPAGATION RATES FOR WROUGHT DUPLEX IN 1M NaCl IN L-T AND T-L ORIENTATIONS AT +0.3 Vsce. 10 J I I L W R O U G H T DUPLEX SS 1M NaCl -0.4 Vsce T - L ORIENTATION o L-T ORIENTATION u 0 q>° <$>o o oo a m • a n o o Jo 8 1 1 1— i — r 10 STRESS INTENSITY, MPaVKfl 3 0 . o 6 10 -7 10 - 8 1 0 -9 1 0 10 1 0 II 30 Figure 31.FATIGUE CRACK PROPAGATION RATES FOR WROUGHT DUPLEX IN 1M NaCl IN L-T AND T L ORIENTATIONS AT -0.4 Vsce. W H < o H < < a, O a, u u 10 -6 1 1 0 -7 10 -8 10 10 10 10 II 10 • 1 I I I L W R O U G H T D U P L E * 1M NaCl FREQ = 85 H Z +1.2 Vsce T - L ORIENTATION o L - T ORIENTATION • • • • n D o • • O • • Q a i i 3 0 io6 10 -7 10 - 8 10 10 10 10 n 30 i — i i i i ' 10 STRESS INTENSITY, MPa/M Figure 32.FATIGUE CRACK PROPAGATION RATES FOR WROUGHT DUPLEX IN 1M NaCl IN L-T AND T-L ORIENTATIONS AT -1.2 Vsce. 78 conditions. The fatigue behaviour of the wrought steel in the L-T orientation is shown in Figures 28 and 29. However, in this orientation there was no significant difference in behaviour between -0.4 Vsce and -1.2 Vsce (Figure 28) and a less pronounced difference between -0.4 VSce and +0.3 VSce (Figure 29). The two different orientations are compared at the same potentials in Figures 30 to 32. The L-T and T-L behaviour characteristics were reasonably similar at +0.3 Vsce and -1.2 Vsce (Figures 31 and 32), whereas the L-T orientation exhibits higher crack propagation rates at -0.4 Vsce (Figure 30). It was somewhat surprising that crack propagation rates parallel to the rolling direction (T-L orientation) were lower than those normal to the rolling direction (L-T orientation) at -0.4 Vsce-Overall, the near-threshold crack propagation behaviour in all tests was considered to be reasonably free from errors due to crack closure effects. The accuracy of crack length measurements and consequently da/dN and AK calculations are shown in Appendix-3. Crack closure stress intensities determined from residual strains when the applied load was decreased to zero were found to be no higher than 0.11 MP&jTn. See Appendix 1. Consequently, under all test conditions represented in Figures 25 to 32, AK^h was never higher than 6 MPa\/nT and frequently less than this value, depending on the potential. These threshold values are very similar to those reported by Ait Bassidi et al. 131. However, precise comparison with this work is not possible because of their lack of control of electrochemical variables (potential and pH) and differences in R-ratio. 5.2.2.Solution Mixing Test Upon cessation of the solution mixing test during fatigue, and subsequent opening of the fatigue crack immediately following the detection of a red colouration in the bulk solution, it was found that the fatigue crack front had just intercepted the drilled hole containing the red dye. This is shown in the Figures 33a (light micrograph) and 33b (SEM fractograph). The conclusion from these observations is that the moment the crack , Crack front (A) OPTICAL MICROGRAPH (B) SEM MICROGRAPH Figure 33.FRACTURE SURFACE FROM THE "MIXING TEST" SAMPLE 80 81 intercepted the hole, some of the dye diffused into the crack solution and was immediately pumped into the bulk solution by the cyclic motion of the opposing crack surfaces. Therefore, good mixing was achieved between the crack solution and bulk solution during fatigue testing. Some additional indirect evidence was obtained after fractographic examination of the fatigue fracture surfaces in a cast steel specimen that had been tested at -1.2 Vsce- Regions were noticed where the fracture surface was pitted (Eg., Figure 34). The position of these regions were later related to the position of the crack when the test was stopped overnight and potential control discontinued. Because of the lack of potential control and the absence of any pumping action, the crack acted as a freely corroding crevice that resulted in localized corrosion of the fracture surface. The pitting studies on the cast steel have already confirmed the susceptibility of this steel to localized corrosion. No localized corrosion was observed on fracture surfaces that had not been subjected to interrupted fatigue testing. Therefore, it must be concluded that good mixing occurred during fatigue and prevented the crack solution from undergoing localized changes in solution chemistry. 5.2.3 Rapid Scratching Tests The time-dependent anodic current transients following the commencement of rapid scratching at time zero all exhibited an initial rise in current that was succeeded by a period of current decay. The peak current and shape of the transient were dependent upon the test environment and electrochemical potential. In all cases, the rise time to peak current was larger than the duration of the scratching period (2.3 ms) and larger than the response time of the storage oscilloscope (1.6 ms), indicating that the rising current transient was a real phenomenon associated with the initial oxidation behavior of the scratched surface. Details of the current transients are shown in Figure 35 for the cast steel in 1M NaCl and synthetic white water. The transients were obtained at different potentials CAST DUPLEX SS I M N a C l TIME.ms Figure 35.CURRENT TRANSIENTS FOR CAST DUPLEX IN 1M NaCi AND SYN. WHITE WATER WROUGHT DUPLEX SS 1M NaCl 0 2 4 6 8 10 12 14 16 T IME, msec Figure 36.CURRENT TRANSIENTS FOR WROUGHT DUPLEX IN 1M NaCl AND SYN. WHITE WATER CAST DUPLEX / 1M NaCl CUMULATIVE CHARGE TIME, msec Figure 37.CUMULATIVE CHARGE PASSED IN CAST DUPLEX TESTED IN IMNaCI CAST DUPLEX/SYN.WHITE WATER CUMULATIVE CHARGE TIME, ms Figure 38.CUMULATIVE CHARGE PASSED IN CAST DUPLEX TESTED IN SYNTHETIC WHITE WATER 86 ranging from -0.4Vsce to 0.3 Vsce. The highest peak current (1520 ua) was observed at +0.3Vsce in 1M NaCl. Much lower peak currents (<300 |ia) were observed at lower potentials in this solution. The inset in Figure 35 clearly shows that the peak currents were quite small in the thiosulphate-containing white water solution at all potentials. Figure 36 shows the details of current transients obtained in the wrought duplex SS at different potentials. A peak current of 1159uA was obtained at +0.3 Vsce. Similar to the cast steel current transients obtained in synthetic white water were much lower than those in 1M NaCl (Inset figure). The cumulative charge passed during the current transient increased with time, as expected. Generally, there was an initial period of relatively rapid increase in charge followed by a longer period of less rapid increase. The extent of these periods and the magnitude of the charge passed were dependent upon the environment and test potential. Figures 37 and 38 clearly show these differences for tests done on cast duplex SS in 1M NaCl and white water solution, respectively. The inset in each of these two figures shows the cumulative charge passed in the first 12 ms after commencement of scratching. This time period corresponds to the cyclic loading period during the high frequency fatigue tests and is relevant to possible oxidation events at the crack tip during one load cycle. Figures 39 and 40 show the results of the cumulative charge passed in the case of wrought duplex SS in 1M NaCl and synthetic white water respectively. Similar to cast duplex SS in 1M NaCl solution, the charge passed is dependent upon the potential and the environment. WROUGHT DUPLEX SS 1 M NaCl 30 i 28 H 0 2 4 6 8 10 12 14 16 F i g u r e 39.CUMULATIVE CHARGE PASSED IN WROUGHT DUPLEX TESTED IN lMNaCl WROUGHT DUPLEX SS SYN.WHITE WATER TIME.ms Figure 40.CUMULATIVE CHARGE PASSED IN WROUGHT DUPLEX TESTED IN SYN.WHITE WATER. 61 6 6 6 £ 28KV X£ I 88K ' 1 5 I 8urn | 8 1 1 1 67 20KV X i ! 8 8 K ' ' ' 3 8 i (A)+0.3 Vsce (B)-0.4 Vsce da/dN = 7.62 E- l 1 m/cycle da/dN = 4.54 E-l 1 m/cycle AK = 4.29 MPaJTT AK = 6.08 MPaVftT Figure 41.TYPICAL APPEARANCE OF FRACTURE SURFACE AT POTENTIALS -0.4 & +0.3 Vsce. 611379 5.0KV X5.00K 6.0um Cast duplex 1M NaCl +0.3 Vsce da\dN = 7.62E-H m/cycle AK = 4.29 MPaJffT (a) Figure 42.TRANSGRANULAR CRACKING OF CAST DUPLEX IN 1M NaCl FAINT RIDGES ON OPPOSING FACES (MIRROR IMAGE) Cast duplex 1M NaCl +0.3 Vsce da\dN = 7.62E-H m/cvcle AK = 4.29 MPaTrn (b) Figure 42.1 K A N K A N U L A K LKACKJNU OF CAST DUPLEX IN 1M NaCl FAINT RIDGES ON OPPOSING FACES(MIRROR IMAGE) Cast duplex 1M NaCl +0.3 Vsce da\dN = 7.62E-H m/cycle AK = 4.29 MPaJm Figure 43.TRANSGRANULAR CRACKING OF CAST DUPLEX IN 1M NaCl FAINT RIDGES ON OPPOSING FACES (MIRROR IMAGE) 93 5.2.4 FRACTOGRAPHY 5.2.4.1 Near-Threshold Fractography (Stage I) Fatigue cracks followed a predominantly transgranular path in the near-threshold region at potentials of +0.3 Vsce and -0.4 Vsce> irrespective of the environment. Fracture surfaces obtained at -1.2 Vsce exhibited more evidence of interfacial failure. Typical examples of cast steel fractography produced in 1M NaCl at +0.3 Vsce and -0.4 Vsce are shown in Figure 41. Both specimens showed transgranular cracking through the austenite islands and ferrite matrix. At higher magnification, facets and ledges were observed on both phases. The ledges (river lines) were coincident with the local direction of crack propagation and are essentially shear ridges that formed between segments of crack front that were not propagating in the same plane. Examples are shown in Figure 42(a) and 42(b). These fractographs show that the shear ridges formed on the opposing faces of the fractured surface. Such crystallographic cracking is a common feature of near-threshold fatigue 121. Careful examination revealed that in the near-threshold region very faint fine ridges traversed the river lines. The ridges were visible on both opposing fracture surfaces, as shown in Figure 43. At very high magnification, the fine ridges were more clearly visible, with a spacing of =< 2 X 10-7 m. Careful matching of opposing fracture surfaces confirmed that the ridges on opposing faces matched peak to valley (interlocking surfaces). Eg.,Figure 44. Fatigue fracture surfaces produced on the cast steel at -1.2 VSce showed evidence of transgranular cracking, via the presence of crystallographic facets, and numerous interfacial fracture features. These characteristics are seen in Figure 45, sequence a,b and c. Fine ridges traversing the transgranular facets (river lines) were not observed. They may be absent or too fine to be detected. The wrought steel showed very similar near-threshold fractography of the Cast duplex 1M NaCl +0.3 Vsce daMN = 7.62E-H m/cycle AK = 4.29 MPaJnr Figure 44.TRANSGRANULAR CRACKING OF CAST DUPLEX IN 1M NaCl FAINT RIDGES ON OPPOSING FACES (HIGH MAGNIFICATION) 95 Cast duplex 1M NaCl -1.2 Vsce daW = 9.14E-10 m/cycle AK = 6.6 MPa<m Figure 45(A,B,C).EXAMPLES OF INTERFACIAL FRACTURE IN CAST DUPLEX AT-1.2 Vsce Cast duplex B lMNaCl C -1.2 Vsce da^ dN = 9.14E-10 m/cycle AK = 6.6 MPamT Figure 45 ( A , B , C ) . E X A M P L E S O F I N T E R F A C I A L F R A C T U R E IN C A S T D U P L E X A T-1 . 2 Vsce KD 010860 20KV X 1 0 . 0 K 3 . 0 u 010898 £0KV X 2 . 0 0 K 15.0um Wrought duplex 1M NaCl +0.3 Vsce daVdN = 4.06E-H m/cycle AK = 3.575 MPa m B Figure 46(A,B,C).TRANSGRANULAR CRACKING WITH FAINT RIDGES IN WROUGHT DUPLEX IN 1M NaCl AT +0.3 Vsce. 98 C Wrought duplex 1M NaCl +0.3 Vsce da\dN = 4.06E-11 m/cycle AK = 3.575 MParm Figure 4 6 ( A , B , Q . T R A N S G R A N U L A R C R A C K I N G WITH FAINT RIDGES IN W R O U G H T D U P L E X IN 1M NaCl A T +0.3 Vsce. 99 Wrought duplex 1M NaCl -0.4 Vsce da\dN = 6.77E-11 m/cycle AK = 5.80 MPaVm Figure 47.TRANSGRANULAR CRACKING WITH FAINT RIDGES IN WROUGHT DUPLEX IN 1M NaCl AT -0.4 Vsce. o o 5 1 4 4 3 9 20KV X 3 . 0 0 K 1 0 . 0 u m fx vl 5 1 4 4 7 8 26KV X 3 . 8 0 K 10 .Burn Wrought duplex 1M NaCl -1.2 Vsce da\dN= 1.17E-09 m/cycle AK = 4.796 MPa^ m B Figure 48.(A,B)EXAMPLES OF INTERFACIAL FRACTURE IN WROUGHT DUPLEX AT-1.2 Vsce A Cast duplex 1M NaCl -0.6 Vsce daVdN = 2.11E-07 m/cycle AK = 26.37 MPalffi Figure 49(A,B).CLASSICAL FATIGUE SITUATIONS IN MATCHING SURFACES CAST DUPLEX SS; 1M NaCl; -0.6 Vsce. 012410 £0KV X1!Q9K"""3@um B Cast duplex 1M NaCl -0.6 Vsce daVlN = 2.11E-07 m/cycle AK • 26.37 MPaflfi Figure 49(A,B).CLASSICAL FATIGUE STRIATIONS IN M A T C H I N G SURFACES CAST D U P L E X SS; 1M NaCl; -0.6 Vsce. 103 cast steel. At potentials of +0.3 V s c e (Figure 46) and -0.4 V s c e (Figure 47), transgranular facets were observed with fine ridges traversing the river lines. The ridge spacing was approximately 2X10-7 m, similar to that of the cast steel. At -1.2 Vsce. more interfacial fracture was observed, as seen in Figures 48a and 48b. It was not possible to easily distinguish transgranular ferrite regions from transgranular austenite regions on wrought steel fractography, because of similarities in fractographic appearance and the similarity in the distribution of the ferrite and austenite phases. 5.2.42 Stage II Fatigue Fractography In both steels, at higher stress intensities where there was no obvious effects of either the potential or environment on crack propagation rates, classical stage II fatigue striations were visible on the fracture surfaces. Such striations are usually associated with the simultaneous operation of several slip systems in stage I. An attempt was made to match the stage JJ striations on opposing fracture surfaces. It was possible to match the general features (eg., Figures 49a and 49b), suggesting that they formed at more or less the same time in the crack advance process and that their formation was related to each other. However, it was not possible to clearly determine whether the striations matched peak to valley (interlocking fracture surfaces) or peak to peak (mirror image surfaces). 104 5.3 DISCUSSION 5.3.1 General Behaviour The results confirmed that good solution mixing occurred between the crack solution and bulk solution, and that there was a definite effect of electrochemical potential on near-threshold fatigue behaviour in 1M NaCl and a negligible effect in synthetic white water. Also, the rapid scratch tests showed a pronounced effect of potential on transient current behaviour in 1M NaCl and, consistent with fatigue behaviour, a less pronounced effect of potential on current transients in white water. The fractographic observations confirmed that near-threshold transgranular cracks contained interlocking features as small as -0.1 um on opposing fracture surfaces and that fatigue cracks produced at -1.2 Vsce contained a significant proportion of interfacial fracture regions. It is worth noting that Bassidi et al. 105,129-131 did not report observing such fine scale interlocking features. This is attributed to the fact that they did not employ the necessary SEM conditions needed for their detection.(ie. low excitation voltages to improve contrast coupled with high magnification.) 5.3.2 Consequences of Solution Mixing The observed good solution mixing between the crack solution and bulk solution implies that the crack solution chemistry and pH were reasonably similar to the bulk solution. Under these circumstances, it is possible to define the electrochemical conditions under which hydrogen may be evolved at the crack tip via the reduction of hydrogen ions, 2H+ + 2e- = H2 (9) The Nernst equation defines the thermodynamic conditions of potential below which hydrogen is evolved. At a temperature of 25 deg.C (298 K), and one atmospheric pressure, the Nernst equation is given by, Vshe = -2.303 R T (pH) / F , Volts (10) where R is the gas constant (8.314 J.K-1 mol-1), T is the temperature (K), F is the 105 Faraday (96500 Coulombs), and she refers to the Standard Hydrogen Electrode scale. The pH of the 1M NaCl was 6 and that of synthetic white water was 4.7. Therefore, from equation (10), hydrogen may only be evolved below -0.355 Vshe in 1M NaCl and below -0.278 Vshe in white water. These potentials may be placed on SCE scale used in the tests by the conversion, Vsce = Vshe - 0.246, V....(ll) This shows that hydrogen evolution at the crack tip was unlikely during fatigue experiments above -0.601 Vsce in 1M NaCl and above -0.524 Vsce in synthetic white water. Therefore, it is possible to separate the fatigue tests into conditions where hydrogen embrittlement processes were possible (eg., tests at -1.2 Vsce) and conditions where only oxidation-related processes were possible (eg., -0.4 Vsce and +0.3 Vsce). This assists interpretation of the results on a mechanistic basis and allows the anodic current transients obtained during the rapid scratching tests to be related more meaningfully to the fatigue behaviour. 5.3.3 Scratch Tests - Repassivation Kinetics The scratch tests showed that the anodic potential at which the highest crack growth rates (da/dN) were observed at low AK coincided with the potential at which the highest current transient was obtained. This potential also produced the highest passage of coulombic charge during repassivation of the scratch. At the highest anodic potential (+0.3 Vsce) in 1M NaCl, most of the charge was passed during the time period spent by the crack tip during one complete load cycle (12 ms). The anodic transient curves comprised of two regions. A region of rapidly rising current (region I) followed by a region of decreasing current (region JJ). In region I the current continued to rise after the completion of the scratch (-2.3 ms). The increasing current is attributed principally to the nucleation and lateral spreading of oxide nuclei on the bare surface coupled with some thickening of the nuclei. There may be some dissolution of bare surface but it is believed to be small because the potential conditions 106 were merrnodynamically favourable for film (oxide) formation. As soon as a bare surface is created, oxide molecules must arrange themselves in clusters of minimum critical size before growth of oxide can take place. This process is analogous to nucleation and growth process during phase transformation of metals. The critical size (radius, R) of the nucleus after which growth can take place is give by 122 R=lS/ZFh .(12) where 1 is the boundary free energy/unit length, S is the area covered by 1 mole in a mono-atomic layer, Z is the atomic number, F is the Faraday and h is the over-voltage (electrochemical driving force). The critical sized clusters have the same probability to grow or dissolve. Large clusters will tend to grow while small ones will tend to dissolve. At any instant of time after the completion of the scratch, far more critical sized nuclei will be present at a higher over-voltage (because the critical size is smaller) than at a lower over-voltage. The growth of these nuclei is responsible for the increase in current. As expected, the rate of current increase is determined by the number of critical nuclei, the number being higher at the higher over-voltage. The growth of nuclei can take place either at the edge of the nucleus or at the top of nucleus. Ion migration through the film has to take place for the top to grow (i.e. for the film to thicken). Therefore thickening is a relatively slow process compared to lateral growth at the edge of the nuclei. The rate of lateral growth of a circular nucleus is linear and hence a linear increase in current can be expected at the end of the scratch, as seen in Figures 35 and 36. The peak in the current transient corresponds to a point when there is complete coverage of the surface. Island formation of thin oxide films has been observed!23. These islands eventually grow together to form a continuous oxide film. At the point of complete coverage the film will not be uniformly thin, because the slower thickening processes will be accompanying spreading (lateral growth). Hence, the early formed regions will be thicker than the later regions of laterally formed film. The current tends to decrease once the surface is completely covered (region 107 II) because film thickening or growth occurs by ion migration through the already formed film. The kinetics of very thin film growth has been known to follow the logarithmic law of the forml23, 8 = Aln (1+Bt) (13) where 8 is the thickness of the film , t is the time in milliseconds and both A and B are constants. Fehlner and Mottl23 have given a differential form of the logarithmic growth rate law as (d8/dt) = A e[-(W)/kT] (14) where W = (WO+p.8) is the activation energy. WO is defined as the activation energy at zero thickness and 'p.' is a constant characteristic of the oxide formed. They have proposed that as the oxide thickens (ie., increase in 8), the growth rate decreases due to decrease in ionic mobility. The decrease in ionic mobility was attributed to either blocking of pores or elimination of channels in the film. However, since oxide formation and thickening occurs by electrochemical processes in aqueous solutions, as shown in the following generalized equation (15), x M + yH20 = M x O y + 2yH+ + 2ye"....(15) then it is obvious that the charge passed (Q),during film formation is proportional to the film thickness (8). Therefore equation (14) can be rewritten as, (dQ/dt) = A ' e[-(W0+uQ)/kT]....(16) ln(dQ/dt) = InA' -(W0+uQ)/kT....(17) ln(dQ/dt) = C + DQ....(18) where C = (InA' -WO/kT) and D = (u/kT) Consequently, a plot of ln(dQ/dt) versus Q in region II, where Q is the cumulative charge passed during the scratch test, should be a straight line if all the charge is consumed by film formation. Also, both C and D should vary with potential. Figures 50a and 50b show a linear plot of ln(dQ/dt) versus Q for the cast and wrought steels at +0.3 Vsce in the synthetic white water. Similar plots were observed at 108 LN(dQ/dt) VS Q Cast duplex/syn.white watcr/+0.3Vscc 0 20 40 60 LN(dQ/dt) VS Q Wrought duplex/syn.white water/0 Vsce -2.2 --2.4 -•2.6 --2.8 --3 --3.2 --3.4 --3.6 --3.8 --4 --4.2 -4.4 -4.6 -4.8 " -5 --5.2 - -10 Q 50.FIGURE SHOWING LOGARITHMIC FILM GROWTH RATES 109 CAST DUPLEX SS 1M NaCl POTENTIAL C 0 R2 No.or POINTS + 0.3 7.1 -0.0857 0.965 16 0 3.4 -0.208 0.86 6 -0.2 4.32 -0.365 0.86 3 -0.4 3.53 -0.742 0.87 3 -0.6 3.19 -2.042 0.78 4 CAST DUPLEX SS SYN.WHITE WATER POTENTIAL C O R2 No.or POINTS + 0.3 -0.38 -0.0389 0.9735 197 0 0 . 391 -0.221 0.845 200 -0.2 -0 .65 -0.133 0.967 195 -0.4 -2.095 -0.305 0.997 195 -0 . 6 -2 .57 -1.56 0.95 197 WROUGHT DUPLEX SS 1M NaCl POTENTIAL C D R2 No.or POINTS + 0.3 10.62 - 0 .345 0.745 3 0.20 5.69 - 0 .165 0.596 3 -0.2 0.986 - 0 .365 0.86 3 WROUGHT DUPLEX SS SYN.WHITE WATER POTENTIAL c D R2 NO.or POINTS + 0.3 0 .345 -0.1626 0.888 161 0.2 0 .381 -0.735 0.93 158 0 -0.9 -0.427 0.974 151 Table 7 Constants C and D for oxide growth rates 110 other potentials and the values of the constants C and D changed with potential, as shown in Table 7. However, as good a linear fit could not be obtained for scratch tests done in 1M NaCl compared to that in synthetic white water. This was because the region II current transients decreased very rapidly and only a small number of data points (~3 in some cases) were obtained before the current became immeasurably small. Consequently, accurate analyses of region II were not possible. Overall, the rapid scratching tests and analyses of the current transients suggested that a predominant portion of the current, if not all the current, was used in film nucleation and growth processes. In other words film rupture was followed by repassivation (film formation) and there was very little (if any) metal dissolution. In the fatigue studies, the rising part of the load cycle occupied about 5.6 ms. During the initial part of the rising load-time curve, the crack tip strain may be insufficient to rupture a crack tip film. Therefore, it may be argued that film rupture occurs only during the final portion of the load-time cycle (eg., the last 1-2 ms). Comparison with the scratch tests show that 2 ms was insufficient time for complete film coverage of the bare surface to occur (ie., time to peak current) in any of the test conditions. There was only time for film nucleation and spreading. 5.3.4 Dissolution Mechanism The analyses of the scratch tests on the wrought and cast duplex steels have shown that current transients following rupture of a passive film at the crack tip are predominantly due to repassivation (film formation) phenomena and not metal dissolution. Thus, it is highly improbable that dissolution processes contribute in any significant way to the propagation of fatigue cracks. Nevertheless a simple analysis was undertaken on the assumption that the peak current density (ip) obtained during the rapid scratching tests were maintained continuously at the crack tip of a propagating crack. The current was converted to metal dissolution rate (R) via the Faraday's law R = ip W/F p (19) I l l where W is the averaged equivalent weight of the alloy and p is the density. The predicted crack growth rate per cycle, Rp, based on dissolution controlled crack advance then becomes, Rp = R/f....(20) where f is the cyclic loading frequency. The predicted rates are listed for all test conditions in Table 8. Comparison with observed crack growth rates in Figures 20-23,26-27 shows that dissolution cannot account for near-threshold crack propagation rates except at +0.3 Vsce. Other factors that negated the importance of dissolution during corrosion fatigue cracking of the duplex SS were the absence of fractographic features that could be identified with dissolution phenomena at the advancing crack tip. 5.3.5 Film-Related models of Corrosion Fatigue Models of corrosion fatigue that involve the generation of bare metal and the subsequent formation of oxide films at the crack tip are oxidation-related mechanisms of cracking. In the context of the present study, they are most relevant to those potentials where hydrogen embrittlement effects can be discounted on thermodynamic grounds (eg., fatigue tests at +0.3 Vsce and -0.4 Vsce). The analyses of the rapid scratch tests provided strong support for film-related models because at these same potentials it was concluded that film formation, rather than dissolution, was the predominant oxidation process following film rupture. Other factors that must be considered are the thickness of the films that form during repassivation of the crack tip and the time scale of the loading cycle (12 ms) relative to the time required for nucleation and growth of films. Estimates of the charge required to form a monolayer of oxide, assuming a spinel-type oxide, indicate that -8.4 C/m2 are necessaryl24 (See Appendix-2). The two models of most interest are those involving film-induced cleavage and restricted slip reversibility. 112 CAST DUP LEX SS Pot lMNaCl SYN.WHITE WATER Peak Current Density Predicted Velocity Peak Current Density Predicted Velocity Vsce A/m2 m/sec m/cycle A/m2 m/sec m/cycle +0.3 +0.2 0 -0.2 -0.4 -0.6 24675 9935 2649 2155 974 337 9.08X10-7 3.65X107 9.75X10* 7.93X10'* 3.58X10* 1.24X1Q-* 1.07X10' 4.30Xia9 l.isxia' 9.33X1010 4.22X1010 1.46X1010 538.9 292.2 266.2 96.1 41.55 15.58 1.98x1a1 1.07X10-' 9.79X109 3.65X10^  1.53X109 5.73X109 2.33X10" 1.26X10" 1.15X10" 4.16X10" 1.83X10" 6.74X10" WROUGHT DUPLEX SS Pot 1M NaCl SYN.WHITE WATER Peak Current Density Predicted Velocity Peak Current Density Predicted Velocity Vsce A/m2 m/sec m/cycle A/m2 m/sec m/cycle +0.3 +0.2 0 -0.2 -0.4 16994 9501 744.8 6.30X1O7 3.53X107 2.76X10* 7.41X109 4.14X10' 3.25X1010 346 167 130 20 1.28X10* 6.22X10' 4.83X10' 7.3X1010 1.51X1010 7.31X10" 5.68X10" 8.70X1O12 Table 8 Predicted crack velocities by dissolution mechanism 113 5.3.5.1 Film-Induced cleavage (FIC) Crack propagation due to repetitive film-induced micro-cleavage of a coherent brittle film has been proposed as a mechanism of crack growth for transgranular stress corrosion crackingl25. The initial thin oxide film is assumed to be capable of sustaining large elastic strains. As the film thickens with time it becomes brittle (i.e., incapable of sustaining large strains). Cracks eventually form in the film at the highly strained crack tip and propagate into the substrate, thereby creating fresh metal surface on which a new film forms. The FIC model may be applied in similar fashion to fatigue crack growth. A crack will form in the film at the crack tip and propagate a small finite increment into the ductile matrix. The newly exposed metal surface then repassivates and the FIC cycle is repeated. At high loading frequencies, as used in the present work, it may be argued that several load cycles are required for the newly formed crack tip film to thicken to the condition where it becomes brittle. Consequently, if the FIC model is operative, then one should see series of fine ridges on the fracture surface corresponding to micro-cleavage events. The ridges will be perpendicular to the river lines and parallel to the crack front. The distance between the ridges should represent the total jump distance of the micro-cleavage event and not simply the thickness of the film. The transgranular near-threshold fatigue crack fractography exhibited fine ridges that interlocked on opposing fracture surfaces, consistent with micro-cleavage cracking. However, the distance between the fine ridges was constant near ~2X10~7 m (i.e. ,2000A) independent of the potential and independent of near-threshold crack velocities ranging from 7.62X10-11 m/cycle upto ~8X10-10rn/cycle. At the lowest crack velocities, the ridge spacing corresponded to the passage of 2625 load cycles. The constancy of the ridge spacing was not entirely inconsistent with the FIC model provided that the ridges corresponded to individual micro-cleavage events and that the predominant cleavage increment occurred in the matrix. For example, the scratch 114 tests showed that film nucleation and growth was most rapid at the highest test potential of +0.3 Vsce, especially in 1M NaCl. Hence, the time period (i.e., number of cycles) required to achieve a critical film thickness for triggering the micro-cleavage event should be lower at +0.3 Vsce than at -0.4 Vsce. Hence, micro-cleavage events will occur more rapidly at +0.3 Vsce, giving rise to higher crack velocities, as observed in 1M NaCl. However, there was an inconsistency between the FIC model and the constancy of the ridge spacing at constant potential and different crack rates. At the higher near-threshold crack rates in each test, corresponding to higher AK values, it is expected that the cleavage crack increments in the matrix should increase with AK, indicating that the ridge spacing should increase with rising AK in direct contrast to the observed behaviour. It was considered highly unlikely that the ridge spacing (-2000 A) corresponded solely to a micro-cleavage event in an oxide film of same thickness (ie., no cleavage in the substrate). For example, a film thickness of 2000 A consists of -200 monolayers of oxide, each monolayer requiring a charge of 8.4 C/m2 124(se  Appendix-2), for a total charge requirement of -1680 c/m2. Inspection of the cumulative charge plots in Figures 37,38,39 and 40 shows that the accumulated charge during film formation in 1M NaCl did not increase significantly after 15 ms, and that the total charge in all cases, even at +0.3 Vsce was significandy less than -1680 C/m2. Hence crack tip films of -2000 A should not form. Overall, a sufficient number of inconsistencies exist between the predicted fatigue behaviour, based on FIC, and the observed behaviour to suggest that the FIC model is not applicable to the present studies. 5.3.5.2 RESTRICTED SLIP REVERSIBILITY(RSR) Slip reversibility is an old established concept that has been used to describe the process of fatigue crack nucleation. Fong and Tromansl21 recently extended the concept to corrosion fatigue crack propagation. The main elements of their RSR model are presented in Figure 51, where a sequence of events (a) to (h) are depicted. At position Figure 51.Schematic diagram of stage I fatigue crack propagation by restricted slip reversibility mechanism 116 51a, the crack is at minimum load of the load cycle and slip reversal is occurring on only one favorably oriented variant of the possible slip systems along two parallel slip planes, SI and S2. Forward slip occurs solely on SI during the rising load cycle (51b), rupturing any surface films that may be present, and produces a slip step of length x. During the decreasing load cycle (51c) and (5Id), an increment of slip reversal, (x-Ax), occurs on SI and a final increment of slip reversal, Ax, occurs on S2. At minimum load (51d), a permanent increment of crack advance, Ax, has been produced. This process may be repeated over several or many load cycles to produce a final summated increment of crack advance, Ax, after which a similar process may occur on another (conjugate) favorably oriented slip system variant along another pair of parallel slip planes, S3 and S4, as shown in Figures (51e) through to (51h). Repetition of periods of propagation on the conjugate systems (Figures 51(a) through to 51(h)) will produce a fracture surface composed of a series of fine scale ridges that interlocked on opposing fracture surfaces. This prediction of the RSR model was consistent with the fractographic observations. The RSR model predicts that the average crack growth increment per cycle (Ax) is dependent upon factors that control slip reversibility on the forward slipping planes. Such factors are expected to be the degree of work hardening and recovery on the slip planes, and the rate of oxide nucleation and growth on the newly exposed forward slipping surface. The formation of the oxide will now be examined in relation to the fatigue testing conditions employed. The period of rising load during each fatigue cycle was ~5.6 ms. Initial strains at the crack tip during the rising load are expected to be elastic, followed by plastic strains (slip) as the peak load is approached. Hence, it is anticipated that formation of newly exposed metal surface by slip on the forward slipping plane occurred for only 2-3 ms. Inspection of the rising current transients (region I) of the scratch tests show that 3 ms were insufficient for complete film coverage of the exposed slip step under any of the fatigue testing conditions. There was only sufficient time for some oxide nucleation and 117 lateral growth of these nuclei. However, the scratch tests showed convincingly that the rate of oxide nucleation (eg., rate of rising current transient and cumulative charge) was dependent on the environmental testing conditions. For example, the rate of oxide nucleation was higher at +0.3 Vsce in 1M NaCl than at -0.4 Vsce. Hence, during fatigue, a larger area of newly exposed surface on the forward slipping surface will be covered by patches of oxide than at -0.4 Vsce. This will cause a greater restriction on slip reversibility at +0.3 Vsce, raise the average crack growth increment per cycle (Ax) and produce higher crack propagation rates, as observed. The lower fatigue crack growth rates at +0.3 Vsce in synthetic white water, relative to 1M NaCl, were also consistent with the RSR model. The scratch tests showed that the rates of oxide nucleation and growth were lower in white water, indicating a smaller degree of oxide coverage on the forward slipping surface at the crack tip and a lower crack growth increment per cycle. Also, the oxide nucleation and growth rates in white water were relatively insensitive to changes in potential, indicating that the crack growth increment per cycle should be independent of potential, consistent with the observations. Overall, the predictions of the RSR model are in qualitative agreement with the observations, indicating that it is a reasonable mechanism for near-threshold fatigue behaviour. 5.3.6 CORRELATION BETWEEN SCRATCH TESTS AND FCP RATES Irrespective of the detailed mechanism of cracking, the results indicated a useful empirical correlation between scratch tests and FCP rates at anodic potentials where hydrogen evolution does not occur. For example Figures 35 and 36 show that the current transients at +0.3 Vsce in 1M NaCl were quite high for both cast(1520 mA) and wrought duplex SS(1160 mA). FCP rates obtained under the same conditions were also correspondingly high (see Figures 20,26). In the same solution, at -0.4 Vsce, both the current transients and FCP rates were low. In the synthetic white water environment, the 118 peak current transients were low at all potentials and were similar to each other. Correspondingly, the FCP rates were low and comparable to each other. These observations suggest that those alloy/environment combinations, which exhibit large peak current transients and rapid repassivation kinetics should also be the combinations which show fast FCP rates. Although more alloy/environment combinations need to be tested to see whether the correlation is generally applicable , it appears that rapid scratch tests can be used as an indicator to rank alloys for CF susceptibility in the environment of interest. This is an attractive route because fatigue tests are costly and time-consuming, whereas scratch tests are simple and can be performed quite easy. 5.3.7 HYDROGEN EMBRITTLEMENT Thermodynamic considerations have shown that conditions were favourable for hydrogen evolution within cracks during fatigue testing at -1.2 Vsce. Therefore, the possibility of hydrogen embrittlement must be considered. Other crack propagation processes may still occur (eg., by the RSR mechanism). However, a contribution to crack propagation by a hydrogen embrittlement process may account for the higher crack rates at -1.2 Vsce, relative to -0.4 Vsce, in 1M NaCl. Also, hydrogen-related effects may account for the significant proportions of interfacial fracture at -1.2 Vsce. Bassidi et al.105 also suggested that hydrogen embrittlement was responsible for the fatigue crack growth behaviour in duplex SS. However, their conclusions were not supported on the basis of thermodynamics (ie. under well defined conditions of potential and pH at the crack tip) and purely rested on fractographic observations of transgranular fracture regions. Although many mechanisms have been proposed for the hydrogen embrittlement of 81661876-79, identification of the possible mechanism responsible for fatigue cracking in the present work was beyond the scope of the study. Whatever the mechanism, the crack velocity is expected to be influenced by the rate at which hydrogen accumulates in the region ahead of the crack tip. The factors controlling transport of 119 120 atomic hydrogen through the two phases of a duplex steel will be discussed briefly. Transport of atomic hydrogen by conventional lattice diffusion processes at room temperature is much higher in ferrite than austenite 126. Consequently, hydrogen can accumulate more quickly in the ferrite phase, thereby contributing to the well established greater sensitivity of the bcc phase to hydrogen embrittlement. However, propagation of a hydrogen embrittlement crack through a duplex structure requires embrittlement of the austenite phase. Therefore, more unconventional hydrogen transport processes must be operative in order for sufficient hydrogen to accumulate in the austenite. Such processes are usually considered to be dislocation transport controlled!27,128. Hydrogen atoms segregate to the highly strained regions of dislocations and migrate with the moving dislocations in the crack tip region. In this manner, small discrete regions of austenite phase at the crack tip may become sufficiently embrittled by hydrogen to sustain a small increment of crack advance. It should be recognized that the tip of a fatigue crack represents a unique situation. There is considerable sustained dislocation activity, particularly at high loading frequencies, involving forward and reverse slip. Under these circumstances, it is possible that dislocation transport of hydrogen into the local metal lattice can be quite significant. Dislocations may transport hydrogen to the interfaces between phases, thereby promoting interfacial fatigue fracture. Significant interfacial fracture regions were observed at -1.2 Vsce on SEM fractographs. It was not possible to determine from the fractographs whether the interfacial fractures were a-a boundaries or a-y boundaries. However, the optical micrograph in Figure 52 shows a crack propagating along cx-y interfaces. Other workers have shown that hydrogen assisted fracture can occur along interphase boundaries 126. 5.3.8 EFFECT OF SPECIMEN ORIENTATION ON CRACK PROPAGATION The slower rates of near-threshold fatigue crack growth (da/dN) in T-L orientations of the wrought steel, relative to the L-T orientation, at the potentials of -0.4 121 Vsce and +0.3 Vsce were difficult to explain. Fatigue cracking was transgranular in all cases, exhibiting crystallographic features and river lines. Consequently, as the advancing crack traverses an interface (or boundary) into a new phase (or grain), the crack front has to readjust to the orientation of the newly encountered phase (grain). Such an interaction is expected to momentarily retard crack advance. Hence, the more interfaces a crack encounters, the slower the anticipated overall crack growth rate. In L-T orientation, an advancing crack should encounter more interfaces than T-L orientation, due to the elongated nature of the microstructure (see Figure-8b). Therefore, L-T orientation should crack more slowly than T-L orientation, contrary to the actual observations. At this point, there is no possible explanation to account for the differences in fatigue behaviour between T-L and L-T orientations at -0.4 Vsce and +0.3 Vsce. 5.3.9 Summary High frequency fatigue testing of a cast and wrought duplex SS was conducted to effect good mixing between the crack tip solution and the bulk solution. Because of this, meaningful potential control was applied to separate the oxidation related mechanisms from the hydrogen embrittlement mechanism. The results of the scratch tests along with fractographic evidence were used to make meaningful comparison between the possible oxidation-related mechanisms responsible for FCP at potentials where hydrogen evolution was impossible on thermodynamic grounds. Both dissolution and film-induced cleavage models were discounted because it does not support all the observations. It was concluded that FCP at anodic potentials were consistent with a restricted slip reversibility model of cracking. At potentials where hydrogen is evolved during fatigue, it was concluded that hydrogen transport to the crack tip region and subsequent embrittlement of the region ahead of the crack tip was responsible for increase in FCP rates. 122 6.C0NCLUSI0NS The pitting and corrosion fatigue studies on a cast duplex SS and wrought duplex SS lead to the following conclusions regarding the behaviour of duplex SS in aqueous chloride solutions: 1.Preferential pitting of either a-phase or y-phase may occur. This is related to partitioning of the alloy elements between the two phases. There is an enrichment of Cr and Mo in the a-phase and of nitrogen (when present) in the y-phase. 2.1n the absence of alloyed nitrogen (cast steel), enrichment of Cr and Mo in the a-phase causes the y-phase to be most susceptible to pitting. When alloyed nitrogen is present (wrought steel), enrichment of nitrogen in the y raises the pining resistance of this phase. 3.It is proposed that the beneficial effect of nitrogen on the pitting resistance of the y-phase is due to a surface enrichment of nitrogen, possibly because of the irreversibility of electrochemical reduction of alloyed nitrogen. 4. Near-threshold fatigue crack propagation studies are affected by the applied potential in 1M NaCl solution. In the range of the potentials where hydrogen is not produced, the increased crack propagation rates at +0.3Vsce compared to -0.4Vsce are due to the higher nucleation and growth rate of oxide at +0.3Vsce and the resulting effect on restricted slip reversibility on the newly slipped surfaces at the crack tip. 5. There is not much effect of potential on crack propagation rates in the synthetic white water solution. This is attributed to a lesser degree of oxidation (lower peak anodic current transient and slow repassivation rate). Consequently, slip reversibility is not restricted to the same degree as compared to +0.3Vsce in 1M NaCl solutions. 6. The rising part of the scratch test current transient is attributed principally to the nucleation and spreading of oxide nuclei. The peak currents occurred when the film coverage is complete. The decaying current is due to the growth and thickening of the film. Thickening of the film followes logarithmic kinetics. 123 7.At cathodic potentials where hydrogen is produced, the increased cracking rate is attributed to embrittlement of the crack tip by transport of atomic hydrogen into the metal lattice and to interfaces. The fractography indicated interfacial cracking. 7.The Scratch test could provide an easy test to assess the susceptibility of an alloy to corrosion fatigue in the environment of interest. A large current transient and a very rapid repassivation behaviour should be indicative of faster fatigue crack growth rates at anodic potentials. 124 7. SUGGESTIONS FOR FUTURE WORK l.In stage I fatigue, where crystallographic cracking is possible, etch pitting studies of fracture surfaces in austenitic SS121 have shown evidence in support of the RSR mechanism. However, in the present study, it was difficult to pit the ferrite areas preferentially. High frequency fatigue testing of a coarse grained ferritic steel followed by etch pitting of the fracture surfaces should be performed to check whether the observations are in support of RSR mechanism. 2.High frequency fatigue testing should be performed in bulk solutions having a composition (pH) corresponding to a crevice solution that may develop under low frequency loading. Concurrently, rapid scratch tests in the same solution should be done to check the correlation between current transients and FCP rates. 125 REFERENCES 1. Metals Handbook, Vol 13, Corrosion, ASM, Metals Park, Oh. 2. P.H.Thorpe, Third Intnl. Symp. on corrosion in the Pulp and Paper Industry,pp.l84-190, NACE, Houston,1982. 3. N.Sridhar, J.Kolts, Corrosion, Vol.43, No.l 1, pp.646-651, Nov. 1987. 4. H.Tsuge,Y.Tarutani and T.Kudo, Corrosion, Vol.44, No.5, pp.305-314, May 1988. 5. "Binary Alloy Phase Diagrams",T.B.Massalski, Ed-in-Chief, ASM, Metals Park, Oh; 1:1081,1986. 6. C.C.Nathan and AJ.Piluso, Pulp & Paper Industry Corrosion Problems, NACE, Houston, TX, Vol.1, pp.126-132, 1977. 7. D.F.Bowers, Paper No. 144, Corrosion 86, Houston, NACE, 1986. 8. R.Yeske, Proc. 1987 Engg.Conf, pp.l7-25,TAPPI Press Atlanta, Georgia, 1987. 9. D.Craig Reed and Donald A Wensley, Tappi Journal, pp.57-61, Mar 1989. 10. A.Castillo, "Alloy Development based on Past and Present Experiences", Tappi Engg.Conf., Tappi Press, pp.601-604, Atlanta, 1984 11. R.Yeske,TAPPI Journal,pp.47-54, March 1988. 12. C.Fong and D.Tromans, Met.Trans.A, Vol.l9A, pp.2753-2764, 1988. 13.S.J.Hudak,Jr., D.L.Davidson & R.A.Page, "Embrittlement by the localized crack Environment", R.P.Gangloff, Ed., TMS-AIME, Warrendale, Pa, pp. 173-98, 1984. 14. H.Spaehn, "Sub-critical crack growth due to Fatigue, Stress Corrosion and Creep, L.H.Larsson, Ed., Elsevier Applied Science Publishers, London, U.K.,pp.275-303, 1984. 15. D.A.Jones, Met.Trans.A, V 0 I . I 6 A , pp.1133-41, 1985. 16A.Boetang, J.A.Begley and R.W.Staehle, Corrosion, Vol.36, pp.633-38, 1980. 17. A.Boetang, J.A.Begley and RW.Staehle, Met.Trans.A, Vol.lOA, pp.1157-64,1979. 18. C.Amzallag, P.Rabbe and A.Desestret, "Corrosion fatigue behaviour of some special stainless steels", Corrosion fatigue Technology, ASTM STP 642, H.L.Craig,Jr., T.W.Crooker and D.W.Hoeppner, Eds., pp.117-132, ASTM, Warrendale, Pa.1978. 19. Corrosion Engineering, M.G.Fontana and N.D.Greene, McGraw Hill Book Co., 1978. 20. W.H.Hartt, J.S.Tennant and W.C.Hooper, "Solution chemistry modification within corrosion fatigue cracks", Corrosion fatigue technology, ASTM STP 642, H.L.Craig,Jr., T.W.Crooker and D.W.Hoeppner,Eds., pp.5-18, ASTM Warrendale, Pa. 1978. 126 21. R.Sriram and D.Tromans,paper submitted to Corrosion 22. RI.Marshall and G.T.Burstein, Corrosion Science, Vol.24, No.5, pp.463-478,1984. 23. J.W.01dfield, International Metals Review,V.32,No.3, pp. 153-170,1987. 24. R.J.Brigham, Mater.Perform., Vol.13, p.29, Nov 1974. 25. Y.M.Kolotyrkin, J.Electrochem.Soc.,Vol.l08,pp.209-216, March 1961. 26. H.P.Leckie and H.H.Uhlig, J.Electrochem Soc., Vol.113, No. 12, pp.1262-1267, Dec. 1966. 27.I.L.Rosenfield and I.S.Danilov, Corrosion Science, 7,129,1967. 28.T.P.Hoar,D.C.Mears and G.P.Rothwell, Corr.Science, Vol.5, No.4, pp.279-289, April 1965. 29.S.Maitra and E.D.Verink, "Passivity of Metals", R.P.Frankenthal and J.Kruger Eds.,The Electrochemical Soc.,USA,p.309,1978. 30J.Augustynski,ibid,989,1978. 31. M.F. Abd Raboo, G.C.Wood,J.A.Richardson and C.K.Jackson,Corrosion Sci., 14,645,1974. 32. N.Sato, Electrochimica. Acta, Vol.16, pp.1683-1692,,1971. 33. T.RHoar, Corr.Science, Vol.7, No.6, pp.341-355, June 1967. 34. Corrosion of Stainless steels.A.John Sedriks, Wiley-Interscience Publication,pp.63-87,1979. 35J.W.01dfield and W.H.Sutton, Br.Corr.J. Vol.13, No.l, pp. 13-22, Jan 1978. 36. J.W.01dfield and W.H.Sutton, Br.Corr.J., Vol. 13, No.3, pp. 104-111, Mar 1978. 37. J.M.Kolotyrkin, Corrosion, Vol.19, No.8, pp.261t-268t, Aug 1963. 38. E.A.Lizlovs & A.M.Bond, J.Electrochem. Soc., V.116, No.5, pp.574-579, May 1969. 39. R.Steigerwald, Corrosion, Vol.22, No.4, pp. 107-112, April 1966. 40. N.D.Stolica, Corros.Sci. Vol.9, No.4, pp.205-216, Apr 1969. 41. J.Horvath & H.H.Uhlig, J.Electrochem. Soc., V.115, No.8, pp.791-795, Aug 1968. 42. M.A.Streicher, J.Electrochem.Soc., Vol. 103, No.7, pp.375-390,Aug 1956. 43. R.Bandy and D.Van Rooyen, Corrosion, Vol.39, No.6, pp.227-236,June 1983. 44. A.Moskovitz Et.al., Effect of Residual elements on properties of Stainless steel, ASTM STP 418, pp.3, ASTM, Philadelphia, Pa,1969. 127 45. Z.Szklarska Smialowska, "Pitting Corrosion of Metals", NACE, Houston, TX, 1986. 46. M.Henthome, Corrosion, Vol.26, No.12, pp.511-528, Dec 1970. 47. H.J.Dundas and A.P.Bond, Paper No.159, Corrosion 75,1975. 48. M.A.Streicher, J.Electrochem.Soc., Vol.l03,pp.375-390,1956. 49. Z.Szklarska-Smialowska & MJanik-Czachor, Corros.Sci.,Vol 7,No.2, pp.65-72, Feb 1967. 50. R.M.Forbes-Jones and R.M.Kain, Paper No.67, Corrosion 75, 1975. 51. R.C.Newman, H.P.Wong, H.Ezuber and A.Garner, Paper No.202, Corrosion 87, San Francisco, 1987. 52. P.T.Gilbert, Metallurgical Reviews,l,379-417, 1956. 53. J.Congleton and I.H.Craig, Corrosion Processes; Ed:R.N.Parkins, pp.209-269, Applied Science Publishers,1982. 54.S.T.Wolfe and J.M.Barsom, Fracture and Fatigue Control lin Structures, Eds:N.M.Newmark and W.J.Hall, Prentice Hall Inc., New Jersey (1977). 55. L.F.Coffin, Fatigue at high temperatures, ASTM STP 520, pp.5-34,1973. 56. H.Ishii and J.Weertman, Scripta Met.,Vol.3,pp.229-232, April 1969. 57. R.N.Wright and A.S.Argon, Met.Trans.,Vol.1, pp.3065-3074, Nov.1970. 58. D.Broek, Elementary Engineering Fracture Mechanics, Noordhoff, Leyden, 1976. 59. R.W.Hertzberg, Deformation and Fracture Mechanics of Engineering Materials, John Wiley & Sons Inc., N.Y., N.Y. 1983. 60. T.Pyle,V.Rollins and D.Howard.J.Electrochem. Soc., Vol. 122, pp. 1445-1453,1975. 61. C.Laird and D.J.Duquette, Corrosion Fatigue: chemistry, mechanics and microstructure, pp.88-113,NACE, Houston, 1972. 62. D.J.Duquette and H.H.Uhlig, Corrosion 25, pp.839,1969. 63. C.Garcia and D.J.Duquette in "Modelling Environmental Effects on Crack Growth Processes., Proc. Symp. Toronto, Canada, 1985, Eds: R.H.Jones and W.W.Gerberich, pp.343-354, Metallurgical Soc. of AIME, Warrendale, Pa., 1986. 64. D.J.Duquette and H.H.Uhlig, Trans.ASM,62,839-845, 1969. 65J.Moskovitz and R.M.Pelloux, Proc. of ICMII, Spl.Volume, pp.295-312, ASM 1976. 66. Paris.P.and Erdogen F, J.Basic Engg., ASME Trans., 85, 528-534, Dec. 1963. 67. R.O.Ritchie, International Metal Reviews, No.5 & 6,pp.205-230,1979. 128 68.R.P.Wei and Landes, Mats. Res.Standards,9,25-27,1969. 69.I.M.Austin and E.F.Walker, Proc. Int. Conf. on the effect of environment on Fatigue, Inst, of Mech Engg. Conf. Publication 4, pp. 1-10,1977. 70. D.Rhodes,J.K.Musuva and J.C.Radon, Engg.Fracture Mechanics, Vol.15, No.3-4, pp.407-419, 1981. 71. F.P.Ford, Corrosion,Vol.35, No.7, pp.281-287, July, 1979. 72. A.J.McEvily and R.C.Boettner, Acta Met., 1963, Vol2, pp.725-743. 73. H.Ishii and J.Weertman, Met.Trans., Vol.2,pp.3441-52,1971. 74. G.A.Miller,D.H.Avery, and W.A.Backofen, Trans. TMS-AIME, Vol.236, pp.1667-73, 1966. 75. G.R.Caskey,Jr., Hydrogen Degradation of Ferrous Alloys, Eds: R.A.Oriani, J.P.Hirth and M.Smialowski,pp.822-862, Noyes Publications, New Jersey, 1985. 76. C.Zaffe and C.Sims, Trans AIME, p.225, V.145, 1941. 77. N.J.Petch and P.Stables, Nature, V. 169, p.842,1952. 78. A.R.Troiano, Trans. ASM, V.52, p.54, 1960. 79. C.D.Beachem, Metall.Trans., V.3, p.437-451, 1972. 80. A.Turnbull, A.S.Dolphin and F.A.Rackley, Corrosion, Vol.44, No.l, pp.55-61, Jan 1988. 81. A.Turnbull and D.H.Ferriss, Corr.Science, Vol.26, No.8, pp.601-628, Aug.1986. 82. A.Turnbull, Corr.Science, Vol.23, No.8, pp.833-870, 1983. 83. B.F.Brown, Corrosion Fatigue, Chemistry, Mechanics and Microstructure, NACE, 2,25-29,1972. 84. D.A.Verrnilyea and C.S.Tedmon, J.Electrochem.Soc., Vol.117, No.4, pp.437-439, Apr. 1970. 85. B.G.Ateya & H. W.Pickering, J.Electrochem.Soc., Vol. 122, No.8, pp.1018-1026, Aug. 1975. 86. P.Doig and P.E.J.Flewitt, Proc. Roy.Soc., 357A, 439-452,1977. 87. R.N.Parkins, Metals Sci.J.,13,381-386,1979. 88. R.O.Ritchie, Metal Science, Vol.11, pp.368-381,1977. 89. Paris.P.C, Bucci R.J., Wessel E.T., Clark.W.G. and Magers T.R., Stress Analysis and Growth of Cracks, ASTM STP 513, pp.141-176, 1972. 90. Bucci.R.J., Clark.W.G.Jr. & Paris.P.C, in Stress Analysis and Growth of Cracks, 129 ASTM STP 513, pp.177-195,1972. 91. Cook R.J., Irving.P.E., Booth.G.S.,Beevers.C.J., Engg.Fracture Mechanics, Vol.7, No.l,pp.69-77,1975. 92. Cook.R.J.,Beevers.C.J., Engineering Fracture Mechanics, Vol.5, pp. 1061-1079,1973. 93. Cook.RJ., Beevers.C.J., Materials Science & Engg., Vol 13, pp.201-210, 1974. 94. Mckittrick J., Liaw P.K., Kwun S J. and Fine M.E., Metallurgical Transactions, Vol.l2A, No.8, pp.1535-1539, Aug 1981. 95. Petit.J. & Mailard J.L., Scripta Metallurgica, Vol.14, No.l, pp.163-166, Jan 1980. 96. Jrving.P.E. & Kurzfeld.A., Metal Science, pp.495-502,1978. 97. E.K.Priddle, F.Walker and C.Wiltshire, Proc. Conf. on Influence of Environment on Fatigue,137,1977, London,. Institution of Mechanical Engineers). 98. W.Elber, Damage Tolerance in Aircraft Structures; STP 486, pp.230-242, Philadelphia, pa., ASTM, 1971. 99.S.Suresh & R.O.Ritchie, Fatigue Crack Growth Threshold Concepts, Symp.Proc., Ed., by D.L.Davidson and S.Suresh, AIME, Warrendale, Pa., 1984. 100. J.D.Atkinson & T.C.Lindley, The influence of Environment on Fatigue, Inst.Mech.Engg.Publications,4,65-74,1977. 101. J.P.Gallaghar,J.Materials, JMLSA,6,941-964, 1971. 102. R.V.Ramage, K.V.Jata, GJ.Shiflet and E.A.StarkeJr., Med.Trans., V0I.I8A, No.7, pp. 1291-1298, July 1987. 103. K.S.Ravichandran, H.C.Venkata Rao, E.S.DwarakaDasa, C.G.KrishnaDas Nair, MeU.Trans., Vol. 18A, No.5, pp.865-876, May 1987. 104. K.Fukaura,H.Izumi and H.Kawabe,J.Soc.Matl.Science,35,393, pp.48-54, 1986. 105. M.Ait BassidiJ.Masounave and J J.Dickson, Duplex Stainless Steel, Conf .Proc., Ed:R.A.Lula, pp.445-463, ASM, Metals park, Ohio, 1983. 106. J.Moskovitz and Pelloux R, Corrosion Fatigue Technology, ASTM STP 642, pp. 133-145, ASTM, Philadelphia, 1978. 107. H.D.Solomon and T.M.Devine, "Duplex Stainless Steel", pp.693-756, Ed., R.A.Lula, ASM, Metals Park, OH, 1983. 108. H.D.Solomon, "Duplex Stainless Steel", Ed.R.A.Lula, pp.41-69, ASM, Metals Park, OH., 1983. 109. "Binary Alloy Phase Diagrams", T.B.MassalskiJEd.in Chief, Vol.1, p.1081, ASM, Metals Park, OH, 1986. 1 lO.AJ.Sedriks, Corrosion, Vol.42, pp.376-389,1986. 130 111. J.E.Truman and M.J.Coleman and K.R.Pirt, Br.Corr.J., Vol.12, pp.236-238,1977. 112. P.Guha and C.A.Clark, "Duplex Stainless Steel", Ed.R.A.Lula, pp. 355-369, ASM, Metals Park, OH, 1983. 113. A.Poznansky,C.S.Nalbone and J.D.Crawford, "Duplex Stainless Steel", Ed.R.A.Lula, pp.431-443, ASM, Metals Park, OH., 1983. 114. K.Osozowa and N.Okato, "Passivity and its breakdown on Iron and Iron base alloys", Ed.R.Staehle and H.Okada, pp. 135-139, NACE, Houston, TX., 1976. 115. R.F.A.Jargelius and T.Wallin, "The Effect of Nitrogen Alloying on the Pitting and Crevice Corrosion Resistance of CrNi and CrNiMo Austenitic Stainless Steels" 10th Scandinavian Corrosion Congress, Stockholm, 161-164,1986. 116. M.Pourbaix, "Atlas of Electrochemical Equilibria in Aqueous Solutions", Pergamon Press, London, 1966. 117. R.C.Newman, Corr.Science, Vol.25, pp.331-339,1985. 118. The stress analysis of cracks handbook, Del Research Corp., Hellertown, Pa., 1973. 119. ASTM Standard E399-78:Test for Plane Strain Fracture Toughness of Metallic Materials, Annual Book of ASTM Standards-Part 10. 120.1nstruction Bulletein B-141-3 .Measurements Group, Inc., Raleigh, NC.1983. 121. C.Fong and D.Tromans, Met.Trans.A, Vol.l9A, pp.2765-2773,1988. 122. J.C.Scully, "Atomistics of Fracture", Ed.R.M.Latinision and J.R.Pickens, pp.637-667, Plenum Press, N.Y. 1983. 123. F.P.Fehlner and N.F.Mott, Oxidatiob of Metals and Alloys, p.37-62, ASM, Metals Park, Ohio, 1971. 124. D.Tromans, Unpubished work, Univ. of British Columbia, 1989. 125. K.Sieradzki and R.C.Newman, Phil. Mag,A, Vol.51, No.l, pp.95-132, 1985. 126. G.R.Caskey Jr, Hydrogen Degradation of Ferrous Alloys, Eds. R.A.Oriani, J.P.ffirth, Michael Smialowski, pp.822-862, Noyes Publications, New Jersey, 1985. 127. M.R.Louthan, G.R.Caskey Jr., J.A.Donovan, and D.E.RawlJr., Mat.Sc. and Eng., Vol.10, p.357, 1972. 128.S.V.Nair, R.R.Jensen, and J.K.Tien, Metl.Trans A., Vol. 14, pp.385-393,1983. 129. M.AitBassidi and J.Masounave and J.I.Dickson, ASM Metals Congress, October 1982, St.Louis, ASM Metals Park, Ohio 44073. 130. M.AitBassidi, J.Masounave, J.I.Dickson and J.P.Bailon, Can.Met.Q, Vol.23, No.l, pp. 17-24, 1984. 131. M.AifBassidi, J.Masounave, J.I.Dickson and J.P.Bailon, ICF6, New Delhi. 131 APPENDTY.1 Calibration of Back Face Strain Gauge to Measure Crack Length To measure the crack length and consequentiy the stress intensity at the crack tip, load Vs.back face strain curves were plotted for different crack lengths. A compact tension specimen containing a fatigue pre-crack was prepared and a strain gauge of 3mm length and 120 ohms resistance was attached to the back face. A bridge amplified meter (Model BAM-1) was used to measure the signals from the gauge. In order to find the calibration factor the following formula was used: ^ ... . _ „ 400 . Calibration Setting (\iinch\ Calibration Factor - — * ————— — Gage Factor No.ofWorbngArms { inch ) In the present case the calibration knob was set at 1 and the no. of working arms was 1. The gauge factor of the strain gauge was 2.05 which gives a calibration factor of 195 uin/in strain. Using the GAIN control knob the meter was set such that each division of the scale reads 1.95 uin/in strain. However, in order to read the strain in volts the output of the meter was connected to a analog to digital converter (Data Translation Board DT 2801) and further to a personal computer. The board was set to receive uni-polar voltages (fatigue was tension-tension) In order to calibrate the output voltage in terms of crack length, the compact tension specimen was cracked incrementally and the surface crack lengths were measured using a travelling microscope.For each crack length, the output voltage was recorded for different applied loads. At the end of the test, the specimen was broken open. An inspection of the fractured surfaced revealed that the crack had propagated with a reasonably straight crack front. This showed that the surface crack length measurement was indicative of the true crack length. Table 9 gives compliances (which is the ratio of strain to load) obtained for different crack lengths.(a or a/W where W is the width of the specimen). 132 a crack,(mm) a/W Compliance(C) (volts/KN) 12.73 12.89 13.69 15.43 6.14 17.16 20.49 23.13 25.43 25.88 26.72 0.286 0.289 0.3079 0.3472 0.3632 0.3861 0.461 0.5204 0.5722 0.5822 0.6012 0.0356 0.037129 0.039625 0.042381 0.045668 0.050641 0.076273 0.107993 0.123269 0.136774 0.14954 A non-linear curve fitting routine UBC *OLSF was used to fit a polynomial to the compliance data. a/W = -0.4657694+35.98919*C-543.2037*C2+3739.491*C3-93149.196*C4...(21) To obtain the crack length and stress intensity at the crack tip during a test, the strain in volts was measured and compliance calculated. Using the polynomial the crack length was calculated and hence stress intensity. Measurement of crack closure stresses In order to determine the crack closure stresses near threshold, the fatigue cycling was temporarily stopped. The applied load was decreased step wise and the corresponding strain measured. On complete unloading (zero load), the strain gauge showed a residual voltage corresponding to residual strain. The procedure is shown schematically in Figure 53. From the measured crack length (a) and the compliance polynomial (equation 21), the specimen compliance (C) was determined. Using the measured residual voltage and specimen compliance (C), which is volts/load (and is proportional to strain/load), the load (closure load) was calculated. From the closure load and crack length, the closure stress intensity was computed using the formula shown in Figure 16. The measured closure stress intensities were low consistent with low values expected at a R-ratio of 0.3399. Load Figure 53 Schematic of the procedure used to measure crack closure stress 134 APPENDLX-2 H^ry* r e q u i r e d to form Fft.0. film o n o n e nnif r*\\ t h i c l m * ^ (aj Fe,04 =» 8.3965X10'cm No. of cells/cm2 = 1/area of the face of the cell.= 1/C8.3965X10**)2 No. of atoms/unit cell*56 No. of molecules of Fe,0«/unit cell»56V7«8 No. of molecules of F^Cycm2*^ 1/(8.3965 XltvV)X8 No. of moles of Fe 30 4/cm 2»(( 1/(8.3965 X10*)2)X8)X( 1/6.02X10°) Each molecule contains 8 charges (two Fe*,one Fe2*) *8 charges No. of coulombs required/cm2 to form unit cell thickness of oxide* ((1/(8.3965 X10,)2)X8)X(1/6.02X1023)X8XF, where F is the Faraday * 96500 coulombs No. of coulombs* (1.46 X 10'3)Ocm2 However, for a monolayer of {111} Fe,04 the charge calculation is slightly different. Z X site contains 2Fe O 3 4 Area of triangle=(2 ao.sin60/2)2. a^  = (ao)2. sin60 No. of sites/triangle =3/6 +3/2 = 2 sites No. of molecules of Fe304/cm=(2X2)/(ao)2 sin 60) No. of moles=(2X2)/(ao)2 sin 60)X( 1/6.02X 1023) Each molecule contains 8 charges (two Fe**,one Fe2*) =8 charges No. of coulombs=(2X2)/(a0)2 sin 60)X(1/6.02X1023)X8XF No. of coulombs for Fe304 {111} =8.4 C/m2 135 APPENDIX-3 Accuracy of EDS Microanalysis The accuracy of x-ray determination of an element concentration (C) is determined by the x-ray counting statistics and a precise knowledge of the film thickness (t). Counting statistics alone produce an error of +5% in C*. If the film thickness is uncertain (tiAt) a further error is introduced which causes the overall accuracy to be further decreased. The overall accuracy may be written simply as Concentration = [l±f(At)][C±0.05C] If the element concentration is measured in adjacent areas, then the term [l±f(At)] may be considered to be constant. Therefore concentration of an element in two adjacent regions may be written as [Concentration]! = Constant [Ci±0.05Ci]...(l) [Concentration^ = Constant [C2±0.05C2]...(2) where Cl and C2 are measured concentrations. Therefore, the difference in concentration.AC, between 1 and 2 then becomes AC=Constant x {[CiiO.05Cil-tC2iO.05C2]} If AC is > 5% of either (1) or (2), then AC is significant and meaningful. Accuracy of Crack Measurements in Fatigue Crack Propagation Studies: Resolution of travelling microscope = 0.01 mm. Smallest crack increment measured (eg.+300 mv SCE in 1M NaCl) = 0.0305 mm. Therefore error in measurement in the threshold region is ~30% Largest crack increment measured (eg.+300 mv SCE in 1M NaCl) = 0.345 mm. Therefore error in measurement ~ 3%. Error in crack length measurements decreases from ~30% in the threshold region to ~ 3% in stage II FCP studies. For a fatigue test running at 85 Hz the cycle counter reads with a resolution of 100 cycles. Therefore the error in measurements of 'No. of elapsed cycles' is less than 1%. Consequently the error in the values of 'da/dN' is approximately 30% in the near-threshold region and approximately 3% in the stage II measurements. The resulting error in stress intensity calculations due to errors in crack length measurements are less than 1% in all stages of FCP. *D.B.Williams, Norelco Reporter, Vol.29, No.3, Dec. 1982. 136 APPENDIX-4 LOAD 5 4.6 4.2 3.8 3.6 3.4 3.2 3 2 2 2 .8 .6 .2 2 .8 .6 1.6 CRACKmm 12.93368 13.27912 13.53312 13.69568 13.83792 14.0208 14.18336 14.30528 14.87424 15.01648 15.1892 15.21968 15.27048 15.31112 15.3416 CAST DUPLEX STAINLESS I M N a C I +0.3 Vsce STR.INT.CUM.CYC 11.63255 STEEL 10.91725 10.11609 9.239965 8.826954 8.426809 8.007724 7.561765 7.302315 6.839295 5.848115 5.326345 4.808580 4.284909 4.292896 10500 20500 30500 45500 70500 110500 175500 275500 335500 388000 438000 538000 788000 1188000 da dN 0 3.45E-04 10500 2.54E-04 10000 1.63E-04 10000 1.42E-04 15000 1.83E-04 25000 1.63E-04 40000 1.22E-04 65000 5.69E-04 100000 1.42E-04 60000 2.54E-04 52500 3.05E-05 50000 5.08E-05 100000 4.06E-05 250000 3.05E-05 400000 da/dN ERR 3.29E-08 2.54E-08 1.63E-08 9.48E-09 7.32E-09 4.06E-09 1.88E-09 5.69E-09 2.37E-09 4.84E-09 6.10E-10 5.08E-10 1.63E-10 7.62E-11 137 LOAD CRACKmm 4 20.58416 3.6 21.27504 3.2 21.7932 2.8 22.33168 2.4 22.57552 2 22.73808 1.8 22.8092 1.6 23.28672 1.6 23.4696 1.8 23.53056 1.8 23.68296 1.8 23.79472 1.8 23.9268 1.8 24.1808 1.8 24.3332 1.8 24.39416 1.8 24.55672 2 24.67864 2 24.77008 2 24.90216 2 25.81656 2 26.01976 2 26.2128 2 26.65984 2 27.5336 2 28.55976 2 33.83712 CAST DUPLEX STAINLESS STEEL I M N a C I +0 Vsce STR.INT.CUM.CYC 12.95219 0 12.25897 10000 11.32583 20000 10.32417 30000 9.017397 40000 7.610130 70000 6.887293 130000 6.357466 530000 6.451270 2130000 7.293380 2184000 7.383807 2284000 7.451205 2334000 7.532068 2384000 7.691367 2484000 7.789418 2534000 7.829172 2584000 7.936704 2634000 8.909820 2684000 8.979212 2734000 8.947368 2784000 9.691244 2934000 9.869410 2984000 10.04336 3024000 10.46478 3064000 11.37128 3094000 12.60046 3114000 24.12231 3124000 da ERR 6.91 E-04 5.18E-04 5.38E-04 2.44E-04 1.63E-04 7.11E-05 4.78E-04 1.83E-04 6.10E-05 1.52E-04 1.12E-04 1.32E-04 2.54E-04 1.52E-04 6.10E-05 1.63E-04 1.22E-04 9.14E-05 1.32E-04 9.14E-04 2.03E-04 1.93E-04 4.47E-04 8.74E-04 1.03E-03 5.28E-03 dN 10000 10000 10000 10000 30000 60000 400000 2E+06 54000 100000 50000 50000 100000 50000 50000 50000 50000 50000 50000 150000 50000 40000 40000 30000 20000 10000 da/dN 6.91 E-08 5.18E-08 5.38E-08 2.44E-08 5.42E-09 1.19E-09 1.19E-09 1.14E-10 1.13E-09 1.52E-09 2.24E-09 2.64E-09 2.54E-09 3.05E-09 1.22E-09 3.25E-09 2.44E-09 1.83E-09 2.64E-09 6.10E-09 4.06E-09 4.83E-09 1.12E-08 2.91 E-08 5.13E-08 5.28E-07 138 LOADCRACKIn 5 0 . 7 4 5 2 4.6 0 . 7 5 4 4.3 0 . 7 6 1 2 4 0 . 7 7 4 4 3.6 0 . 7 8 9 6 3.3 0 . 7 9 7 2 3 0 . 8 0 5 6 2.6 0 . 8 1 5 6 2.3 0 . 8 2 9 6 2 0 . 8 5 4 8 1.6 0 . 8 8 3 6 1.4 0 . 8 8 9 6 3.6 0 . 9 3 2 8 3 0 . 9 6 4 2.6 0 . 9 7 2 8 2.4 0 . 9 7 8 2.2 0 . 9 9 0 4 2 0 . 9 9 7 2 1.8 1 . 0 0 2 1.6 1 . 0 1 2 1.4 1 . 0 1 7 2 1.2 1.02 1.6 1 . 0 2 5 6 1.6 1 . 0 3 3 6 1.6 1 . 0 4 1 2 1.6 1 . 0 4 7 6 1.6 1 . 0 5 2 2 1.06 2 1 . 0 9 9 6 2 1 . 1 0 8 4 2 1 . 1 2 0 8 2 1 . 1 2 7 6 2 1 . 1 4 7 2 2 1 . 1 8 4 8 2 1 . 2 0 4 8 2 1 . 2 6 2 4 CAST DUPLEX STAINLESS STEEL I M N a C I -0 .2 Vsce STR.INT. CUM.CYC da 1 3 . 9 2 2 6 1 1 3 . 0 0 5 8 4 1 2 . 3 1 1 4 6 1 1 . 7 2 1 9 4 1 0 . 8 3 9 6 0 1 0 . 0 7 3 2 3 9 . 2 9 8 0 7 7 8 . 2 0 7 0 5 7 7 . 5 4 7 6 3 6 6 . 7 9 3 4 4 8 5 . 7 4 5 7 3 2 5 . 0 8 7 1 5 0 1 4 . 2 7 1 1 9 1 2 . 6 9 6 4 9 1 1 . 2 1 3 1 0 1 0 . 4 6 7 4 6 9 . 8 5 8 3 1 0 9 . 0 9 7 5 6 1 8 . 2 7 5 5 6 7 7 . 5 2 2 7 5 1 6 . 6 6 0 2 6 2 5 . 7 4 5 2 3 2 7 . 7 5 8 9 0 9 7 . 9 0 3 1 8 2 8 . 0 4 4 0 9 6 8 . 1 6 5 7 8 0 8 . 1 2 7 9 3 9 1 0 . 3 5 5 2 9 1 1 . 4 1 3 5 4 1 1 . 6 7 1 4 5 1 2 . 0 5 0 5 1 1 2 . 2 6 6 5 5 1 2 . 9 2 4 1 7 1 4 . 3 5 0 2 3 1 5 . 2 1 1 7 3 1 8 . 2 0 5 4 5 0 7 3 0 0 1 9 2 0 0 2 9 2 0 0 4 2 9 0 0 5 7 9 0 0 8 2 9 0 0 1 4 2 9 0 0 2 7 2 9 0 0 4 7 2 9 0 0 6 7 2 9 0 0 8 7 2 9 0 0 5 3 4 3 8 0 0 5 3 5 5 9 0 0 5 3 6 1 1 0 0 5 3 8 4 9 0 0 5 4 1 2 5 0 0 5 4 1 7 0 0 0 5 5 0 7 0 0 0 5 6 5 1 8 0 0 5 8 5 4 8 0 0 6 0 7 7 1 0 0 6 3 3 9 8 0 0 6 4 1 2 0 0 0 6 5 1 2 0 0 0 6 6 1 2 0 0 0 6 7 1 2 0 0 0 6 9 1 2 0 0 0 7 1 1 2 0 0 0 7 2 1 2 0 0 0 7 2 4 1 3 0 0 7 2 7 1 3 0 0 7 3 2 3 8 0 0 7 3 6 2 8 0 0 7 3 6 9 0 0 0 7 3 7 4 0 0 0 1 8 E - 0 6 5 1 E - 0 6 3 8 E - 0 5 1 . 5 9 E - 0 5 7 . 9 3 E - 0 6 8 . 7 6 E - 0 6 1 . 0 4 E - 0 5 1 . 4 6 E - 0 5 2 . 6 3 E - 0 5 3 . 0 0 E - 0 5 6 . 2 6 E - 0 6 4.51 E - 0 5 3 . 2 6 E - 0 5 9 . 1 8 E - 0 6 5 . 4 3 E - 0 6 1 . 2 9 E - 0 5 7 . 0 9 E - 0 6 5.01 E - 0 6 1 . 0 4 E - 0 5 5 . 4 3 E - 0 6 2 . 9 2 E - 0 6 5 . 8 4 E - 0 6 8 . 3 5 E - 0 6 7 . 9 3 E - 0 6 6 . 6 8 E - 0 6 4 . 5 9 E - 0 6 8 . 3 5 E - 0 6 4 . 1 3 E - 0 5 9 . 1 8 E - 0 6 1 . 2 9 E - 0 5 7 . 0 9 E - 0 6 2 . 0 4 E - 0 5 3 . 9 2 E - 0 5 2 . 0 9 E - 0 5 6.01 E - 0 5 dN 7 3 0 0 1 1 9 0 0 1 0 0 0 0 1 3 7 0 0 1 5 0 0 0 2 5 0 0 0 6 0 0 0 0 1 3 0 0 0 0 2 0 0 0 0 0 2 0 0 0 0 0 2 0 0 0 0 0 9 0 0 1 2 1 0 0 5 2 0 0 2 3 8 0 0 2 7 6 0 0 4 5 0 0 9 0 0 0 0 1 4 4 8 0 0 2 0 3 0 0 0 2 2 2 3 0 0 1 2 5 9 0 0 7 2 2 0 0 1 0 0 0 0 0 1 0 0 0 0 0 1 0 0 0 0 0 2 0 0 0 0 0 2 0 0 0 0 0 1 0 0 0 0 0 2 9 3 0 0 3 0 0 0 0 5 2 5 0 0 3 9 0 0 0 6 2 0 0 5 0 0 0 da/dN 1 . 2 6 E - 0 9 6.31 E - 1 0 3 8 E - 0 9 . 1 6 E - 0 9 . 2 9 E - 1 0 5 1 E - 1 0 . 7 4 E - 1 0 . 1 2 E - 1 0 3 1 E - 1 0 . 5 0 E - 1 0 3 . 1 3 E - 1 1 5.01 E - 0 8 2 . 6 9 E - 0 9 1 . 7 7 E - 0 9 2 . 2 8 E - 1 0 4 . 6 9 E - 1 0 1 . 5 8 E - 0 9 5 . 5 6 E - 1 1 7 . 2 1 E - 1 1 2 . 6 7 E - 1 1 1 . 3 1 E - 1 1 4 . 6 4 E - 1 1 1 . 1 6 E - 1 0 7 . 9 3 E - 1 1 6 . 6 8 E - 1 1 4 . 5 9 E - 1 1 4 . 1 7 E - 1 1 2 . 0 7 E - 1 0 9 . 1 8 E - 1 1 4 . 4 2 E - 1 0 2 . 3 6 E - 1 0 3 . 9 0 E - 1 0 1 . 0 1 E - 0 9 3 . 3 7 E - 0 9 1 . 2 0 E - 0 8 139 CAST DUPLEX STAINLESS STEEL I M N a C I -0.4 Vsce LOADCRACKmm STR.INT. CUM.CYC da dN da/dN 4r 16.8808 10.39721 0 ERR 3.6 17.0332 9.449215 35800 1.52E-04 35800 4.26E-09 3.2 17.32784 8.560285 105800 2.95E-04 70000 4.21 E-09 3 17.40912 8.067646 165800 8.13E-05 60000 1.35E-09 2.8 17.57168 7.609835 265800 1.63E-04 100000 1.63E-09 2.6 17.7444 7.146513 465800 1.73E-04 200000 8.64E-10 2.4 17.80536 6.623217 823800 6.10E-05 358000 1.70E-10 2.2 17.846 6.087517 1718800 4.06E-05 895000 4.54E-11 2 18.5572 5.802074 2408800 7.11E-04 690000 1.03E-09 1.8 18.98392 5.374924 2808800 4.27E-04 400000 1.07E-09 1.2 23.25112 4.904579 6597600 4.27E-03 3788800 1.13E-09 1.6 23.3832 6.608809 6682600 1.32E-04 85000 1.55E-09 2 23.51528 8.349124 7345600 1.32E-04 663000 1.99E-10 2 23.7388 8.501517 7495600 2.24E-04 150000 1.49E-09 2 24.06392 8.730828 7595600 3.25E-04 100000 3.25E-09 2 24.22648 8.849025 7645600 1.63E-04 50000 3.25E-09 2 24.3992 8.977297 7695600 1.73E-04 50000 3.45E-09 2 24.63288 9.155396 7745600 2.34E-04 50000 4.67E-09 2 24.77512 9.266454 7795600 1.42E-04 50000 2.84E-09 2 24.958 9.412289 7845600 1.83E-04 50000 3.66E-09 2 25.0596 9.494827 7895600 1.02E-04 50000 2.03E-09 2 25.24248 9.646209 7945600 1.83E-04 50000 3.66E-09 2 25.65904 9.715114 7995600 4.17E-04 50000 8.33E-09 2 25.8724 9.901365 8025600 2.13E-04 30000 7.11 E-09 2 26.16704 10.16771 8045600 2.95E-04 20000 1.47E-08 2 27.40656 11.41853 8230600 8.84E-04 35000 2.53E-08 2 27.96536 12.06170 8255600 5.59E-04 25000 2.24E-08 2 28.54448 12.78947 8280600 5.79E-04 25000 2.32E-08 2 29.0728 13.51475 8305600 5.28E-04 25000 2.11 E-08 140 CAST DUPLEX STAINLESS STEEL I M N a C I -0.6 Vsce LOADCRACKmm STR.INT. CUM.CYC 4 14.2136 4 14.2836 3.6 14.7136 3.4 14.7936 3.4 15.0636 2.8 17.8668 2.6 17.9582 2.2 18.141 2 18.2019 2 18.3746 2 18.66924 2 18.99436 2 19.06548 2 19.21788 2 19.34996 2 19.4414 2 19.53284 2 19.63444 2 19.7462 2 19.82748 2 19.99004 2 20.07152 2 20.26436 2 20.33548 2 20.48788 2 20.9146 2 21.09748 2 21.29052 2 21.4226 2 21.60548 2 21.74772 2 21.97124 2 22.11348 2 22.2354 2 22.57068 2 22.87548 2 28.05186 2 28.36682 2 28.4989 2 29.04754 2 29.98226 2 31.28274 2 33.39602 10.27339 10.31616 9.526199 9.040370 9.189114 8.903636 8.317600 7.123614 6.502261 6.577555 6.709012 6.858620 6.892004 6.964356 7.027974 7.072524 7.117493 7.167958 7.224086 7.265318 7.348836 7.391237 7.493048 7.531124 7.613693 7.852207 7.957857 8.071701 8.151005 8.262753 8.351265 8.493260 8.585518 8.540210 8.763625 8.974467 13.96223 14.41308 14.60891 15.46833 17.12599 19.94062 26.37995 0 21600 43600 66000 96000 126000 146000 196000 296000 696000 1096000 1246000 1321000 1396000 1471000 1521000 1571000 1621000 1671000 1721000 1771000 1821000 1871000 1921000 1971000 2021000 2071000 2171000 2221000 2271000 2321000 2421000 2471000 2521000 2571000 2621000 2927000 2947000 2967000 3.00E+06 3.03E+06 3.05E+06 3.06E+06 da ERR 7.00E-05 4.30E-04 8.00E-05 2.70E-04 2.44E-04 9.14E-05 1.83E-04 6.09E-05 1.73E-04 2.95E-04 3.25E-04 7.11 E-05 1.52E-04 1.32E-04 9.14E-05 9.14E-05 1.02E-04 1.12E-04 8.13E-05 1.63E-04 8.15E-05 1.93E-04 7.11 E-05 .52E-04 .27E-04 .83E-04 .93E-04 .32E-04 .83E-04 .42E-04 2.24E-04 1.42E-04 1.22E-04 3.35E-04 3.05E-04 5.18E-03 3.15E-04 1.32E-04 5.49E-04 9.35E-04 1.30E-03 2.11E-03 1. 4. 1  1  1  1  1 dN 21600 22000 22400 30000 30000 20000 50000 100000 400000 400000 150000 75000 75000 75000 50000 50000 50000 50000 50000 50000 50000 50000 50000 50000 50000 50000 100000 50000 50000 50000 100000 50000 50000 50000 50000 306000 20000 20000 30000 30000 20000 10000 da/dN 3.24E-09 1.95E-08 3.57E-09 9.00E-09 8.13E-09 4.57E-09 3.66E-09 6.09E-10 4.32E-10 7.37E-10 2.17E-09 9.48E-10 2.03E-09 1.76E-09 1.83E-09 1.83E-09 2.03E-09 2.24E-09 1.63E-09 3.25E-09 1.63E-09 3.86E-09 1.42E-09 3.05E-09 8.53E-09 3.66E-09 1.93E-09 2.64E-09 3.66E-09 2.84E-09 2.24E-09 2.84E-09 2.44E-09 6.71 E-09 6.10E-09 1.69E-08 1.57E-08 6.60E-09 1.83E-08 3.12E-08 6.50E-08 2.11E-07 141 LOADCRACKmm 5 13.573 4.8 14.782 4.6 15.9504 4.4 16.48888 4.2 16.99688 4 17.322 3.8 17.53536 3.6 17.78936 3.4 18.10432 3.2 18.2364 3 18.36848 2.8 18.76472 2.6 18.96792 2.4 19.21176 2.2 19.36416 2 19.45556 1.8 19.5064 2 19.9128 2 20.02456 2 20.17696 2 20.24808 2 20.32936 2 20.41064 2 20.48176 2 20.54272 2 20.61384 2 20.7256 2 21.386 2 21.87368 2 22.03624 2 22.10736 2 22.59504 2 22.88968 2 23.418 2 23.63136 2 24.00728 2 24.4848 2 24.7388 2 24.91152 2 25.07488 2 25.18584 2 25.28744 2 25.46016 2 25.62272 2 25.77512 2 26.0088 2 26.37456 2 26.68952 2 27.29912 2 27.8884 2 29.52416 CRACKin 0.534370 0.581968 0.627968 0.649168 0.669168 0.681968 0.690368 0.700368 0.712768 0.717968 0.723168 0.738768 0.746768 0.756368 0.762368 0.765966 0.767968 0.783968 0.788368 0.794368 0.797168 0.800368 0.803568 0.806368 0.808768 0.811568 0.815968 0.841968 0.861168 0.867568 0.870368 0.889568 0.901168 0.921968 0.930368 0.945168 0.963968 0.973968 0.980768 0.9872 0.991568 0.995568 1.002368 1.008768 1.014768 1.023968 1.038368 1.050768 1.074768 1.097968 1.162368 CAST DUPLEX STAINLESS I M N a C I -0.8 Vsce STR.INTCUM.CYC. 11.64593 12.01046 12.36281 12.23103 12.05847 11.72722 11.29624 10.72546 10.34223 9.819638 9.287365 8.902325 8.381308 752716 181625 570136 933993 783140 836763 910907 945918 6.986255 7.026942 7.062834 7.093816 7.130219 7.187989 7.544088 7.824294 7.921176 7.964130 8.268349 8.460718 8.823010 8.976010 9.255550 9.630141 9.838873 9.984777 10.12583 10.22339 10.31398 10.47083 10.62181 10.76640 10.99405 11.36549 11.70104 12.39542 13.12908 15.33155 0 25000 40000 48000 58000 68000 78000 88000 98000 108000 128000 158000 180000 210000 251000 326000 476000 909500 1009500 1409500 1509500 1609500 1709500 1809500 1909500 2009500 2109500 2994500 3194500 3344500 3394500 3444500 3474500 3506500 3516500 3527100 3532100 3537100 3542100 3547100 3552100 3557100 3562100 3567100 3572100 3577100 3582100 3587100 3595100 3597600 3600900 STEEL da ERR 1.21E-03 1.17E-03 5.38E-04 5.08E-04 3.25E-04 2.13E-04 2.54E-04 3.15E-04 1.32E-04 1.32E-04 3.96E-04 2.03E-04 2.44E-04 1.52E-04 9.14E-05 5.08E-05 4.06E-04 1.12E-04 1.52E-04 7.11E-05 8.13E-05 8.13E-05 7.11E-05 6.10E-05 7.11E-05 1.12E-04 6.60E-04 4.88E-04 1.63E-04 7.11E-05 4.88E-04 2.95E-04 5.28E-04 2.13E-04 3.76E-04 4.78E-04 2.54E-04 1.73E-04 1.63E-04 1.11E-04 1.02E-04 1.73E-04 1.63E-04 1.52E-04 2.34E-04 3.66E-04 3.15E-04 6.10E-04 5.89E-04 1.64E-03 dN 25000 15000 8000 10000 10000 10000 10000 10000 10000 20000 30000 22000 30000 41000 75000 150000 433500 100000 400000 100000 100000 100000 100000 100000 100000 100000 885000 200000 150000 50000 50000 30000 32000 10000 10600 5000 5000 5000 5000 5000 5000 5000 5000 5000 5000 5000 5000 8000 2500 3300 da/dN 4.84E-08 7.79E-08 6.73E-08 5.08E-08 3.25E-08 2.13E-08 2.54E-08 3.15E-08 1.32E-08 6.60E-09 1.32E-08 9.24E-09 8.13E-09 3.72E-09 1.22E-09 3.39E-10 9.37E-10 1.12E-09 3.81E-10 7.11E-10 8.13E-10 8.13E-10 7.11E-10 6.10E-10 7.11E-.12E-.46E-.44E-09 .08E-09 .42E-09 .75E-09 9.82E-09 1.65E-08 2.13E-08 3.55E-08 9.55E-08 5.08E-08 3.45E-08 3.27E-08 2.22E-08 2.03E-08 3.45E-08 3.25E-08 3.05E-08 4.67E-08 7.32E-08 6.30E-08 7.62E-08 2.36E-07 4.96E-07 1 7. 2. 1. 1 9. -10 -09 -10 142 LOADCRACKmm 5 13.0454 4.8 13.34004 4.6 13.79724 4.6 13.6044 4.4 13.87872 4.4 14.08192 4 14.40704 3.8 14.48832 3.6 14.69152 3.4 14.8236 3.2 14.99632 3 15.27064 2.8 15.39256 2.6 15.57544 2.4 15.8904 2.2 16.16477 3 16.57098 2.8 16.6929 2.6 16.77418 2.4 16.946 2.2 17.11962 2 17.23138 2 17.4041 1.8 17.51586 1.8 17.66826 1.8 17.87146 1.8 18.0137 1.8 18.08482 1.8 18.27786 1.8 18.33882 1.8 18.81634 1.8 19.03986 1.8 19.34466 1.8 19.51738 1.8 19.75106 1.8 19.87298 1.8 20.1473 1.8 20.3505 1.8 20.6045 1.8 20.79754 1.8 20.79754 2 21.03122 2 21.28522 2 22.18946 2 22.60602 CRACKin 0.513598 0.525198 0.543198 0.535606 0.546406 0.554406 0.567206 0.570406 0.578406 0.583606 0.590406 0.601206 0.606006 0.613206 0.625606 0.636408 0.652400 0.657200 0.660400 0.667165 0.674000 0.678400 0.685200 0.689600 0.695600 0.703600 0.709200 0.712000 0.719600 0.722000 0.740800 0.749600 0.761600 0.768400 0.777600 0.782400 0.793200 0.801200 0.811200 0.818800 0.818800 0.828000 0.838000 0.873600 0.890000 CAST DUPLEX STAINLESS I M N a C I -1.2 Vsce STR.INTCUM.CYC. STEEL 11.29523 11.02961 10.85574 10.73391 10.43356 10.55929 9.786407 9.208827 8.830983 8.406915 7.995383 7.621692 7.166775 6.632432 6.242562 5.821012 8.143222 7.659315 7.149074 6.671910 6.184416 5.662822 5.726459 5.191376 5.243249 5.313605 5.363683 5.388983 5.458546 5.480789 5.659740 5.746482 5.867973 5.938514 6.035973 6.087766 6.206745 6.297123 6.412880 6.502992 6.502992 7.349587 7.488061 8.014152 8.275399 0 5000 10000 35000 58000 70000 139000 159000 199000 239000 319000 479000 679000 879000 1179000 3229000 116000 163000 187000 217000 311200 511200 594200 749200 1049200 1449200 1649200 1849200 2249200 3449200 4049200 4249200 4349200 4399200 4479200 4519200 4679200 4759200 4839200 5159200 6199200 6299200 6349200 6374200 6384200 da ERR 2.95E-04 4.57E-04 ********* 2.74E-04 2.03E-04 3.25E-04 8.13E-05 2.03E-04 1.32E-04 1.73E-04 2.74E-04 1.22E-04 1.83E-04 3.15E-04 2.74E-04 2.95E-04 1.22E-04 8.13E-05 1.72E-04 1.74E-04 1.12E-04 1.73E-04 1.12E-04 1.52E-04 2.03E-04 1.42E-04 7.11 E-05 1.93E-04 6.10E-05 4.78E-04 2.24E-04 3.05E-04 1.73E-04 2.34E-04 1.22E-04 2.74E-04 2.03E-04 2.54E-04 1.93E-04 0.00E+00 2.34E-04 2.54E-04 9.04E-04 4.17E-04 dN 5000 5000 25000 23000 12000 69000 20000 40000 40000 80000 160000 200000 200000 300000 2050000 116000 47000 24000 30000 94200 200000 83000 155000 300000 400000 200000 200000 400000 1200000 600000 200000 100000 50000 80000 40000 160000 80000 80000 320000 1040000 100000 50000 25000 10000 da/dN 5.89E-08 9.14E-08 ********* 1.19E-08 1.69E-08 4.71 E-09 4.06E-09 5.08E-09 3.30E-09 2.16E-09 1.71 E-09 6.10E-10 9.14E-10 1.05E-09 1.34E-10 2.54E-09 2.59E-09 3.39E-09 5.73E-09 .84E-09 .59E-10 .08E-09 .21E-10 .08E-10 5.08E-10 7.11E-10 .56E-10 .83E-10 .08E-11 .96E-10 .12E-09 .05E-09 3.45E-09 2.92E-09 3.05E-09 1.71 E-09 2.54E-09 3.18E-09 6.03E-10 0.00E+00 2.34E-09 5.08E-09 3.62E-08 4.17E-08 1 5  2  7  5  3  4  5  7  1 3  143 LOADCRACKmm 5 16.09 5 16.1301 4.8 16.201 4.6 16.3638 4.4 16.3841 4.2 16.5162 4 16.6686 3.8 16.81088 3.6 16.99376 3.4 17.2884 3.2 17.45096 2.8 17.4916 2.4 17.68464 2 17.68464 2 17.7557 2 17.81672 2 17.88784 2 17.9488 2 18.52792 2 18.62952 2 18.80224 2 18.87336 2 18.92416 2 18.99528 2 19.22896 2 19.38136 2 19.65568 2 19.84872 2 20.05192 2 20.16368 2 20.3567 2 20.55992 2 20.6412 2 20.7936 2 20.9053 2 21.28128 2 21.3727 2 21.51496 2 21.51496 2 21.76896 2 21.8604 2 21.95184 2 22.09408 2 22.3277 2 22.5716 2 22.75448 2 22.90681 2 23.15072 2 23.29296 2 23.44536 2 23.56728 CRACKIn 0.633464 0.635043 0.637834 0.644244 0.645043 0.650244 0.656244 0.661845 0.669045 0.680645 0.687045 0.688645 0.696245 0.696245 0.699043 0.701445 0.704245 0.706645 0.729445 0.733445 0.740245 0.743045 0.745045 0.747845 0.757045 0.763045 0.773845 0.781445 0.789445 0.793845 0.801444 0.809445 0.812645 0.818645 0.823043 0.837845 0.841444 0.847045 0.847045 0.857045 0.860645 0.864245 0.869845 0.879043 0.888645 0.895845 0.901842 0.911445 0.917045 0.923045 0.927845 CAST DUPLEX STAINLESS SYN.WHITE WATER +0.3 Vsce STR.INTCUM.CYC. 13.00133 13.03393 12.56819 12.16821 11.65403 11.21736 10.78680 10.34061 9.911528 9.540086 9.073989 7.960742 6.909960 5.758300 5.785163 5.808377 5.835607 5.859096 6.089361 6.131130 6.203116 6.233120 6.165331 6.195271 6.295176 6.361623 6.483875 6.572006 6.666713 6.719670 6.812618 6.912563 6.953150 7.030216 7.087514 7.285597 7.335013 7.412915 7.412915 7.555127 7.607327 7.660071 7.743216 7.882753 8.032510 8.147632 8.245436 8.405773 8.501462 8.605828 8.564772 0 15000 55000 145000 175000 235000 295000 355000 415000 535000 595000 655000 899000 1619000 1859000 2219000 2579000 2939000 3839000 4039000 4239000 4339000 4439000 4539000 4739000 5139000 5339000 5439000 5639000 5739000 5839000 5939000 6039000 6139000 6239000 6339000 6439000 6539000 6639000 6739000 6914000 7014000 7114000 7214000 7314000 7364000 7404000 7444000 7484000 7524000 7564000 STEEL da 1.61 E-02 4.01 E-05 7.09E-05 1.63E-04 2.03E-05 1.32E-04 1.52E-04 1.42E-04 1.83E-04 2.95E-04 1.63E-04 4.06E-05 1.93E-04 0.00E+00 7.11 E-05 6.10E-05 7.11 E-05 6.10E-05 5.79E-04 1.02E-04 1.73E-04 7.11 E-05 5.08E-05 7.11 E-05 2.34E-04 1.52E-04 2.74E-04 1.93E-04 2.03E-04 1.12E-04 1.93E-04 2.03E-04 8.13E-05 1.52E-04 1.12E-04 3.76E-04 9.14E-05 1.42E-04 0.00E+00 2.54E-04 9.14E-05 9.14E-05 1.42E-04 2.34E-04 2.44E-04 1.83E-04 1.52E-04 2.44E-04 1.42E-04 1.52E-04 1.22E-04 dN 0 15000 40000 90000 30000 60000 60000 60000 60000 120000 60000 60000 244000 720000 240000 360000 360000 360000 900000 200000 200000 100000 100000 100000 200000 400000 200000 100000 200000 100000 100000 100000 100000 100000 100000 100000 100000 100000 100000 100000 175000 100000 100000 100000 100000 50000 40000 40000 40000 40000 40000 da/dN ERR 2.67E-09 1.77E-09 1.81 E-09 6.77E-10 2.20E-09 2.54E-09 2.37E-09 3.05E-09 2.46E-09 2.71 E-09 6.77E-10 7.91 E-10 0.00E+00 2.96E-10 1.69E-10 1.98E-10 1.69E-10 6.43E-10 5.08E-10 8.64E-10 7.11E-10 5.08E-10 7.11E-10 1.17E-09 3.81 E-10 1.37E-09 1.93E-09 1.02E-09 1.12E-09 1.93E-09 2.03E-09 8.13E-10 1.52E-09 1.12E-09 3.76E-09 9.14E-10 1.42E-09 0.00E+00 2.54E-09 5.23E-10 9.14E-10 1.42E-09 2.34E-09 2.44E-09 3.66E-09 3.81 E-09 6.10E-09 3.56E-09 3.81 E-09 3.05E-09 144 2 23.6892 2 23.8416 2 24.05496 2 24.31912 2 24.63408 2 25.22336 2 25.53832 2 25.80248 2 26.00568 2 26.3308 2 27.00136 2 27.20456 2 27.37728 2 27.48904 2 27.58048 2 27.68208 2 27.77352 2 27.83448 2 27.99704 2 28.12912 2 28.28152 0.932645 8.649695 0.938645 8.757658 0.947045 8.781214 0.957445 8.975563 0.969845 9.215968 0.993045 9.692940 1.005445 9.963511 1.015845 10.19946 1.023845 10.38684 1.036645 10.69784 1.063045 11.38629 1.071045 11.60844 1.077845 11.80257 1.082245 11.93088 1.085845 12.03746 1.089845 12.15763 1.093445 12.26738 1.095845 12.34139 1.102245 12.54219 1.107445 12.70909 1.113445 12.90599 7604000 1.22E-04 7644000 1.52E-04 7684000 2.13E-04 7724000 2.64E-04 7754000 3.15E-04 7784000 5.89E-04 7804000 3.15E-04 7824000 2.64E-04 7844000 2.03E-04 7864000 3.25E-04 7884000 6.71 E-04 7894000 2.03E-04 7904000 1.73E-04 7909000 1.12E-04 7914000 9.14E-05 7919000 1.02E-04 7924000 9.14E-05 7929000 6.10E-05 7934000 1.63E-04 7939000 1.32E-04 7944000 1.52E-04 40000 3.05E-09 40000 3.81 E-09 40000 5.33E-09 40000 6.60E-09 30000 1.05E-08 30000 1.96E-08 20000 1.57E-08 20000 1.32E-08 20000 1.02E-08 20000 1.63E-08 20000 3.35E-08 10000 2.03E-08 10000 1.73E-08 5000 2.24E-08 5000 1.83E-08 5000 2.03E-08 5000 1.83E-08 5000 1.22E-08 5000 3.25E-08 5000 2.64E-08 5000 3.05E-08 145 LOADCRACKmm 5 15.03658 5 15.33892 4.8 15.40129 4.6 15.52436 4.4 15.64505 4.2 15.87962 4 15.87962 3.8 16.08311 3.6 16.46935 3.4 16.71217 3.2 16.29573 3 16.67878 2.8 17.03835 2.6 16.82933 2.4 17.20049 2.2 17.46018 2 18.08171 2 18.61249 2 19.27088 2 19.46046 2 19.63775 2 19.88382 2 20.07368 2 20.11026 2 20.25248 2 20.25248 2 20.48860 2 20.48860 2 20.68186 2 20.77646 2 21.02568 2 21.11899 2 21.33872 2 21.62973 2 21.69625 2 21.93543 2 22.18589 2 22.72413 2 23.01204 2 23.05415 2 23.26804 2 23.66579 2 23.75600 2 24.12163 2 24.85825 2 25.71806 2 26.08309 2 26.11607 CAST CRACKin 0.591991 0.603894 0.606350 0.611195 0.615946 0.625182 0.625182 0.633193 0.648399 0.657959 0.641564 0.656645 0.670801 0.662572 0.677184 0.687408 0.711878 0.732775 0.758696. 0.766160 0.773139 0.782827 0.790302 0.791742 0.797341 0.797341 0.806638 0.806638 0.814246 0.817971 0.827782 0.831456 0.840107 0.851564 0.854183 0.863599 0.873460 0.894650 0.905986 0.907643 0.916064 0.931724 0.935275 0.949670 0.978671 1.012522 1.026893 1.028191 DUPLEX STAINLESS STEEL SYN.WHITE WATER 0 Vsce STR.INTCUM.CYC. 12.18567 12.41200 11.96107 11.54954 11.12973 10.77907 10.26578 9.735674 9.449542 9.062848 8.308301 7.979690 7.620596 6.982291 6.600413 6.152935 5.826357 6.036616 6.313352 6.396524 6.475780 6.588226 6.676979 6.694281 6.762196 6.762196 6.877242 6.877242 6.973589 7.021490 7.150050 7.199097 7.316589 7.476672 7.514002 7.539690 7.684602 8.010555 8.073055 8.099947 8.238653 8.506330 8.568876 8.829658 9.393106 10.12313 10.45962 10.49086 0 10000 20000 40000 60000 90000 120000 150000 180000 220000 260000 340000 500000 700000 900000 1100000 1500000 1900000 2200000 2400000 2500000 2600000 2650000 2700000 2750000 2800000 2850000 2900000 2950000 3000000 3050000 3100000 3150000 3190000 3230000 3270000 3310000 3350000 3380000 3400000 3420000 3440000 3460000 3480000 3500000 3520000 3540000 3560000 da 0 3.02E-04 6.24E-05 1.23E-04 1.21E-04 2 ;35Ej04 2.03E-04 3.86E-04 2.43E-04 ********* 3.83E-04 3.60E-04 dN 10000 10000 20000 20000 30000 30000 30000 30000 40000 40000 80000 160000 ********* 200000* 3.71 E-04 2.60E-04 6.22E-04 5.31 E-04 6.58E-04 1.90E-04 1.77E-04 2.46E-04 1.90E-04 3.66E-05 1.42E-04 0.00E+00 2.36E-04 0.00E+00 1.93E-04 9.46E-05 2.49E-04 9.33E-05 2.20E-04 2.91 E-04 6.65E-05 2.39E-04 2.50E-04 5.38E-04 2.88E-04 4.21 E-05 2.14E-04 3.98E-04 9.02E-05 3.66E-04 7.37E-04 8.60E-04 3.65E-04 3.30E-05 200000 200000 400000 400000 300000 200000 100000 100000 50000 50000 50000 50000 50000 50000 50000 50000 50000 50000 50000 40000 40000 40000 40000 40000 30000 20000 20000 20000 20000 20000 20000 20000 20000 20000 da/dN 3.02E-08 6.24E-09 6.15E-09 6.03E-09 7.82E-09 ********* 6.78E-09 1.29E-08 6.07E-09 ********* 4.79E-09 2.25E-09 1.86E-09 1.30E-09 1.55E-09 1.33E-09 2.19E-09 9.48E-10 1.77E-09 2.46E-09 3.80E-09 7.31E-10 2.84E-09 0.00E+00 4.72E-09 0.00E+00 3.87E-09 1.89E-09 4.98E-09 1.87E-09 4.39E-09 7.28E-09 1.66E-09 5.98E-09 6.26E-09 1.35E-08 9.60E-09 2.11 E-09 1.07E-08 1.99E-08 4.51 E-09 1.83E-08 3.68E-08 4.30E-08 1.83E-08 1.65E-09 146 LOADCRACKmm 5 4.8 4.6 4.4 4.2 4 3.8 3.6 3.4 3.2 3 2.6 1.8 .8 .8 .8 .8 .8 .8 1.8 1.8 1.8 1.8 1.8 1.8 1.8 1.8 .8 .8 .8 .8 .8 1.8 1.8 1.8 18.06274 18.20988 18.64285 18.93337 19.17706 19.40159 19.59102 20.07943 20.85760 21.36436 22.26347 23.67684 25.95430 25.3348 25.41608 25.51768 25.56848 26.12728 26.32032 26.51336 26.67592 26.9096 26.99088 27.0112 27.17376 27.3668 27.6208 28.11864 28.30152 28.4336 28.56568 29.88648 30.37416 30.53672 30.70944 31.0752 31.634 32.18264 32.68048 33.158 33.7168 34.5296 35.22048 35.55576 35.92152 36.24664 CRACKin 0.711131 0.716924 0.733970 0.745408 0.755002 0.763842 0.771300 0.790528 0.821165 0.841116 0.876514 0.932159 1.021822 0.997433 1.000633 1.004633 1.006633 1.028633 1.036233 1.043833 1.050233 1.059433 1.062633 1.063433 CAST DUPLEX STAINLESS SYN.WHITE WATER -0.2 Vsco STR.INTCUM.CYC. .069833 .077433 .087433 .107033 .114233 1.119433 1.124633 1.176633 1.195833 1.202233 1.209033 1.223433 1.245433 1.267033 1.286633 1.305433 1.327433 1.359433 1.386633 1.399833 1.414233 1.427033 16.33970 15.83999 15.62652 15.24418 14.79559 14.31059 13.77480 13.50459 13.48617 13.17363 13.21546 12.80270 10.76332 10.18909 10.26172 10.35363 10.40006 10.77637 10.96731 11.16351 11.33295 11.58355 11.67272 11.69518 11.87727 12.09925 12.40123 13.02801 13.27065 13.45028 13.63372 15.70449 16.59554 16.91025 17.25501 18.02258 19.30355 20.70674 21.80429 23.29304 25.25064 28.18433 31.58326 33.47603 35.75726 38.00264 0 7500 18100 28100 43100 63100 88100 136100 208100 245100 378100 382500 802500 825500 835500 845500 855500 945500 965500 985500 1005500 1025500 1045500 1065500 1085500 1105500 1125500 1185500 1205500 1225500 1245500 1265500 1275500 1280500 1285500 1295500 1305500 1315500 1323000 1330500 1335500 1340500 1343000 1344500 1345500 1346500 STEEL da 1.81 E-02 1.47E-04 4.33E-04 2.91 E-04 2.44E-04 2.25E-04 1.89E-04 4.88E-04 7.78E-04 5.07E-04 8.99E-04 1.41 E-03 2.28E-03 2.53E-02 8.13E-05 1.02E-04 5.08E-05 7.11 E-05 1.93E-04 1.93E-04 1.63E-04 2.34E-04 8.13E-05 2.03E-05 63E-04 93E-04 54E-04 83E-04 83E-04 32E-04 32E-04 5.08E-04 4.88E-04 1.63E-04 1.73E-04 66E-04 59E-04 5.49E-04 4.98E-04 4.78E-04 5.59E-04 8.13E-04 6.91 E-04 35E-04 66E-04 25E-04 1, 1, 2. 1. 1. 1. 1. 3. 5. dN 0 7500 10600 10000 15000 20000 25000 48000 72000 37000 133000 4400 420000 825500 10000 10000 10000 20000 20000 20000 20000 20000 20000 20000 20000 20000 20000 20000 20000 20000 20000 20000 10000 5000 5000 10000 10000 10000 7500 7500 5000 5000 2500 1500 1000 1000 da/dN ERR .96E-08 08E-08 91 E-08 62E-08 .12E-08 58E-09 02E-08 .08E-08 1.37E-08 6.76E-09 3.21 E-07 5.42E-09 8.13E-09 1.02E-08 5.08E-09 3.56E-09 9.65E-09 9.65E-09 8.13E-09 1.17E-08 4.06E-09 1.02E-09 8.13E-09 9.65E-09 1.27E-08 9.14E-09 9.14E-09 60E-09 60E-09 54E-08 88E-08 25E-08 45E-08 3.66E-08 5.59E-08 5.49E-08 6.64E-08 6.37E-08 1.12E-07 1.63E-07 2.76E-07 2.24E-07 3.66E-07 3.25E-07 147 LOADCRACKmm CAST DUPLEX STAINLESS SYN.WHITE WATER -0.4 Vsce STR.INTCUM.CYC. STEEL 5 5 4.6 4.4 4.2 4 3.8 3.6 3.4 3.2 3 2.8 2.6 2.2 2.2 2.2 2.2 2.2 2.2 2.2 2.2 2.2 2.2 2.2 2.2 2.2 2.2 2.2 2.2 2.2 2.2 2.2 2.2 2.2 2.2 2.2 2.2 2.2 2.2 2.2 2.2 2.2 2.2 2.2 2.2 2.2 2.2 2.2 2.2 2.2 2.2 17.15571 17.20049 17.41799 17.58438 17.78327 17.76771 17.81419 17.90548 18.03845 17.86012 17.97255 18.08172 17.54339 17.78327 18.08171 18.35748 18.55057 18.79122 18.84853 18.84853 19.11970 19.27088 19.41427 19.55053 19.72223 19.92282 19.99922 20.18212 20.32141 20.65006 20.74503 20.96356 21.11899 21.27550 21.37047 .49886 .66289 .66289 .76352 .90062 22.07710 22.11312 22.29696 22.44866 22.60452 22.64414 22.72413 23.05415 23.13903 23.22485 23.39894 21, 21, 21 21 21 CRACKin 0.675421 0.677184 0.685747 0.692298 0.700128 0.699516 0.701346 0.704940 0.710175 0.703154 0.707580 0.711879 0.690684 0.700128 0.711878 0.722735 0.730337 0.739811 0.742068 0.742068 0.752744 0.758696 0.764341 0.769706 0.776465 0.784363 0.787370 0.794572 0.800055 0.812994 0.816733 0.825337 0.831456 0.837618 0.841357 0.846411 0.852869 0.852869 0.856831 0.862229 0.869177 0.870595 0.877833 0.883805 0.889941 0.891501 0.894650 0.907643 0.910985 0.914364 0.921218 13.90996 13.95014 13.01597 12.58556 12.17083 11.57944 11.03408 10.36607 9.876487 9.186818 8.676681 8.156908 7.311123 6.284124 6.408992 6.527650 6.612668 6.720931 6.747099 6.747099 6.872988 6.944687 7.013721 7.080269 7.165466 7.266941 7.306153 7.401320 7.475030 7.653360 7.706072 7.829448 7.919007 7.894643 7.950269 8.026408 8.005795 8.005795 8.066455 8.150219 8.259976 8.282655 8.399855 8.498460 8.601596 8.628116 8.682028 8.909942 8.970023 9.031388 9.157850 0 10000 30000 70000 110000 150000 190000 230000 310000 470000 570000 670000 770000 2350000 2550000 2750000 2935400 3026400 3076400 3126400 3226400 3326400 3426400 3526400 3626400 3726400 3826400 3926400 4026400 4126400 4176400 4216400 4256400 4286400 4316400 4336400 4356400 4376400 4396400 4416400 4436400 4446400 4456400 4466400 4476400 4486400 4496400 4506400 4516400 4526400 4536400 da 0 4.48E-05 2.17E-04 1.66E-04 1.99 E-04 ********* 4.65E-05 9.13E-05 1.33E-04 ********* 1.12E-04 1.09 E-04 7.84E-05 2.98E-04 2.76E-04 1.93E-04 2.41 E-04 5.73E-05 0.00E+00 2.71 E-04 1.51 E-04 1.43E-04 1.36E-04 1.72E-04 2.01 E-04 7.64E-05 1.83E-04 1.39E-04 3.29E-04 9.50E-05 2.19E-04 1.55E-04 1.57E-04 9.50E-05 1.28E-04 1.64E-04 0.00E+00 1.01 E-04 1.37E-04 1.76E-04 3.60E-05 1.84E-04 1.52E-04 1.56E-04 3.96E-05 8.00E-05 3.30E-04 8.49E-05 8.58E-05 1.74E-04 dN 10000 20000 40000 40000 40000 40000 40000 80000 160000 100000 100000 100000 130000 200000 200000 185400 91000 50000 50000 100000 100000 100000 100000 100000 100000 100000 100000 100000 100000 50000 40000 40000 30000 30000 20000 20000 20000 20000 20000 20000 10000 10000 10000 10000 10000 10000 10000 10000 10000 10000 da/dN 4.48E-09 1.09E-08 4.16E-09 4.97E-09 ********* 1.16E-09 2.28E-09 1.66E-09 ********* 1.12E-09 1.09E-09 ********* 6.03E-10 1.49E-09 1.38E-09 1.04E-09 2.64E-09 1.15E-09 0.00E+00 2.71 E-09 1.51 E-09 1.43E-09 1.36E-09 1.72E-09 2.01 E-09 7.64E-10 1.83E-09 1.39E-09 3.29E-09 1.90E-09 5.46E-09 3.89E-09 5.22E-09 3.17E-09 6.42E-09 8.20E-09 0.00E+00 5.03E-09 6.86E-09 8.82E-09 3.60E-09 1.84E-08 1.52E-08 1.56E-08 3.96E-09 8.00E-09 3.30E-08 8.49E-09 8.58E-09 1.74E-08 148 2.2 2.2 2.2 2.2 2.2 2.2 2.2 2.2 2.2 2.2 2.2 23. 23. 24. 24, 24 24 25 25 25 26 26 48720 75600 12163 21382 35238 76702 03892 .59797 .94434 40146 .62995 0.924692 0.935275 0.949670 0.953300 0.958755 0.975079 0.985784 1.007794 1.021431 1.039427 1.048423 9.222990 9.425764 9.712624 9.787047 9.900553 10.25243 10.49364 11.01766 11.36270 11.84406 12.09651 4546400 4556400 4566400 4576400 4586400 4596400 4606400 4616400 4626400 4636400 4646400 8.83E-05 2.69E-04 3.66E-04 9.22E-05 1.39E-04 4.15E-04 2.72E-04 5.59E-04 3.46E-04 4.57E-04 2.28E-04 10000 10000 10000 10000 10000 10000 10000 10000 10000 10000 10000 8.83E-09 2.69E-08 3.66E-08 9.22E-09 1.39E-08 4.15E-08 2.72E-08 5.59E-08 3.46E-08 4.57E-08 2.28E-08 149 CAST DUPLEX STAINLESS STEEL SYN.WHITE WATER -0.6 Vsce Kin STR.INTCUM.CYC. 5 11.92784 0.4696 5 12.23264 0.4816 4.8 12.26312 0.4828 4.6 12.40536 0.4884 4.4 12.52728 0.4932 4.2 12.56792 0.4948 4 12.66952 0.4988 3.8 12.83208 0.5052 3.6 13.07592 0.5148 3.4 13.11656 0.5164 3.2 13.208 0.52 3 13.32992 0.5248 2.8 13.40104 0.5276 2.6 13.42136 0.5284 2.8 13.45184 0.5296 3 13.50264 0.5316 3 13.68552 0.5388 3 13.89888 0.5472 3 14.15288 0.5572 3 14.21384 0.5596 3 14.29512 0.5628 3 14.33576 0.5644 3 14.36624 0.5656 3 14.43736 0.5684 3 14.59992 0.5748 3 14.69136 0.5784 3 14.74216 0.5804 3 14.89456 0.5864 3 15.00632 0.5908 3 15.1384 0.596 3 15.24 0.6 3 15.32128 0.6032 3 15.42288 0.6072 3 15.45336 0.6084 3 15.57528 0.6132 3 15.67688 0.6172 3 15.81912 0.6228 3 15.98168 0.6292 3 16.73352 0.6588 3 16.89608 0.6652 3 17.39392 0.6848 3 17.99336 0.7084 3 18.09496 0.7124 3 18.16608 0.7152 3 18.288 0.72 3 18.36928 0.7232 3 18.48104 0.7276 3 18.61312 0.7328 3 18.7452 0.738 3 18.98904 0.7476 3 19.1008 0.752 10.16741 10.34322 9.946596 9.609184 9.255224 8.854976 8.482269 8.133377 7.813992 7.39719 6.998917 6.607962 6.192939 5.757379 6.211252 6.674616 6.746247 6.83122 6.934397 6.959493 6.993158 7.010078 7.022808 6.951887 7.019778 7.058388 7.079971 7.145294 7.193752 7.251636 7.296622 7.332903 7.378622 7.392419 7.44798 7.494745 7.560936 7.637628 8.007488 8.090887 8.354351 8.68861 8.747208 8.788572 8.860156 8.908357 8.975268 0 20000 30000 112000 172000 232000 432000 732000 1132000 1302000 1602000 2002000 2402000 3002000 3202000 3602000 4202000 4602000 4772000 4872000 4972000 5072000 5472000 5632000 5932000 6232000 6732000 7032000 7282000 7532000 7732000 8132000 8332000 8667000 8817000 9017000 9217000 9417000 9617000 9717000 9817000 9892000 9922000 9952000 9982000 10012000 10042000 9.055303 10072000 9.136397 10102000 9.288954 10132000 9.360137 10162000 da 0.000305 0.000030 0.000142 0.000122 0.000040 0.000102 0.000163 0.000244 0.000040 0.000091 0.000122 0.000071 0.000020 0.000030 0.000050 0.000183 .000213 .000254 .000061 .000081 .000040 .000030 .000071 .000163 .000091 .000050 .000152 .000112 .000132 0.000102 0.000081 0.000102 0.000030 0.000122 0.000102 0.000142 0.000163 0.000752 0.000163 0.000498 0.000599 0.000102 0.000071 0.000122 0.000081 0.000112 0.000132 0.000132 0.000244 0.000112 0.000102 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. dN 20000 10000 82000 60000 60000 200000 300000 400000 170000 300000 400000 400000 600000 200000 400000 600000 400000 170000 100000 100000 100000 400000 160000 300000 300000 500000 300000 250000 250000 200000 400000 200000 335000 150000 200000 200000 200000 200000 100000 100000 75000 30000 30000 30000 30000 30000 30000 30000 30000 30000 30000 da/dN 1.52E-08 3.05E-09 1.73E-09 2.03E-09 6.77E-10 5.08E-10 5.42E-10 6.10E-10 2.39E-10 3.05E-10 3.05E-10 1.78E-10 3.39E-11 1.52E-10 1.27E-10 3.05E-10 1.49E-09 6.10E-10 8.13E-10 4.06E-10 7.62E-11 4.45E-10 5.42E-10 3.05E-10 1.02E-10 5.08E-10 4.47E-10 5.28E-10 5.08E-10 2.03E-10 5.08E-10 9.10E-11 8.13E-10 5.08E-10 7.11E-10 8.13E-10 3.76E-09 1.63E-09 4.98E-09 99E-09 39E-09 37E-09 06E-09 71 E-09 73E-09 4.40E-09 4.40E-09 8.13E-09 3.73E-09 3.39E-09 150 3 3 19.2024 19.26336 LOADCRACKmm 0.756 9.425553 10192000 0.000061 0.7584 9.465127 10222000 0.000071 CAST DUPLEX STAINLESS STEEL SYN.WHITE WATER -0.8 Vsce STR.INTCUM.CYC. 30000 30000 11. 10 9.2 8.4 7.6 7 6.4 5.8 5.4 5 4.6 4.2 4 3.8 3.6 3.4 3.2 3 2.8 2.4 2.4 4 .8 .6 .4 .2 3 2.6 2.4 2.2 2 2.2 2.2 2.2 2.2 2.2 2.2 2.2 2.2 2.2 2.2 2.2 2.2 2.2 2.2 2.2 2.2 2.2 2.2 3. 3  3  3. 13.0691 15.28398 15.54814 15.64974 15.89358 16.05614 16.27966 16.48286 16.64542 16.7267 16.90958 17.04166 17.1331 17.29566 17.47854 17.70206 17.78334 18.05766 18.48438 18.98222 19.27686 19.38862 19.58166 19.6731 19.7747 20.00838 20.17094 20.27254 20.36398 20.41478 20.42494 20.49606 20.6891 20.75006 20.82118 20.95326 21.10566 21.1971 24.82422 24.97662 25.04774 25.09854 25.2103 25.33222 25.44398 25.54558 25.59638 25.7183 25.8199 CRACKIn 0.514531 0.601731 0.612131 0.616131 0.625731 0.632131 0.640931 0.648931 0.655331 0.658531 0.665731 0.670931 0.674531 0.680931 0.688131 0.696931 0.700131 0.710931 0.727731 0.747331 0.758931 0.763331 0.770931 0.774531 0.778531 0.787731 0.794131 0.798131 0.801731 0.803731 0.804131 0.806931 0.814531 0.816931 0.819731 0.824931 0.830931 0.834531 0.977331 0.983331 0.986131 0.988131 0.992531 0.997331 1.001731 1.005731 1.007731 1.012531 1.016531 23.56881 23.57478 22.04257 20.25210 18.60180 17.30739 16.04671 14.72951 13.85546 12.89551 12.00299 11.05234 10.58807 10.16468 9.744562 9.338378 8.836008 8.434877 8.099600 7.180206 7.326592 12.30545 11.84764 11.29579 10.74425 10.27966 9.748661 8.510071 7.906844 7.274311 6.617820 7.316803 7.419219 7.452005 7.490530 7.562871 7.647646 7.699195 10.25646 10.39100 10.45471 10.50059 10.60260 10.71562 10.82085 10.91790 10.96692 11.08596 11.18667 0 13000 16000 17800 19900 22400 26400 36400 46400 56400 76400 96400 136400 176400 226400 276400 326400 608400 819700 970700 1048200 1092200 1242200 1342200 1442200 1542200 1642200 1842200 2342200 2692200 3292200 3512700 3712700 3812700 4012700 4112700 4212700 4312700 8600100 8800100 8970100 9070100 9170100 9260100 9310100 9410100 9460100 9560100 9660100 da 2.21 E-03 2.64E-04 1.02E-04 2.44E-04 1.63E-04 2.24E-04 2.03E-04 1.63E-04 8.13E-05 1.83E-04 1.32E-04 9.14E-05 1.63E-04 1.83E-04 2.24E-04 8.13E-05 2.74E-04 4.27E-04 4.98E-04 2.95E-04 1.12E-04 1.93E-04 9.14E-05 1.02E-04 2.34E-04 1.63E-04 1.02E-04 9.14E-05 5.08E-05 1.02E-05 7.11 E-05 1.93E-04 6.10E-05 7.11 E-05 1.32E-04 1.52E-04 9.14E-05 3.63E-03 1.52E-04 7.11 E-05 5.08E-05 1.12E-04 1.22E-04 1.12E-04 1.02E-04 5.08E-05 1.22E-04 1.02E-04 3.05E-04 dN 13000 3000 1800 2100 2500 4000 10000 10000 10000 20000 20000 40000 40000 50000 50000 50000 282000 211300 151000 77500 44000 150000 100000 100000 100000 100000 200000 500000 350000 600000 220500 200000 100000 200000 100000 100000 100000 4287400 200000 170000 100000 100000 90000 50000 100000 50000 100000 100000 209000 03E-09 37E-09 da/dN 1.70E-07 8.81 E-08 5.64E-08 1.16E-07 6.50E-08 5.59E-08 2.03E-08 1.63E-08 8.13E-09 9.14E-09 6.60E-09 2.29E-09 4.06E-09 3.66E-09 4.47E-09 1.63E-09 9.73E-10 2.02E-09 3.30E-09 3.80E-09 2.54E-09 1.29E-09 9.14E-10 1.02E-09 2.34E-09 1.63E-09 5.08E-10 1.83E-10 1.45E-10 69E-11 23E-10 65E-10 10E-10 56E-10 32E-09 52E-09 14E-10 8.46E-10 7.62E-10 4.18E-10 5.08E-10 1.12E-09 1.35E-09 2.24E-09 1.02E-09 1.02E-09 1.22E-09 1.02E-09 1.46E-09 151 2.2 2.2 2.2 2.2 2.2 2.2 2.2 2.2 2.2 2.2 LOAD 12 11.6 11.2 10.8 10.4 10 9.6 9 8 26.1247 1.028531 11.49739 9869100 2.34E-04 70000 3 34E-09 26.35838 1.037731 11.74468 9939100 3.05E-04 100000 3 05E-09 26.66318 1.049731 12.07972 10039100 2.24E-04 50000 4 47E-09 26.8867 1.058531 12.33487 10089100 1.93E-04 40000 4.83E-09 27.07974 1.066131 12.56197 10129100 1.32E-04 50000 2.64E-09 27.21182 1.071331 12.72109 10179100 1.63E-04 50000 3.25E-09 27.37438 1.077731 12.92124 10229100 1.83E-04 40000 4.57E-09 27.55726 1.084931 13.15229 10269100 1.93E-04 40000 4.83E-09 27.7503 1.092531 13.40318 10309100 2.54E-04 40000 6.35E-09 28.0043 1.102531 13.74475 10349100' -1E+07 2.71E-09 CAST DUPLEX STAINLESS STEEL SYN.WHITE WATER -1.2 Vsce CRACKmmCRACKInSTR.INTCUM.CYC. da dN 2 4 8 7.6 7 6.6 6 5.6 5.2 5 4.8 4.6 4.4 4.2 4 3.8 3.6 3.4 3.2 3 2.8 2.6 2.4 2.2 2.2 2.2 2.2 2.6 2.6 2.6 2.6 2.6 2.6 2.6 10.24128 10.70864 10.86104 11.16584 11.2776 11.5824 11.76528 11.97864 12.08024 12.16152 12.41552 12.60856 12.99464 13.17752 14.33576 14.45768 14.56944 14.70152 14.7828 14.90472 15.0876 15.22984 15.30096 15.41272 15.46352 15.65656 15.76832 15.82928 16.13408 16.256 16.34744 16.43888 16.51 16.60144 16.92656 17.13992 17.2212 17.34312 17.6276 17.74952 17.91208 da/dN 0.4216 22.569 0 0 0.4216 21.817 72004.67E-04 7200 6.49E-08 0.4276 21.240 102001.52E-04 3000 5.08E-08 0.4396 20.827 132003.05E-04 3000 1.02E-07 0.444 20.180 162001.12E-04 3000 3.73E-08 0.456 19.734 212003.05E-04 5000 6.10E-08 0.4632 19.138 243001.83E-04 3100 5.90E-08 0.4716 18.561 27300 2.13E-04 3000 7.11 E-08 0.4756 17.044 303001.02E-04 3000 3.39E-08 0.4788 16.307 353008.13E-05 5000 1.63E-08 0.4888 15.716 42300 2.54E-04 7000 3.63E-08 0.4964 14.635 523001.93E-04 10000 1.93E-08 0.5116 14.107 72300 3.86E-04 20000 1.93E-08 0.5188 12.960 1023001.83E-04 30000 6.10E-09 0.5644 12.946 1423001.16E-03 40000 2.90E-08 0.5692 12.109 1473001.22E-04 5000 2.44E-08 0.5736 11.721 1543001.12E-04 7000 1.60E-08 0.5788 11.342 1693001.32E-04 15000 8.81 E-09 0.582 10.922 179300 8.13E-05 10000 8.13E-09 0.5868 10.524 1933001.22E-04 14000 8.71 E-09 0.594 10.158 2083001.83E-04 15000 1.22E-08 0.5996 9.7587 2243001.42E-04 16000 8.89E-09 0.6024 9.3111 244300 7.11 E-05 20000 3.56E-09 0.6068 8.8815 2743001.12E-04 30000 3.73E-09 0.6088 8.4143 3143005.08E-05 40000 1.27E-09 0.6164 8.0139 4243001.93E-04 110000 1.75E-09 0.6208 7.5650 5593001.12E-04 135000 8.28E-10 0.6232 7.0874 7193006.10E-05 160000 3.81 E-10 0.6352 6.7072 802300 3.05E-04 83000 3.67E-09 0.64 6.2387 9043001.22E-04 102000 1.20E-09 0.6436 5.7517 1239300 9.14E-05 335000 2.73E-10 0.6472 5.7849 1599300 9.14E-05 360000 2.54E-10 0.65 5.8109 2034300 7.11 E-05 435000 1.63E-10 0.6536 5.8446 2444300 9.14E-05 410000 2.23E-10 0.6664 7.0517 2814300 3.25E-04 370000 8.79E-10 0.6748 7.1489 3829300 2.13E-04 1015000 2.10E-10 0.678 7.1864 4080300 8.13E-05 251000 3.24E-10 0.6828 7.2433 42398001.22E-04 159500 7.64E-10 0.694 7.3785 4389800 2.84E-04 150000 1.90E-09 0.6988 7.4376 45548001.22E-04 165000 7.39E-10 0.7052 7.5175 46268001.63E-04 72000 2.26E-09 152 2.6 1 8 . 0 2 3 8 4 0 . 7 0 9 6 7 . 5 7 3 2 4 6 8 6 8 0 0 1 . 1 2 E - 0 4 6 0 0 0 0 1 8 6 E - 0 9 2.6 1 8 . 1 6 6 0 8 0 . 7 1 5 2 7 .6449 4 7 3 6 8 0 0 1 . 4 2 E - 0 4 5 0 0 0 0 2 8 4 E - 0 9 2.6 1 8 . 2 5 7 5 2 0 . 7 1 8 8 7 . 6 9 1 5 4 7 8 6 8 0 0 9 . 1 4 E - 0 5 5 0 0 0 0 1 .83E-09 2.6 1 8 . 3 2 8 6 4 0 . 7 2 1 6 7.7281 4 8 3 6 8 0 0 7.11 E - 0 5 5 0 0 0 0 1 .42E-09 2.6 1 8 . 5 2 1 6 8 0 . 7 2 9 2 7 .8286 4 8 8 6 8 0 0 1 . 9 3 E - 0 4 5 0 0 0 0 3 . 8 6 E - 0 9 2.6 1 8 . 7 4 5 2 0 . 7 3 8 7 .9475 4 9 8 6 8 0 0 2 . 2 4 E - 0 4 1 0 0 0 0 0 2 . 2 4 E - 0 9 2.6 1 8 . 9 5 8 5 6 0 .7464 8 .0634 5 1 4 6 8 0 0 2 . 1 3 E - 0 4 1 6 0 0 0 0 1 .33E-09 2.6 1 9 . 2 3 2 8 8 0 . 7 5 7 2 8 .2162 5 2 4 6 8 0 0 2 . 7 4 E - 0 4 1 0 0 0 0 0 2 . 7 4 E - 0 9 2.6 1 9 . 3 5 4 8 0 .762 8 . 2 8 5 5 5 3 4 6 8 0 0 1 . 2 2 E - 0 4 1 0 0 0 0 0 1 .22E-09 2.6 1 9 . 5 9 8 6 4 0 . 7 7 1 6 8 .4266 5 4 2 6 8 0 0 2 . 4 4 E - 0 4 8 0 0 0 0 3 . 0 5 E - 0 9 2.6 1 9 . 7 6 1 2 0 .778 8 .5228 5 5 2 6 8 0 0 1 . 6 3 E - 0 4 1 0 0 0 0 0 1 .63E-09 2.6 1 9 . 9 7 4 5 6 0 .7864 8 .6514 5 5 8 6 8 0 0 2 . 1 3 E - 0 4 6 0 0 0 0 3 . 5 6 E - 0 9 2.6 2 0 . 0 6 6 0 .79 8 .7074 5 6 4 6 8 0 0 9 . 1 4 E - 0 5 6 0 0 0 0 1 .52E-09 2.6 2 0 . 3 2 0.8 8 .8659 5 6 9 7 8 0 0 2 . 5 4 E - 0 4 5 1 0 0 0 4 . 9 8 E - 0 9 2.8 2 0 . 5 4 3 5 2 0 .8088 9 . 7 0 1 9 5 7 9 7 8 0 0 2 . 2 4 E - 0 4 1 0 0 0 0 0 2 . 2 4 E - 0 9 2 .8 2 0 . 9 6 0 0 8 0 . 8 2 5 2 9 .9990 5 8 5 8 8 0 0 4 . 1 7 E - 0 4 6 1 0 0 0 6 . 8 3 E - 0 9 2.8 2 1 . 1 9 3 7 6 0 .8344 10.171 5 9 1 8 8 0 0 2 . 3 4 E - 0 4 6 0 0 0 0 3 . 8 9 E - 0 9 2.8 2 1 . 2 8 5 2 0 . 8 3 8 10 .240 6 0 3 8 8 0 0 9 . 1 4 E - 0 5 1 2 0 0 0 0 7 . 6 2 E - 1 0 2.8 2 1 . 5 3 9 2 0 .848 1 0 . 4 3 5 6 0 9 8 8 0 0 2 . 5 4 E - 0 4 6 0 0 0 0 4 . 2 3 E - 0 9 2.8 2 1 . 6 5 0 9 6 0 . 8 5 2 4 10 .522 6 1 5 8 8 0 0 1 . 1 2 E - 0 4 6 0 0 0 0 1 .86E-09 2.8 2 1 . 8 8 4 6 4 0 . 8 6 1 6 10 .709 6 2 1 8 8 0 0 2 . 3 4 E - 0 4 6 0 0 0 0 3 . 8 9 E - 0 9 2.8 2 2 . 2 8 0 8 8 0 . 8 7 7 2 11 .036 6 2 7 8 8 0 0 3 . 9 6 E - 0 4 6 0 0 0 0 6 . 6 0 E - 0 9 2.8 2 2 . 6 0 6 0 .89 1 1 . 3 1 7 6 3 3 8 8 0 0 3 . 2 5 E - 0 4 6 0 0 0 0 5 . 4 2 E - 0 9 2.8 2 3 . 2 4 6 0 8 0 . 9 1 5 2 11.901 6 3 8 8 8 0 0 6 . 4 0 E - 0 4 5 0 0 0 0 1 .28E-08 2.8 2 3 . 4 1 8 8 0 .922 12 .066 6 4 2 3 8 0 0 1 . 7 3 E - 0 4 3 5 0 0 0 4 . 9 3 E - 0 9 2.8 2 3 . 5 4 0 7 2 0 .9268 12 .185 6 4 4 3 8 0 0 1 . 2 2 E - 0 4 2 0 0 0 0 6 . 1 0 E - 0 9 2.8 2 3 . 7 1 3 4 4 0 . 9 3 3 6 12 .357 6 4 5 3 8 0 0 1 . 7 3 E - 0 4 1 0 0 0 0 1 .73E-08 2.8 2 3 . 9 1 6 6 4 0 . 9 4 1 6 12 .563 6 4 6 3 8 0 0 2 . 0 3 E - 0 4 1 0 0 0 0 2 . 0 3 E - 0 8 2.8 2 4 . 1 3 0 .95 12 .786 6 4 7 3 8 0 0 2 . 1 3 E - 0 4 1 0 0 0 0 2 . 1 3 E - 0 8 2.8 2 4 . 6 3 8 0 . 9 7 13.341 6 4 8 3 8 0 0 5 . 0 8 E - 0 4 1 0 0 0 0 5 . 0 8 E - 0 8 2.8 2 5 . 0 2 4 0 8 0 . 9 8 5 2 13 .787 6 4 9 3 8 0 0 3 . 8 6 E - 0 4 1 0 0 0 0 3 . 8 6 E - 0 8 2.8 2 5 . 3 7 9 6 8 0 . 9 9 9 2 1 4 . 2 1 9 6 4 9 8 8 0 0 3 . 5 6 E - 0 4 5 0 0 0 7 . 1 1 E - 0 8 2.8 2 5 . 6 3 3 6 8 1.0092 14 .540 6 5 0 3 8 0 0 2 . 5 4 E - 0 4 5 0 0 0 5 . 0 8 E - 0 8 2.8 2 6 . 1 5 1 8 4 1.0296 15 .230 6 5 0 8 8 0 0 5 . 1 8 E - 0 4 5 0 0 0 1 .04E-07 2.8 2 6 . 3 2 4 5 6 1.0364 15 .472 6 5 1 3 8 0 0 1 . 7 3 E - 0 4 5 0 0 0 3 . 4 5 E - 0 8 WROUGHT DUPLEX STAINLESS STEEL 1M NaCl -1.200 Vsce LOADCRACKmmCRACKin STR.INTCUM.CYC. da dN da/dN 1 2 1 0 . 5 1 4 8 0 . 4 1 3 9 6 2 0 . 3 4 1 7 9 0 0 11 1 0 . 9 8 2 1 6 0 . 4 3 2 3 6 1 9 . 1 2 8 0 9 5 0 0 0 4 . 6 7 E - 0 4 5 0 0 0 9 . 3 5 E - 0 8 10. 1 1 . 1 9 5 5 2 0 . 4 4 0 7 6 1 7 . 9 4 6 1 8 1 5 0 0 0 2 . 1 3 E - 0 4 1 0 0 0 0 2 . 1 3 E - 0 8 9 1 2 . 2 0 1 3 6 0 . 4 8 0 3 6 1 6 . 7 4 8 0 7 3 5 0 0 0 1 .01E-03 2 0 0 0 0 5 . 0 3 E - 0 8 7.8 1 2 . 6 8 9 0 4 0 . 4 9 9 5 6 1 4 . 9 2 2 2 2 5 0 0 0 0 4 . 8 8 E - 0 4 1 5 0 0 0 3 . 2 5 E - 0 8 7.2 1 2 . 8 4 1 4 4 0 . 5 0 5 5 6 13 .89491 5 5 0 0 0 1 .52E-04 5 0 0 0 3 . 0 5 E - 0 8 6.6 1 3 . 0 1 4 1 6 0 . 5 1 2 3 6 1 2 . 8 6 3 9 2 6 5 0 0 0 1 .73E-04 1 0 0 0 0 1 .73E-08 6 1 3 . 1 4 6 2 4 0 . 5 1 7 5 6 1 1 . 7 8 3 8 2 7 5 0 0 0 1 .32E-04 1 0 0 0 0 1 .32E-08 5.6 1 3 . 2 4 7 8 4 0 . 5 2 1 5 6 1 1 . 0 6 2 9 9 8 5 0 0 0 1 .02E-04 1 0 0 0 0 1 .02E-08 5.2 1 3 . 4 6 1 2 0 . 5 2 9 9 6 1 0 . 4 0 0 7 3 1 0 5 0 0 0 2 . 1 3 E - 0 4 2 0 0 0 0 1 .07E-08 4 .8 1 3 . 7 1 5 2 0 . 5 3 9 9 6 9 . 7 4 4 0 5 9 1 2 0 0 0 0 2 . 5 4 E - 0 4 1 5 0 0 0 1 .69E-08 4.4 1 3 . 8 3 7 1 2 0 . 5 4 4 7 6 8 . 9 9 6 1 4 3 150000 1 .22E-04 3 0 0 0 0 4 . 0 6 E - 0 9 4 1 3 . 9 1 8 4 0 . 5 4 7 9 6 8 .217488 2 0 0 0 0 0 8 . 1 3 E - 0 5 5 0 0 0 0 1 .63E-09 3.8 1 4 . 1 5 2 0 8 0 . 5 5 7 1 6 7 .915051 2 5 0 0 0 0 2 . 3 4 E - 0 4 5 0 0 0 0 4 . 6 7 E - 0 9 3.6 1 4 . 5 7 8 8 0 . 5 7 3 9 6 7 . 6 9 1 4 3 6 3 5 0 0 0 0 4 . 2 7 E - 0 4 1 0 0 0 0 0 4 . 2 7 E - 0 9 3.4 1 4 . 7 3 1 2 0 . 5 7 9 9 6 7 . 3 3 0 8 6 3 4 0 0 0 0 0 1 .52E-04 5 0 0 0 0 3 . 0 5 E - 0 9 153 3.2 3 2.8 2.6 2.4 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 14.85312 15.036 15.2392 15.37128 15.65576 16.22552 16.45918 16.52014 16.7335 16.84526 17.01798 17.20086 17.37358 17.61742 17.87142 18.14574 18.43022 18.69438 19.06014 19.53766 19.77134 20.09646 20.30982 20.52318 20.77718 21.01086 21.31566 21.62046 22.03702 22.2707 22.45358 22.63646 0.58476 0.59196 0.59996 0.60516 0.61636 0.6388 0.64799 0.65039 0.65879 0.66319 0.66999 0.67719 0.68399 0.69359 0.70359 0.71439 0.72559 0.73599 0.75039 0.76919 0.77839 0.79119 0.79959 0.80799 0.81799 0.82719 0.83919 0.85119 0.86759 0.87679 0.88399 0.89119 6.950480 6.588532 6.225721 5.827841 5.474360 4.726414 4.796357 4.814853 4.880414 4.915277 4.969876 5.028660 5.085118 5.166418 5.253151 5.349249 5.451665 5.549376 5.688995 5.879236 5.975789 6.114090 6.207454 6.302958 6.419539 6.529661 6.677597 6.830610 7.048386 7.175153 7.276772 7.380575 450000 530000 630000 730000 930000 930000 1130000 1230000 1430000 1630000 1830000 2030000 2230000 2430000 2630000 2830000 3030000 3230000 3430000 3630000 3730000 3830000 3880000 3930000 3980000 4030000 4080000 4130000 4180000 4205000 4230000 4255000 5. 2. 6. 2. 1. 1. 1.22E-04 1.83E-04 2.03E-04 1.32E-04 2.84E-04 70E-04 34E-04 10E-05 13E-04 12E-04 .73E-04 1.83E-04 1.73E-04 2.44E-04 2.54E-04 2.74E-04 2.84E-04 2.64E-04 3.66E-04 4.78E-04 2.34E-04 3.25E-04 2.13E-04 2.13E-04 2.54E-04 2.34E-04 3.05E-04 3.05E-04 4.17E-04 2.34E-04 1.83E-04 1.83E-04 50000 80000 100000 100000 200000 0 200000 100000 200000 200000 200000 200000 200000 200000 200000 200000 200000 200000 200000 200000 100000 100000 50000 50000 50000 50000 50000 50000 50000 25000 25000 25000 2.44E-09 2.29E-09 03E-09 32E-09 42E-09 ERR 17E-09 6.10E-10 1.07E-09 5.59E-10 8.64E-10 9.14E-10 8.64E-10 1.22E-09 1.27E-09 1.37E-09 1.42E-09 1.32E-09 1.83E-09 2.39E-09 2.34E-09 3.25E-09 4.27E-09 4.27E-09 5.08E-09 4.67E-09 6.10E-09 6.10E-09 8.33E-09 9.35E-09 7.32E-09 7.32E-09 154 LOADCRACKmm WROUGHT DUPLEX STAINLESS STEEL 1M NaCL -0.400 Vsce STR.INTCUM.CYC. da 1 2 9 . 3 1 6 7 1 0 9 . 7 3 3 2 6 9.2 9 . 8 5 5 1 8 8.4 1 0 . 0 0 7 5 8 7.8 1 0 . 3 3 2 7 7.2 1 0 . 4 3 4 3 6.6 1 0 . 6 7 8 1 4 6 1 0 . 7 5 9 4 2 5.6 1 0 . 8 5 0 8 6 5.2 1 0 . 9 8 2 9 4 4.8 1 1 . 0 3 3 7 4 4.4 1 1 . 1 4 5 5 4 1 1 . 4 7 0 6 2 4 1 1 . 5 2 1 4 2 3.8 1 1 . 6 1 2 8 6 3.6 1 1 . 6 9 4 1 4 3.4 1 1 . 7 3 4 7 8 3.2 1 1 . 7 4 4 9 4 4 1 1 . 8 0 5 9 4 1 2 . 0 4 9 7 4 4 1 2 . 3 7 4 8 6 4 1 2 . 5 8 8 2 2 4 1 2 . 9 3 3 6 6 4 1 3 . 2 4 8 6 2 4 1 3 . 3 7 0 5 4 4 1 3 . 8 9 8 8 6 4 1 5 . 2 2 9 8 2 4 1 5 . 6 7 6 8 6 4 1 6 . 1 3 4 0 6 4 1 6 . 4 1 8 5 4 4 1 6 . 8 5 5 4 2 4 1 7 . 0 5 8 6 2 4 1 7 . 1 9 0 7 4 1 7 . 4 1 4 2 2 4 1 7 . 6 1 7 4 2 4 1 7 . 8 5 1 1 4 1 8 . 1 1 5 2 6 4 1 8 . 3 4 8 9 4 4 1 8 . 6 7 4 0 6 4 1 9 . 0 0 9 3 4 4 1 9 . 6 9 0 0 6 4 2 0 . 0 8 6 3 4 2 0 . 5 7 3 9 8 4 2 0 . 8 6 8 6 2 4 2 1 . 3 6 6 4 6 4 2 1 . 6 5 0 9 4 4 2 1 . 9 8 6 2 2 4 2 2 . 3 2 1 5 4 2 2 . 7 8 8 8 6 4 2 3 . 2 1 5 5 8 4 2 3 . 7 3 3 7 4 CRACKin 0 . 3 6 6 7 9 9 0 . 3 8 3 1 9 9 0 . 3 8 7 9 9 9 0 . 3 9 3 9 9 9 0 . 4 0 6 7 9 9 0 . 4 1 0 7 9 9 0 . 4 2 0 3 9 9 0 . 4 2 3 5 9 9 0 . 4 2 7 1 9 9 0 . 4 3 2 3 9 9 0 . 4 3 4 3 9 9 0 . 4 3 8 7 9 9 0 . 4 5 1 5 9 9 0 . 4 5 3 5 9 9 0 . 4 5 7 1 9 9 0 . 4 6 0 3 9 9 0 . 4 6 1 9 9 9 0 . 4 6 2 3 9 9 0 . 4 6 4 7 9 9 0 . 4 7 4 3 9 9 0 . 4 8 7 1 9 9 0 . 4 9 5 5 9 9 0 . 5 0 9 1 9 9 0 . 5 2 1 5 9 9 0 . 5 2 6 3 9 9 0 . 5 4 7 1 9 9 0 . 5 9 9 5 9 9 0 . 6 1 7 1 9 9 0 . 6 3 5 1 9 9 0 . 6 4 6 3 9 9 0 . 6 6 3 5 9 9 0 . 6 7 1 5 9 9 0 . 6 7 6 7 9 9 0 . 6 8 5 5 9 9 0 . 6 9 3 5 9 9 0 . 7 0 2 7 9 9 0 . 7 1 3 1 9 9 7 2 2 3 9 9 7 3 5 1 9 9 7 4 8 3 9 9 7 7 5 1 9 9 7 9 0 7 9 9 0 . 8 0 9 9 9 9 0 . 8 2 1 5 9 9 0 . 8 4 1 1 9 9 0 . 8 5 2 3 9 9 0 . 8 6 5 5 9 9 0 . 8 7 8 7 9 9 0 . 8 9 7 1 9 9 0 . 9 1 3 9 9 9 0 . 9 3 4 3 9 9 0. 0. 0. 0. 0. 1 9 . 0 7 7 4 6 1 6 . 2 5 3 5 2 1 5 . 0 5 0 9 4 1 3 . 8 5 4 8 2 1 3 . 0 9 2 4 2 1 2 . 1 5 1 9 2 1 1 . 2 8 7 7 6 1 0 . 3 0 7 1 7 9 . 6 6 8 1 8 6 9 . 0 4 2 7 5 8 8 . 3 7 0 4 5 7 7 . 7 2 0 1 9 8 7 . 1 4 5 5 4 6 7 . 1 6 5 7 1 3 6 . 8 4 2 1 0 6 6 . 5 1 1 3 9 8 6 . 1 6 3 6 0 5 5 . 8 0 4 3 3 0 7 . 2 8 0 1 5 2 7 . 3 8 0 3 1 7 7 . 5 1 6 9 4 5 7 . 6 0 8 5 7 2 7 . 7 6 0 3 3 1 7 . 9 0 2 4 9 7 7 . 9 5 8 5 3 1 8 . 2 0 8 0 4 5 8 . 8 8 8 8 0 5 9 . 1 3 5 8 1 5 9 . 3 9 8 8 8 3 9 . 5 6 8 1 6 9 9 . 8 3 6 9 2 9 9 . 9 6 5 7 0 4 1 0 . 0 5 0 7 3 1 0 . 1 9 7 0 7 1 0 . 3 3 2 8 3 1 0 . 4 9 2 2 6 1 0 . 6 7 6 8 8 1 0 . 8 4 4 2 2 1 1 . 0 8 3 5 3 1 1 . 3 3 8 5 9 1 1 . 8 8 3 8 8 1 2 . 2 1 9 3 9 1 2 . 6 5 2 0 4 1 2 . 9 2 4 5 9 1 3 . 4 0 5 4 8 1 3 . 6 9 2 4 0 1 4 . 0 4 2 5 4 1 4 . 4 0 6 3 3 1 4 . 9 3 7 5 8 1 5 . 4 4 8 9 1 1 6 . 1 0 6 2 3 0 3 5 0 0 0 4 5 0 0 0 6 5 0 0 0 8 5 0 0 0 1 0 5 0 0 0 1 5 0 0 0 0 1 9 0 0 0 0 2 5 0 0 0 0 3 5 0 0 0 0 4 5 0 0 0 0 7 5 0 0 0 0 2 8 5 0 0 0 0 3 1 5 0 0 0 0 3 5 5 0 0 0 0 4 1 5 0 0 0 0 4 7 5 0 0 0 0 5 2 5 0 0 0 0 5 5 5 0 0 0 0 5 9 5 0 0 0 0 6 3 6 0 0 0 0 6 6 6 0 0 0 0 7 0 6 0 0 0 0 7 4 6 0 0 0 0 7 6 8 5 0 0 0 8 3 8 5 0 0 0 8 6 8 5 0 0 0 8 7 3 5 0 0 0 8 7 7 5 0 0 0 8 8 0 5 0 0 0 8 8 3 5 0 0 0 8 8 4 5 0 0 0 8 8 5 5 0 0 0 8 8 6 5 0 0 0 8 8 7 5 0 0 0 8 8 8 5 0 0 0 8 8 9 5 0 0 0 8 9 0 5 0 0 0 8 9 1 5 0 0 0 8 9 2 5 0 0 0 8 9 4 5 0 0 0 8 9 5 5 0 0 0 8 9 6 5 0 0 0 8 9 7 0 0 0 0 8 9 7 5 0 0 0 8 9 8 0 0 0 0 8 9 8 5 0 0 0 8 9 9 0 0 0 0 8 9 9 5 0 0 0 9 0 0 0 0 0 0 9 0 0 5 0 0 0 4. 1 1 . 1 7 e - 0 4 . 2 2 e - 0 4 . 5 2 e - 0 4 3 . 2 5 e - 0 4 1 . 0 2 e - 0 4 2 . 4 4 e - 0 4 8 . 1 3 e - 0 5 9 . 1 4 e - 0 5 1 . 3 2 e - 0 4 5 . 0 8 e - 0 5 1 . 1 2 e - 0 4 3 . 2 5 e - 0 4 5 . 0 8 e - 0 5 9 . 1 4 e - 0 5 8 . 1 3 e - 0 5 4 . 0 6 e - 0 5 1 . 0 2 e - 0 5 6 . 1 0 e - 0 5 2 . 4 4 e - 0 4 3 . 2 5 e - 0 4 2 . 1 3 e - 0 4 3 . 4 5 e - 0 4 3 . 1 5 e - 0 4 1 . 2 2 e - 0 4 5 . 2 8 e - 0 4 1 . 3 3 e - 0 3 4 . 4 7 e - 0 4 4 . 5 7 e - 0 4 2 . 8 4 e - 0 4 4 . 3 7 e - 0 4 2 . 0 3 e - 0 4 1 . 3 2 e - 0 4 2 . 2 4 e - 0 4 2 . 0 3 e - 0 4 2 . 3 4 e - 0 4 2 . 6 4 e - 0 4 2 . 3 4 e - 0 4 3 . 2 5 e - 0 4 3 . 3 5 e - 0 4 6.81 e - 0 4 3 . 9 6 e - 0 4 4 . 8 8 e - 0 4 2 . 9 5 e - 0 4 . 9 8 e - 0 4 . 8 4 e - 0 4 . 3 5 e - 0 4 . 3 5 e - 0 4 . 6 7 e - 0 4 4 . 2 7 e - 0 4 5 . 1 8 e - 0 4 4. 2. 3. 3. 4. dN 3 5 0 0 0 1 0 0 0 0 2 0 0 0 0 2 0 0 0 0 2 0 0 0 0 4 5 0 0 0 4 0 0 0 0 6 0 0 0 0 1 0 0 0 0 0 1 0 0 0 0 0 3 0 0 0 0 0 2 1 0 0 0 0 0 3 0 0 0 0 0 4 0 0 0 0 0 6 0 0 0 0 0 6 0 0 0 0 0 5 0 0 0 0 0 3 0 0 0 0 0 4 0 0 0 0 0 4 1 0 0 0 0 3 0 0 0 0 0 4 0 0 0 0 0 4 0 0 0 0 0 2 2 5 0 0 0 7 0 0 0 0 0 3 0 0 0 0 0 5 0 0 0 0 4 0 0 0 0 3 0 0 0 0 3 0 0 0 0 1 0 0 0 0 1 0 0 0 0 1 0 0 0 0 1 0 0 0 0 1 0 0 0 0 1 0 0 0 0 1 0 0 0 0 1 0 0 0 0 1 0 0 0 0 2 0 0 0 0 1 0 0 0 0 1 0 0 0 0 5 0 0 0 5 0 0 0 5 0 0 0 5 0 0 0 5 0 0 0 5 0 0 0 5 0 0 0 5 0 0 0 da/dN 1.19E-08 1.22E-08 7.62E-09 . 6 3 e - 0 8 0 8 e - 0 9 4 2 e - 0 9 . 0 3 e - 0 9 1 . 5 2 e - 0 9 1 . 3 2 e - 0 9 5 . 0 8 e - 1 0 3 . 7 3 e - 1 0 1 . 5 5 e - 1 0 1 . 6 9 e - 1 0 2 . 2 9 e - 1 0 1 . 3 5 e - 1 0 6 . 7 7 e - 1 1 2 . 0 3 e - 1 1 2 . 0 3 e - 1 0 6 . 1 0 e - 1 0 7 . 9 3 e - 1 0 7.11 e - 1 0 8 . 6 4 e - 1 0 7 . 8 7 e - 1 0 5 . 4 2 e - 1 0 7 . 5 5 e - 1 0 4 . 4 4 e - 0 9 8 . 9 4 e - 0 9 1 . 1 4 e - 0 8 9 . 4 8 e - 0 9 1 . 4 6 e - 0 8 2 . 0 3 e - 0 8 1 . 3 2 e - 0 8 2 . 2 4 e - 0 8 2 . 0 3 e - 0 8 2 . 3 4 e - 0 8 2 . 6 4 e - 0 8 2 . 3 4 e - 0 8 3 . 2 5 e - 0 8 3 . 3 5 e - 0 8 3 . 4 0 e - 0 8 3 . 9 6 e - 0 8 4 . 8 8 e - 0 8 5 . 8 9 e - 0 8 9 6 e - 0 8 6 9 e - 0 8 7 1 e - 0 8 7 1 e - 0 8 3 5 e - 0 8 8 . 5 3 e - 0 8 1 . 0 4 e - 0 7 9. 5. 6. 6. 9. 155 LOADCRACKmmCRACKIn 12 10 9.2 8.4 7.6 6.8 6.2 5.6 5 4.6 4.2 3.8 3.6 3.4 3.2 3 2.8 2.6 2.4 2.2 2 1.8 2 2 2 2.2 2.2 2.2 2.2 2.2 2.2 2.2 2.2 2.2 2.2 2.2 2.2 2.2 2.2 2.2 2.2 2.2 2.2 2.2 2.2 2.2 2.2 2.2 2.2 2.2 2.2 10.414 10.65784 10.78992 10.85088 10.95248 11.08456 11.30808 11.51128 11.76528 11.97864 12.18184 12.27328 12.37488 12.446 12.63904 12.83208 12.94384 13.06576 13.17752 13.26896 13.32992 13.34008 13.45184 13.54328 13.62456 13.72616 13.8684 14.06144 14.224 14.46784 14.58976 14.75232 14.91488 15.07744 15.22984 15.40256 15.61592 15.80896 16.01216 16.21536 16.4084 16.6624 16.85544 17.1196 17.35328 17.60728 17.84096 18.16608 18.31848 18.50136 18.7452 WROUGHT DUPLEX STAINLESS STEEL 1M NaCL +0.300 Vsce STR.INT CUM.CYC. 0.41 0.4196 0.4248 0.4272 0.4312 0.4364 0.4452 0.4532 0.4632 0.4716 0.4796 0.4832 0.4872 0.49 0.4976 0.5052 0.5096 0.5144 0.5188 0.5224 0.5248 0.5252 0.5296 0.5332 0.5364 0.5404 0.546 0.5536 0.56 0.5696 0.5744 0.5808 0.5872 0.5936 0.5996 0.6064 0.6148 0.6224 0.6304 0.6384 0.646 0.656 0.6636 0.674 0.6832 0.6932 0.7024 0.7152 0.7212 0.7284 0.738 20.23096 17.08378 15.83065 14.50229 13.19429 11.89129 10.97641 10.02635 9.079567 8.453546 7.807171 7.100180 6.765258 6.415222 6.104505 5.786445 5.435408 5.082681 4.722048 4.351493 3.969899 3.575015 3.998106 4.019450 4.038560 4.468894 4.506346 4.557899 4.601972 4.669241 4.703408 4.749528 4.796304 4.843749 4.888848 4.940699 5.005858 5.065892 5.130222 5.195748 5.259132 5.344264 5.410317 5.502657 5.586277 5.679299 5.766902 5.892125 5.952205 6.025500 6.125317 0 6000 10000 14000 20000 26000 34000 58000 98000 128000 158000 188000 218000 268000 348000 428000 508000 608000 728000 868000 1048000 1298000 1498000 1698000 2098000 2248000 2448000 2648000 2848000 3048000 3198000 3348000 3498000 3648000 3798000 3948000 4098000 4248000 4398000 4548000 4698000 4848000 4998000 5148000 5298000 5448000 5598000 5748000 5848000 5948000 6048000 da 0 2.44E-04 1.32E-04 6.10E-05 1.02E-04 1.32E-04 2.24E-04 2.03E-04 2.54E-04 2.13E-04 2.03E-04 9.14E-05 1.02E-04 7.11 E-05 1.93E-04 1.93E-04 1.12E-04 1.22E-04 1.12E-04 9.14E-05 6.10E-05 1.02E-05 1.12E-04 9.14E-05 8.13E-05 1.02E-04 1.42E-04 1.93E-04 1.63E-04 2.44E-04 .22E-04 .63E-04 .63E-04 .63E-04 .52E-04 .73E-04 .13E-04 .93E-04 2.03E-04 2.03E-04 1.93E-04 2.54E-04 1.93E-04 2.64E-04 2.34E-04 2.54E-04 2.34E-04 3.25E-04 1.52E-04 1.83E-04 2.44E-04 1 1 1. 1. 1. 1. 2  1 dN 6000 4000 4000 6000 6000 8000 24000 40000 30000 30000 30000 30000 50000 80000 80000 80000 100000 120000 140000 180000 250000 200000 200000 400000 150000 200000 200000 200000 200000 150000 150000 150000 150000 150000 150000 150000 150000 150000 150000 150000 150000 150000 150000 150000 150000 150000 150000 100000 100000 100000 da/dN 4.06E-08 3.30E-08 1.52E-08 1.69E-08 2.20E-08 2.79E-08 8.47E-09 6.35E-09 7.11 E-09 6.77E-09 3.05E-09 3.39E-09 1.42E-09 2.41 E-09 2.41 E-09 1.40E-09 1.22E-09 9.31 E-10 6.53E-10 3.39E-10 4.06E-11 5.59E-10 4.57E-10 2.03E-10 6.77E-10 7.11E-10 9.65E-10 8.13E-10 1.22E-09 8.13E-10 1.08E-09 1.08E-09 1.08E-09 1.02E-09 1.15E-09 1.42E-09 1.29E-09 1.35E-09 1.35E-09 1.29E-09 1.69E-09 1.29E-09 1.76E-09 1.56E-09 1.69E-09 1.56E-09 2.17E-09 1.52E-09 1.83E-09 2.44E-09 156 2.2 18.9484 2.2 19.1516 2.2 19.38528 2.2 19.6088 2.2 19.84248 2.2 20.15744 2.2 20.48256 2.2 20.828 2.2 21.20392 2.2 21.44776 2.2 21.67128 2.2 21.8948 2.2 22.15896 2.2 22.42312 2.2 22.72792 2.2 23.04288 2.2 23.27656 2.2 23.5204 2.2 24.01824 2.2 24.30272 2.2 24.72944 2.2 25.05456 2.2 25.62352 2.2 26.30424 2.2 26.57856 2.2 26.90368 2.2 27.23896 2.2 27.66568 2.2 28.1432 2.2 28.40736 0.746 6.210375 0.754 6.297191 0.7632 6.399269 0.772 6.499222 0.7812 6.606215 0.7936 6.754615 0.8064 6.913069 0.82 7.087576 0.8348 7.285065 0.8444 7.417603 0.8532 7.542297 0.862 7.670163 0.8724 7.825523 0.8828 7.985666 0.8948 8.176661 0.9072 8.381358 0.9164 8.538277 0.926 8.706808 0.9456 9.066939 0.9568 9.282982 0.9736 9.622042 0.9864 9.893217 1.0088 10.39671 1.0356 11.05251 1.0464 11.33489 1.0592 11.68418 1.0724 12.06208 1.0892 12.57101 1.108 13.18099 1.1184 13.53838 6148000 2.03E-04 6248000 2.03E-04 6348000 2.34E-04 6448000 2.24E-04 6548000 2.34E-04 6648000 3.15E-04 6748000 3.25E-04 6848000 3.45E-04 6948000 3.76E-04 7018000 2.44E-04 7068000 2.24E-04 7118000 2.24E-04 7168000 2.64E-04 7218000 2.64E-04 7268000 3.05E-04 7298000 3.15E-04 7328000 2.34E-04 7358000 2.44E-04 7408000 4.98E-04 7438000 2.84E-04 7468000 4.27E-04 7488000 3.25E-04 7518000 5.69E-04 7548000 6.81 E-04 7558000 2.74E-04 7568000 3.25E-04 7578000 3.35E-04 7588000 4.27E-04 7598000 4.78E-04 7603000 2.64E-04 100000 2.03E-09 100000 2.03E-09 100000 2.34E-09 100000 2.24E-09 100000 2.34E-09 100000 3.15E-09 100000 3.25E-09 100000 3.45E-09 100000 3.76E-09 70000 3.48E-09 50000 4.47E-09 50000 4.47E-09 50000 5.28E-09 50000 5.28E-09 50000 6.10E-09 30000 1.05E-08 30000 7.79E-09 30000 8.13E-09 50000 9.96E-09 30000 9.48E-09 30000 1.42E-08 20000 1.63E-08 30000 1.90E-08 30000 2.27E-08 10000 2.74E-08 10000 3.25E-08 10000 3.35E-08 10000 4.27E-08 10000 4.78E-08 5000 5.28E-08 157 WROUGHT DUPLEX STAINLESS STEEL 1M NaCL -1.200 Vsce Transverse direction LOADCRACKmmCRACKIn STR.INTCUM.CYC. 10 10 9.2 8.4 7.6 7 6.4 5.8 5.4 5 4.6 4.2 3.8 3.6 3.4 3.2 3 2.8 2.6 2.4 2.2 2 1.8 2 10.5156 10.85088 11.16584 11.35888 11.5824 11.8872 12.05992 12.20216 12.3444 12.446 12.53744 12.63904 12.87272 13.04544 13.26896 13.53312 13.84808 14.0716 14.21384 14.36624 14.54912 14.6812 14.75232 14.85392 0.414 0.4272 0.4396 0.4472 0.456 0.468 0.4748 0.4804 0.486 0.49 0.4936 0.4976 0.5068 0.5136 0.5224 0.5328 0.5452 0.554 0.5596 0.5656 0.5728 0.578 0.5808 0.5848 16.95223 17.26463 16.16031 14.91308 13.66105 12.79833 11.81526 10.79369 10.13039 9.434151 8.724614 8.012163 7.346584 7.029328 6.725035 6.427313 6.137684 5.804445 5.435418 5.062954 4.691979 4.299337 3.885978 4.344260 0 4000 9000 14000 22000 32000 42000 52000 62000 72000 92000 122000 182000 242000 302000 354000 414000 514000 614000 764000 964000 1164000 1364000 1564000 da 3.35E-04 3.15E-04 1.93E-04 2.24E-04 3.05E-04 1.73E-04 1.42E-04 1.42E-04 1.02E-04 9.14E-05 1.02E-04 2.34E-04 1.73E-04 2.24E-04 2.64E-04 3.15E-04 2.24E-04 1.42E-04 1.52E-04 1.83E-04 1.32E-04 7.11 E-05 1.02E-04 dN 4000 5000 5000 8000 10000 10000 10000 10000 10000 20000 30000 60000 60000 60000 52000 60000 100000 100000 150000 200000 200000 200000 200000 da/dN 8.38E-08 6.30E-08 3.86E-08 2.79E-08 3.05E-08 1.73E-08 1.42E-08 1.42E-08 1.02E-08 4.57E-09 3.39E-09 3.89E-09 2.88E-09 3.73E-09 5.08E-09 5.25E-09 2.24E-09 1.42E-09 1.02E-09 9.14E-10 6.60E-10 3.56E-10 5.08E-10 158 LOADCRACKmm 12 9.9058 11.2 10.03788 10.4 10.22076 9.6 10.39348 8.8 10.5662 8 10.72876 7.4 10.93196 6.8 11.10468 6.2 11.21644 5.8 11.3282 5.4 11.39932 5 11.51108 4.6 11.55172 4.2 11.61268 4 11.77524 3.8 11.87684 3.6 12.05972 3.4 12.20196 3.2 12.25276 3 12.395 2.8 12.4966 2.6 12.5982 2.4 12.65916 2.2 12.72012 2 12.903 2 12.99444 2.2 13.04524 2.4 13.13668 2.4 13.32972 2.4 13.5126 2.4 13.59388 2.4 13.64468 2.4 13.72596 2.6 13.82756 2.6 13.94948 2.6 14.04092 2.6 14.2238 2.6 14.36604 2.6 14.54892 2.6 14.7826 2.6 14.97564 2.6 15.06708 2.6 15.42268 2.6 15.82908 2.6 16.23548 2.6 16.43868 2.6 16.78412 2.6 16.98732 2.6 17.18036 2.6 17.41404 WROUGHT DUPLEX STAINLESS 1M NaCL -0.400 Vsce Transverse direction STR.INTCUM.CYC. STEEL CRACKIn 0.389992 0.395192 0.402392 0.409192 0.415992 0.422392 0.430392 0.437192 0.441592 0.445992 0.448792 0.453192 0.454792 0.457192 0.463592 0.467592 0.474792 0.480392 0.482392 0.487992 0.491992 0.495992 0.498392 0.500792 0.507992 0.511592 0.513592 0.517192 0.524792 0.531992 0.535192 0.537192 0.540392 0.544392 0.549192 0.552792 0.559992 0.565592 0.572792 0.581992 0.589592 0.593192 0.607192 0.623192 0.639192 0.647192 0.660792 0.668792 0.676392 0.685592 19.68491 18.50316 17.35140 16.16680 14.95900 13.71993 12.83263 11.90445 10.92104 10.27968 9.608459 8.952002 8.254447 7.562252 7.267694 6.943638 6.646013 6.327266 5.972172 5.644145 5.298322 4.948383 4.583634 4.216308 3.873344 3.893740 4.295651 4.710928 4.763824 4.814725 4.837598 4.851973 4.875100 5.312919 5.351150 5.380084 5.438630 5.484802 5.545000 5.623317 5.689224 5.720834 5.846208 5.994429 6.148175 6.227209 6.365007 6.448155 6.528626 6.628021 0 10000 23000 33000 43000 53000 63000 73000 83000 93000 103000 118000 133000 163000 213000 263000 363000 463000 544000 744000 944000 1144000 1344000 1744000 3124000 3724000 4044000 4444000 5044000 5494000 5644000 5994000 6394000 6794000 7194000 7544000 7844000 8094000 8494000 8808000 9008000 9138000 9538000 9738000 9888000 9988000 10088000 10138000 10188000 10238000 1 1 1 1. 1 2 1. 1. 1. 7. 1. 4 6 1 1. 1. 1. 5. 1. 1. 1. 6. 6. 1. 9. 5. 9. 1. 1. 8. 5. 8. 1. 1. 9. 1, 1 1 2 1 9 3 4 4 2 3 2 1 2 da 32E-04 83E-04 73E-04 73E-04 63E-04 03E-04 73E-04 12E-04 12E-04 11 E-05 12E-04 06E-05 10E-05 63E-04 02E-04 83E-04 42E-04 08E-05 42E-04 02E-04 02E-04 10E-05 .10E-05 .83E-04 14E-05 .08E-05 .14E-05 93E-04 83E-04 .13E-05 .08E-05 .13E-05 .02E-04 .22E-04 .14E-05 .83E-04 .42E-04 .83E-04 .34E-04 .93E-04 .14E-05 .56E-04 .06E-04 06E-04 03E-04 45E-04 .03E-04 93E-04 .34E-04 dN 10000 13000 10000 10000 10000 10000 10000 10000 10000 10000 15000 15000 30000 50000 50000 100000 100000 81000 200000 200000 200000 200000 400000 1380000 600000 320000 400000 600000 450000 150000 350000 400000 400000 400000 350000 300000 250000 400000 314000 200000 130000 400000 200000 150000 100000 100000 50000 50000 50000 da/dN 1.32E-08 1.41 E-08 1.73E-08 1.73E-08 1.63E-08 2.03E-08 1.73E-08 1.12E-08 1.12E-08 7.11 E-09 7.45E-09 2.71 E-09 2.03E-09 3.25E-09 2.03E-09 1.83E-09 1.42E-09 6.27E-10 7.11E-10 5.08E-10 5.08E-10 3.05E-10 1.52E-10 1.33E-10 1.52E-10 1.59E-10 2.29E-10 3.22E-10 4.06E-10 5.42E-10 1.45E-10 2.03E-10 2.54E-10 3.05E-10 2.61E-10 6.10E-10 5.69E-10 4.57E-10 7.44E-10 9.65E-10 7.03E-10 8.89E-10 2.03E-09 2.71 E-09 2.03E-09 3.45E-09 4.06E-09 3.86E-09 4.67E-09 159 2.6 2.6 2.6 2.6 2.6 2.6 2.6 2.6 2.6 2.6 2.6 2.6 2.6 2.6 2.6 2.6 2.6 2.6 2.6 2.6 2.6 2.6 17.6782 17.85092 18.11508 18.29796 18.51132 18.76532 18.95836 19.16156 19.36476 19.58828 19.89308 20.015 20.18772 20.35028 20.51284 20.6754 20.89892 21.11228 21.34596 21.55932 21.86412 22.12828 0.695992 0.702792 0.713192 0.720392 0.728792 0.738792 0.746392 0.754392 0.762392 0.771192 0.783192 0.787992 0.794792 0.801192 0.807592 0.813992 0.822792 0.831192 0.840392 0.848792 0.860792 0.871192 6.743074 6.819894 6.939895 7.024802 7.125809 7.248873 7.344514 7.447218 7.552064 7.669940 7.835133 7.902693 7.999897 8.093018 8.187763 8.284173 8.419528 8.551838 8.700350 8.839348 9.043770 9.226735 10288000 10328000 10368000 10398000 10428000 10458000 10478000 10498000 10518000 10538000 10558000 10568000 10578000 10588000 10598000 10608000 10618000 10628000 10638000 10648000 10658000 10668000 2.64E-04 1.73E-04 2.64E-04 1.83E-04 2.13E-04 2.54E-04 1.93E-04 2.03E-04 2.03E-04 2.24E-04 3.05E-04 1.22E-04 1.73E-04 1.63E-04 1.63E-04 1.63E-04 2.24E-04 2.13E-04 2.34E-04 2.13E-04 3.05E-04 2.64E-04 50000 40000 40000 30000 30000 30000 20000 20000 20000 20000 20000 10000 10000 10000 10000 10000 10000 10000 10000 10000 10000 10000 5.28E-09 4.32E-09 6.60E-09 6.10E-09 7.11 E-09 8.47E-09 9.65E-09 1.02E-08 1.02E-08 1.12E-08 1.52E-08 1.22E-08 1.73E-08 1.63E-08 1.63E-08 1.63E-08 2.24E-08 2.13E-08 2.34E-08 2.13E-08 3.05E-08 2.64E-08 160 LOADCRACKmm 12 10.6476 11 10.86096 10 10.98288 9.2 11.17592 8.4 11.308 7.6 11.47056 7 11.55184 6.4 11.6636 5.8 11.78552 5.4 11.95824 5 12.13096 4.6 12.24272 4.2 12.36464 4 12.5272 3.8 12.63896 3.6 12.80152 3.4 12.89296 3.2 13.00472 3 13.09616 2.8 13.27904 2.6 13.47208 2.4 13.71592 2.2 13.80736 2 13.85816 1.8 13.92928 2 14.06136 2 14.2036 2 14.34584 2 14.4576 2 14.63032 2 14.83352 2 15.02656 2 15.2704 2 15.50408 2 15.7784 2 16.10352 2 16.2864 2 16.45912 2 16.66232 2 16.8452 2 17.02808 2 17.2516 2 17.47512 2 17.7088 2 17.9628 2 18.2168 2 18.49112 2 18.73496 2 19.06008 2 19.21248 WROUGHT DUPLEX STAINLESS 1M NaCL +0.300 Vsce Transverse direction STR.INTCUM.CYC. STEEL CRACKIn 0.419196 0.427596 0.432396 0.439996 0.445196 0.451596 0.454796 0.459196 0.463996 0.470796 0.477596 0.481996 0.486796 0.493196 0.497596 0.503996 0.507596 0.511996 0.515596 522796 530396 539996 543596 545596 548396 0.553596 0.559196 0.564796 0.569196 0.575996 0.583996 0.591596 0.601196 0.610396 0.621196 0.633996 0.641196 0.647996 0.655996 0.663196 0.670396 0.679196 0.687996 0.697196 0.707196 0.717196 0.727996 0.737596 0.750396 0.756396 0. 0. 0. 0. 0. 0. 20.48912 19.00157 17.38986 16.16928 14.87120 13.57649 12.56119 11.55613 10.54420 9.912378 9.267645 8.580132 7.888225 7.582205 7.249066 6.931596 6.580893 6.233668 5.874914 5.541498 5.203662 4.872235 4.490219 4.094213 3.700238 4.143525 4.178552 4.214008 4.242171 4.286232 4.338919 4.389842 4.455412 4.519589 4.596653 4.690484 4.744494 4.796339 4.858398 4.915258 4.973096 5.045150 5.118743 5.197378 5.284879 5.374564 5.473965 5.564615 5.688972 5.748683 0 3000 6000 10000 14000 19000 24000 32000 42000 57000 72000 87000 107000 132000 157000 187000 217000 257000 307000 377000 477000 577000 627000 677000 777000 977000 1177000 1377000 1577000 1777000 1977000 2127000 2327000 2527000 2727000 2927000 3027000 3127000 3227000 3327000 3427000 3527000 3627000 3727000 3827000 3927000 4027000 4127000 4227000 4277000 2. 1 1 1, 1 8 1 1 1 1 1 1. 1 1 1 9 1. 9 1 1 2 9 5. 7 1 1 1 1 1 2 1 2, 2 2, 3 1, 1 2 1. 1 2 2. 2. 2 2. 2 2 3. 1 da 13E-04 22E-04 93E-04 32E-04 63E-04 13E-05 12E-04 22E-04 73E-04 .73E-04 12E-04 22E-04 63E-04 12E-04 63E-04 14E-05 12E-04 14E-05 83E-04 93E-04 44E-04 14E-05 08E-05 .11 E-05 32E-04 42E-04 42E-04 12E-04 73E-04 03E-04 93E-04 44E-04 34E-04 74E-04 25E-04 83E-04 73E-04 03E-04 83E-04 83E-04 24E-04 24E-04 34E-04 54E-04 54E-04 74E-04 44E-04 25E-04 52E-04 dN 3000 3000 4000 4000 5000 5000 8000 10000 15000 15000 15000 20000 25000 25000 30000 30000 40000 50000 70000 100000 100000 50000 50000 100000 200000 200000 200000 200000 200000 200000 150000 200000 200000 200000 200000 100000 100000 100000 100000 100000 100000 100000 100000 100000 100000 100000 100000 100000 50000 da/dN 7.11E-08 4.06E-08 4.83E-08 3.30E-08 3.25E-08 1.63E-08 1.40E-08 1.22E-08 1.15E-08 1.15E-08 7.45E-09 6.10E-09 6.50E-09 4.47E-09 5.42E-09 3.05E-09 2.79E-09 1.83E-09 2.61 E-09 1.93E-09 2.44E-09 1.83E-09 1.02E-09 7.11E-10 6.60E-10 7.11E-10 7.11E-10 5.59E-10 8.64E-10 1.02E-09 1.29E-09 1.22E-09 1.17E-09 1.37E-09 1.63E-09 1.83E-09 1.73E-09 2.03E-09 1.83E-09 1.83E-09 2.24E-09 2.24E-09 2.34E-09 2.54E-09 2.54E-09 2.74E-09 2.44E-09 3.25E-09 3.05E-09 161 2 19.37504 2 19.55792 2 19.77128 2 19.95416 2 20.16752 2 20.41136 2 20.64504 2 20.9092 2 21.07176 2 21.2648 2 21.45784 2 21.69152 2 21.80328 2 21.99632 2 22.11824 2 22.30112 2 22.45352 2 22.84976 2 23.37808 2 23.88608 2 24.39408 2 25.4812 0.762796 5.813404 0.769996 5.887515 0.778396 5.975763 0.785596 6.052978 0.793996 6.144951 0.803596 6.252634 0.812796 6.358492 0.823196 6.481410 0.829596 6.558825 0.837196 6.652565 0.844796 6.748327 0.853996 6.867039 0.858396 6.924925 0.865996 7.026651 0.870796 7.092061 0.877996 7.191907 0.883996 7.276738 0.899596 7.504492 0.920396 7.825307 0.940396 8.153784 0.960396 8.503660 1.003196 9.333777 4327000 1.63E-04 4377000 1.83E-04 4427000 2.13E-04 4477000 1.83E-04 4527000 2.13E-04 4577000 2.44E-04 4627000 2.34E-04 4677000 2.64E-04 4707000 1.63E-04 4737000 1.93E-04 4767000 1.93E-04 4797000 2.34E-04 4817000 1.12E-04 4837000 1.93E-04 4857000 1.22E-04 4877000 1.83E-04 4897000 1.52E-04 4937000 3.96E-04 4987000 5.28E-04 5027000 5.08E-04 5067000 5.08E-04 5117000 1.09E-03 50000 3.25E-09 50000 3.66E-09 50000 4.27E-09 50000 3.66E-09 50000 4.27E-09 50000 4.88E-09 50000 4.67E-09 50000 5.28E-09 30000 5.42E-09 30000 6.43E-09 30000 6.43E-09 30000 7.79E-09 20000 5.59E-09 20000 9.65E-09 20000 6.10E-09 20000 9.14E-09 20000 7.62E-09 40000 9.91 E-09 50000 1.06E-08 40000 1.27E-08 40000 1.27E-08 50000 2.17E-08 33 3.SC0PE OF THE PROJECT It has been pointed out in the literature that corrosion pits on the surface could act as crack initiation sites for fatigue cracking to occur. These fatigue cracks can eventually propagate causing failure of suction roll shells. Although some work has been done to understand fatigue crack propagation of duplex steels in synthetic white water environment, crack tip variables like pH and chemistry were either unknown or uncontrolled. For example, Bassidi et al. 105,129-30 have published several studies on the fatigue of duplex stainless steels but the electrochemical conditions at the crack tip (eg., pH and electrochemical potential) were neither controlled nor reported.Further, they did not consider the possible changes that could occur in the localized solution chemistry within the crack. This leads to ambiguities in ascertaining the mechanism responsible for cracking. The main goals of this study are listed below: (i) To understand the effect of phase composition on pitting behaviour of two duplex steels, one cast and other wrought, in chloride solutions. (ii) To generate near-threshold fatigue crack propagation data in chloride solutions by performing corrosion fatigue tests under well defined crack tip conditions (crack tip pH and potential). Therefore, the potentials at which hydrogen evolution is possible can be defined unambiguously. In addition, high R-ratio will be used to eliminate crack tip closure effects. This will minimise crack retardation effects. At potentials where hydrogen evolution is impossible and oxidation mechanisms are possible, the kinetics of oxide growth rate on bare metals surface will be studied (scratch tests). The results of the scratch tests and fractography of the failed sample will be used to ascertain the mechanism of cracking. At potentials where hydrogen evolution is possible, no attempt will be made to ascertain the mechanism of hydrogen embrittlement. 3 4 In accordance with Section 1.5 the main objectives of the present work are summarized below: 1. To find the effect of phase composition and partitioning effects on the pitting behaviour of duplex SS in chloride environments. 2. To determine region I (near-threshold) crack propagation rates of cast and wrought duplex SS in simple chloride solutions and synthetic white water as a function of applied stress intensity, controlled electrochemical potentials (anodic and cathodic) and rolling direction (wrought steel). 3. To measure crack closure stresses during the unloading part of the fatigue cycle. 4. To determine the repassivation kinetics of the steels in the environments of interest, including nucleation and growth of oxide films, and relate it to possible oxidation related models of crack growth at anodic potentials. 5. To determine whether good mixing occurs between the bulk and crack tip solutions at high frequency. 6. To determine the mechanistic model of cracking based on the effect of potential on FCP, crack fractography and repassivation kinetics. 

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:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0078527/manifest

Comment

Related Items