FORMATION AND BREAKDDOWN OF CHROMATE CONVERSION COATINGS ON Al-Zn-Mg-Cu 7X75 ALLOYS
DISSERTATION
Presented in Partial Fulfillment of the Requirements for the Degree Doctor of Philosophy in the Graduate School of The Ohio State University
By
Yuhchae Yoon, M.S.
* * * * *
The Ohio State University
2004
Dissertation Committee:
Dr. Rudolph G. Buchheit, Adviser Approved by
Dr. Gerald S. Frankel
______Dr. Michael J. Mills Adviser
Graduate Program in Mat. Sci. & Eng. Dr. Philip J. Grandinetti
ABSTRACT
The objective of this study was to characterize the formation and breakdown of
chromate conversion coatings (CCCs) on aluminum alloys Al-Zn-Mg-Cu 7075 and 7475
with a focus on the effect of alloy temper, alloy purity and selected coating processing
variables. Overall, results consistently pointed to a slight temper effect. Conversion
coated AA7475-T7 was significantly more corrosion resistant than conversion coated
AA7475-T6. In AA7075, there was only a slight difference in corrosion resistance
between the two tempers. This was attributed to the effect of constituent particles on coating formation and breakdown, which are present to a much greater extent in AA7075 than in AA7475. The difference in the corrosion resistance between the T6 and T7 tempers in the coated and uncoated conditions is about the same suggesting that the origin of any “temper effect” in conversion coated materials is ultimately due to the intrinsic change in corrosion susceptibility of the alloy itself. Thicker coating formed on
AA7475-T7 has the effect of increasing corrosion resistance, which could be associated with the 860 cm-1 Raman intensity band. Studies were also conducted with alloys in
retrogression and reaged tempers and the W temper. Results with these tempers were
mixed and no general conclusions could be drawn. In terms of electrochemically derived
measures of corrosion resistance (electrochemical impedance, and pitting potential
ii measurements), the magnitude of the temper effect was about the same as the effect due
to the purity difference between AA7075 and AA7475. The temper effect was less
significant than effects due to increasing coating time from 1 to 3 minutes, withholding
certain substrate precleaning steps, or withholding key ingredients from the coating bath.
Scanning probe microscopy, scanning Kelvin probe force microscopy (SKPFM)
and scanning electron microscopy were used to characterize coating formation in the vicinity of constituent intermetallic particles (IMPs) present in the alloys. Coatings formed on IMPs exhibited different morphologies and were much thinner that coatings formed on the matrix. Coatings were thinnest on S phase particles, and in post-coating exposure to aggressive chloride environments coating breakdown was almost always associated with these particles. The difference in coating thickness between particles and the surrounding matrix was established within the first tens of seconds of coating. The thickness differential was then observed to remain constant for the remainder of the coating immersion time. Although coatings were thinner on IMPs, they did confer some level of protection. SKPFM measurements showed that the difference in the Volta potential measured between the matrix and the particles was strongly reduced by the conversion coating suggesting that intact coatings reduced any tendencies for microgalvanic coupling.
Thin film analogs of two important intermetallic compound particles, η (MgZn2) and S (Al2CuMg), were synthesized and analyzed using quartz crystal microbalance
weight change measurements, potentiodynamic polarization and Raman spectroscopy.
MgZn2 compounds were prepared with different levels of Cu to simulate the
compositional change that occurs when 7X75 alloys are artificially aged. Results showed
iii that these phases are electrochemically active and the conversion coatings reduce activity
only slightly. The coatings that do form on these compounds are thin and may have lower
hexavalent Cr contents than coatings that form elsewhere. With increasing amounts of Cu dissolved in the η phase, coatings are even thinner and possibly more defficient in Cr6+.
Multi-element electrode array experiments were used to electrochemically
characterize coating formation and breakdown. Coating formation was characterized by
an episode of measurable electrochemical activity on the array lasting for about 30
seconds. This was followed by a period of little measurable electrochemical activity.
During the episode of electrochemical activity, individual electrode elements in the array
exhibited transients that were predominantly anodic, predominantly cathodic or mixed. In
subsequent potentiodynamic polarization of these coated electrodes in a chloride
environment there was a weak dependence of the pitting potential on the character of the
current transient with electrodes exhibiting cathodic transients having slightly more noble
pitting potentials than electrodes that had anodic or mixed transients.
iv
Dedicated to
My loving wife, Chimi Woo,
My beloved son, John Hyunsung Yoon,
My parents and parents-in-law,
And Elohim
v
ACKNOWLEDGMENTS
I would like to express my heartfelt thanks to my advisor, Dr. Rudy Buchheit, who has given me profound guidance, inspiring motivation, excellent advice, and support over past five years. I deeply appreciate not only his knowledge for research but also his demeanor as a scholar. I would like to thank specially to Dr. Jerry Frankel and Dr.
Michael Mills for their helps as academic advisory committee for comments and suggests on my research.
I would like to give special thanks to all members of Fontana Corrosion Center:
Dr. Patrick Leblanc, Dr. Valerie Laget, Dr. Christian Pagila, Dr. Thodla Ramgopal, Dr.
Weilong Zhang, Dr. Wenping Zhang, Dr. Qingjiang Meng, Mr. Rob Fecke, Mr. Ryan
Leard, Mr. Younghoon Baek and many current FCC members. I owe a special thanks to
Jiho Kang who gave me much help for EQCM, flash evaporation instrument, and computer and also to Ms. Belinda Hurley from Department of Chemistry for Raman
Spectroscopy. I also thank to Ms. Cindy Flores, Ms. Christine Putnam, Ms. Dena
Bruedigam, and Ms. Susan Meager for their administrative help and assistance. To all of the faculty and staff of the Materials Science and Engineering I wish to give my thanks for their teaching and help.
vi I wish to extend my thanks to all church members of the Korean Mission Lane
Ave Baptist Church and to my friends in the Christian Graduate Student Alliance for their
love and prayer, especially to Pastor Chun and Bob Trube. Thanks also go to the staffs of the Korean Students Abroad in USA. In addition, I really appreciate all of those who deserve mention by name but are not named here.
I sincerely thank my parents and parents-in-law from my inmost heart and feelings for their endless love and care. Without their sacrificing love and encourage- ment, I couldn’t be here.
Most of all, I would like to give my wholehearted thanks to my loving wife,
Chimi who has been with me from freshman year at Yonsei University as life partner for her emotional supports and devoted love. I also thank my beloved son, Johnny who has been joy to me after his birth and forever.
Lastly, I wish to thank God who created and saved me.
vii
VITA
July 16, 1970...... Born – Pusan, Korea
1996...... B.S. Metallurgical Engineering, Yonsei University, Seoul, Korea
1998...... M.S. Metallurgical Engineering, Yonsei University, Seoul, Korea
1999 - present ...... Graduate Research Associate The Ohio State University
2002...... M.S. Materials Science and Engineering, The Ohio State University
PUBLICATIONS
1. Y. Yoon, V. Laget, and R.G. Buchheit. "The Effect of Artificial Heat Treatment on the Chromate Conversion Performance of Al-Zn-Mg-Cu alloys,” p.363, Tri- Service Corrosion, San Antonio, TX (2002). 2. Y. Yoon and R.G. Buchheit, “The Formation and Breakdown of Chromate Conversion Coatings in Al-Zn-Mg-Cu alloy using Electrode Arrays,” Proceedings of the 204th Meeting of the Electrochemical Society, Philadelphia, PA (2003).
FIELDS OF STUDY
Major Field: Materials Science and Engineering
viii
TABLE OF CONTENTS
page Abstract...... ii Acknowledgments...... vi Vita…...... viii List of Tables ...... xii List of Figures...... xiv
Chapters:
1. Introduction...... 1
2. Literature review...... 8 2.1 7xxx Aluminum Alloys ...... 9 2.1.1 Historical Development of 7xxx series ...... 9 2.1.2 Physical Metallurgy of Pure Aluminum and Aluminum Oxide...... 11 2.1.3 Physical Metallurgy of 7xxx series Aluminum Alloys...... 13 2.2 Localized Corrosion...... 21 2.2.1 Pitting Corrosion ...... 21 2.2.2 Intergranular Corrosion ...... 26 2.2.3 Stress Corrosion Cracking...... 29 2.3 Chromate Conversion Coatings...... 31 2.3.1 Role of Chromate Ion...... 32 2.3.2 Coating Processes and Formation of CCCs...... 36 2.3.3 Protection Mechanisms of CCCs...... 44 2.4 Key Unresolved Issues ...... 46 2.4.1 Formation and breakdown of CCCs on heterogeneous 7xxx aluminum alloys ...... 46 2.4.2 Influence of Temper on the Formation and Breakdown of CCCs ...... 47 2.4.3 Electrochemical activities of important IMCs ...... 48 2.5 Objective of Dissertation...... 48
ix 3. The Effect of Microstructural heterogeneity of Chromate Conversion Coating Formation on Al Alloy 7x75...... 76 3.1 Introduction ...... 76 3.2 Experimental...... 79 3.2.1 Materials and chemicals...... 79 3.2.2 Atomic force microscopy...... 80 3.3 Results ...... 82 3.3.1 Identification of the main constituent particles in AA7x75...... 82 3.3.2 AFM characterization of intermetallic compounds in as-polished AA7075. 83 3.3.3 AFM characterization of 3-minute chromate conversion coated surfaces on intermetallic compounds of AA7x75 ...... 84 3.3.4 Formation of chromate conversion coating at short times...... 87 3.4 Discussion...... 88 3.4.1 Heterogeneity in CCC formation...... 88 3.4.2 Coating formation on Al-Fe-Cu IMC particles ...... 89 3.4.3 Coating formation on Al2CuMg particles...... 90 3.4.4 Coating formation on Mg2Si particles ...... 92 3.4.5 Formation of trenches around constituent particles...... 92 3.4.6 CCC formation on heterogeneous AA7x75 surface...... 93 3.5 Summary...... 94
4. Corrosion behavior and Chromate Conversion Coating Formation on Thin Film analogs of η (MgZn2) and S (Al2CuMg) phases...... 119 4.1 Introduction ...... 119 4.2 Experimental...... 123 4.2.1 Thin film analogs of IMCs ...... 123 4.2.2 Quartz substrate working electrode...... 125 4.2.3 Electrochemical quartz crystal microbalance measurement...... 126 4.2.4 Raman spectroscopy ...... 127 4.3 Results and Discussion...... 128 4.3.1 Effect of Cu on the anodic polarization behavior of the compositional analogs ...... 128 4.3.2 Effect of Cu content in compositional analogs on CCC formation ...... 129 4.3.3 Effect of ferricyanide on CCC formation on η ...... 129 4.3.4 Reactivity of Al2CuMg in 0.5 M NaCl...... 130 4.3.5 Reactivity of Al2CuMg during chromate conversion coating formation .... 131 4.3.6 Reactivity of Al2CuMg after chromate conversion coating formation...... 132 4.4 Summary...... 134
x 5. Chromate Conversion Coating Formation and Breakdown on 7x75 aluminum Alloys ...... 152 5.1 Introduction ...... 152 5.2 Experimental...... 156 5.2.1 Materials and heat-treatment ...... 156 5.2.2 Microelectrode arrays ...... 157 5.2.3 Raman spectroscopy ...... 158 5.2.4 X-ray absorption near edge spectroscopy ...... 159 5.3 Results and Discussion...... 159 5.3.1 Coating thickness determined by Raman spectroscopy...... 159 5.3.2 Cr6+ content in the coating determined by XANES...... 161 5.3.3 Formation of CCCs on electrode array...... 161 5.3.4 Breakdown of CCCs on electrode array...... 163 5.3.5 The effect of coating bath chemistry on CCC formation and breakdown .. 165 5.4 Summary...... 167
6. Effect of Alloy Temper on Chromate Conversion Coating Performance of Al-Zn- Mg-Cu alloys ...... 194 6.1 Introduction ...... 194 6.2 Experimental...... 197 6.2.1 Materials and Heat-treatment...... 197 6.2.2 Chemicals and chromate conversion coatings...... 197 6.2.3 Electrochemical measurements...... 198 6.2.4 X-ray absorption near edge spectroscopy ...... 199 6.2.5 Salt spray exposure...... 200 6.3 Results ...... 200 6.3.1 EIS of conversion-coated samples ...... 200 6.3.2 Cr6+ contents in the coating...... 202 6.3.3 Salt spray testing...... 204 6.4 Discussion...... 206 6.5 Summary...... 208
7. Conclusions and Future work ...... 240 7.1 Conclusions ...... 240 7.2 Future work...... 242 Appendix A...... 245 Bibliography ...... 262
xi
LIST OF TABLES
Table page
2.1: Aluminum alloy and temper designation systems [11, 13] ...... 59
2.2: Compositions of 7xxx series aluminum alloys [11, 13] ...... 60
2.3: Zinc and magnesium contents and Zn/Mg ratio in 7xxx aluminum alloys [11]...... 61
2.4: Characteristics of microstructures for various heat treatment conditions [53]...... 61
3.1: T6 and T7 alloy temper schedule for AA7075 and AA7475...... 98
3.2: Comparison of typical Volta potential in air between AA7075-T7 and AA7075-T6...... 98
3.3: Comparison of surface height and Volta potential differences on different IMCs before and after 3-minute CCC...... 99
3.4: Surface height and Volta potential difference for 3-minute and 5-second CCC on Al- Fe-Cu IMC...... 99
4.1: Dissolution rates of thin film analog of S phase in different conditions...... 137
5.1: Chemical composition of AA7475 and AA7075 alloys using ICP-MS...... 172
6.1: Composition of the AA7475 and AA7075 alloys by ICP-MS...... 211
6.2: Chromate conversion coating process...... 211
6.3: Impedance parameter values of 1-minute chromate coated AA7475 after exposure to 0.5 M NaCl solution...... 212
6.4: Impedance parameter values of 3-minute chromate coated AA7475 after exposure to 0.5 M NaCl solution...... 213
6.5: Comparison of Cr6+ content of XANES for 3-minute and 1-minute CCCs of AA7475-T7 and AA7475-RRA in 0.5 M NaCl after 7 days...... 214
xii 6.6: Rankings of the degree of corrosion in salt spray test at 651 hr and 1310 hr visual inspection...... 215
A.1: Corrosion potentials, breakdown potentials, and repassivation potentials for AA7475 of all tempers and three different planes...... 253
xiii
LIST OF FIGURES
Figure page
2.1. Relationships of tensile strength, yield strength and SCC resistance in the 7xxx series (From Ref. [10])...... 62
2.2. Pourbaix diagram (Potential versus pH diagram) for aluminum with Al2O3•3H2O film at 25 °C (From Ref. [12])...... 63
2.3. Schematic showing the structure of the oxide film formed on unalloyed aluminum in dry air, Al = Aluminum, 1 = Surface layer, 2 = Mixed oxides, 3 = Pores, 4 = Barrier layer, and 5 = Heterogeneous components (From Ref. [13])...... 63
2.4. Schematic view of aluminum oxide film on rolled product (From Ref. [1])...... 64
2.5. Schematic of a galvanic corrosion cell involving cathodic activity at a Cu-containing IMCs (From Ref. [17])...... 64
2.6. Redox activity on IMCs. (a) SECM images of AA2024. (b) SEM images of the same region as in (a) (From Ref. [17])...... 65
2.7. The aluminum corner of the Al-Mg-Zn equilibrium phase diagram (From Ref. [3])...... 65
2.8. Typical metastalbe pit transients observed on 302 stainless steel in 0.1 M NaCl at 420 mVSCE (From Ref. [68])...... 66
2.9. Influence of potential on the pit growth, pit depth, and pit current density of Al foil in 0.01 M NaCl at pH 11 (From Ref. [73])...... 66
2.10. Regimes of crevice corrosion and inhibition of iron in 0.25 mm crevices in various chromate/chloride solutions (From Ref. [118])...... 67
2.11. Schematic diagram showing various regions in the section of the passive film developed on aluminum after immersion in chromate/dichromate solution (From Ref. [125])...... 67
xiv 2- 2.12. Time series showing the effect of CrO4 inhibitor on aluminum of the anodic current spikes associated with metastable pitting at potentiostatic potential (-0.5 VSCE) (From Ref. [129])...... 68
2.13. AFM image of AA2024-T3 after 3 s to a chromate solution. A: Al-Cu-Fe-Mn IMC (From Ref. [138, 141])...... 68
2.14. The measured net current density, the corrosion current density, and the true cathodic current density for (a) pure Al, (b) Al-4%Cu, and (c) Al-Cu IMC as a function of dichromate concentration (From Ref. [143])...... 69
2.15. Schematic of CCC inhibition mechanism over IMC and matrix. Chemisorption of cyanide or ferricyanide might inhibit redox mediation and reduce chromate reduction (From Ref. [144])...... 69
2.16. Schematic of CrIII film formation on copper (From Ref. [145])...... 70
2.17. Uniform formation and growth model for CCCs by Katzman (From Ref. [108]). 70
2.18. Non-uniform formation and growth model for CCCs by Brown and Wood (From Ref. [148])...... 71
2.19. CCC formation showing polymerization of Cr3+ hydroxide and adsorption of Cr6+ by Xia and McCreery (From Ref. [135])...... 71
2.20. Structure model of CCCs on Al by Treverton and Davis (From Ref. [133])...... 72
2.21. Structure model of CCCs on Al by Townsend and Hart (From Ref. [155])...... 72
2.22. Structure model of CCCs on AA2024 by Hughes (From Ref. [107])...... 73
2.23. A bipolar membrane model for passivity of metals by Sato (From Ref. [158])..... 73
2.24. Model for dynamic adsorption of Cr6+ by insoluble Cr3+ oxy-hydroxide by Xia (From Ref. [162])...... 74
2.25. A new duplex model incorporating both the Sato-bipolar model and McCreery-pit repassivation model by Clayton (From Ref. [153, 163])...... 74
2.26. Schematic depiction of condensation and dehydration reactions in Cr(OH)3 (From Ref. [164])...... 75
2.27. Schematic drawing of the gel near the Al surface (From Ref. [165])...... 75
3.1. Schematic description of surface morphology and Volta potential. (a) Tapping mode traces the topographic image of the sample surface and (b) Lift mode measures the Volta potential about 100 nm height from sample surface...... 100
xv 3.2. Schematic diagram for Atomic force microscopy. The surface morphology and Volta potential can be measured by cantilever deflection with constant resonant frequency and nullifying the DC voltage difference between the tip and the sample respectively...... 101
3.3. BSE images of AA7075 and AA7475. (a) AA7075 has many IMCs including Al-Fe- Cu (white), Al2CuMg (white), and Mg2Si (black) IMCs but (b) AA7475 has few IMCs, which are mainly Al-Fe-Cu IMCs (white)...... 102
3.4. SEM and EDS of AA7075-T6. (a) BSE image of Al-Fe-Cu particles (A~D), Mg2Si particles (1~4), and Al2CuMg particles(a~d). (b) Typical EDS spectrum of Al-Fe-Cu particles, Mg2Si particles, and Al2CuMg particles...... 103
3.5. BSE and AFM images of Al-Fe-Cu IMC, Al2MgCu and Mg2Si of as-polished AA7075-T7. (a) BSE image, (b) topographic images with a scale range of 500 nm, (c) Volta potential map with 0.5 V scale...... 104
3.6. BSE and AFM images of Al-Fe-Cu IMC and Mg2Si of 3-minute CCC coated AA7075-T7. (a) BSE image, (b) topographic images with a scale range of 200 nm, (c) Volta potential map with 0.1 V scale...... 105
3.7. 3-D surface height of IMCs in 3-minute chromate coated AA7075-T7...... 106
3.8. Surface height and line profile of Al-Fe-Cu IMC in 3-minute CCC on AA7075-T7...... 107
3.9. Surface height and line profile of Mg2Si IMC in 3-minute CCC on AA7075-T7.. 108
3.10. Surface height and line profile of Al2CuMg IMC in 3-minute CCC on AA7075-T7...... 109
3.11. AFM images of Mg2Si of as-polished AA7075-T6. (a) topographic images and (b) Volta potential map with line profiles...... 110
3.12. Surface height and line profile of Mg2Si IMC in 3-minute CCC on AA7075-T6.111
3.13. Maximum trench depth on Mg2Si and Al-Fe-Cu IMCs as a function of coating time...... 112
3.14. AFM image of Al-Fe-Cu IMCs in AA7475-T6 (a) before and (b) after 5-second CCCs at room temperature, 40×40 µm...... 113
3.15. High Magnification of 3.14(b), Al-Fe-Cu IMCs for 5-second coating in AA7475- T6...... 115
3.16. BSE image of 3-minute CCC AA7075-T6 after 24hr and 48 hr exposures to 0.5 M NaCl...... 116
xvi 3.17. Schematic of trenches formation at different intermetallic compounds for 5 sec coatings. (a) Al-Fe-Cu IMC, (b) Mg2Si, and (c) Al2CuMg. Symbol • indicates chromium oxide and symbol ♦stands for adsorbed ferricyanide. Coating thicknesses are exaggerated about 20x compared to IMCs...... 117
3.18. Schematic of CCC formation on different IMCs in 7xxx aluminum alloy. (a) CCC formation on IMCs at early stage. Absorbed ferricyanide inhibited chromate reduction reaction on Al2CuMg and Al-Fe-Cu IMCs. Al2CuMg and Mg2Si IMCs showed anodic dissolution of Mg around IMCs. (b) CCC formation on IMCs during commercial coating time (3-minute). The surface height difference between IMCs and matrix was not significantly changed after early coating. Coating thicknesses are exaggerated about 20x compared to IMCs for comparison...... 118
4.1. Schematic description of flash evaporation technique. Grounded cast bulk alloys are fed onto W boat and then evaporated from the W boat to Si wafer homogeneously...... 138
4.2. Schematic description of quartz crystal as working electrode, showing (a) front view and (b) side view...... 139
4.3. Schematic representation of typical EQCM system, showing electrochemical parts of potentiostat and mass change measurement such as oscillator and frequency counter...... 140
4.4. Polarization curves of MgZn2, Mg3Zn5Cu2, and MgZnCu at a scan rate of 0.2 mV/s in deaerated 0.1 M NaCl ...... 141
4.5. Raman spectra of 3-minute CCCs on thin film Mg(Zn,xCu)2, x=0, 17, 25 and 35 atom %...... 142
4.6. Raman spectra of 3-minute CCCs on thin film Mg(Zn,xCu)2, x=0, 25 and 35 atom % in Cr+F synthesized bath and 3-minute CCCs on thin film MgZn2 in Alodine 1200S bath. Cr and F represented CrO3 and NaF respectively...... 143
4.7. OCPs of S phase in aerated 0.5 M NaCl during 1 hr immersion...... 144
4.8. OCP and mass change of thin film analog of Al2CuMg in aerated 0.5 M NaCl. ... 145
4.9 OCP and EDS of thin film analogs of S phase. (a) Open circuit potential of S phase in aerated 0.5 M NaCl. (b) EDS spectra of S phase in different immersion time. The intensities are offset vertically by 2000...... 146
4.10. CCC formation on S phase during 3-minute in Alodine 1200S...... 147
4.11. CCC formation on thin film of pure Al during 3-minute in Alodine 1200S...... 148
xvii 4.12. Surface height and Volta potential of thin film analog of (a) before chromate coating and (b) after chromate coating on S phases...... 149
4.13. 3-minute chromate coated S phase immersion to aerated 0.5 M NaCl for 1 hr.... 150
4.14 EDS of thin film analogs of S phase. (a) and (b) are EDS spectra of S phase before and after CCC. (c) and (d) are EDS spectra of 3-minute CCC on S phase under OCP for 1 hr in 0.5 M NaCl. The intensities are offset vertically by 1800...... 151
5.1. Photograph of 5×5 electrodes array configuration. Electrodes spacing was 2.5 mm spacing within low-viscosity epoxy...... 173
5.2. Raman spectra of 3-minute CCC on AA7475-T7. The spectra were obtained three different positions of AA7475-T7. Three spectra are identical. The intensity of 860 cm-1 indicates the thickness in the coating. The units are arbitrary intensity to relative Raman shift...... 174
5.3. Raman spectra for 860 cm-1 intensity of CCC on AA7075-T6 at different coating times...... 175
5.4. Raman spectra showing 860 cm-1 intensity of 3-minute CCC on AA7475 in different tempers...... 176
5.5. The 860 cm-1 intensity of Raman spectra on CCCs for both AA7075 and AA7475 in T6 and T7 tempers...... 177
5.6. Scanning electron micrographs of the morphology of 3-minute CCCs on (a) AA7075-T7 and (b) AA7475-T7...... 178
5.7. XANES spectra for Cr metal, Cr(III)2O3 and Na2Cr(VI)O4 showing normalized absorbance against energy relative to Cr K edge (E0 = 5989 eV). The energy of absorption edge increased with the valence state of chromium, i.e. from Cr0, Cr3+ to Cr6+. The pre-edge peak of Cr6+ is at 5994 eV...... 179
5.8. XANES spectra for standard Cr2O3 and Na2CrO4 with mixture between 39% Na2CrO4 and 61% Cr2O3 showing the spectrum from a 3-minute CCC on AA7475- T7, and the spectra are offset vertically by 0.5...... 180
5.9. XANES spectra for 3-minute CCC on AA7475-T6 and -T7 showing absolute absorbance with with relative energy...... 181
5.10. Typical current vs. time plots for anodic behavior of CCC formation on AA7475- T6. The peak current sustained about 30 sec after immersion...... 182
5.11. Typical current vs. time plot for cathodic behavior of CCC formation on AA7475- T6. The peak current sustained about 30 sec after immersion...... 183
xviii 5.12. Typical current vs. time plot for cathodic and anodic mixed-behavior of CCC formation on AA7475-T6. The peak current sustained about 30 sec after immersion...... 184
5.13. Typical breakdown behavior for AA7075-T7 during anodic polarization with and without 3-minute CCC in 0.5 M NaCl solution...... 185
5.14. Cumulative probability with breakdown potential distribution for AA7075 and AA7475 in the T6 and T7 tempers in 0.5 M NaCl solution...... 186
5.15. BPDs for 7075-T6 and T7 with and without 3-minute CCC...... 187
5.16. BPDs for 7475-T6 and -T7 with and without 3-minute CCC...... 188
5.17. Effect of polarity on breakdown behavior for 3-minute CCC of AA7475-T7 as by anodic polarization determined in aerated 0.5 M NaCl solution...... 189
5.18. Current evolution of net anodic behavior for CCC formation on AA7475-T6 Cr+F, Cr+F+Fe, Alodine 1200S solutions. The currents are offseted vertically by 5×10-7 A between plots...... 190
5.19. Current evolution of net cathodic behavior for CCC formation on AA7475-T6 in Cr+F, Cr+F+Fe, Alodine 1200S solutions. The currents are offseted vertically by 5×10-7 A between plots...... 191
3- 5.20. Effect of Fe(CN)6 on (a) coating breakdown potential distribution and (b) 860 cm-1 intensity of AA7075-T7...... 192
5.21. Effect of ferricyanide, alloy temper and alloy purity on the 860 cm-1 intensity... 193
6.1. Schematic illustration of heat treatment on AA7075 and AA7475. W temper of AA7075 is 490 °C and W temper of AA7475 is 515 °C. WQ and AC indicate water quenching and air-cooling respectively...... 217
6.2. Physical model and equivalent circuit model of chromate conversion coating...... 218
6.3. Simplified equivalent circuit model of chromate conversion coating analysis. Roxide is an electrical open due to high resistance and Rp is a short due to low resistance. Ccoat represents the summation of Coxide and Cd. CPE stand for a constant phase element for Ccoat...... 219
6.4. Typical Nyquist and Bode plots for 3-minute CCC on AA7475-T7 with complex nonlinear least-square fitting in 0.5 M NaCl solution for 6 hours...... 220
6.5. (a) Nyquist and (b) Bode plots of impedance for 3-minute CCCs on AA7475-T7 of different immersion time in 0.5 M NaCl solution...... 221
xix 6.6. Nyquist plots of impedance for 1 and 3-minute CCCs on AA7475-T7 in different immersion time of 0.5 M NaCl solution...... 222
6.7. (a) Nyquist and (b) Bode plots of impedance for 3-minute CCC on AA7475-W, T6, T7, and RRA tempers in 0.5 M NaCl solution for 72 hours...... 223
6.8. (a) Nyquist and (b) Bode plots of impedance for 1-minute CCC on AA7475-W, T6, T7, and RRA tempers in 0.5 M NaCl solution for 68 hours...... 224
6.9. Corrosion resistance variation with tempers for (a) 3-minute CCC and (b) 1-minute CCCs of AA7475 in 0.5 M NaCl solution...... 225
6.10. (a) Nyquist and (b) Bode plots of impedance for 3-minute CCCs on AA7075-T6 of different immersion time in 0.5 M NaCl solution...... 226
6.11. (a) Nyquist and (b) Bode plots of impedance for 3-minute CCCs on AA7075-T7 of different immersion time in 0.5 M NaCl solution...... 227
6.12. Corrosion resistance variation of 3-minute CCCs on AA7075 with T6 and T7 tempers in 0.5 M NaCl solution...... 228
6.13. Corrosion resistance variation of 1-minute CCCs on AA7475 with all tempers in 0.5 M NaCl solution. Only 1 µm mechanical polishing without pretreatment...... 229
6.14. Corrosion resistance variation of 3-minute CCCs on AA7475 with all tempers in 0.5 M NaCl solution. Only 1 µm mechanical polishing without pretreatment...... 230
6.15. XANES spectra for AA7475-T7 with 3-minute CCC and 1-minute CCC...... 231
6.16. XANES spectra for AA7475 of 3-minute CCC showing (a) absolute absorbance (b) normalized absorbance with W, T6, T7, and RRA tempers...... 232
6.17. XANES spectra for AA7475-T7 and AA7475-RRA with (a) 3-minute CCC and (b) 1-minute CCC exposed to 0.5 M NaCl solution after 7 days...... 233
6.18. Rankings of the degree in salt spray test according to surface morphology...... 234
6.19. Rankings of the degree of corrosion at 651 hr visual inspection for chromate coated 7xxx with / without pretreatment. (A: alkaline pretreatment, B: acid pretreatment, C: chromate conversion coating)...... 235
6.20. Rankings of the degree of corrosion at 1310 hr visual inspection for chromate coated 7xxx with / without pretreatment. (A: alkaline pretreatment, B: acid pretreatment, C: chromate conversion coating)...... 236
6.21. Optical photographs of salt spray test for pretreatment of AA7075-T6 samples at 1310 hr exposure. (a) AA7075-T6 with (A+B+C), (b) AA7075-T6 with (A+C),
xx (c)AA7075-T6 with (C), and (d)AA7075-T6 with (B+C). A: Alkaline pretreatment, B: Acidic pretreatment, and C: Chromate conversion coating...... 237
6.22. Optical photographs of salt spray test for AA7075 at comparison 651 hr and 1310 hr exposure. (a) AA7075-T6 @651 hr, (b) AA7075-T7 @651 hr, (c)AA7075-T6 @1310 hr, and (d)AA7075-T7 @1310 hr...... 238
6.23. Optical photographs of salt spray test for AA7475 at 651 hr and 1310 hr exposure. (a) AA7475-T6 @651 hr, (b) AA7475-T7 @651 hr, (c) AA7475-T6 @1310 hr, and (d)AA7475-T7 @1310 hr...... 239
A.1. Three-dimensional representation of polished and etched grain morphology for AA7475-T7 alloy. L, T, and ST indicate longitudinal, transverse and short transverse directions respectively. The rolling plane consists of L and T directions...... 254
A.2. Scanning electron micrograph of AA7475-T7 showing the Al-Fe-Cu and Al2CuMg intermetallic compounds (white) and Mg2Si (black) in (a) SE and (b) BSE images...... 255
A.3. Anodic potentiodynamic polarization curves for all tempers of AA7475 in deaerated 0.1 M NaCl (⎯) and 0.5 M NaCl (---) at a scan rate of 0.2 mV/s. (a) W, solution heat-treatment, (b) T6, artificial peak-aging treatment (c) T7, over-aging treatment, and (d) RRA, retrogression and reaging temper...... 256
A.4. Open circuit potential (OCP) against time curves in deaerated 0.5 M NaCl for different heat treatments on rolling planes (●: W temper, ▼: T6 temper, ■: T7 temper, and ¡: RRA temper)...... 257
A.5. Characteristic potentials of AA7475-T7: Corrosion potential: Ecorr, Breakdown nd potential: Ebr, Repassivation potential: Erepass, and 2 zero current potential: Ecorr’ on reverse scan...... 258
A.6. Anodic polarization curves of AA7475 after 2hr OCP in deaerated 0.5 M NaCl for different heat treatments on rolling planes...... 259
A.7. Corrosion potential (Ecorr), breakdown potential (Ebr), repassivation potential nd (Erepass), and 2 zero current potential (Ecorr’ on reverse scan) in deaerated 0.5 M NaCl as a function of the heat-treatment for (a) longitudinal plane, (b) transverse plane, and (c) short transverse plane...... 260
xxi
CHAPTER 1
INTRODUCTION
Heat treatable aluminum alloys are used extensively as aircraft structural
materials due to their high strength and low density [1]. These alloys, however, are
susceptible to localized corrosion such as pitting corrosion, crevice corrosion and
intergranular corrosion (IGC), hydrogen embrittlement as well as stress corrosion cracking (SCC). Because localized corrosion and environmental fracture can lead to accelerated failures of structures and systems, the Air Force spends about one billion dollars per year respectively on corrosion remediation and prevention [2].
In order to improve corrosion resistance, aluminum alloys are usually treated with a protective coating. Chromate inhibitors are an important part of these coating systems because they are extremely effective in the corrosion protection of aluminum [3-5].
Chromate conversion coatings (CCCs) are an important element in these systems. They inhibit corrosion and serve to increase adhesion of subsequently applied paints [3-4]. The formation and growth of coatings on aluminum alloys is complicated by the heterogeneous microstructure inherent in high strength alloys.
1 In the case of 7xxx Al alloys, the primary alloying elements are Zn, Mg and
sometimes Cu. These alloying elements are used for age hardening and damage tolerance,
but they form a range of second phase particles that lower corrosion resistance [6-7].
Chromate conversion coating formation can be affected by these particles in unexpected
ways. Though the chromate provides excellent corrosion protection to aluminum alloys,
it is toxic and carcinogenic, and its use is being restricted by environmental regulations.
Extensive studies have been directed at characterizing CCCs and developing chromate-free coatings based on environmentally benign chemistries. The exact mechanism of corrosion inhibition by chromates has not been fully explained. No material has been found that can replace chromate with equivalent performance in the most demanding applications.
The objective of this study is to investigate the coating formation and breakdown on 7xxx aluminum alloys with a specific focus on the influence of artificial aging and alloy purity on CCC formation and breakdown. It is well known that the temper can affect the mechanical property of tensile strength and the corrosion property of stress corrosion cracking. However, the effect of temper on the performance of chromate conversion coatings has not been investigated rigously [8]. The effect of alloy purity in
7xxx alloys is uncharacterized. This study, therefore, aims to acquire a basic understanding of corrosion protection by chromates, to lead a better understanding of the coating formation and breakdown and to develop a successful replacement for chromate conversion coatings.
This dissertation consists of seven chapters with one appendix. Chapter 1 is the current chapter giving the brief description of background and objective of this study.
2 Chapter 2 reviews the metallurgy of the 7xxx series, which is associated with the
heterogeneous microstructures of aluminum alloys. The chapter examines various aspects
of localized corrosion to which aluminum alloys are susceptible, and address as chromate
inhibition of localized corrosion in terms of the role of chromate, coating process and formation of chromate conversion coatings. In the end of the chapter, the key unresolved issues and the objectives of this study are given.
Chapters 3 through 6 contain the details of the technical findings with individual introduction, experimental, results, discussion, and summary sections essentially as stand-alone papers. Each chapter is related to others as appropriate.
Chapter 3 describes that the formation of chromate conversion coatings (CCC) on aluminum alloys 7075 and 7475 was characterized using scanning probe microscopy approaches, scanning electron microscopy (SEM) and energy dispersive spectroscopy
(EDS). Overall, corrosion resistance is improved by application of CCCs, but CCC formation and breakdown was found to be strongly affected by the dispersion of coarse
0.5 to 20µm diameter Al2CuMg, Mg2Si, and Al-Fe-Cu-type intermetallic (IMC) particles.
Surface profiles of the coated alloy surface produced by tapping mode atomic force
microscopy showed that characteristic coating defect structures developed near IMC
particles. A combination of particle dissolution and inhibited coating formation resulted
in depressed sites on the coated surface over the IMCs. Depression depths were
characteristic of IMC chemical type and the coatings were of a different morphology and
chemistry than the coating over the matrix phase. Depressions over IMCs were evident
within the first 5 seconds of coating. Although the coating thickened in consequent
immersion time, the difference in coating thickness over particles and the surrounding
3 matrix remained essentially constant. Preferential dissolution at the peripheries of Mg2Si and Al-Fe-Cu-type particles occurred during coating resulting in trench-like defect.
Trenches were not observed at Al2CuMg particles because particles were quite small.
Nonetheless, Al2CuMg particles appeared to be initiation sites for corrosion when
conversion coated surfaces were exposed to chloride solutions. Volta potential difference
measurements made using the SPM showed that the electrochemical potential differences
between IMCs and the surrounding matrix were extinguished by CCCs suggesting that
one element of CCC corrosion protection is minimization of microgalvanic activity on
the alloy surface and suppression of cathodic corrosion near IMC particles. Coating
thickness differentials and Volta potential differences between particles and the matrix
were not dependent on alloy chemistry or temper, though the defect structure of CCCs on
7475 was affected by the reduced number of particles in the alloy.
Chapter 4 shows that thin film analogs of the intermetallic compound Al2CuMg and η phases were fabricated using the sputtering and flash evaporation techniques for use in electrochemical quartz crystal microbalance (EQCM) mass loss and open circuit potential (OCP) measurements. In aerated 0.5M NaCl solutions, thin films analogs of
Al2CuMg demonstrated active and passive behavior. Active dissolution rates of 0.08
µg/cm2⋅s were measured. Analogs were passivated by a Cu-rich surface film and mass
loss rates were essentially zero. Net mass loss rates, which included mass loss from
dissolution and mass gain from film formation, were detected during immersion in a
ferricyanide-accelerated chromate conversion coating bath. In contrast, pure Al thin films
exhibited mass gains under similar conditions. Dissolution of the Al2CuMg analog was
inhibited by conversion coating with measured rates of 0.001 µg/cm2⋅s. This result
4 suggests that although conversion coatings on Al2CuMg particles may be thinner than on
the surrounding matrix, corrosion resistance is increased. Energy dispersive spectroscopy
measurements showed that Cu enrichment occurred on conversion coated analogs during
exposure to aerated 0.5 M NaCl solution suggesting that CCCs cannot completely suppress Cu surface enrichment and deposition corrosion in Al2CuMg-forming Al alloys.
The open circuit potentials and breakdown potentials of η phase increased with increasing Cu content of η phase. It was found that the coating thicknesses of η phase
were decreased with increasing Cu content after conversion coating formation in Alodine
solution using Raman spectroscopy. However, higher Cu containing η phases had no Cr6+ intensity after coating formation in chromate bath with or without ferricyanide. This ferricyanide acted as an accelerator for the chromate conversion coating formation on η phase.
Chapter 5 examines that the chromate conversion coating formation and breakdown on arrays of 7x75 aluminum alloy electrodes was examined and characterized using a multi-channel microelectrode analyzer (MMA), Raman spectroscopy, and X-ray absorption near edge spectroscopy (XANES). MMA was used to provide the current during coating formation and the coating breakdown potential. Raman spectroscopy was used to characterize coating thickness, and XANES was used to characterize the Cr6+ content of coating. During chromate conversion coating formation on electrode arrays observed three distinct current transient behaviors showing coating formation is not uniform. The length of the transient was measured about 30 s until the net current decayed though coating thickness continuously grew up to 3-minute of immersion. The breakdown behavior of the chromate-coated alloy showed higher breakdown potentials 5 and gradual increased of current. Coating thickness had a more significant effect on breakdown potential than alloy temper and Cr6+ content. As coating time increased, 860
cm-1 intensity indicating coating thickness increased during chromate conversion coating
formation. The coating thickness is dependent on alloy temper and alloy purity though
Cr6+ content is not significantly dependent on alloy temper. The presence of ferricyanide
in coating bath considerably increased the coating thickness.
Chapter 6 discusses the effect of alloy temper on the chromate conversion coating
performance. CCC performance was also characterized by electrochemical impedance
spectroscopy (EIS), XANES, and salt spray testing (SST). The coating resistance of
AA7475-T7 was better than that of AA7475-T6 in 1-minute and 3-minute CCCs
with/without acid and alkaline pretreatment while the coating resistance of AA7075 was almost independent of alloy temper in 3-minute CCCs. 1310 hr salt spray test revealed that AA7475-T7 had less corrosion products and pits than AA7475-T6, whereas
AA7075-T6 had less corrosion products and pits than AA7075-T7.
Chapter 7 summarizes the findings of this work and makes conclusions on the formation and breakdown of chromate conversion coating and the effects of alloy temper and alloy purity on the coating performance in 7xxx aluminum alloys. The chapter suggests several unresolved issues that are appropriate for future work.
6
REFERENCES
1. R. W. Revie, "Uhlig's corrosion handbook,” New York: Wiley (2000).
2. R. G. Buchheit, Critical Factors for the Transition from Chromate to Chromate- Free Corrosion. 1999, Strategic Environmental Research and Development Program, Contract No. DACA72-99-C-0002.
3. K. A. Korinek, Chromate Conversion Coatings, in Metals Handbook, Vol.13, Corrosion, ASM: Metals Park, OH. p. 389 (1987).
4. H. G. Seiler, "Handbook on Toxicity of Inorganic Compounds,” New York: Marel Dekker (1988).
5. P. L. Hagans and C. M. Haas, Chromate Conversion Coatings, in ASM Handbook, Vol. 5, Surface Engineering, ASM International: Metals Park, OH. p.405 (1994)
6. R.P. Wei, M. Gao, and P.S. Pao, Scripta Met., 18, 1195 (1984).
7. R. P. Wei, C.M. Liao, and M. Gao, Met. Trans. A, 29A, 1153 (1998).
8. L. J. Bailin, P. Fitzpatrick, and M. J. Joyce, Evaluation of Unpainted Alodine Chromate Conversion Coatings for Corrosion Resistance and Electrical Conductivity. 1985, Lockheed Missiles and Space Co., Report F035575.
7
CHAPTER 2
LITERATURE REVIEW
There has been significant interest in the corrosion behavior of 7xxx high strength aluminum alloys, which are of importance for aircraft structural applications. These alloys provide high strength and stiffness. Because of their high alloying element content, they are prone to pitting, crevice corrosion, intergranular corrosion (IGC), exfoliation corrosion, hydrogen embrittlement, and stress corrosion cracking (SCC).
Chromate conversion coatings (CCCs) are generally an effective means of protection against the localized corrosion and other forms of corrosion. A large amount of research has been devoted to understanding corrosion protection by CCCs. Nonetheless certain details remain to be resolved.
The purpose of this literature review is to provide an overview of CCCs from the metallurgical and localized corrosion point of view. This chapter contains metallurgy of aluminum and 7xxx aluminum alloys, localized corrosion behavior, CCC formation and breakdown. Several chromate inhibition mechanisms and CCC structures are described.
8 2.1 7XXX ALUMINUM ALLOYS
Heat treatable 7xxx aluminum alloys are extensively used for aircraft structural
materials due to their high strength and low density [1].
2.1.1 Historical Development of 7xxx series
Al-Zn-Mg alloys were first introduced in 1901 [2]. In 1913, Eger reported the structure of the ternary Al-Zn-Mg alloys [3, 4]. However, Sander and Meissener examined the first real high strength obtained by heat treatment of the wrought Al-Zn-Mg alloys in 1926 [2-4]. Zinc and magnesium were the main strengthening additions, but also led to high susceptibility to stress corrosion cracking [5]. In 1932, Weber developed Al-
10%Zn-2%Mg-2%Cu-1%Mn, which exhibited improved resistance to stress corrosion cracking (SCC) due to additions of copper and manganese [2]. After considerable investment in research, Alcoa introduced AA7075 in 1943. This alloy contained satisfactory resistance to SCC and good mechanical properties from copper and chromium additions. AA7178 was a higher strength modification introduced in 1951.
AA7079 was a forging alloy introduced in 1954 and AA7001 was a very high strength alloy released in 1963 [2]. Cordier et al. suggested an optimum Zn/Mg ratio of 3:1 and a limit of 7% of both elements for a weldable Al-Zn-Mg alloy with good SCC resistance
[6].
Until the 1960’s, one-step aging, i.e., the T6 temper, which was a single treatment at temperature in the range 120~135 °C, was applied to the Al-Zn-Mg-Cu alloy system.
Although the T6 temper has maximum strength, it also has a relatively low resistance to
SCC. For this reason, multi-step aging processes such as the T7x tempers have been
9 developed for improving the stress corrosion cracking resistance for alloys with more
than 1% copper. In these tempers, there is a trade-off in strength. Since that time, T7
tempers have been developed for later generation 7xxx alloys such as AA7475, AA7079, and AA7050 [7, 8]. More recently, retrogression and re-aging tempers (RRAs) have been developed, which enable alloys to exhibit the mechanical properties expected of the T6 condition combined with SCC resistance equal to the T7 condition [9]. Figure 2.1 shows the strength, corrosion resistance, and fracture toughness among commonly used alloys in
the 7xxx series [10].
Aluminum alloys are classified and standardized for annotation by International
Alloy Designation System (IADS) in Table 2.1 [11]. In aluminum alloy identification, the
first digit indicates the main alloying elements, the second digit modifies alloy or
impurity limits (e.g., AA7475 has higher purity than AA7175 and AA7075), and the last
two digits identify the aluminum alloy or indicate the aluminum purity (e.g., AA1145 has
a minimum purity of 99.45% and AA1200 has a minimum purity of 99.00%). Temper
designations involving more than one digit are assigned to indicate stress relieving
tempers of wrought products. Tx51 indicates stress relief by stretching after quenching,
Tx52 by compressing to 2.5% strain, and Tx53 by thermal treatment (x refers to
appropriate second digit such as 3, 4, 6, or 8). Compositions of representative 7xxx series
alloys are shown in Table 2.2 [11].
10 2.1.2 Physical Metallurgy of Pure Aluminum and Aluminum Oxide
Since aluminum is the main element of the matrix in 7xxx series alloys, it is very
important to know the characteristics of pure aluminum for understanding of 7xxx series
alloy systems better.
Pure Aluminum and Aluminum Oxide: The element aluminum belongs to group III of the periodic system and is relatively reactive. It has a very low standard
potential of –1.66VSHE. The density of pure aluminum (99.65~99.99%) is of the order of
2.7 g/cm3, one-third that of steel [1].
Because aluminum is an active metal, its resistance to corrosion depends on the
passivity produced by a protective oxide film. For this reason, aluminum and its alloys
are relatively stable in most environments due to the rapid formation of a natural oxide
film on the surface that kinetically inhibits dissolution [11]. In aqueous solutions, the
potential-pH diagram in Figure 2.2 expresses the thermodynamic conditions under which
the film develops [12]. As this diagram shows, aluminum is passive in the pH range of 4 to 9. The protective oxide film formed in water and atmosphere at ambient temperature is only few nanometers thick and amorphous as shown in Figure 2.3 and Figure 2.4 [1, 13].
The oxide films in aqueous solution consist of a thin amorphous Al2O3 barrier layer and a
thicker aluminum hydroxide layer. The thin amorphous Al2O3 barrier layer is virtually
free from pores. The aluminum hydroxide, which is gel-like and not stable, crystallizes
with time to convert into the rhombohedral monohydrate (Al2O3⋅H2O or boehmite), then
into the monoclinic trihydrate (Al2O3⋅3H2O or bayerite), and finally into another
monoclinic trihydrate (Al2O3⋅3H2O or hydrargillite) under aqueous solution [1].
11
Corrosion Behavior of Pure Aluminum and Intermetallic Compounds:
Aluminum is passive between pH ranges of 4 to 9. Usually, the oxide film is readily
3+ - soluble in strong acids and alkalis to yield Al ions in the acidic conditions and AlO2
(aluminate) ions in the alkaline conditions as shown in Figure 2.2. However, there are a few exceptions, either where the oxide film is not soluble in specific acidic solution (such as acetic acid and nitric acid) or alkaline solution (such as sodium disilicate and ammonium hydroxide) [1, 8]. In addition, the Pourbaix potential-pH diagram is based on only theoretical thermodynamic considerations and does not provide information on corrosion rate but predicts oxide film stability and, thus, resistance to general dissolution
[1].
Alloying elements such as Mg and Li, which are more active than aluminum, oxidize first, forming poorly protecting oxides at the outer layer as formation of Mg(OH)2 and Li2CO3. On the other hand, alloying elements nobler than aluminum, present in solid solution or in the form of small coherent precipitates, produce a oxide film. In contrast, the largest precipitates, such as Al3Fe in Figure 2.4, are not coherent with the matrix [1].
Resistance to corrosion can be improved considerably as purity is increased, but
the oxide film on even the purest aluminum contains a few defects where minute
corrosion can develop. The presence of impurity-type second phase particles in aluminum
alloys are produced primarily from iron and silicon. Intentionally formed precipitates of
compounds are produced from soluble alloying elements. Most of these phases, such as
Al2Cu, Al3Fe, Al20Cu2(Fe,Mn)3 and AlxCuxMg, are cathodic to aluminum, but a few such
as Mg2Al3, MgZn2, Al2CuMg and AlxZnxMg are anodic [1, 14, 15]. Buchheit summarized
12 the corrosion potentials for IMCs in aluminum alloys [16]. In either case, these can lead to galvanic cells because of the potential difference between these second phases and the aluminum matrix. Specifically, some of the Cu- and Fe-containing IMCs are recognized
cathodic sites at which oxygen reduction is supported and drive dissolution of the matrix
(Figure 2.5) [17]. Seegmiller and Buttry observed the high redox reactivity on Cu- containing IMCs using scanning electrochemical microscopy (SECM) as shown in Figure
2.6 [17].
2.1.3 Physical Metallurgy of 7xxx series Aluminum Alloys
Wrought alloys are divided into seven major classes according to their principal alloying elements [8, 10, 11]. Alloy classes can also be divided into two types according to whether they are strengthened by work hardening or precipitation hardening. Non-heat treatable alloys are 1xxx (pure Al), 3xxx (Al-Mn and Al-Mn-Mg), 4xxx (Al-Si), and
5xxx (Al-Mg) series. Precipitation hardening alloys include 2xxx (Al-Cu and Al-Cu-Mg),
6xxx (Al-Mg-Si), and 7xxx (Al-Zn-Mg and Al-Zn-Mg-Cu) series [8, 10, 11].
The 7xxx series can be classified into three subgroups: those that have medium- strength and are readily weldable (Al-Zn-Mg), the high-strength weldable Al-Zn-Mg alloys, and the high-strength Al-Zn-Mg-Cu alloys that have been developed primarily for aircraft construction. Most of the latter have very limited weldability. Compositions of representative weldable Al-Zn-Mg alloys and high strength Al-Zn-Mg-Cu alloys are shown in Table 2.3 [11]. It is generally accepted that the Zn + Mg content should be less than 6% in order for a welded alloy to have a satisfactory resistance to cracking [11].
High Zn:Mg ratios produce the best strength and response to heat treatment, but also have
13 the highest susceptibility to stress corrosion. Alloys with low Zn:Mg ratios produce the
best weldability and the lowest quench sensitivity [5]. Gruhl has proposed that the Zn:Mg
ratio between 2.7 to 2.9 lead to the best SCC resistance [11, 18].
Precipitation Hardening: In heat-treatable alloys, strengthening is produced by
precipitation hardening. The mechanism of precipitation hardening is based on particles
impeding the movement of dislocations dispersed in a ductile matrix. Depending on the
size of a particle, the dislocation can either cut through it [19] or bend round it, leaving behind a dislocation loop (Orowan mechanism)[20]. In addition to the size of the
particles, the nature of the phase boundary with the matrix is also important for the
hardening [21]. Coherency is an atom-to-atom registry between the lattice planes in the
particles and those in the matrix; the particle is said to be coherent. Similarly, with partially coherent particles, the atomic planes in the particles are partially continuous with those in the matrix, but with incoherent particles, there is no such continuity. This leads to the lattice being subjected to different degrees of strain. Obstacles to the motion of dislocations in age-hardened alloys are the internal strains around precipitates, notably
GP (Guinier-Preston) zones, and the actual precipitates. This strain is greatest when the phase boundary is coherent because of a mismatch in size [21]. The moving dislocations
encounter a strong stress around coherent precipitates, which leads to a marked increase
in strength. Consequently, the presence of the precipitate particles and the strain fields in
the matrix surrounding the coherent particles provide higher strength by obstructing and
retarding the movement of dislocations. The characteristic that determines whether a
precipitate phase is coherent or incoherent is the closeness of match or degree of
14 disregistry between atomic spacings on the lattice of the matrix and on that of the
precipitate.
Age hardening comprises three stages: Solution heat treatment, quenching, and
aging [3, 8, 11]. The purpose of solution heat treatment is to take the maximum practical amounts of the solute hardening elements in the alloy into homogenous solid solution; that of quenching is to produce a supersaturated solution at room temperature; that of aging is to precipitate solute atoms out of solid solution at room temperature (natural aging) or elevated temperature (peak aging: T6 temper or over-aging: T7 temper) [3, 8,
11, 13, 21].
The aging of Al-Zn-Mg-(Cu) alloys from room temperature to a single aging treatment about 120 °C is accompanied by the generation of GP zones having an approximately spherical shape. This is the T6 temper in which the highest strengths can be developed. A very small transition precipitate η’, which is partially coherent on
(111)Al planes, can also be detected in the T6 temper [21]. A duplex aging treatment has a
higher temperature of 160~170 °C by a followed low temperature. It is designated by the
T7 temper in which a dispersion of the η’ or η precipitate is obtained through nucleation
on pre-existing GP zones. In order words, GP zones which form at low temperature are
stable at higher temperature and then transform to the intermediate η’ precipitates and
finally to the equilibrium η (MgZn2) phase during over-aging [21]. T73, T74 and T76 tempers are variants of T7 tempers applicable to alloys that contain more than 1.25% copper. The T73 temper results in greatly minimized SCC susceptibility and the T76 increases the resistance to exfoliation corrosion. The T74 temper is an intermediate condition between T73 and T76 tempers. Hunsicker showed that tensile properties of 15 7075-T73 are about 15% below those for the T6 temper but alloy 7075 aged to the T73
temper remained uncracked at 300 MPa, whereas the same alloy failed at 50 MPa in the
T6 condition [22].
Retrogression and reaging (RRA) heat treatments were originally developed by
Cina to improve the combination of strength and resistance to stress-corrosion cracking
of the aluminum alloy 7075 [9]. The method consists of a three step-aging treatment, i.e.,
T6 temper treatment, retrogression, and reaging. Retrogression is a very short treatment
at a higher temperature (200~240 °C) and reaging is similar to a T6 treatment. The evolution of strength occurs during retrogression. The strength first decreases then
steeply increases with temperature after exhibiting a minimum. The change of strength is
strongly affected by aging time and temperature. Longer treatments produce a gradual
reduction in the strength. Park and Ardell suggested that the initial decrease in the
strength should be caused by the dissolution of the precipitate phases formed during the
T6 treatment. However, the overall RRA process is complex, and it seems probable that
the relative importance of each step of heat treatment and heterogeneous microstructures
will be involved [23-25].
Al-Zn-Mg Alloys: Aging of Al-Zn-Mg alloys starts in most cases with a natural
aging after solution treatment and quenching. The formation of GP zones takes place by
the decomposition of the supersaturated solid solution near room temperature [3, 5, 26-
34]. The intermediate phase forms at somewhat higher aging temperatures while the
formation of the stable (or equilibrium) phases can be observed at still higher
temperatures. The structure and composition of the metastable and stable precipitates is
16 strongly dependent on the alloy composition and on the aging temperature, and the
morphology and crystallography in the Al-Zn-Mg system has been the subject of much
research. Generally, in the age-hardenable Al-Zn-Mg alloys the decomposition of the
supersaturated solid solution (SSSS) takes place in the following sequence [3, 5, 26-34]:
SSSS → GP zone → intermediate phase → final equilibrium precipitate phase.
Equilibrium Precipitate Phase: The phase diagram of the Al-Zn-Mg system is shown in Figure 2.7 [3]. Two types of equilibrium precipitate phase can be seen: the binary MgZn2 phase (η) and the ternary Mg3Zn3Al2 phase (T).
The MgZn2 phase is hexagonal with 12 atoms in the unit cell and space group
P63/mmc. The lattice parameters of the MgZn2 were found with a range of a = 0.516 to
0.522 nm, c = 0.849 to 0.855 nm [35]. The equilibrium precipitates of the MgZn2 phase
have incoherent interphase boundaries with the aluminum matrix. A coherent precipitate
particle forming a preferred crystallographic plane and direction in the lattice of the
particle is parallel with specific preferred planes and directions in the matrix lattice.
There are nine characteristic orientation relations between the precipitate MgZn2 lattice
and the matrix lattice. Most common are: (10.0)η//(001)Al; (00.1)η//(110)Al and
(00.1)η//(111)Al; (10.0)η//(110)Al [31].
The composition of the ternary T phase varies between relatively wide limits: from 20 to 35% Mg and from 22 to 65% Zn. The formula Mg32(AlZn)49, or Mg3Zn3Al2, is used to characterize its composition [36]. The T phase is cubic with 162 atoms in the unit cell and space group Im3. The particles of the T phase are incoherent with the matrix 17 and the orientation relationships are: (110)T//(112)Al; (001)T//(1 1 0)Al [26]. From the equilibrium diagram, the formation range of the T phase becomes narrower than that of the η phase with decreased temperatures. Schmalzried and Gerold observed that the T phase is only precipitated at temperatures higher than 200 °C [26]. Additionally, in the alloys with Mg/Zn ratios between about 1:2 to 1:3, the η phase is the only stable precipitate at lower temperatures, while the T phase is the only stable phase at the higher temperatures.
Intermediate Precipitate Phases: The sequence of steps in the decomposition process strongly depends on the composition of the alloy, on the quenching conditions, and on the aging temperatures. In particular, the decomposition process is highly dependent on the excess vacancy concentration and is thereby sensitive to the quenching condition [37].
The supersaturated solid solution is obtained by a solution treatment followed by quenching to a lower temperature, then the decomposition commences with the formation of GP zones. Graf found that GP zones consist of layers parallel to the (100) matrix planes enriched and impoverished in zinc alternately [38, 39]. Schmalzried and Gerold also observed that during GP zone formation the (100) planes became enriched in zinc or in magnesium alternately, and hence the lattice spacing of the matrix remained unchanged [26].
Mondolfo et al. suggested the existence of a metastable form of MgZn2 that followed initial GP zones in an aged Al-Zn-Mg alloy [27]. Since then, the nature of this transition phase has been the subject of controversy. Graf observed that the characteristic of GP zones changes at about 80 to 100 °C. In other words, the (100) reflection 18 disappears gradually while the (200) reflection becomes more and more pronounced. The structure of the precipitates becomes similar to that of the η phase but it is not same. The author determined the structure of η’ transition phase, which is hexagonal with a = 0.496 nm, c = 1.403 nm [39]. Mondolfo et al. determined the lattice parameters of the η’ precipitates to be a = 0.496 nm, c = 0.868 nm [27]. Thomas and Nutting confirmed the values of Mondolfo et al., whereas Auld and Cousland confirmed the value of Graf [28,
40]. Gjonnes and Simensen interpreted η’ in terms of a monoclinic structure with a = b =
0.497 nm, c = 0.554 nm and γ = 120 °. In general, the composition of the platelike η’ particles MgZn2 is accepted, but according to Auld and Cousland it is better described by the formula Mg4Zn11Al [40]. They found orientation relationships such as
(00.1)η’//(111)Al; (10.0)η’//(110)Al. According to Mondolfo et al. the formation of the η’ transition phase starts by the segregation of alloying elements on stacking faults [27]. As a consequence of the differences in atomic sizes, coherency is gradually lost and the ordered structure of the η phase develops. This would mean that there is a continuous transition from the supersaturated solid solution to the precipitation of the η phase.
Some authors have proposed that the intermediate η’ phase could be formed directly from stable clusters, which are probably GP zones, but not necessarily [41].
Thackery suggested that GP zones could transform directly to the equilibrium η phase in an aged Al-6%Zn-2%Mg [42]. Ferragut et al. found that at the early stage of artificial aging at 150 °C (which is above the temperature to GP zone formation), the solute clusters and/or GP zones dissolve and simultaneously η’ precipitates are formed [43].
19 Al-Zn-Mg-Cu Alloys: For the commercial 7xxx series, less work has been done on the characterization of the microstructure than 2xxx series aluminum alloys. Adler et al. [44-46] found that GP zones are the predominant precipitates in the T651 temper and
η’ phases are dominant in T73 temper of AA7075 using hot stage TEM with differential scanning calorimetry (DSC). Park and Ardell [24, 47, 48] suggested a bimodal precipitate distribution in AA7075-T7 i.e., plate-shaped and coarse η phases and a finer η phase.
The microstructure of T651 temper contains predominantly the η’ transition phase, which is plate-shaped, with smaller amounts of the η phase. The fine coherent dispersion of small η’ transition particle causes AA7075 the maximum hardness. The results of Park and Ardell are different from those of Adler [44-46] and the other researchers [49-52] who concluded that GP zone was the main constituent of the T6 temper microstructure.
More recently, Tsai and Chuang quantitatively investigated the microstructure of
AA7x75 as shown in Table 2.4 [53]. They found that the the microstructure of T7 temper had more coarse precipitate particles at grain boundary and matrix that that of T6 temper.
Microstructural Features in Alloys: Mechanical behavior can corrosion properties can be controlled by intermetallic compounds. The coarse intermetallic compounds (constituents) form interdentritically during solidification and usually contain iron and silicon impurities [8, 11]. They include the virtually insoluble compounds
Al6(Fe,Mn), Al3Fe, α-Al(Fe,Mn,Si), Al2Cr2Fe and the more soluble phases Al2Cu, Mg2Si and Al2CuMg. Smaller submicron particles (dispersoids) are typically 0.05-0.5 µm and intermetallic compounds that usually contain the transition metals chromium, manganese or zirconium such as Al20Cu2Mn3, Al18Mg3Cr2 and Al3Zr. The dispersoids are 20 precipitated during subsequent pre-heating and serve to retard recrystallization and grain growth during processing and heat treatment. Fine precipitates (up to 0.1 µm), e.g.,
MgZn2, Al2Cu (θ’), and GP zones, that form during age hardening promote a high level of strengthening in alloys. Porosity and inclusions can also influence the behavior of aluminum alloys [8, 11].
2.2 LOCALIZED CORROSION
Most high strength and heat treatable aluminum alloys are susceptible to localized corrosion, which includes pitting, crevice, intergranular corrosion, and stress corrosion cracking.
2.2.1 Pitting Corrosion
Pitting and crevice corrosion can cause rapid local penetration. Aluminum and aluminum alloys depend on passive films for corrosion resistance. Such alloys are susceptible to localized corrosion, when the breakdown of the passive film occurs.
Aluminum goes into solution according to the following anodic reaction. The reactions at the anode decrease pH because of the consumption of hydroxide ions. The chloride ions, which are very small, mobile and frequently present in service, migrate into the pit to maintain charge neutrality and sometimes form aluminum chlorides. Aluminum hydoroxide forms by hydrolysis [54, 55].
Al → Al3+ + 3e- Eqn. 2-1
3+ + Al + 3H2O → Al(OH)3 + 3H Eqn. 2-2
21 The electrons released migrate to the local cathode. Hydrogen evolution (Eqn.
2-3) and oxygen reduction reaction (Eqn. 2-4) occur at cathodic sites. The corrosion can be prevented by removal of the reducible species required for a cathodic reaction such as oxygen.
+ - 2H + 2e → H2 Eqn. 2-5
- - O2 + 2H2O + 4e → 4OH Eqn. 2-6
There are two characteristic potentials for pitting corrosion [56, 57]. The pitting potential or breakdown potential is related to the initiation of stable pits above which the current density increases quickly by several magnitudes from the passive current density.
Wall and Martinez studied the pit initiation behavior on pure Al using a statistical approach. It was shown that the average pitting potential is proportional to log(1/area) and log(scan rate) [56]. The second one is the repassivation potential below which pits repassivate and stop growing. Alloys having higher pitting and repassivation potentials are generally recognized as having higher resistance to localized corrosion. However, pitting corrosion is also strongly dependent on alloy composition, microstructure and cathodic reactivity. For example, Alloy 2024 (Al-Cu alloy) has a higher pitting and repassivation potential than pure aluminum in chloride solutions, but the former accumulates corrosion damage more rapidly than the latter because of the microstructural heterogeneity; mainly intermetallic particles. Pitting corrosion occurs preferentially at these intermetallic particles in the aluminum alloys [57].
22 Pit Initiation: Generally, there are three stages for pitting corrosion, that is, pitting initiation, metastable pitting, and stable pit growth [58]. A number of pit initiation mechanisms have been proposed and including as the adsorption model, the film breakdown model, and the anion penetration and ion migration model [58].
The adsorption model describes the competitive adsorption of chloride ions and oxygen for the formation of pits at sites where oxygen adsorbed on the local surface site was displaced by chloride ions [59-62]. This model addresses the importance of the adsorbed ion for film thinning and breakdown for pitting initiation. However, it is generally considered that adsorption of the aggressive ions is the first step for pitting. The film breakdown model considers the film pressure for pit initiation [63]. This model addresses the mechanically stressed oxide film damaged by pores and flaws. The limitation of this model is that it cannot explain the need for specific anions for the pit initiation. The anion penetration and ion migration model not only address the penetration of aggressive anions as an important step for pit initiation, but also explain the migration of adsorbed aggressive anions through the passive film for the film breakdown when these anions reach bare metal [64]. This penetration mechanism was supported by the existence of an induction time for pitting and incorporation of chloride in the passive film [65, 66]. However, this model did not explain either the difference of induction time between the observed value and the estimated one or the difficulty of
- 2- 4- diffusing large anions such as NO3 , SO4 , and ClO for pitting [67]. In addition, intermetallic compounds (IMCs) in alloys could be nucleation sites of pitting. Wei et al. observed that the pitting initiated around Fe-rich IMCs due to galvanic interaction between IMCs and the matrix [170, 171].
23 Metastable Pitting: Metastable pitting is recognized as an important phenomenon for better understanding of pitting corrosion, especially the pitting initiation and pit growth, as an intermediate state [68]. Metastable pits are formed at potentials lower than the pitting potential and/or before stable pits appear. Metastable pit lifetimes are typically on the order of seconds or less [58]. These metastable pits experience initiation, growth, and repassivation during that short time.
Metastable pits have been investigated by several electrochemical methods such as potentiostatic method (anodic current transient under an applied anodic potential) and galvanostatic method (potential transient under open circuit or applied anodic current).
Stainless steel has been examined by the potentiostatic method between the corrosion potential (100 mVSCE) and the pitting potential (380 mVSCE) of stainless steel [69]. Small currents transients were observed on of the stainless steel in 0.5 M H2SO4 and 0.5 M
NaCl have been attributed to metastable pitting. Frankel et al. systematically investigated metastable pitting of stainless steel by using the potentiostatic method and used the terminology, “metastable pitting” for the first time [68]. Figure 2.8 is the typical metastable pit transient on stainless steel in chloride solution. The authors suggested that the average metastable pit current density increased with increasing applied potential and the metastable pit growth was under either ohmic control or ohmic and charge transfer control as a result of the resistance associated with the porous pit cover of passive film
[68]. They pointed out that stable pits survive the metastable stage and continue to grow, whereas metastable pits repassivate and stop growing because of the rupture of passive film cover [68]. The reason for the rupture of the passive film cover could be related to
24 hydrogen evolution, which causes the passive film to distort and leads to blister formation in aluminum alloy 7075 [70].
Pit Growth: Pit growth can lead to to crack initiation, which eventually leads to accelerated failure of structure. Pit growth is dependent on material composition, pit electrolyte concentration, and pit bottom potential, which can be limited by any of the following factors: charge transfer control, mass transport control, and ohmic drop control.
Each of these factors or some combinations of them can be a rate-determining step for pit growth [58]. For a simple analysis for hemispherical pit growth, different rate controlling factors would lead to specific relationships between current I, current density i, pit radius r, time t, and potential E [58].
Charge transfer control is related to the electrochemical dissolution of metals, which is suggested for the early stage of pit growth [71, 72]. Charge transfer control is based on the free energy changes that are associated with the electrochemical reaction.
Tafel behavior (i ∝ exp E) is characteristic of charge transfer control.
Under ohmic potential drop (IR drop), the rate-controlling step is determined by
Ohm’s law, i=E/R, where the ohmic resistance, R, can come from electrolyte conductivity and pit cover resistance. Under a fixed applied potential, the pit current, I, should be proportional to the pit depth, I ∝ r, but i ∝ I/r2 ∝ 1/r [58]. From Faraday’s law, i ∝ dr/dt. Therefore, r ∝ t1/2, thus i ∝ t-1/2 [58]. Potential is linearly related to current density by Ohm’s law (i ∝ E). Figure 2.9 shows that ohmic potential drop controls the pit growth depth and pit current density of Al foil in 0.01 M NaCl at pH 11 [73]. As can be seen, the pit growth is dependent on the applied potential as well as time [73]. 25 Under mass transport control, the controlling rate would be the diffusion of the enriched dissolved metal cations inside the pit to the bulk electrolyte according to Fick’s law, i ∝ 1/r and also i ∝ t-1/2. Potential will have no effect on the pit growth rate, i.e., i is independent of E [58].
The i-E relationship, however, cannot be identified by the non-steady state nature of pit growth and inaccurate measurement of the pit current density [58]. In measuring the i-E relationship under the potentiostatic method, the current comes from several pits with unknown active surface areas and different initiation and different growth rates.
Therefore, the study of a single pit is important [58].
Pit stability has been studied for decades. Local concentration of ions [74, 75], pit size [69], geometry [76], local pH [77-79], salt film precipitation [80-84], and local potential [85] can affect the stability of pit growth. The local environment of a pit is very critical for pit initiation, growth and repassivation, but the effect of these factors is still the subject of much study.
2.2.2 Intergranular Corrosion
Intergranular corrosion (IGC) in aluminum alloys is the selective attack of grain boundaries and/or adjacent solute depleted zones without appreciable attack of the grain body or matrix [14]. Most mechanisms proposed to explain the phenomenon of IGC are based on the development of a preferential anodic path along grain boundaries resulting from localized precipitation [1, 14, 86, 87]. The precipitates have a different corrosion potential and dissolution kinetics than the adjacent depleted solid solution. Aged Al-Cu alloy is described as a system having anodic and cathodic zones. The copper-depleted
26 zone along the grain boundary is the anode, while the grain bodies and the intermetallic
Al2Cu and AlxCuxMg are cathodes. In other alloys, such as Al-Mg and Al-Zn-Mg-Cu alloys, the precipitates Mg2Al3, MgZn2 and AlxZnxMg are more anodic than the adjacent solid solution. In either case, selective attack of the grain boundary region occurs [1].
IGC in Al-Cu alloys: Galvele and De Micheli proposed that IGC resulted not from differences in corrosion potentials but from differences in breakdown potentials
[88]. They correlated the susceptibility to IGC with pitting corrosion in Al-4%Cu alloy, which had two breakdown potentials in the peak-aged condition. They explained that the precipitate free zone (PFZ) on the grain boundaries has a lower breakdown potential than grain matrix and grain boundary precipitates such as noble Al2Cu. When potential is higher than PFZ breakdown potential but lower than the matrix breakdown potential, localized attack on the PFZ results in IGC. In the over-aged condition, Al-Cu alloy had just one breakdown potential and more resistance to IGC because the PFZ disappears [67,
88].
IGC in Al-Mg-Zn-Cu alloys: Maitra and English tried to explain the susceptibility to IGC in Al-Mg-Zn-Cu alloy [89, 90]. They found that aluminum alloy
7075 plate in T6 temper (peak-aged condition) had two breakdown potentials. Between the active and noble breakdown potentials, the alloy was susceptible to IGC, and above the noble breakdown potential there was pitting corrosion in matrix [89]. They suggested that the active breakdown potential was related to the breakdown of the anodic zone in the vicinity of grain boundaries where Zn and Mg were segregated [89]. In T7 temper
27 (over-aged condition), the alloy had only one breakdown potential which was associated with pitting in the matrix. They thought that a single breakdown potential in the T7 temper was associated with the incorporation of Cu into the grain boundary precipitates in spite of the segregation of Mg and Zn to the grain boundary [89].
Ramgopal et al. recently investigated the electrochemical behavior of PFZ and
MgZn2 precipitates on the grain boundaries in AA7150 [91]. They made thin film analogs of the PFZ by using the flash evaporation method, which had a similar electrochemical behavior as the Al matrix. However, there was a significant difference in the composition of grain boundary precipitates in T6 and T7 tempers. In T7 temper, high Cu enriched grain boundary precipitates such as Mg(Zn,Cu)2 or Mg(Zn,Cu,Al)2 due to Cu incorporation into MgZn2 raised breakdown potential to more noble potential and reduced the difference of breakdown potential, which caused the IGC and SCC resistance in AA7150 to improve [91].
Meng et al. also suggested that the breakdown potential of 7xxx depended on the
Cu content in the alloy and the first breakdown potential was associated with transient dissolution of fine hardening particles. They suggested that the first breakdown potential was associated with the Cu content in these precipitate hardening particles and the second breakdown potential was associated with IGC and selective grain attack [172].
Microchemistry of Al-Zn-Mg-Cu alloys: Doig et al. have shown the effects on the electrochemical properties of the grain boundary regions where the concentration of solute atoms was higher than in the grain interior [92, 93]. They found that Mg solute segregated at the grain boundaries in quenched alloys and Mg depleted in PFZ due to the
28 formation of MgZn2 precipitates on the grain boundaries in over-aged alloys [92-96].
Park and Ardell investigated the solute composition of PFZ in the AA7075 of T6, T7, and
RRA tempers [25]. They found that Zn depleted in PFZ at grain boundaries in three tempers of AA7075, and Zn depletion was relatively insensitive to the tempers. The depletion of Cu was more sensitive to the aging, being larger in both the T7 and RRA tempers than the T6 temper. They suggested that the reduction of the Cu concentration led to an increase in the SCC resistance of AA7075 in the T7 and RRA tempers [25].
Exfoliation corrosion is a type of attack in which the grain boundary regions of the alloy are preferentially dissolved, leading to the precipitation of corrosion products, which act to wedge open the boundaries and thereby allowing further solution ingress and corrosion to take place. Wrought and high-strength aluminum alloys, which contain alloying elements such as Mg and Zn, are particularly susceptible to this form of corrosion [97].
2.2.3 Stress Corrosion Cracking
Stress corrosion cracking (SCC) of aluminum alloys is the intergranular cracking in an aqueous, mostly chloride-containing, corrosive medium under the influence of tensile mechanical stresses, which can either be internal stresses or external-applied stresses [13]. SCC has accounted for many failures of 7xxx alloys, hence particular attention has been given to the development of modified compositions and temper practices such, as T73 and RRA tempers to minimize this problem.
Studies of stress corrosion attack in Al-Mg-Zn type alloys have focused on three principal microstructural features: the precipitate-free-zone (PFZ) which forms adjacent
29 to high angle grain boundaries [28, 98-100], the matrix precipitate structure [101-103], and the grain boundary precipitate structure [45, 92, 93, 104, 105]. Nutting and Speidel suggested that there is a preferential plastic flow across the PFZ which has a lower yield strength than the matrix [28, 98]. However, Hall suggested that the matrix precipitates have an important role in determining the susceptibility to SCC and found that the matrix microstructure changes to more homogeneous deformation from the peak-aged to over- aged conditions [103]. Maitra and Galvele suggested that the susceptibility to the intergranular and stress corrosion is ascribed to the more anodic breakdown potential of the region near the grain boundaries when compared with the breakdown potential of the matrix. They proposed that the decrease in the SCC susceptibility is due to depletion of
Cu in the matrix solid solution which causes the breakdown potential of the matrix solid solution to approach the breakdown potential of the grain boundaries [89, 90]. Doig et al. suggested that the initiation of SCC was not related to the precipitate free zone width but rather to the width of the solute depletion associated with the grain boundary precipitation [92, 93].
The cause of SCC in Al-Mg-Zn type alloys, which can be attributed to atomic hydrogen, has also studied. The atomic hydrogen from a corrosion reaction, which diffuses to the grain boundaries under the influence of tensile stress, causes embrittlement
[1]. Kim et al. suggested that Mg segregation to the grain boundary plays a critical role in the hydrogen embrittlement of the SCC in Al-Zn-Mg-Cu alloy [106]. The relative importance of these features is still controversial and may depend upon the particular combination of alloy and environment.
30 From the electrochemical point of view, susceptibility to IGC is closely related to
SCC, and heat treatment of aluminum alloy to improve SCC resistance also improves
IGC resistance [1, 14]
2.3 CHROMATE CONVERSION COATINGS
Chromate conversion coatings (CCCs) are extremely effective and widely used to enhance the localized corrosion resistance of high strength aluminum alloys, to improve the adhesion of paint, and to provide the metallic surface with a decorative finish [107-
109].
CCCs are formed on metal surfaces as a result of a chemical and/or an electrochemical process in an aqueous solution of chromic acid, chromium salts such as sodium or potassium chromate or dichromate which contain hexavalent chromium (Cr6+) as a coating formation agent, hydrofluoric acid or hydrofluoric acid salts as an activator, and ferricyanides as an accelerator [109, 110]. The process results in the formation of an amorphous protective coating composed of the substrate, complex chromium compounds, and other components of the processing bath. Recently, there has been considerable interest in the mechanism of protection offered by CCCs because of environmentally driven pressures to replace chromate as a corrosion inhibitor. Chromate is toxic and carcinogenic [111]. Therefore, understanding the exact mechanism of corrosion inhibition by chromate can promote the development of chromate-free conversion coatings that are environmentally benign.
31 2.3.1 Role of Chromate Ion
The coordination chemistry of the compounds of hexavalent chromium (Cr6+) is
2- 2- mainly the chemistry of anions such as chromate ion (CrO4 ), dichromate ion (Cr2O7 ),
- - - bichromate ion (HCrO4 ), HCr2O7 , CrO3X (X=F, Cl), and chromic acid (H2CrO4) [112].
The oxo-anions (mainly chromate and dichromate ions) have slightly different
2- characteristics as follows: the chromate ion (CrO4 ) is a slightly more negative ion than
2- 2- the dichromate ion (Cr2O7 ) with basis on ionic reactivity; the dichromate ion (Cr2O7 ) is the most abundant species at 2 ≤ pH ≤ 6 and for higher concentrations of total hexavalent
- 6+ chromium; the bichromate ion (HCrO4 ) dominates for low total Cr concentrations
2- (µM) and at generally acidic environments (pH 2~4); whereas the chromate ion (CrO4 )
2- 2- is dominant above pH 7 [113-115]. The chromate ion (CrO4 ), dichromate ion (Cr2O7 )
- and bichromate ion (HCrO4 ) are strong oxidizing agents and have high reduction potentials [116]. When added to chloride-containing solutions, chromates affect the pitting potential, the behavior of metastable pits, pit growth and repassivation, which will be discussed below.
The Effects of Chromate Ion on Pitting Corrosion: It is well established that inhibitors can raise the pitting potentials of Al base alloys [61]. Bohni et al. [61] observed that a linear relationship exists between the logarithm of the chloride ion activity and the logarithm of the minimum activity of the chromate ion required for protection of the Al passive film. Heine and Pryor [117] suggested that passive film formed on aluminum in chromate solution showed lower ionic resistance, which was attributed to the formation
32 of γ-Al2O3 crystalline at the oxide solution interface and higher electronic resistance, which was included some protons and removal of some Al3+ ions in the passive film.
McCafferty [118] and Matsuda [119] also confirmed the linear relationship between chloride ion and chromate ion activities for protection against crevice corrosion and pitting corrosion of Fe (Figure 2.10). McCafferty proposed that the inhibition of crevice corrosion and pitting can be described by a competitive adsorption model in
- 2- which aggressive ions (Cl ) and inhibitive ions (CrO4 ) compete for sites on the metal surface. If the ratio of the surface coverage of aggressive to inhibitive ions exceeds a certain critical value, then breakdown of passivity occurs as crevice corrosion on occluded surfaces or as pitting on open surfaces [118]. Hackerman et al. found that an early step enroute to the formation of Cr2O3 or Cr(OH)3, which is the barrier layer for
2- corrosion, involves the chemisorption of CrO4 ions [120]. Thus, if chromate adsorption predominates within the crevice or on the open surface, passive film formation of Cr2O3 or Cr(OH)3 can follow. However, if the adsorption of chloride ions predominates, a complete passive film cannot be formed; and crevice corrosion then ensues for the shielded surfaces or pitting corrosion for open surfaces [118].
Several studies have emphasized the Cr6+/Cr3+ reduction reaction (which is the
6+ 3+ 3+ Cr ion reduction to hydrated Cr2O3 or mixed Cr /Al oxide film) rather than competitive adsorption for the corrosion inhibition mechanism of aluminum [117, 121-
127]. Wainright et al. suggested the following steps for the enhanced corrosion protection
2- of hexavalent chromium [124]: (i) Adsorption: the strong specific adsorption of CrO4 or
2- Cr2O7 ions on bare metal surfaces as a reservoir of oxidizing species necessary to
2- passivate flaws in the oxide film; the strong oxidizing action of redox couples CrO4
33 2- /Cr2O3 or Cr2O7 /Cr2O3. (ii) Neutralization: the formation of highly insoluble and impervious surface deposits of hydrated Cr2O3 at cathodic sites; and the growth of thin passivating oxide film at anodic sites, arising from surface redox reactions involving hexavalent chromium and metal atoms.
Xu and Wood [125, 126] also examined the formation of steady-state passive film in chromate and dichromate solution for corrosion protection. They found that this passive film comprised a duplex film, an inner thin region of relative pure alumina
(Al2O3) at the metal/film interface and an outer region of disaggregated alumina with adsorbed chromate/dichromate species. They suggested that the inner film is continuously penetrated and reforms in a relatively uniform at the film/solution interface over the macroscopic metal surface supported by the available cathodic process at flaws where hydrated Cr2O3 develops as shown in Figure 2.11 [125].
Akiyama and Frankel [128] investigated the influence of dichromate ions on aluminum dissolution kinetics by using artificial crevice electrodes, which formed a cylindrical pit similar to the real pit. They found that the addition of dichromate ions did not suppress the active dissolution so the inhibition of a localized corrosion by dichromate was not the anodic inhibition of Al dissolution in the pit or crevice environment. The charge, however, of local cathodic reaction on aluminum electrode increased with the addition of dichromate because of the dichromate reduction [128].
The Effects of Chromate ion for Metastable Pitting: Pride et al. [129] investigated the metastable pitting behavior of aluminum by adding chromate ions.
Figure 2.12 shows the effect of the chromate inhibitor on the properties of metastable pits.
34 Chromate ions dramatically decreased metastable nucleation rate, peak pit current density, and the cumulative number of pitting events per unit area more than one order magnitude at the same time, which can be attributed to either incorporation of the chromate inhibitor into the passive film [123-126] or its competitive adsorption with Cl- for defect sites in the passive oxide [118]. The chromate inhibitor also reduced the pit growth rate by measuring apparent pit radii at peak pit current for metastable pitting transients, which can be attributed to competitive adsorption with Cl- on bare metal at the base of the pit and supporting electrolyte. However, the apparent rate of repassivation of metastable pits was not enhanced [129].
The Effects of Chromate ion for Pit Growth and Repassivation: Hunkeler and
Bohni [73] examined the effect of the chromate ion for the pit growth inhibition by using foil penetration technique, which was the first direct method of measuring pit growth rate.
Sehgal and Frankel [130, 131] also used the foil penetration technique for the pit growth study. They found that the corrosion and repassivation potential shifted in the noble direction in the dichromate containing solution [130]. While showing little influence on the pit growth rate at controlled anodic potentials, dichromate ions effectively inhibited pitting at the open circuit. Additionally, polarization curves of AA2024-T3 in aerated 1M
NaCl showed a large effect of dichromate ions in the cathodic region and no effect in the anodic region, which indicated that chromate ions act as a cathodic inhibitor at cathodic sites of Al and Al alloys [131].
35 2.3.2 Coating Processes and Formation of CCCs
CCCs can be applied by an immersion, spray or rolling method for the coating process. The corrosion inhibition of CCCs depends on the details of the coating process such as alloy composition, microstructure, pretreatment procedure, and chromating bath chemistry.
CCC Processes: The chromate coating process is the chemical conversion between the surface of the aluminum alloy and chromate containing solution. Before the chromate coating process, the metal is usually cleaned in an alkaline solution to get rid of contaminates such as grease, oil, soils, and metallic impurities on the surface. It is then immersed in an acidic solution to remove the oxide film and activate the metal surface
[109]. These pretreatments are important steps for the subsequent chromating process in order to develop good properties [132].
The chromate coating process is the most important step for the property of CCCs and includes several critical factors including coating time, temperature, pH, and composition of chromating solution. Chromating solutions typically consist of the following: chromate as a coating forming agent, fluoride as an activator, and ferricyanide as an accelerator. For example, the Alodine 1200S (Henkel Corp.) is widely used for chromate conversion coating. It includes 50-60 wt.% CrO3, 5-10 wt.% NaF, 10-15 wt.%
K3Fe(CN)6, 20-30 wt.% KBF4 and 5-10 wt.% K2ZrF6. After the chromating process, the final steps in the formation of chromate coating are rinsing and drying.
The presence of a fluoride ion as an activator is important for the formation of
CCCs. Film growth of CCCs is extremely slow without fluoride ion. Fluoride dissolves
36 6 3+ the aluminum oxide (Al2O3) initially present on the surface and allows the Cr /Cr redox reaction and coating deposition to proceed. Fluoride also dissolves a part of the growing film (AlOOH and CrOOH), which prevents further coating growth. Partial dissolution of film allows penetration of the electrolyte to the surface and ion transport from the surface for more coating growth. The reactions are as follows [108, 110]:
Al2O3 + 6HF → 2AlF3 (soluble) + 3H2O Eqn. 2-7
AlOOH + 3HF → AlF3 (soluble) + 2H2O Eqn. 2-8
CrOOH + 3HF → CrF3 (soluble) + 2H2O Eqn. 2-9
3- The presence of ferricyanide [Fe(CN)6 ] as an accelerator is also important for the growing coating [110, 116, 133, 134]. Treverton et al. suggested that the enhanced corrosion resistance of CCCs is attributed to the increase in coating weight due to the incorporation of the accelerator into the growing coating [133, 134]. Di Quarto et al. also showed that the coating for 3-minute coating time was 2.5 times thicker with ferricyanide than without although the Cr(VI)/Cr total ratio was similar [136]. Hagans and Haas indicated that ferricyanide was distributed on the copper-containing intermetallic particles such as CuAl2, which were known to accelerate the corrosion rate of AA2024-
T3, and the corrosion of these particles was inhibited by the formation of copper- ferricyanide on the intermetallic particles [110, 116]. However, the mechanism of acceleration was not clearly articulated. Recently, Xia and McCreery proposed that ferricyanide was a redox mediation mechanism for CCC formation [135]. They found that CCC formation rate greatly decreased by removing ferricyanide from the coating
37 bath. Furthermore, they observed that ferricyanide rapidly oxidized AA2024-T3 while it
4- was reduced to ferrocyanide [Fe(CN)6 ]. In this mechanism, ferricyanide is reduced to ferrocyanide, while aluminum oxidized. Ferrocyanide is rapidly oxidized back to ferricyanide while Cr6+ is reduced Cr3+ [135].
CCC Formation on Matrix: The formation of CCCs is complex due to the chemistry of the chromating solution bath, the heterogeneous microstructure of aluminum alloy, and the pretreatment of CCCs. However, the formation of CCCs is generally described as a redox reaction between chromate or dichromate in solution and aluminum as follows [137, 138]:
2Al → 2Al3+ + 6e- (in the presence of F- [139]) Eqn. 2-10
3+ + 2Al + 3H2O → Al2O3 + 6H Eqn. 2-11
2- + - Cr2O7 + 8H + 6e → 2Cr(OH)3 + H2O Eqn. 2-12
2- + Overall reaction: 2Al + Cr2O7 + 2H + 2H2O → Al2O3 + 2Cr(OH)3 Eqn. 2-13
The chromate or dichromate oxidizes the aluminum in the presence of fluoride and the hexavalent chromium (Cr6+) is reduced to trivalent chromium (Cr3+). These
Cr6+/Cr3+ redox reactions consume protons and increase the local pH. The pH increase results in the precipitation of an amorphous mixture of hydrated aluminum and chromium oxides [110]. Hydrogen evolution and/or oxygen reduction reaction occurs at cathodic sites.
Other possible reactions have been suggested [107, 108, 133, 140]: 38
- + 3+ 2Al + 2HCrO4 + 8H → 2Al + Cr2O3⋅2H2O↓ + 3H2O Eqn. 2-14
2- + 2Al + Cr2O7 + 2H → 2AlOOH ↓ +2CrOOH↓ Eqn. 2-15
More recently, CCC formation on matrix has been suggested by multi-step formation, i.e., hydrolysis, condensation and polymerization. The redox reation forms
Cr3+ rapidly and makes Cr3+ hydroxides which condense and precipitate on matrix by polymerization in the solution. These condensation and polymerization continue in air which accompany with dehydration during drying [135, 152].
CCC Formation on Intermetallic Compounds: The formation of CCC described above is based on coating formation on pure Al. Aluminum-based precipitation age hardened alloys, such as 2xxx and 7xxx, contain intermetallic compounds which complicate CCC formation. Among the intermetallic compounds, copper-containing
IMCs, such as Al-Cu, Al-Cu-Mg and Al-Cu-Fe-Mn compounds, are mainly studied.
Waldrop and Kendig studied the nucleation and early growth of CCCs on aluminum 2024-T3 by using AFM. Al-Cu-Fe-Mn IMCs, which are cathodically active compared to the matrix, showed rapid nucleation and conversion coating growth while
Al-Cu-Mg IMCs, less cathodically active than the matrix, supported slow growth [138].
This result indicates that Al-Cu-Fe-Mn IMCs accelerate cathodic reaction of chromate but Al-Cu-Mg IMC inhibits this reaction (Figure 2.13)[138, 141]. Brown and Kobayashi also observed that the growth of CCCs of Al-Cu-Mg IMCs was sustained by anodic
39 dissolution, while the film growth of Al-Cu-Fe-Mn IMCs was enhanced with localized dissolution of a copper-depleted zone at the periphery of Al-Cu-Fe-Mn IMCs [142].
Ilevbare and Scully investigated oxygen reduction reaction (ORR) kinetics on conversion coated bulk-synthesized analogs of Al-Cu, Al-Cu-Mg, and Al-Cu-Fe-Mn
IMCs [15]. The OCPs of Al-Cu and Al-Cu-Mg IMCs after chromate coating were not changed while that of Al-Cu-Fe-Mn IMCs was slightly decreased. They suggested that
Al-Cu and Al-Cu-Mg IMCs could not alter corrosion of AA2024-T3 while Al-Cu-Fe-Mn could be expected to have some effect. It was observed that ORR kinetics on conversion coated Al-Cu-Fe-Mn decreased due to suppression of corrosion at OCP [15]. However, these buk-synthesized analogs could not be representative of IMCs in AA2024 since these analogs exclude the galvanic coupling effect between IMCs and matrix [174].
Recently, Baek and Frankel measured the cathodic corrosion rate on the thin-film analogs of pure Al, Al-4%Cu, and Al-Cu IMCs with and without chromate by electrochemical quartz crystal microbalance [143]. The true cathodic corrosion rates, calculated from net current density and dissolution rate, of pure Al and Al-4%Cu were large while that of Al-Cu was almost identical to the net current density. They suggested that the presence of dichromate effectively decreased the cathodic reaction rate on Al-Cu
IMC (Figure 2.14) [143].
McGovern et al. monitored the CCC formation on bulk synthesized Al-Cu-Mg
IMCs and AA2024-T3 using Raman spectroscopy [144]. It was shown that the bulk- synthesized analogs of Al-Cu-Mg IMCs had slower CCC formation with higher copper content and also found that CCC formation on Al-Cu-Mg and Al-Cu-Fe-Mn IMCs was suppressed compared to the matrix. The authors suggested the possible inhibition
40 mechanism of CCC in Figure 2.15 [144]. Hurley and McCreery also studied the interactions of CrVI solutions with copper by using surface enhanced Raman scattering.
They suggested chromate inhibition by monolayers of Cr3+ hydroxide on copper in Figure
2.16 [145].
Meng et al. observed that coatings on intermetallic compounds, such as, Al2CuMg,
Mg2Si, and Al3(Fe Cu) were thinner than on matrix by TEM cross-section. They suggested that the chromate coating formation on these intermetallics is dependent on several factors, such as electrochemical reactivity of the intermetallics, local pH, and interaction between intermetallics and ingredient of coating bath [173].
Physical Models for CCC formation and Growth: Katzman et al. proposed a uniform formation and growth model for CCCs as shown in Figure 2.17 [108]. They suggested that HF dissolved the initial oxide film and the bare aluminum reacted with dichromate resulting in the formation of hydrated Al3+/Cr3+ mixed oxide. AlOOH in the mixed oxide, was dissolved by HF and less soluble CrOOH was left on the surface. As the coating continues to grow, aluminum is oxidized at the aluminum-coating interface and Al3+ diffuses through the coating. Dichromate is reduced to CrOOH at coating and solution interface and forms hydrated chromium oxide [108].
Brown and Wood proposed another model for CCC formation and growth [146-
150]. They found that the dissolution of Al occurred only at the bottom of holes and suggested that tunneling of electrons from the metal through the oxide played an important role in the formation and growth of CCCs (Figure 2.18). The initial oxide film on the aluminum surface was reduced by the attack of HF. The reduced oxide film
41 allowed electron tunneling. During coating growth, cathodic reduction of Cr6+ took place at segregated flaw sites or impurities to form the hydrated Cr3+ oxide as a result of electron tunneling through the residual oxide film, while the anodic reaction, i.e., dissolution of the Al oxide film and formation of Al oxide film by oxidizing Al, occurred generally over the Al surface. As a result, the morphology was a relatively uniform conversion coating with some holes from the outer coating surface down to the Al substrate [148].
Arrowsmith et al. proposed a layered structure of spherical particles for the CCC formation and growth based on the SEM observation [151]. The initial oxide film was dissolved and anodic dissolution of Al occurred. Cathodic reduction of Cr6+ to Cr3+ proceeded to precipitate strands of hydrated chromium oxide, which grew into spherical particles. These spheres grew and merged to form a monolayer of particulates, which had gaps between sites of anodically dissolved aluminum, and hydrogen evolution. These gaps served as paths for transportation of fresh solution to the metal and dissolved aluminum to solution. Spherical particles were nucleated to form successive layers with continued immersion.
Xia and McCreery proposed a multi-step formation and growth model for CCCs as shown in Figure 2.19 [135, 152-154]. The Al oxide was etched by fluoride. During
6+ 3+ 3-/4- CCC formation, Cr was reduced to Cr by Fe(CN)6 , which went through a series of
3+ 3+ condensation reactions to produce a hydrated Cr oxy-hydroxide [Cr x(OH)y] coating to a Cr3+ polymerization reaction. Cr6+ was adsorbed to the Cr3+ oxy-hydroxy polymer and this binding was reversible. Release of Cr6+ from the coating was resulted in self-healing
[153].
42 Structures for CCCs: By using various surface sensitive characterization techniques, CCCs are known to be amorphous films and multi-layer structures. Treverton and Davies proposed a CCC structure (Figure 2.20) [133]. The surface film consisted of a hydrated form of chromium oxide (Cr2O3⋅xH2O) with smaller quantities of chromium ferricyanide [CrFe(CN)6]. The bulk film was a hydrated form of chromium oxide
(Cr2O3⋅xH2O). Aluminum oxides (Al2O3) and fluorides (AlF3, AlOF) lie at the interface between the film and the metal substrate. The authors suggested that Cr6+ was undetectable due to conversion from Cr6+ to Cr3+. Townsend and Hart proposed chromate passivation film which contained a layer of metallic chromium between coating and metal and Cr6+ in the top layer (Figure 2.21) [155]. They suggested the chromium metal came from the complete reduction of chromic acid or sputtering artifact. Hughes et al. suggested the structure model of CCCs on AA2024 by using XPS, SEM and EDX
(Figure 2.22) [107]. The external layer was composed of CrOOH, with significant levels
3- 6+ of Fe(CN)6 and smaller amounts of Cr . The bulk of the coating was made up of
- 3- Cr2O3⋅CrOOH, with F and Fe(CN)6 anions within the coating. Some of the fluorides were presented as (Cr, Al)OF. The interface region with the alloy appeared to have Al2O3 as well as increased levels of Cu. Both flouride and ferricyanide with Cr2O3⋅CrOOH were present at the interface. The authors remarked that the fluorides and ferricyanides were present throughout the coating. Di Quarto et al. observed a duplex structure of the 30 s coating time. The external region of duplex structure at the solution/CCC interface showed a slightly coarser texture about 25 nm where the only iron species existed [136].
43 2.3.3 Protection Mechanisms of CCCs
As addressed above, the chromium oxides are generally recognized to form a barrier layer as the mechanism of corrosion protection. This barrier layer is impermeable and amorphous, which comes from the insoluble Cr3+ species.
Sato suggested a bipolar membrane model for the passivity of metals [156-158].
This bipolar membrane consists of an anion-selective layer on the metal side and a cation-selective layer on the solution side as shown in Figure 2.23. As metal dissolves, the hydrated metal oxide film precipitates on the metal surface. When this hydrated oxide
2- charge is reversed from positive to negative with adsorption of CrO4 , the inner layer remains anion-selective due to an excess of metal cation. Both anion-selective and cation- selective layers are therefore bipolar membranes, which are ion-permeable irrespective of the direction of ionic current. The ion transport through these bipolar membranes is made asymmetric and allowed to occur only in the forward direction. Anionic ion migration is hindered and hence the passivation can be accelerated. The cation-selective layer discourages the adsorption of aggressive chloride. The CCC structure can be appropriate for a bipolar membrane, i.e., the inner anion-selective layer is the Al3+/Cr3+ mixed oxide and the outer cation-selective layer is Cr6+ adsorbed layer. Kendig et al. confirmed that the presence of dilute chromate substantially lowers the surface potential within the anodized layer on sulfuric acid anodized aluminum [159].
An important component of corrosion protection of CCCs is self-healing. When damage occurs, the soluble hexavalent chromate preserved in CCCs moves to the active corrosion sites through diffusion and is reduced to form insoluble trivalent chrominum species, which suffocates the cathodic reactions. The corrosion at the damaged area,
44 therefore, is inhibited [108, 110, 135, 160, 161]. Zhao et al. confirmed the self-healing effect using an artificial scratch cell. A CCC treated sample and bare AA2024-T3 sample were placed face to face in a sodium chloride solution to determine if corrosion protection of the untreated alloy resulted from the CCC sample [161]. The initially untreated sample was protected from corrosion by migration from the CCC treated sample by monitoring the Cr species on the untreated sample. The polarization resistance and pitting potential also increased. For the release of Cr6+ into the defect site, Xia et al. investigated the storage and release of soluble Cr6+ species from CCCs as shown in
Figure 2.24 [162]. They suggested that Cr6+ in solution was in equilibrium with Cr6+ in the CCC. Solution Cr6+ adsorbed to a porous and insoluble Cr3+ hydroxide and formed a
Cr3+-O-Cr6+ mixed oxide. Equilibrium Cr6+ concentrations were analogous to a Langmuir adsorption mechanism. The release of Cr6+ into solution was pH dependent, with higher pH shifting the equilibrium in the direction of the desorptive reaction. Under the low pH and relatively high Cr6+ such as in the Alodine treatment, adsorption of Cr6+ was favored.
The adsorption and desorption were reversible with repeated pH excursions. This observation may explain why Cr6+ adsorbs during the Alodine treatment, but desorbs and migrates in the field in order to bring about self-healing [162].
Clayton et al. proposed a duplex model, which incorporated both the Sato-bipolar model and McCreery-pit repassivation model, for the protection mechanism of CCCs
(Figure 2.25) [153, 163]. Aluminum hydroxide [Al(OH)3] was first formed and then exposed to chromate and chloride solutions. Chromate adsorption appeared to induce deprotonation to the hydrated aluminum anodic film (Al2O3) consistent with the bipolar
45 mode. Cr6+ enrichment in the valleys was consistent with the possible formation of an
Al3+-Cr6+ compound according to the McCreery pit repassivation.
Laget et al. observed the loss of corrosion protection due to dehydration, which was associated with the structural changes [164]. It is shown that the extended X-ray fine structure measurement indicates the shortening of Cr(III)-Cr(III) during dehydration, which causes to shrinkage cracking and immobilization of Cr(VI) (Figure 2.26) [164].
Recently, Campestrini et al. proposed the formation of CCCs in three distinct stages by monitoring OCP and AES surface analysis [165]. The first stage is the activation of the surface within 2 second; the second stage is the initiation of the layer formation within 5-second, and finally the third stage is the growth of the conversion layer. They suggested that the growth of the chromate film characterized by different layers of spherical particles, through which the transport of the different reagents from and to the Al surface (Figure 2.27).
2.4 KEY UNRESOLVED ISSUES
2.4.1 Formation and breakdown of CCCs on heterogeneous 7xxx aluminum alloys
Over the past several decades, research has been carried out in order to find replacements of chromate which are environmentally friendly. However, no replacement has been developed that is as effective as chromate for corrosion protection of high strength Al alloys. Chromates seem to have a unique ability to inhibit the corrosion of high strength Al alloys. Additionally, the heterogeneity of Al alloys has resulted in un expected complications that have delayed finding the non-toxic inhibitors.
46 As mentioned earlier, the aluminum alloys contain different kinds of intermetallic phases such as coarse constituent phases, dispersoid phases, and precipitation hardening phases. It is not clear which of these phases has the greatest influence on the formation, growth and breakdown of CCCs on the alloys. It is assured that the presence of these intermetallic phases leads to corrosion susceptiblity, while the formation and breakdown on these intermetallic phases of 7xxx aluminum alloys has not been rigorously studied.
The coating formation kinetics are still far from being clear.
2.4.2 Influence of Temper on the Formation and Breakdown of CCCs
Although a subtle change in the heat treatment can significantly change the SCC resistance, there is no mechanism that fully explains the beneficial effect of T7 temper.
Most of the studies have focused on the role of grain boundary precipitates, precipitate free zone (PFZ), and matrix precipitates on the susceptibility to SCC. There have not been many studies on the formation and breakdown of CCCs on Al alloys at different tempers.
The exact size, shape, and distribution of precipitates depend on the alloys in which they form and on the heat treatment and eventually these microstructures can affect the mechanical properties. It is essential to fully characterize the coating as a function of temper in order to understand the relationship between microstructures and the performance of CCCs.
47 2.4.3 Electrochemical activities of important IMCs
As this literature review has indicated, several mechanisms and structures of
CCCs have been proposed for coating processes and formation of CCCs by chromate.
However, most research focused on the coating behavior on pure Al or 2024-T3. The important IMCs, such as MgZn2 and Al2CuMg particles have not been fully examined for their corrosion behavior and coating formation characteristics.
In order to achieve the chromate-free corrosion inhibition, it is very important to gain a complete understanding of chromate coating formation and breakdown on each particle type and to understand the relationship between the chromate inhibitor and heterogeneous microstructure from the metallurgical to electrochemical points of view.
2.5 OBJECTIVE OF DISSERTATION
The previous results showed that chromate conversion performance of T7 temper was more effective than that of T6 temper in AA7475 using electrochemical impedance spectroscopy (EIS) [167, 168]. However, industrial reports indicated that corrosion protection by CCC of AA7075-T6 was better than that of AA7075-T7 using salt spray test [169]. The reason why this change in CCC protection occurs has not been clearly explained. For this reason, we have characterized the effect of 7xxx alloy temper on CCC formation and breakdown. There are several possibilities for that difference. AA7475 and
AA7075 have different levels of impurities and intermetallic compounds. The chemicals of the chromate bath for corrosion protection are different. Finally, the experimental methods are different, i.e., salt spray test and EIS.
48 Therefore, the objective of this dissertation is to investigate the CCC formation and breakdown, and the effect of temper on CCC performance. In order to study the formation and breakdown of CCCs, atomic force microscopy (AFM) was carried out for the investigation of the initial formation of CCCs on the heat-treated samples, especially the effects of nucleation and surface Volta potential change on the CCCs performance.
Coating formation and breakdown on different intermetallic phases were also investigated using AFM. The multichannel microelectrode analyzer (MMA) allowed measurement of the coating formation and breakdown. The effects of impurities were studied by comparing AA7475 and AA7075. The chromate content was measured by X- ray absorption near edge spectroscopy (XANES) and Raman spectroscopy as a function of heat treatment and coating time. Thin film analogs of Mg(Zn,xCu)2 and Al2CuMg phases, prepared by the flash evaporation technique and sputtering, were studied for the understanding of the corrosion behavior and coating formation of the η and S phases with chromate. The behavior of chromate coating formation was investigated using electrochemical quartz microbalance (EQCM), which allows measurement of non- electrochemical mass changes as well as electrochemical current and potential changes simultaneously.
49 REFERENCES
1. R.W. Revie, "Uhlig's Corrosion Handbook,” New York: Wiley (2000).
2. D.O. Sprowls and R.H. Brown. "Stress Corrosion Mechanisms for Aluminum Alloys,” R.W. Stahle et al. eds., NACE, Houston, Texas (1969).
3. L.F. Mondolfo, Metallurgical Reviews, 153, 95 (1971).
4. P.L. Cabot, F. Centellas, J.A. Garrido, R.M. Rogriguez, E. Brillas, E. Perez, A.V. Benedetti, and P.T.A. Sumodjo, J. Appl. Electrochem, 22, 541 (1992).
5. L.F. Mondolfo, "Aluminium alloys: Structure and properties,” Boston: Buttenworths (1976).
6. H. Cordier, C. Dumont, and W. Gruhl, Aluminium, 55, 777 (1979).
7. J.T. Staley, H.Y. Hunsicker, and R.H. Brown, U.S. Patent 3,881,966 (1975).
8. J.E. Hatch, "Aluminum properties and physical metallurgy,” Metals Park, Ohio: American Society for Metals (1984).
9. B. Cina, U.S. Patent 3,856,584 (1974).
10. J.R. Davis, "Alloying : understanding the basics,” Materials Park, OH: ASM International (2001).
11. I.J. Polmear, "Light Alloys: Metallurgy of the Light Metals,” New York: J. Wiley & Sons (1996).
12. M. Pourbaix, "Atlas of Electrochemical Equilibrium in Aqueous Solutions,” Oxford: Pergamon Press (1966).
13. C. Kammer, "Aluminum Handbook,” Oldenburg, Germany: Aluminium-Zentrale (1999).
14. E.H. Hollingsworth and H.Y. Hunsicker, "Corrosion and Corrosion Protection Handbook,” New York: Marcel Dekker (1983).
15. G.O. Ilevbare and J.R. Scully, J. Electrochem. Soc., 148, B196 (2001).
16. R.G. Buchheit, J. Electrochem. Soc., 142, 3994 (1995).
17. J.C. Seegmiller and D.A. Buttry, J. Electrochem. Soc., 150, B413 (2003).
18. W. Gruhl. "International Congress on Aluminium Alloys in the Aircraft Industry,” eds., Turin (1976). 50 19. M. Conserva, Alumino E. Nuova Metallurgia, 39, 515 (1970).
20. A.S. Argon, "Physics of Strength and Plasticity,” Cambridge: MIT Press (1969).
21. ASM International, "ASM handbook: Vol. 4 Heat Treating,” Metal Parks: American Society for Metals (1991).
22. H.Y. Hunsicker. eds., Rosenhain Centenary Conference on the Contribution of Physical Metallurgy to Engineering Practice, London (1976).
23. J.K. Park and A.J. Ardell, Metall. Trans. A, 15A, 1531 (1984).
24. J.K. Park and A.J. Ardell, Scripta Metall., 22, 1115 (1988).
25. J.K. Park and A.J. Ardell, Acta Metall. Mater., 39, 591 (1991).
26. H. Schmalzried and V. Gerold, Z. Metallkd., 49, 291 (1958).
27. L.F. Mondolfo, N.A. Gjostein, and D.W. Levinson, Trans. Amer. Inst. Min. Met. Eng., 206, 1378 (1956).
28. G. Thomas and J. Nutting, J. Inst. Metals, 88, 81 (1959-60).
29. R.B. Nicholson, G. Thomas, and J. Nutting, J. Inst. Metals, 87, 429 (1958-59).
30. R.B. Nicholson, G. Thomas, and J. Nutting, Brit. J. Appl. Phys., 9, 25 (1958).
31. J. Gjonnes and C.J. Simensen, Acta Met., 18, 881 (1970).
32. J.D. Embury and R.B. Nicholson, Acta Met., 13, 403 (1965).
33. G.W. Lorimer and R.B. Nicholson, Acta Met., 14, 1009 (1966).
34. G.W. Lorimer and R.B. Nicholson, "The Mechanism of Phase Transformations in Crystalline Solids,” London: Inst. Metals (1968).
35. Pearson, "Handbook of Lattice Spacing and Structure of Metals and Alloys,” Oxford: Pergamon Press (1958).
36. G. Bergmann, L.T. Waugh, and L. Paulng, Acta Cryst., 10, 254 (1957).
37. K.H. Westmacott, R.S. Barnes, D. Hull, and R.E. Smallman, Phil. Mag., 6A, 929 (1961).
38. R. Graf, Compt. Rend., 242, 1311 (1956).
39. R. Graf, Compt. Rend., 244, 337 (1957).
51 40. J.H. Auld and S. Mck. Cousland, Scripta Metall., 5, 765 (1971).
41. D.W. Pashley, M.H. Jacobs, and J.T. Veitz, Phil. Mag., 16, 51 (1967).
42. P.A. Thackery, J. Inst. Metals, 96, 228 (1968).
43. R. Ferragut, A. Somoza, and A. Tolley, Acta mater, 47, 4355 (1999).
44. P.N. Adler and R. DeIasi, Metall. Trans. A, 8A, 1185 (1977).
45. P.N. Adler, R. DeIasi, and G. Geschwind, Metall. Trans., 3, 3191 (1972).
46. R. DeIasi and P.N. Adler, Metall. Trans. A, 8A, 1177 (1977).
47. J.K. Park and A.J. Ardell, Metall. Trans. A, 14A, 1957 (1983).
48. J.K. Park and A.J. Ardell, Mater. Sci. Engng., A114, 197 (1989).
49. N. Danh, K. Rajan, and W. Wallace, Metall. Trans., 14A, 1843 (1983).
50. N. Danh, K. Rajan, and W. Wallace, Metall. Trans., 16A, 2068 (1985).
51. D.J. Lloyd and M.C. Chaturvedi, J. Mater. Sci., 17, 1819 (1982).
52. A. Baldantoni, Mater. Sci. Engr., L5, (1985).
53. T.C. Tasi and T.H. Chuang, Metall. Trans., 27A, 2617 (1996).
54. G. Faita, F. Mazza, and G. Bianchi. "Role of Water and Ionic Solvation in Localized Corrosion Phenomena,” R.W. Stahle et al. eds., NACE, Houston (1974).
55. J. Newman. "Mass Transport and Potential Distribution in the Geometries of Localized Corrosion Phenomena,” R.W. Stahle et al. eds., NACE, Houston (1974).
56. F.D. Wall and M.A. Martinez, J. Electrochem. Soc., 150, B146 (2003).
57. J.R. Scully, D.E. Peebles, A.D. Romg, D.R. Frear Jr, and C.R. Hills, Metallurgical Transaction A, 23A, 2641 (1992).
58. G.S. Frankel, J. Electrochem. Soc., 145, 2186 (1998).
59. H.P. Lekie and H.H. Uhlig, J. Electrochem. Soc., 113, 1262 (1966).
60. H.H. Uhlig, J. Electrochem. Soc., 97, 215C (1950).
61. H. Bohni and H.H. Uhlig, J. Electrochem. Soc., 116, 906 (1969).
52 62. H.H. Uhlig. "Competitive Adsorption as a Mechanism for Breakdown of Passivity,” R.W. Stahle et al. eds., NACE, Houston, Texas (1976).
63. N. Sato, Electrochimica Acta, 16, 1683 (1971).
64. T.P. Hoar, D.C. Mears, and G.P. Rothwell, Corrosion Science, 5, 279 (1965).
65. M. Koudelkova, J. Augustynski, and H. Berthou, J. Electrochem. Soc., 124, 1165 (1977).
66. P. Marcus and J.M. Herbelin, Corrosion Science, 34, 1123 (1993).
67. J.R. Galvele, S.M. de De Micheli, I.L. Muller, S.B. de Wexler, and I.L. Alanis. "Critical Potentials for Localized Corrosion of Aluminum Alloys,” R.W. Stahle et al. eds., NACE, Houston, Texas (1974).
68. G.S. Frankel, L. Stockert, F. Hunkeler, and H. Boehni, Corrosion, 43, 429 (1987).
69. Y. Hisamatsu, T. Yoshii, and Y. Matsumura. "Electrochemical and Microscopical Study of Pitting Corrosion of Austenitic Stainless Steel,” R.W. Stahle et al. eds., NACE, Houston, Texas (1974).
70. C.J. Newton and N.J.H. Holroyd. "Time-Lapse Video Technique in the Corrosion Testing of Aluminum Alloys,” V.S. Agarwala et al. eds., ASTM, Philadelphia (1992).
71. H.H. Strehblow and M.B. Ives, Corrosion Science, 16, 317 (1976).
72. H.H. Strehblow and J. Wenners, Electrochem. Acta., 22, (1977).
73. F. Hunkeler and H. Bohni, Corrosion, 37, 645 (1981).
74. Y. Hisamatsu. "Pitting Corrosion of Stainless Steels in Chloride Solutions,” R.W. Stahle et al. eds., NACE, Houston, Texas (1974).
75. N. Sato, J. Electrochem. Soc., 129, 260 (1982).
76. D.E. Williams, J. Stewart, and P. Balkwill, Corrosion Science, 36, 1213 (1994).
77. J.R. Galvele. "Present State of Understanding of the Breakdown of Passivity and Repassivation,” R.P. Frankkethal et al.. eds., The Electrochemical Society-4th International Symposium on Passivity, Princeton, NJ (1978).
78. J.R. Galvele, J. Electrochem. Soc., 123, 464 (1976).
79. J.R. Galvele, Corrosion Science, 21, 551 (1981).
80. T.R. Beck and S.G. Chan, Corrosion, 37, 665 (1981). 53 81. T.R. Beck, J. Electrochem. Soc., 129, 2412 (1982).
82. T.R. Beck, Electrochimica Acta, 29, 485 (1984).
83. H.S. Isaacs. "Potential Scanning of Stainless Steel during Pitting Corrosion,” R.W. Stahle et al. eds., NACE, Houston, TX (1974).
84. N.J. Laycock, M.H. Moayed, and R.C. Newman. "Prediction of Pitting Potential and Critical Pitting Temperatures,” P.M. Natishan et al.. eds., The Electrochemical Society, 95-15, Pennington, NJ (1996).
85. J.L. Luo and M.B. Ives. "Factors Controlling Pit Development,” G.S. Frankel et al. eds., The Electrochemical Society, PV92-9, Princeton, NJ (1992).
86. J.E.H. Dix, AIME, 137, 11 (1940).
87. J.E.H. Dix, ASM, 42, 1057 (1950).
88. J.R. Galvele and S.M. de De Micheli, Corrosion Science, 10, 795 (1970).
89. S. Maitra and G.C. English, Metallurgical Transaction A, 12A, 535 (1981).
90. S. Maitra and G.C. English, Metallurgical Transaction A, 13A, 161 (1982).
91. T. Ramgopal, P. Schmutz, and G.S. Frankel, J. Electrochem. Soc., 148, B348 (2001).
92. P. Doig and J.W. Edington, Corrosion, 31, 347 (1975).
93. P. Doig and J.W. Edington, Metall. Trans. A, 6A, 943 (1975).
94. P. Doig and J.W. Edington, British Corrosion Journal, 9, 22 (1974).
95. P. Doig and J.W. Edington, British Corrosion Journal, 9, 88 (1974).
96. P. Doig and J.W. Edington, British Corrosion Journal, 9, 220 (1974).
97. K.R. Cooper and R.G. Kelly, J. Chromatography A, 739, 183 (1996).
98. M.O. Speidel, Metall. Trans., 6A, 631 (1975).
99. A.J. McEvily, J.B. Clark, and A.P. Bond, Trans. ASM, 60, 661 (1967).
100. G. Thomas, J. Inst. Metals, 89, 287 (1960-61).
101. W. Gruhl and H. Cordier, Aluminum, 44, 403 (1968).
102. W. Gruhl, Aluminum, 38, 775 (1962).
54 103. H.A. Hall, Corrosion, 23, 173 (1967).
104. K.G. Kent, J. Inst. Metals, 97, 127 (1969).
105. P.N.T. Unwin and R.B. Nicholson, Acta Met., 17, (1969).
106. H.P. Kim, R.H. Song, and S.I. Pyun, Br. Corros. J., 23, (1998).
107. A.E. Hughes, R.J. Taylor, and B.R.W. Hinton, Surface and interface analysis, 25, 223 (1997).
108. H.A. Katzman, G.M. Malouf, R. Bauer, and G.W. Stupian, Applications of Surface Science, 2, 416 (1979).
109. K.A. Korinek, "Chromate Conversion Coatings,” Metals Park, OH: ASM (1987).
110. P.L. Hagans and C.M. Haas, "Chromate Conversion Coatings,” Metals Park, OH: ASM International (1994).
111. H.G. Seiler, "Handbook on Toxicity of Inorganic Compounds,” New York: Marel Dekker (1988).
112. P. Gili and P.A. Lorenzo-Luis, Coordination Chemistry Reviews, 193-195, 747 (1999).
113. J. Mestres, M. Duran, P. Matrin-Zarza, E. Medina de la Rosa, and P. Gili, Inorg. Chem., 32, 4708 (1993).
114. J.D. Ramsey, L. Xia, M.W. Kendig, and R.L. McCreery, Corrosion Science, 43, 1557 (2001).
115. M. Kendig and R. Buchheit. "Corrosion Inhibition of Al and Al Alloys by Hexavalent Cr Compounds-A Mechanistic Overview,” R.G. Buchheit et al.. eds., NACE, Houston, TX (2000).
116. P.L. Hagans and C.M. Haas, Surface and Interface Analysis, 21, 65 (1994).
117. M.A. Heine and M.J. Pryor, J. Electrochem. Soc., 114, 1001 (1969).
118. E. McCafferty, J. Electrochem. Soc., 137, 3731 (1990).
119. S. Matsuda and H.H. Uhlig, J. Electrochem. Soc., 111, 156 (1964).
120. N. Hackerman and R.A. Powers, Journal of Physical Chemistry, 57, 139 (1953).
121. H.H. Uhlig and P.F. King, J. Electrochem. Soc., 106, 1 (1959).
122. J.A. Richardson and G.C. Wood, J. Electrochem. Soc., 120, 193 (1973). 55 123. J.K. Hawkins, H.S. Isaacs, S.M. Heald, J. Tranquada, G.E. Thompson, and G.C. Wood, Corrosion Science, 27, 391 (1987).
124. J.S. Wainright, O.J. Murphy, and M.R. Antonio, Corrosion Science, 33, 281 (1992).
125. N. Xu, G.E. Thompson, J.L. Dawson, and G.C. Wood, Corrosion Science, 34, 461 (1993).
126. N. Xu, G.E. Thompson, J.L. Dawson, and G.C. Wood, Corrosion Science, 34, 479 (1993).
127. H. Habazaki, K. Shimizu, P. Skeldon, G.E. Thompson, X. Zhou, J. De Laet, and G.C. Wood, Corrosion Science, 39, 719 (1997).
128. E. Akiyama and G.S. Frankel, J. Electrochem. Soc., 146, 4095 (1999).
129. S.T. Pride, J.R. Scully, and J.L. Hudson, J. Electrochem. Soc., 141, 3028 (1994).
130. A. Sehgal, D. Lu, and G.S. Frankel, J. Electrochem. Soc., 145, 2834 (1998).
131. A. Sehgal, G.S. Frankel, B. Zoofan, and S. Rokhlin, J. Electrochem. Soc., 147, 140 (2000).
132. G.S. Frankel, "Mechanism of Al Alloy corrosion and the role of Chromate Inhibitors,” Air Force Office of Scientific Research, Contract No. F49620-96-1- 0479, Second Annual Report (1998).
133. J.A. Treverton and N.C. Davies, Metals Technology, 10, 480 (1977).
134. J.A. Treverton and N.C. Davies, Surface and Interface Analysis, 3, 194 (1981).
135. L. Xia and R.L. McCreery, J. Electrochem. Soc., 146, 3696 (1999).
136. F. Di Quarto, M. Santamaria, N. Mallandrino, V. Laget, R. Buchheit, and K. Shimizu, J. Electrochem. Soc., 150, B462 (2003).
137. M.W. Kendig, A.J. Davenport, and H.S. Isaacs, Corrosion Science, 34, 41 (1993).
138. J.R. Waldrop and M.W. Kendig, J. Electrochem. Soc., 145, L11 (1998).
139. K. Asami, M. Oki, G.E. Thompson, G.C. Wood, and V. Ashworth, Electrochemica Acta, 32, 337 (1987).
140. J.A. Treverton, M.P. Amor, and A. Bosland, Corrosion Science, 33, 1411 (1992).
141. M. Kendig, S. Jeanjaquet, R. Addison, and J. Waldrop, Surface and Coatings Technology, 140, 58 (2001). 56 142. G.M. Brown and K. Kobayashi, J. Electrochem. Soc., 148, B457 (2001).
143. Y. Baek and G.S. Frankel, J. Electrochem. Soc., 150, B1 (2003).
144. W.R. McGovern, P. Schmutz, R.G. Buchheit, and R.L. McCreery, J. Electrochem. Soc., 147, 4494 (2000).
145. B.L. Hurley and R.L. McCreery, J. Electrochem. Soc., 150, B367 (2003).
146. M.F. Abd Pabbo, J.A. Richardson, and G.C. Wood, Corrosion Science, 18, 117h (1978).
147. G.M. Brown, K. Shimizu, K. Kobayashi, G.E. Thompson, and G.C. Wood, Corrosion Science, 33, 1371 (1992).
148. G.M. Brown, K. Shimizu, K. Kobayashi, G.E. Thompson, and G.C. Wood, Corrosion Science, 34, 1045 (1993).
149. G.M. Brown, K. Shimizu, K. Kobayashi, G.E. Thompson, and G.C. Wood, Corrosion Science, 34, 2099 (1993).
150. G.M. Brown, K. Shimizu, K. Kobayashi, G.E. Thompson, and G.C. Wood, Corrosion Science, 35, 253 (1993).
151. D.J. Arrowsmith, J.K. Dennis, and P.R. Sliwinski, Transactions of the Institute of Metal Finishing, 62, 117 (1984).
152. L. Xia and R.C. McCreery, J. Electrchem. Soc., 145, 3083 (1998).
153. G.S. Frankel, "Mechanism of Al Alloy Corrosion and the role of Chromate Inhibitors,” Air Force Office of Scientific Research, Contract No. F49620-96-1- 0479, Final Report (2001).
154. G.S. Frankel and R.L. McCreery, "Inhibition of Al Alloy Corrosion by Chromates,” Inferface, 10, 34 (2001).
155. H.E. Townsend and R.G. Hart, J. Electrochem. Soc., 131, 1345 (1984).
156. M. Sakashita and N. Sato, Corrosion, 35, 351 (1979).
157. N. Sato, Corrosion, 45, 354 (1989).
158. N. Sato, Corrosion Science, 31, 1 (1990).
159. M. Kendig, E. Addison, and S. Jeanjaquet, J. Electrochem. Soc., 146, 4419 (1999).
160. F.W. Lytle, R.B. Greegor, G.L. Bibbins, K.Y. Blohowiak, R.E. Smith, and G.D. Tuss, Corrosion Science, 37, 349 (1995). 57 161. J. Zhao, G.S. Frankel, and R.L. McCreery, J. Electrochem. Soc., 145, 2258 (1998).
162. L. Xia, E. Akiyama, G. Frankel, and R. McCreery, J. Electrochem. Soc., 147, 2556 (2000).
163. D. Chidambaram, C.R. Clayton, and G.P. Halada, J. Electrochem. Soc., 150, B224 (2003).
164. V. Laget, C.S. Jeffcoate, H.S. Isaacs, and R.G. Buchheit, J. Electrochem. Soc., 150, B425 (2003).
165. P. Campestrini, G. Goeminne, H. Terryn, and J. Vereecken, J. Electrochem. Soc., 151, B59 (2004).
166. W. Zhang, B.L. Hurley, and R.G. Buchheit, J. Electrochem. Soc., 149, B357 (2002).
167. Y. Yoon and R.G. Buchheit. "The Effect of Heat Treatment on the Corrosion behavior and Chromate Conversion Performance of Al-Zn-Mg-Cu Alloy 7475,” NACE, Houston, TX (2001).
168. Y. Yoon, V. Laget, and R.G. Buchheit. "The Effect of Artificial Heat Treatment on the Chromate Conversion Performance of Al-Zn-Mg-Cu alloys,” Tri-Service Corrosion, San Antonio, TX (2002).
169. L.J. Bailin, P. Fitzpatrick, and M.J. Joyce, "Evaluation of Unpainted Alodine Chromate Conversion Coatings for Corrosion Resistance and Electrical Conductivity,” Lockheed Missiles and Space Co., Report F035575 (1985).
170. R. P. Wei, C. M. Liao, and M. Gao, Metall. Mater. Trans. A., 29A, 1153 (1998).
171. D. G. Harlow and R. P. Wei, Engineering Fracture Mechanics, 59, 305 (1998).
172. Q. Meng and G. S. Frankel, J. Electrochem. Soc., 151, B271 (2004).
173. Q. Meng and G. S. Frankel, Surface and Interface Analysis, 36, 30 (2004).
174. P. Campestrini, H. Terryn, J. Vereecken, and J. H. de Wit., J. Electrochem. Soc., 151, B359 (2004).
58 TABLES
Aluminum Alloy Temper designations Identification 4 digit Main Suffix Second suffix First suffix digit series elements letter digit
F (As-fabricated)
O (Annealed-wrought product only) 2 (1/4 Hard) 1 (Cold-worked only) H 4 (1/2 Hard) (Cold 2 (Cold-worked and partially worked- 6 (3/4 Hard) annealed) strain 1xxx Pure Al 8 (Hard) hardened) 3 (Cold-worked and stabilized) 2xxx Cu 9 (Extra Hard) W (Solution heat-treated) 3xxx Mn 1 (Partial solution + natural aging) 4xxx Si 2 (Annealed cast products only) 5xxx Mg 3 (Solution + natural aging) 6xxx Mg & Si 4 (Solution + artificial aging) 7xxx Zn T (Heat 5 (Artificially aged only) 8xxx others treated) 6 (Solution + artificial aging)
7 (Solution + stabilizing, over-aged)
8 (Solution + cold work + artificial aging)
9 (Solution + artificial aging + cold work)
Table 2.1: Aluminum alloy and temper designation systems [11, 13]
59
IADS Zn Mg Cu Si Fe Mn Cr Ti Other AA 6.8- 2.6- 1.6- 0.18- 7001 0.35 0.04 0.20 0.20 8.0 3.4 2.6 0.35 3.8- 1.0- 0.2- 7004 0.05 0.25 0.35 0.05 0.05 0.10-0.20 Zr 4.6 2.0 0.7 4.0- 1.0- 0.2- 0.06- 0.01- 7005 0.10 0.35 0.40 0.08-0.20 Zr 5.0 1.8 0.7 0.2 0.06 5.5- 2.1- 0.6- 0.1- 7009 0.20 0.20 0.10 0.20 0.25-0.40 Ag 6.5 2.9 1.3 0.25 5.7- 2.2- 1.5- 7010 0.10 0.15 0.30 0.05 0.11-0.17 Zr 6.7 2.7 2.0 4.0- 0.8- 0.45- 7016 0.10 0.12 0.03 0.03 5.0 1.4 1.0 4.0- 2.0- 0.05- 0.10-0.25 Zr, 7017 0.20 0.35 0.45 0.35 0.15 5.2 3.0 0.5 0.15min Mn + Cr 3.5- 2.3- 0.1- 0.15- 7039 0.10 0.30 0.40 0.10 4.5 3.3 0.4 0.25 7.2- 2.0- 1.2- 0.1- 7049 0.25 0.35 0.20 0.10 8.2 2.9 1.9 0.22 7.2- 2.0- 1.2- 0.1- 7149 0.15 0.20 0.20 0.10 8.2 2.9 1.9 0.22 5.7- 1.9- 2.0- 7050 0.12 0.15 0.10 0.04 0.06 0.08-0.15 Zr 6.7 2.6 2.6 5.9- 2.0- 1.9- 7150 0.12 0.15 0.10 0.04 0.06 0.08-0.15 Zr 6.9 2.7 2.5 5.1- 2.1- 1.2- 0.18- 7075 0.40 0.50 0.30 0.20 0.25 Zr + Ti 6.1 2.9 2.0 0.28 5.1- 2.1- 1.2- 0.18- 7175 0.15 0.20 0.10 0.10 6.1 2.9 2.0 0.28 5.2- 1.9- 1.2- 0.18- 7475 0.10 0.12 0.06 0.06 6.2 2.6 1.9 0.25 6.3- 2.4- 1.6- 0.18- 7178 0.40 0.50 0.30 0.20 7.3 3.1 2.4 0.35 3.8- 2.9- 0.4- 0.1- 0.1- 7079 0.30 0.40 0.10 4.8 3.7 0.8 0.3 0.25 3.8- 2.9- 0.4- 0.1- 0.1- 7179 0.15 0.20 0.10 4.8 3.7 0.8 0.3 0.25
Table 2.2: Compositions of 7xxx series aluminum alloys [11, 13]
60
Zn Mg Zn + Mg Zn/Mg Alloy (%) (%) (%) ratio 7104 4.0 0.7 4.7 5.7 7008 5.0 1.0 6.0 5.0 Medium-strength 7011 4.7 1.3 6.0 3.7 weldable Al-Zn- 7020 4.3 1.2 5.5 3.6 Mg alloys 7005 4.5 1.4 5.9 3.2 7004 4.2 1.5 5.7 2.8 7051 3.5 2.1 5.6 1.7 Higher strength 7003 5.8 0.8 6.6 7.2 weldable Al-An- 7046 7.1 1.3 8.4 5.5 Mg alloys 7039 4.0 2.8 6.8 1.4 7049 7.7 2.5 10.2 3.1 7050 6.2 2.3 8.5 2.7 7010 6.2 2.5 8.7 2.5 High-strength Al- 7475 5.7 2.3 8.0 2.5 Zn-Mg-Cu alloys 7001 7.4 3.0 10.4 2.5 7075 5.6 2.5 8.1 2.2 7079 4.3 3.3 7.6 1.3
Table 2.3: Zinc and magnesium contents and Zn/Mg ratio in 7xxx aluminum alloys [11]
PFZ Grain Boundary Precipitates (GBPs) Matrix Precipitates Temper Width d λ NA f p p p Type Size (nm) (nm) (nm) (nm) (PPTs/µm2) (Pct) 7475-T6 28 18.8±4.5 25.0 400 13 GP+η’+η 5.4±1.3 7075-T6 29 24.5±5.6 27.1 340 18 GP+η’+η 4.9±1.0 7475-RRA 37 33.1±6.8 34.1 215 21 η’+η 8.9±1.6 7475-T73 45 45.6±12.2 44.7 125 23 η’+η 11.6±2.8 dp: precipitate diameters λp: intraplanar spacings NA: numbers per unit area fp: areal fractions * More than 100 GBPs in five different grain-boundary areas were measured.
Table 2.4: Characteristics of microstructures for various heat treatment conditions [53]
61 FIGURES
Figure 2.1. Relationships of tensile strength, yield strength and SCC resistance in the 7xxx series (From Ref. [10]).
62
Figure 2.2. Pourbaix diagram (Potential versus pH diagram) for aluminum with Al2O3•3H2O film at 25 °C (From Ref. [12]).
Figure 2.3. Schematic showing the structure of the oxide film formed on unalloyed aluminum in dry air, Al = Aluminum, 1 = Surface layer, 2 = Mixed oxides, 3 = Pores, 4 = Barrier layer, and 5 = Heterogeneous components (From Ref. [13]).
63
Figure 2.4. Schematic view of aluminum oxide film on rolled product (From Ref. [1]).
Figure 2.5. Schematic of a galvanic corrosion cell involving cathodic activity at a Cu- containing IMCs (From Ref. [17]).
64
Figure 2.6. Redox activity on IMCs. (a) SECM images of AA2024. (b) SEM images of the same region as in (a) (From Ref. [17]).
Figure 2.7. The aluminum corner of the Al-Mg-Zn equilibrium phase diagram (From Ref. [3]). 65
Figure 2.8. Typical metastalbe pit transients observed on 302 stainless steel in 0.1 M NaCl at 420 mVSCE (From Ref. [68]).
Figure 2.9. Influence of potential on the pit growth, pit depth, and pit current density of Al foil in 0.01 M NaCl at pH 11 (From Ref. [73]).
66
Figure 2.10. Regimes of crevice corrosion and inhibition of iron in 0.25 mm crevices in various chromate/chloride solutions (From Ref. [118]).
Figure 2.11. Schematic diagram showing various regions in the section of the passive film developed on aluminum after immersion in chromate/dichromate solution (From Ref. [125]). 67
2- Figure 2.12. Time series showing the effect of CrO4 inhibitor on aluminum of the anodic current spikes associated with metastable pitting at potentiostatic potential (-0.5 VSCE) (From Ref. [129]).
Figure 2.13. AFM image of AA2024-T3 after 3 s to a chromate solution. A: Al-Cu-Fe- Mn IMC (From Ref. [138, 141]).
68
inet iano icat
Figure 2.14. The measured net current density, the corrosion current density, and the true cathodic current density for (a) pure Al, (b) Al-4%Cu, and (c) Al-Cu IMC as a function of dichromate concentration (From Ref. [143]).
Figure 2.15. Schematic of CCC inhibition mechanism over IMC and matrix. Chemisorption of cyanide or ferricyanide might inhibit redox mediation and reduce chromate reduction (From Ref. [144]). 69
Figure 2.16. Schematic of CrIII film formation on copper (From Ref. [145]).
Figure 2.17. Uniform formation and growth model for CCCs by Katzman (From Ref. [108]).
70
Figure 2.18. Non-uniform formation and growth model for CCCs by Brown and Wood (From Ref. [148]).
Figure 2.19. CCC formation showing polymerization of Cr3+ hydroxide and adsorption of Cr6+ by Xia and McCreery (From Ref. [135]). 71
Figure 2.20. Structure model of CCCs on Al by Treverton and Davis (From Ref. [133]).
Figure 2.21. Structure model of CCCs on Al by Townsend and Hart (From Ref. [155]).
72
Figure 2.22. Structure model of CCCs on AA2024 by Hughes (From Ref. [107]).
Figure 2.23. A bipolar membrane model for passivity of metals by Sato (From Ref. [158]).
73
Figure 2.24. Model for dynamic adsorption of Cr6+ by insoluble Cr3+ oxy-hydroxide by Xia (From Ref. [162]).
Figure 2.25. A new duplex model incorporating both the Sato-bipolar model and McCreery-pit repassivation model by Clayton (From Ref. [153, 163]).
74
Figure 2.26. Schematic depiction of condensation and dehydration reactions in Cr(OH)3 (From Ref. [164]).
Figure 2.27. Schematic drawing of the gel near the Al surface (From Ref. [165]).
75
CHAPTER 3
THE EFFECT OF MICROSTRUCTURAL HETEROGENEITY OF CHROMATE
CONVERSION COATING FORMATION ON Al ALLOY 7X75
3.1 INTRODUCTION
There has been considerable effort to understand the formation and breakdown of chromate conversion coatings (CCCs), which are extremely effective and widely used to enhance the corrosion resistance of high strength aluminum alloys [1-3]. However, there has been pressure to replace chromate as a conversion coating ingredient because it is toxic [4].
CCCs are formed on metal surfaces as a result of chemical and electrochemical processes in an aqueous solution of chromic acid as a coating formation agent, hydrofluoric acid as an activator, and potassium ferricyanide as an accelerator [3, 5]. The formation of CCCs is affected by complex microstructures in high strength Al alloy substrates resulting in a coating whose thickness, composition and protectiveness varies across the alloy surface. In the case of 7xxx Al-Zn-Mg-Cu alloys, one of the main factors affecting coating formation and breakdown is the dispersion of coarse intermetallic
76
particles. Extensive studies have been carried out characterizing the chemistry and structure of CCCs on 2xxx aluminum alloys. Less effort has been given to the characterizing the CCCs on 7xxx aluminum alloys; particularly in the relationship between microstructure and CCC formation [6, 7]. Meng et al. characterized CCCs on
AA7075-T6 using focused ion beam (FIB) sectioning and scanning transmission electron microscopy (STEM). They found that coating thickness on the matrix was much greater than that on Al2CuMg, Mg2Si and Al3(Fe Cu) intermetallics and coating chemistries on
Al2CuMg, Mg2Si and Al3(Fe Cu) intermetallics were a mixed Al/Mg/Cr, Mg/Cr and
Al/Fe/Cr oxides, respectively [8].
CCC formation on AA2024 (Al-4.4Cu-1.5Mg-0.6Mn) has been thoroughly studied and there are several core findings with respect to the relationship between CCC formation and alloy microstructure. CCC formation is strongly affected by the dispersion of constituent particles. These can be classified into two types. One type is a Cu and Mg- containing Al2CuMg. This phase can be electrochemically active with respect to the surrounding matrix. The other type comprises Cu, Fe and/or Mn-containing IMCs such as
Al2Cu, Al2Cu2Fe, Al7Cu2Fe, Al12Si(FeMn)3, Al20Cu2(FeMn)3, and Al20Cu3Mn3 [9-13].
These phases tend to be noble compared to the matrix phase.
On the basis of Auger electron spectroscopy (AES) and X-ray photoelectron spectroscopy (XPS) results, Hagans et al. suggested that the CCC formation rate on IMCs decreases as follows: Alloy matrix > Cu and/or Mg IMCs > Cu, Fe and/or Mn IMCs for a coating time of 3 minutes [5]. However, the results of Waldrop et al. suggest that initial coating formation on some IMCs is more rapid than on the matrix. Using an ex-situ atomic force microscopy (AFM) approach, the initial coating formation rate after less
77
than 10 seconds was found to decrease as: Cu, Fe and/or Mn IMCs > alloy matrix > Cu and/or Mg IMCs [12]. The interpretation of this result was that the rapid nucleation and growth of the coating on Cu, Fe and/or Mn IMCs was due to accelerated cathodic deposition of chromate, while slower CCC growth on Cu and/or Mg IMCs was related to inhibited cathodic reduction of chromate.
McGovern et al. used Raman spectroscopy to characterize the intensity of CCC formation on a range of real and synthetic intermetallic compounds. They found that the integrated 860 cm-1 Raman intensity and coating thickness had a linear relationship, and that the 860 cm-1 Raman intensity decreased as copper content increased in IMCs [14].
After 5-minute immersion, CCC formation rate decreased as follows: Alloy matrix > Cu,
Fe and/or Mn IMCs > Cu and/or Mg IMCs [14]. This result was somewhat at odds with that of coating formation rate on IMCs determined in other studies using AES and XPS
[5, 14].
In the present study, the effect of microstructure on coating formation was systemically investigated using AFM approaches. Variations in coating morphology and
Volta potential were characterized for the major IMC particles found in 7075 and 7475 aluminum alloys. Results confirm that coating formation is inhibited on IMCs on these alloys from the very earliest stages of coating deposition. Nonetheless, galvanic interactions between particles and matrix are shown to be reduced by CCCs, and coating formation on particles is not strongly dependent on alloy temper. Based on the results of this study and information published in relevant literature, coating formation on 7xxx aluminum alloys is described.
78
3.2 EXPERIMENTAL
3.2.1 Materials and chemicals
The alloys used for this study were commercially supplied AA7475 and AA7075.
The compositions of the AA7475 and AA7075 alloys were determined by inductively coupled plasma-mass spectrometry (ICP-MS) using Perkin-Elmer Sciex ELAN 6000. On a weight percentage basis, the composition of AA7475 was Zn 5.64, Mg 2.27, Cu 1.35,
Cr 0.22, Fe 0.05, Si 0.021, Mn 0.001, and Ti 0.044 with the balance being Al. The composition of AA7075 was Zn 5.52, Mg 2.44, Cu 1.41, Cr 0.17, Fe 0.28, Si 0.019, Mn
0.029, and Ti 0.033 with balance being Al. The primary difference in these compositions was the lower impurity level of Fe and Mn in AA7475.
Prior to coating, alloys were heat-treated to the T6 or T7 temper (Table 3.1). In preparation for coating experiments, samples were ground using SiC abrasive paper through 1200 grit and polished using 3, 1, and 0.25 µm diamond pastes on a nylon cloth.
All samples were then degreased ultrasonically with ethyl alcohol. During preparation of samples, non-aqueous lubricating slurry (Blue Lube by Struers) was used to minimize any localized corrosion. All chemicals were reagent grade and were obtained from commercial vendors. All solutions were prepared with deionized water of 18 MΩ⋅cm resistivity. Conversion coatings were applied using Alodine® 1200S (Henkel Surface
Technologies) according to manufacturer’s specification. Alodine® 1200S generally contains chromium oxide, fluoride, and ferricyanide and adjusted 1.5 ~ 1.7 pH by additional HNO3 at room temperature. Samples were coated in chromate bath for 5, 10,
30, 60, 120, or 180 seconds. After coating, specimens were rinsed with deionized water
79
and dried for 24 hours in lab air before any further testing. Samples were not degreased or deoxidized prior to conversion coating.
3.2.2 Atomic force microscopy
Atomic force microscopy (AFM) and scanning Kelvin probe force microscopy
(SKPFM) are used to characterize the surface topography and the Volta potential in regions of interest [15-18, 21, 22]. These characterizations were conducted using a Veeco
DimensionTM 3100 Atomic force microscope. During the first scan, the surface topography was recorded in tapping mode. In the second scan, the Volta potential was recorded in lift mode (Figure 3.1). The tip used in these experiments was a conductive
Pt/Ir-coated silicon cantilever, which was oscillated at its resonant frequency by a piezoelectric element, as it scanned over the sample surface [21, 22]:
1 k resonant frequency = Eqn. 3-1 2π m k: spring constant [N/m] m: mass [kg]
AFM cantilevers exhibit high resonant frequencies due to their high flexibility.
Typical spring constants of cantilevers range from 0.1 to 1 N/m. The typical pyramidal tip has about 10 N/m spring constant and is much stiffer than the cantilever. For this study, the range of resonant frequencies was 72-96 kHz with a 222 µm length cantilever.
This spring constant of cantilever varies with external force gradient. Electronics adjust the distance between tip and sample surface in order to keep resonant frequency constant.
80
By maintaining a constant resonant frequency, the topographic image of the sample surface is mapped in the first scan of tapping mode using beam deflection with position- sensitive photodetector (Figure 3.2).
In order to measure the Volta potential, the cantilever was lifted off the surface to a distance of about 100 nm [21, 22]. To measure the Volta potential, an AC voltage was directly applied to the cantilever tip while turning off the piezoelectric element to suppress the mechanical oscillation of the cantilever. This oscillating voltage created an oscillating electric force at the resonant frequency on the cantilever. When a potential difference between the tip and the sample surface existed, the oscillating force was given by [21, 22]:
dC F = ∆V V Eqn. 3-2 dz dc ac dC/dz: vertical derivative of the tip/sample capacitance
∆Vdc: DC voltage difference between the tip and the sample
Vac: amplitude of the oscillating voltage applied to the cantilever tip
The oscillating electric force on the cantilever depends on the product of the AC drive voltage and the DC voltage difference between the tip and the sample. When the tip and sample are at the same DC voltage (∆Vdc → 0), there is no oscillating force on the cantilever. At this point the applied tip voltage has the same magnitude as the unknown
Volta potential, but an opposite sign. When this process is carried out in a point by point basis, a map of the Volta potential is constructed by recording the voltage applied to the cantilever tip, and inverting it.
81
The Volta potential is typically measured in air and has been used to study localized corrosion susceptibility. The basis for this approach is an empirically derived linear relationship between the Volta potential in air and the corrosion potential of an electrode surface in dilute electrolytes [15-18]. Therefore, the difference of Volta potential indicates the extent of possible galvanic interaction with the surrounding matrix.
3.3 RESULTS
3.3.1 Identification of the main constituent particles in AA7x75
SEM/EDS analysis showed that the main constituent particles in AA7075 and
AA7475 were the same. Changes in particle chemistry upon artificial aging were not observed. The overall volume fraction of constituent particles in AA7075 was higher than in AA7475 (Figure 3.3). Figure 3.4 shows SEM and EDS of as-polished AA7075-T6 surface. The typical microstructures consist of several coarse intermetallic compounds
(IMCs). EDS analysis indicates that there were three main types of IMCs: 1) blocky and irregularly shaped particles containing mainly Al, Fe and Cu, and occasionally Mg, Zn,
Mn, Cr and Si; 2) smaller spherically shaped Al2CuMg particles; and 3) blocky and irregularly shaped Mg2Si particles. These IMCs aligned along the rolling direction.
Al13CuFe3, Al23CuFe4, and (Al,Cu)6(Fe,Cu) were found to be the most dominant intermetallics of the Fe and Cu containing particles [19, 20]. Particles with stoichiometries equivalent to Al7Cu2Fe were not commonly observed in these samples.
Due to the variation in composition observed for these particles, they are designated as a class of particles and are labeled “Al-Fe-Cu” hereafter. In SEM images, Mg2Si had dark contrast while Al-Fe-Cu IMCs and Al2CuMg had bright contrast as shown in Figure 3.4
82
(a). Some Al-Fe-Cu IMCs were small and round in shape. As a result, some Al-Fe-Cu and Al2CuMg IMCs were difficult to distinguish in SEM because the particles exhibited similar shapes and contrasts. The Al-Fe-Cu particles had sizes ranging from 1~20 µm.
The Mg2Si particles had sizes ranging from 1~20 µm and Al2CuMg particles had sizes ranging from 0.5~5 µm.
3.3.2 AFM characterization of intermetallic compounds in as-polished AA7075
The effect of alloy temper on the topography and Volta potentials of IMCs in
AA7075 was characterized. Essentially, there was no significant difference of the morphology and size of these IMCs between AA7075-T6 and AA7075-T7, nor was there a difference in IMC of Volta potentials between these two tempers (Table 3.2). To illustrate the general characteristics, results from AA7075-T7 samples are given.
The three main types of coarse IMC particles in the alloy exhibited characteristic topography and Volta potentials in AFM imaging (Figure 3.5). For the polishing procedures used here, the Al-Fe-Cu particles protruded from the surface by about 70-80 nm due to their higher hardness and lower polishing rate compared to alloy matrix [21,
22]. Al2CuMg and Mg2Si IMCs were difficult to detect in topographic images due to their higher polishing rate than that of alloy matrix [21, 22].
The Volta potential of Al-Fe-Cu particles in AA7075-T7 was about 450 mV (±
100 mV) more positive than the surrounding matrix phase. The Volta potential differences of Al2CuMg and Mg2Si compared to the surrounding matrix were about 290 mV (± 50 mV) and -230 mV (± 50 mV) respectively. The Volta potential of Al-Fe-Cu
IMC was always higher than that of Al2CuMg IMC by about 100 mV. Volta potential
83
map clearly showed that the Al-Fe-Cu IMC and Al2CuMg had much higher potentials than surrounding matrix, while Mg2Si exhibited much lower potentials. This indicates that Al-Fe-Cu IMC and Al2CuMg were more cathodic and Mg2Si was more anodic than aluminum matrix after polishing. Andreatta et al. reported that the Volta potentials of Al-
Fe-Cu IMCs and Mg2Si IMCs with respect to the matrix in AA7075 were 250 ~ 400 mV and -100 ~ -180 mV respectively [23]. Our measurements are consistent with these earlier findings.
S phase in AA7075 can be compared to that in AA2024. In AA2024, the Volta potential of Al2CuMg IMCs was reported to be –500 mV versus Ni (± 50 mV) [21]. To make a comparison to this value, the measured Volta potential compared to matrix was converted to a potential measured against pure Ni [21, 22]. The Volta potential of Ni measured with Pt/Ir coated cantilever used in this study was –0.7 V yielding an offset application to our measurements of VNi = Vmeas + 0.7 V. The Volta potential of matrix measured with Pt/Ir cantilever was –1.8 V indicating that the potential of matrix with respect to pure Ni was –1.1 V and the Volta potential of Al2CuMg IMCs in AA7075 was
-800 to -900 mVNi. These values are somewhat more active than those reported for S phase in AA2024 [21].
3.3.3 AFM characterization of 3-minute chromate conversion coated surfaces on
intermetallic compounds of AA7x75
After 3-minute CCCs were formed on AA7x75 alloy surface, the Volta potential differences and coating thickness differentials on IMC particles took on characteristic values. The Volta potential differences were drastically decreased, which suggests that
84
part of the CCC protection mechanism is due to a reduction in the local variations in surface potential associated with IMCs and suppression of local cathodic corrosion. The coating thickness differentials from AFM surface profile measurement revealed that conversion coating formation on IMCs was strongly suppressed and coatings were thinner although these coatings on IMCs were still protected. Structural defects at the peripheries of Al2CuMg were initiation sites for corrosion.
Figure 3.6 shows the back-scattered electron image, topography, and Volta potential maps of the same region in Figure 3.5 after a 3-minute CCC treatment. After application of a 3-minute CCC on these three IMCs, there was significant alteration of surface height and Volta potential. The CCC on the matrix phase was mud-cracked due to dehydration. However, mud-cracking was not observed on IMC particles. The coating thickness differential ranged from –40 to –70 nm for Al-Cu-Fe particles, 0 to –140 nm for
Mg2Si and -140 to –180 nm for Al2CuMg. Meng et al, also found that coating on matrix was thicker than coatings on intermetallic particles [8]. The coating thicknesses differential were about -270 nm for Al2CuMg, -280 nm for Mg2Si, and -250 to -200 nm for Al3(Fe Cu) particles [8]. These coating thickness differentials include a contribution from particle dissolution and could represent a coating defect structure that can affect subsequent coating protection.
Despite the fact that CCCs are thin on IMCs, the presence of the coating significantly reduces local variations in surface potential suggesting that the potential for particle-matrix galvanic coupling and local cathodic corrosion is minimized. For example the Volta potential difference between Al-Fe-Cu particles and the matrix is reduced from
450 mV to –23 mV by application of a 3-minute CCC. For Mg2Si the Volta potential
85
difference changes from –230 mV to 5 mV. For Al2CuMg the difference of 290 mV is changed to –20 mV. In each of these cases, the potential difference was reduced and in some cases the polarity of the surface potential relationship was reversed.
Three-dimensional surface morphology of 3-minute chromate coated IMCs of
AA7075-T7 is shown in Figure 3.7. It is indicated that there were trenches around Al-Fe-
Cu and Mg2Si IMCs. The evolution of trenching around particle gives potentially valuable, though indirect, information on particle-matrix galvanic interactions and local electrochemical activity during coating formation. Trenches may also be important as sites of coating breakdown and pitting when the coated surface is exposed to an aggressive environment. Figure 3.8, 3.9 and 3.10 show the individual line profile scans for each IMC. Trenches associated with Al-Fe-Cu particles were normally confined to the immediate periphery of the particle (Figure 3.8). Whether the trench exists in the matrix, in the particle or both is not evident from AFM images. In comparison, trenches are broad on Mg2Si particles. In Figure 3.9, the height of the Mg2Si particle is nearly the same as that of the surrounding matrix. In some cases, the surface height decreases gradually from the particles center to the particles periphery in a manner suggestive of particle dissolution. In other cases, trenching was confined to the immediate vicinity of the Mg2Si particle-matrix interface, which will be shown later. Trenching at Al2CuMg particles was not evident in AFM images due to the small size of these particles in this alloy (Figure 3.10).
Conversion coating formation in the vicinity of IMC particles did not depend on alloy temper. No significant differences in the morphology of CCC formation, trenching, surface height differential, or Volta potential difference were observed at IMC particle in
86
AA7075-T6 and –T7 ( Table 3.3). For example, Figure 3.11 shows the AFM images of
Mg2Si of as-polished AA7075-T6. The surface height of Mg2Si was almost flat and the
Volta potential was about -120 mV compared to matrix, which was consistent with those results of AA7075-T7. Figure 3.12 shows the Volta potential maps of the same region in
Figure 3.11 after a 3-minute CCC treatment on AA7075-T6. The line profile of Mg2Si revealed the trench around the particle. Trenches associated with Mg2Si were confined to the immediate periphery of the particle compared to those of Mg2Si in AA7075-T6. This indicates some local particle dissolution, perhaps by selective Mg dissolution as suggested by Meng [8].
Figure 3.13 shows the maximum trench depths on Mg2Si and Al-Fe-Cu IMCs as function of coating time. The trench depth was taken as the difference between center of each IMC, which was usually “high”, and the edge of the IMC, which was usually “low”.
The maximum trench depth was that observed out of 5 line profiles among at least three particles. As coating time increased, the depths of trench on Mg2Si continuously increased. However, trench depths at Al-Fe-Cu IMCs increased rapidly then remained steady. This result suggests that the matrix around Al-Fe-Cu IMCs dissolved actively but then passivated.
3.3.4 Formation of chromate conversion coating at short times
The topographic image and Volta potential map for Al-Fe-Cu particles of
AA7475-T6 before and after 5-second CCC using commercial Alodine 1200S are shown in Figure 3.14. Because the IMCs discussed here are found in both AA7075 and AA7475,
87
the results are expected to apply equally well for understanding CCC formation on both alloys.
A comparison of Figure 3.14 (a) and (b) shows that the combination of particle dissolution and coating formation reduces the particle-matrix height difference from +26 nm to –55 nm for Al-Fe-Cu particles. The Volta potential difference is reduced from
+260 mV to –20mV. This result suggests that particle-matrix galvanic coupling may be reduced even though the coating on the Al-Fe-Cu particles is likely to be thin at this short coating time.
The matrix-particle coating thickness differential after 5 seconds of immersion is essentially the same as that observed after 3 minutes even though the overall coating thickness after 3 minutes is probably much greater ( Table 3.4).
Figure 3.15 shows the surface morphology of Al-Fe-Cu IMCs in Figure 3.14(b) compared to surrounding alloy matrix at the higher magnification. The shapes of chromate reaction product on Al-Fe-Cu IMCs took the form of elongated nodules that were about 50 nm in size. The shapes of chromate reaction product on matrix were also rounded nodules that were much smaller than those on Al-Fe-Cu IMCs. A trench was also observed along part of the particle-matrix interface after this short immersion period.
3.4 DISCUSSION
3.4.1 Heterogeneity in CCC formation
The formation of chromate conversion coating on different intermetallic compounds and the alloy matrix in AA7075 and AA7475 was not uniform. The Volta potential differences were significantly reduced between IMCs and matrix, and coating
88
on IMC particles were thinner than on the matrix. Despite their relative thinness, coatings on IMC particles do provide some measure of corrosion protection.
The formation of CCCs is generally accepted to involve redox reaction between chromate or dichromate in solution and aluminum [12, 24]. The chromate or dichromate oxidizes the aluminum in the presence of fluoride and the hexavalent chromium (Cr6+) is reduced to trivalent chromium (Cr3+). These Cr6+/Cr3+ redox reactions consume protons and increase the local pH. The pH increase triggers the precipitation of an amorphous mixture of hydrated aluminum and chromium oxides [5]. This reaction scheme applies well to coating formation on aluminum.
Because the intermetallic particles present in aluminum alloy may be more anodically or cathodically active than the Al matrix and because their chemistry is different than the matrix, the CCC reaction scheme may not apply as well [10].
3.4.2 Coating formation on Al-Fe-Cu IMC particles
Cathodic sites of Al-Fe-Cu IMCs might be expected to accelerate Cr6+/Cr3+ redox reactions and lead to enhanced chromate deposition compared to the matrix. However,
AFM profiles suggested that CCC is in fact thinner on Al-Fe-Cu IMC particles than on the matrix after a 3-minute conversion coating (Figure 3.8). Campestrini et al. have described the CCC formation as a distinct nucleation and growth process [25]. They explain that CCC formation nucleates preferentially on cathodic IMCs due to more noble potential leading to the initial coating formation on these cathodic sites [25]. The initial compact and dense chromate deposition makes these particles less active and slows further chromate reduction. This suppresses an IMC-stimulated component of CCC
89
growth. Meanwhile, the conversion coating formation on matrix continues due to the porous nature of the incipient coating [25].
In the present study, it was also observed that coatings were thinner on Al-Fe-Cu
IMCs than that on surrounding matrix even after 5-second conversion coating application
(Figure 3.14).
A K3Fe(CN)6 addition in Alodine 1200S solution is present as an accelerator that
3- promotes CCC growth. Xia proposed that ferricyanide [Fe(CN)6 ] acted as a redox
4- mediator that rapidly oxidized aluminum while being reduced to ferrocyanide [Fe(CN)6 ].
Ferrocyanide was then rapidly oxidized back to ferricyanide while Cr6+ was reduced to
Cr3+ [26, 27]. Hagans et al. observed that ferricyanide was distributed on Al-Cu-Mg
IMCs, and the corrosion of these particles was inhibited by the formation of copper- ferricyanide on surface layer [5, 28]. McGovern et al. suggested that the adsorption of ferricyanide on Al-Cu-Mg IMCs inhibited CCC formation [14]. Vasquez et al. suggested that inhibition of CCC growth on Al2Cu and Al2CuMg was related to passivation of surface by cyano species [26]. Meng et al. observed that CCC thickness on Al-Fe-Cu
IMCs in AA7075 was thinner than that on alloy matrix and much more Fe was found on
Al-Fe-Cu IMCs than on matrix [8]. The results of the present study are consistent with the idea that the chromate reduction reaction was inhibited by ferricyanide on Al-Fe-Cu
IMCs [27].
3.4.3 Coating formation on Al2CuMg particles
Al2CuMg is present in AA7075 as a constituent particle. In some respects, S phase in AA7x75 was observed to behave like S phase in AA2024. The Volta potential of
90
Al2CuMg IMCs was always higher than surrounding matrix after polishing. Schmutz et al. have suggested that the nobility of Al2CuMg was related to surface oxide film of altered composition. It was shown that the Volta potential decreased and these particles immediately dissolved in 0.01 M NaCl after removing the surface oxide film by AFM scratching or Ar plasma sputter-etching [21, 29]. Buchheit et al. studied the electrochemical activity of both bulk-synthesized Al2CuMg and Al2CuMg particles in
AA2024-T3 and suggested Cu enrichment by dealloying of Al2CuMg [11, 30, 31].
In a CCC bath, S-phase particles are likely to selectively dissolve Mg and Al.
Dissolution of these elements may occur at a high rates [30, 31]. This leads to Cu enrichment and ennoblement. The Cu-rich IMC remnant then develops a surface coating in a manner consistent with that reported for other Cu-rich IMCs. The coating is thin and compact and may be enriched in Fe and cyano species. This surface layer is passivating and inhibits further coating formation. The thinnest coating of all appears to form on S phase particles perhaps because of a high initial dissolution rate [8].
Volta potential measurements indicate that the coating that does form extinguishes the surface potential difference between ennobled S phase and the matrix.
Nontheless, this does not seem to be enough to prevent S phase from serving as a site of preferential breakdown. As Figure 3.16 suggests, pits are preferentially nucleated at peripheries of S-phase particles when a 3-minute CCC on AA7075-T6 surface is exposed to 0.5 M NaCl for 24 hours. Figure 3.16 shows a number of different particle types. In this example, only S-phase particles appear to have initiated pitting at their periphery.
91
3.4.4 Coating formation on Mg2Si particles
The Volta potential of as-polished Mg2Si IMCs was lower than surrounding matrix (Figure 3.5 and Figure 3.11). This Mg2Si phase is known to be active under a range of environmental conditions [10, 32]. Upon immersion in the chromate bath, the
Mg2Si particles exhibited anodic dissolution probably by dealloying of Mg [8]. The depth of the trench at the periphery of Mg2Si increased continuously during coating formation.
Meng found that ferricyanide was absent on Mg2Si [8]. It is possible that the adsorption of ferricyanide did not occur on Mg2Si and attendant passivation was absent, which could explain why the depth of trench at Mg2Si particles increased as coating immersion time increased.
3.4.5 Formation of trenches around constituent particles
The formation of trenches may be important for understanding the CCC formation and its subsequent protectiveness. Trenches were observed at the peripheries of Al-Fe-Cu
IMCs and Mg2Si (Figure 3.7). Even at the early coating formation, trenching was observed in the vicinity of Al-Fe-Cu IMCs (Figure 3.15). Some authors have suggested that a copper-depleted zone forms during solidification around Cu-containing IMCs [13,
33]. Brown et al. suggested that the anodic dissolution of aluminum at copper-depleted zones supported the cathodic reactions such as chromate reduction and possibly oxygen reduction and/or hydrogen evolution at the cathodic sites. These cathodic sites were the alloy matrix phase and Al-Fe-Cu IMCs [13]. They suggested that the anodic dissolution occurred with high current, which caused the trenches in the vicinity of the Al-Fe-Cu
IMCs due to the small anodic area [13].
92
Buchheit et al. and Meng et al directly measured Al, Cu and Mg line profiles across Al2CuMg particles using X-ray line scans and scanning transmission electron microscopy with nano-electron dispersive spectroscopy line profiling, respectively [8, 11].
It was found that there was no Cu depletion in the matrix around Al2CuMg particles [8,
11]. It is known that Al-Fe-Cu IMCs and Al2CuMg form during solidification and their chemistry or that of the immediately adjacent matrix is not affected by normal solution heat-treatment and procedures. In the present study, results are consistent with a simple local coupling effect between Al-Fe-Cu IMCs and the matrix, which caused the formation of trenches around particles; perhaps initiated at the particle-matrix interface.
Figure 3.13 shows that trenches at Al-Fe-Cu IMCs form rapidly, but then stop growing perhaps due to preferential chromate deposition on Al-Fe-Cu IMCs. In contrast, trenches at Mg2Si particles grew as coating time increased. Al2CuMg and Mg2Si may have initially experienced anodic dissolution by dealloying or incongruent dissolution followed later by a change from anodic to cathodic character relative to the matrix. This anodic dissolution will be shown in thin film analogs of Al2CuMg experiencing mass loss during coating formation using electrochemical quartz crystal microbalance (EQCM) in next chapter 4. After this reversal, the chromate deposition and the trenching may have occurred. The formation of trenches on each IMC during early coating formation (less than 5 seconds) is schematically illustrated in Figure 3.17.
3.4.6 CCC formation on heterogeneous AA7x75 surface
Figure 3.18 schematically illustrates CCC formation on AA7x75 based on the coating thickness differential observations. During early coating formation, the coating
93
thickness differentials on IMCs are thin compared to matrix. Trenches develop around
Mg2Si, Al-Fe-Cu IMCs and possibly around Al2CuMg particles.
Initially, chromate reduction occurs rapidly and forms thin coating on cathodic
Al-Fe-Cu IMCs with perhaps matrix dissolution near particles. Subsequently, ferricyanide adsorption occurs and suppresses further coating formation. CCC formation proceeds on the alloy matrix causing a coating thickness differential to develop at initial stage of coating formation. The anodic particles of Mg2Si and Al2CuMg are initially dissolved by dissolution of Mg and then CCCs form on these anodic particles.
Ferricyanide adsorbed on Al2CuMg inhibits further chromate reduction reaction.
However, at Mg2Si particle sites the chromate reduction reaction occurs unimpeded because ferricyanide does not preferentially absorb there.
Volta potential differences between particles and matrix are dramatically reduced by conversion coating. The reduction in potential difference occurs very early in the coating formation process. After initial coating formation, the subsequent growth of the coating appeared to be more or less uniform across the entire alloy surface.
3.5 SUMMARY
The formation of chromate conversion coating on Al-Fe-Cu, Mg2Si, Al2CuMg
IMCs in 7xxx aluminum alloys was characterized by using AFM and SEM/EDS. The results were as follows:
1. From Volta potential measurements on uncoated (or as-polished) AA7075-T6 and
T7, important galvanic relationships are:
Al2CuMg > aluminum matrix
94
Mg2Si < aluminum matrix
Al-Cu-Fe IMCs > aluminum matrix
After conversion coating, these Volta potential differences are significantly
reduced, suggesting that part of CCC corrosion protection is due to suppression of
local variations in surface potential and any attendant cathodic corrosion.
2. AFM surface profile measurements reveal that coating formation is strongly
affected by constituent particles. Conversion coating formation is strongly
suppressed on IMCs with noticeable differences in coating morphology compared
to the matrix. Trench-like defect appeared to from around particles during coating
formation due to perhaps matrix dissolution and dealloying or incongruent
dissolution. This results in a characteristic defect structure in CCCs on AA7075.
Al2CuMg particles appeared to be the primary initiation point for CCC
breakdown.
3. IMC- matrix coating thickness differentials after 5 and 180-second immersion in
the chromate bath are the same. However, total coating thickness is greater after
180-second immersion. This result suggests that coating formation rates across
the alloy surface are very different initially, but become more uniform as coating
progress.
4. AFM measurements show that alloy temper (T6 or T7) does not affect matrix-
particle coating thickness differences or Volta potential differences.
95
REFERENCES
1. A.E. Hughes, R.J. Taylor, and B.R.W. Hinton, Surface and interface analysis, 25, 223 (1997).
2. H.A. Katzman, G.M. Malouf, R. Bauer, and G.W. Stupian, Applications of Surface Science, 2, 416 (1979).
3. K.A. Korinek, "Chromate Conversion Coatings", ASM: Metals Park, OH (1987).
4. H.G. Seiler, "Handbook on Toxicity of Inorganic Compounds", Marel Dekker:New York (1988).
5. P.L. Hagans and C.M. Haas, "Chromate Conversion Coatings", in ASM Handbook, Vol. 5, Surface Engineering, ASM International: Metals Park, OH (1994).
6. F.W. Lytle, R.B. Greegor, G.L. Bibbins, K.Y. Blohowiak, R.E. Smith, and G.D. Tuss, Corrosion Science, 37, 349 (1995).
7. Q.J. Meng and G.S. Frankel, J. Electrochem. Soc., 151, B271 (2004).
8. Q.J. Meng and G.S. Frankel, Surf. Int. Anal., 36, 30 (2004).
9. G.S. Chen, M. Gao, and R.P. Wei, Corrosion, 52, 8 (1996).
10. R.G. Buchheit, J. Electrochem. Soc., 142, 3994 (1995).
11. R.G. Buchheit, R.P. Grant, P.F. Hlava, B. Mckenzie, and G.L. Zender, J. Electrochem. Soc., 144, 2621 (1997).
12. J.R. Waldrop and M.W. Kendig, J. Electrochem. Soc., 145, L11 (1998).
13. G.M. Brown and K. Kobayashi, J. Electrochem. Soc., 148, B457 (2001).
14. W.R. McGovern, P. Schmutz, R.G. Buchheit, and R.L. McCreery, J. Electrochem. Soc., 147, 4494 (2000).
15. M. Stratmann, Corros. Sci., 27, 869 (1987).
16. M. Stratmann and H. Streckel, Corros. Sci., 30, 681 (1990).
17. M. Stratmann and H. Streckel, Corros. Sci., 30, 697 (1990).
18. M. Stratmann and H. Streckel, Corros. Sci., 30, 715 (1990).
96
19. R. Ayer, J.Y. Koo, J.W. Steeds, and B.K. Park, Metallurgical Transaction A, 16A, 1925 (1985).
20. M. Gao, C.R. Feng, and R.P. Wei, Metallurgical and Materials Transactions A, 29A, 1145 (1998).
21. P. Schmutz and G.S. Frankel, J. Electrochem. Soc., 145, 2285 (1998).
22. P. Leblanc and G.S. Frankel, J. Electrochem. Soc., 149, B239 (2002).
23. F. Andreatta, H. Terryn, and J.H.W. de Wit, Corros. Sci., 45, 1733 (2003).
24. M.W. Kendig, A.J. Davenport, and H.S. Isaacs, Corrosion Science, 34, 41 (1993).
25. P. Campestrini, H. Terryn, J. Vereecken, and J.H.W. de Wit, J. Electrochem. Soc., 151, B359 (2004).
26. M.J. Vasquez, G.P. Halada, C.R. Clayton, and J.P. Longtin, Surf. Int. Anal., 25, 223 (1997).
27. L. Xia and R.L. McCreery, J. Electrochem. Soc., 146, 3696 (1999).
28. P.L. Hagans and C.M. Haas, Surface and Interface Analysis, 21, 65 (1994).
29. P. Schmutz and G.S. Frankel, J. Electrochem. Soc., 145, 2295 (1998).
30. R.G. Buchheit, L.P. Montes, M.A. Martinez, J. Micheal, and P.F. Hlava, J. Electrochem. Soc., 146, 4424 (1999).
31. R.G. Buchheit, M.A. Martinez, and L.P. Montes, J. Electrochem. Soc., 147, 119 (2000).
32. J.R. Davis, "Corrosion of aluminum and aluminum alloys", ASM International:Materials Park, OH (1999).
33. V. Guillaumin and G. Mankowski, Corrosion Science, 41, 421 (1999).
97
TABLES
Solution heat-treatment (W) Quenching Artificial aging
T6 temper *490 / **515°C for 1 hr Water 120°C for 24 hr
T7 temper 490 / 515°C for 1 hr Water 107°C for 6 hr + 163°C for 24 hr *490°C for 1 hr is solution heat treatment for AA7075. **515°C for 1 hr is solution heat treatment for AA7475.
Table 3.1: T6 and T7 alloy temper schedule for AA7075 and AA7475.
Volta potential (V vs. Ni) AA7075-T7 AA7075-T6 Alloy matrix -1.10 -1.10 Al-Fe-Cu IMC -0.65 (± 0.10) -0.74 (± 0.10) Al2CuMg IMC -0.81 (± 0.05) -0.88 (± 0.05) Mg2Si IMC -1.33 (± 0.05) -1.20 (± 0.05)
Table 3.2: Comparison of typical Volta potential in air between AA7075-T7 and AA7075-T6
98
As-polished condition 3-minute Chromate conversion coating AA7075-T7 AA7075-T6 AA7075-T7 AA7075-T6
Al-Fe-Cu Mg2Si Al2CuMg Al-Fe-Cu Mg2Si Al2CuMg Al-Fe-Cu Mg2Si Al2CuMg Al-Fe-Cu Mg2Si Al2CuMg Surface Height 76 -14 5 76 9 5 -52 -106 -164 -53 -78 -137 (nm) Volta potential 450 -230 290 360 -120 220 -23 4 -20 -33 -27 -30 (mV) Surface height and Volta potential indicate the difference from the matrix.
99 Table 3.3: Comparison of surface height and Volta potential differences on different IMCs before and after 3-minute CCC.
Surface height difference (nm) Volta potential difference (mV) 3-minute CCC AA7075-T7 -52 -23 3-minute CCC AA7075-T6 -53 -33 5-second CCC AA7475-T6 -55 -20 Surface height and Volta potential indicate the difference from the matrix.
Table 3.4: Surface height and Volta potential difference for 3-minute and 5-second CCC on Al-Fe-Cu IMC.
99
FIGURES
(b) Lift Mode:
Volta potential (a) Tapping Mode: Surface Morphology
Lift height (100 nm)
Anodic Cathodic Particle Particle
Figure 3.1. Schematic description of surface morphology and Volta potential. (a) Tapping mode traces the topographic image of the sample surface and (b) Lift mode measures the Volta potential about 100 nm height from sample surface.
100
NanoScope Controller High Resolution Oscillator Laser
Scanner Amplitude detector X,Y
Z
Oscillation
Photodiode Piezo detector
Silicon cantilever & Probe tip
Substrate
Figure 3.2. Schematic diagram for Atomic force microscopy. The surface morphology and Volta potential can be measured by cantilever deflection with constant resonant frequency and nullifying the DC voltage difference between the tip and the sample respectively.
101
(a)
(b)
Figure 3.3. BSE images of AA7075 and AA7475. (a) AA7075 has many IMCs including Al-Fe-Cu (white), Al2CuMg (white), and Mg2Si (black) IMCs but (b) AA7475 has few IMCs, which are mainly Al-Fe-Cu IMCs (white).
102
a 4 4 A 3 b 1 B c d
C 2
D 4
(a) BSE image of aligned stringers of Al-Fe-Cu particles (A~D), Mg2Si particles(1~4) and Al2CuMg particles(a~d).
Al Mg B Mg Si Particles: 1~4 B 2 Si
C Al C Cu Al2CuMg Particles: a~d Mg (a.u.) Counts Cu
Al Al-Fe-Cu Particles: A~D
Cu Fe Cu
0246810 Energy (kV)
(b) Typical EDS spectrum of Al-Fe-Cu, Mg2Si, and Al2CuMg particles.
Figure 3.4. SEM and EDS of AA7075-T6. (a) BSE image of Al-Fe-Cu particles (A~D), Mg2Si particles (1~4), and Al2CuMg particles(a~d). (b) Typical EDS spectrum of Al-Fe- Cu particles, Mg2Si particles, and Al2CuMg particles. 103
Al-Fe-Cu IMC
Al2CuMg
Mg2Si
(a) BSE image of Al-Fe-Cu IMC and Mg2Si of as-polished AA7075-T7.
76 nm 0.45 V
< 5 nm 0.29 V
- 14 nm - 0.23 V
0 40 µm 0 40 µm (b) Topographic image (c) Volta potential map
Figure 3.5. BSE and AFM images of Al-Fe-Cu IMC, Al2MgCu and Mg2Si of as-polished AA7075-T7. (a) BSE image, (b) topographic images with a scale range of 500 nm, (c) Volta potential map with 0.5 V scale.
104
Al-Fe-Cu IMC
Al2CuMg
Mg2Si
(a) BSE image of Al-Fe-Cu IMC and Mg2Si of 3-minute CCC coated AA7075-T7.
- 52 nm - 23 mV
-164 nm - 20 mV
4 mV -106 nm
0 40 µm 0 40 µm (b) Topographic image (c) Volta potential map
Figure 3.6. BSE and AFM images of Al-Fe-Cu IMC and Mg2Si of 3-minute CCC coated AA7075-T7. (a) BSE image, (b) topographic images with a scale range of 200 nm, (c) Volta potential map with 0.1 V scale.
105
Trenches
Mg2Si
Al-Fe-Cu IMC
Al2CuMg
Figure 3.7. 3-D surface height of IMCs in 3-minute chromate coated AA7075-T7.
106
0 40 µm
(a) Surface height of Al-Fe-Cu IMC
Matrix Al-Fe-Cu IMC Matrix
(b) Line profile of Al-Fe-Cu IMC in 3-minute chromate coated AA7075-T7
Figure 3.8. Surface height and line profile of Al-Fe-Cu IMC in 3-minute CCC on AA7075-T7.
107
0 40 µm
(a) Surface height of Mg2Si IMC
Matrix Mg2Si Matrix
(b) Line profile of Mg2Si IMC in 3-minute chromate coated AA7075-T7
Figure 3.9. Surface height and line profile of Mg2Si IMC in 3-minute CCC on AA7075- T7.
108
0 40 µm
(a) Surface height of Al2CuMg IMC
Matrix Al2CuMg Matrix
(b) Line profile of Al2CuMg IMC in 3-minute chromate coated AA7075-T7
Figure 3.10. Surface height and line profile of Al2CuMg IMC in 3-minute CCC on AA7075-T7.
109
(a) (b)
0 40 µm 0 40 µm
Figure 3.11. AFM images of Mg2Si of as-polished AA7075-T6. (a) topographic images and (b) Volta potential map with line profiles.
110
0 40 µm
(a) Surface height of Mg2Si IMC
Matrix Mg2Si Matrix
(b) Line profile of Mg2Si IMC in 3-minute chromate coated AA7075-T6
Figure 3.12. Surface height and line profile of Mg2Si IMC in 3-minute CCC on AA7075- T6.
111
200
150 Mg Si 2
100
(nm) Depth Al-Fe-Cu IMC
50
0 0 20 40 60 80 100 120 140 160 180 200 Coating time (sec)
Figure 3.13. Maximum trench depth on Mg2Si and Al-Fe-Cu IMCs as a function of coating time.
112
IMC2 IMC2
IMC1 IMC1
Height:100 nm Volta potential:1.0 V 0 40 µm 0 40 µm
(a) Before chromate conversion coatings
Figure 3.14. AFM image of Al-Fe-Cu IMCs in AA7475-T6 (a) before and (b) after 5- second CCCs at room temperature, 40×40 µm.
(Continued)
113
Figure 3. 14: (Continued)
IMC2 IMC2
IMC IMC 1 1
Height:200 nm Volta potential:0.05 V 0 40 µm 0 40 µm
(b) After 5-second chromate conversion coatings at room temperature
114
Trench: IMC/matrix boundary
(a) (b)
Height: 200 nm Height: 100 nm 0 40 µm 0 10 µm
Matrix Al-Fe-Cu IMC
(c) Matrix (d) Al-Fe-Cu IMC
Height: 50 nm Height: 50 nm 0 500 nm 0 500 nm
Figure 3.15. High Magnification of Figure 3.14(b), Al-Fe-Cu IMCs for 5-second coating in AA7475-T6.
115
Al2CuMg IMCs
(a) BSE image of 3-minute CCC AA7075-T6 after 24 hr exposed to 0.5 M NaCl.
Al2CuMg IMCs
(b) BSE image of 3-minute CCC AA7075-T6 after 48 hr exposed to 0.5 M NaCl.
Figure 3.16. BSE image of 3-minute CCC AA7075-T6 after 24hr and 48 hr exposures to 0.5 M NaCl.
116
(a) Al-Fe-Cu IMC (b) Mg2Si (c) Al2CuMg
Matrix Matrix Matrix Cr(III) Cr(VI) Dealloying Dealloying Cr(III) Cr(VI) Matrix dissolution Cr(III) Cr(VI) Cr(III) Cr(VI) Coating timeinthechromate
e- e- e- e- - - - - e- e- e e e- e- e e e- e-
Cr(III) Cr(VI) Cr(III) Cr(VI) Cr(III) Cr(VI) Dealloying/ Fe(CN) 3- Fe(CN) 4- 3- 4- Matrix dissolution 3- 4- 6 6 Cr(III) Cr(VI) Fe(CN)6 Fe(CN)6 Fe(CN)6 Fe(CN)6 Matrix dissolution Cr(III) Cr(VI) 3+ 3+ Cr(III) Cr(VI) 3+ Al Al Fe(CN)x or -CN Al Al Al Al Fe(CN)x or -CN
bath(lessthan5seconds) ------e e e e - - - - e e 117 e e e e e e Dealloying/ Matrix dissolution
Cr(III) Cr(VI) Cr(III) Cr(VI) Cr(III) Cr(VI)
Dealloying/ 3- 4- 3- 4- 3- 4- Fe(CN)6 Fe(CN)6 Fe(CN) Fe(CN) Fe(CN) Fe(CN) Cr(III) Cr(VI) Matrix dissolution 6 6 Matrix dissolution 6 6 Cr(III) Cr(VI) 3+ 3+ 3+ Cr(III) Cr(VI) Al Al Al Al Al Al Fe(CN) or -CN x Fe(CN) or -CN x
------e e e e - - - - e e e e e e e e Dealloying/ Matrix dissolution
Figure 3.17. Schematic of trenches formation at different intermetallic compounds for 5 sec coatings. (a) Al-Fe-Cu IMC, (b) Mg2Si, and (c) Al2CuMg. Symbol • indicates chromium oxide and symbol ♦stands for adsorbed ferricyanide. Coating thicknesses are exaggerated about 20x compared to IMCs.
117
Cr(III) Cr(VI) Cr(III) Cr(VI) 3- 4- Cr(III) Cr(VI) Cr(III) Cr(VI) Fe(CN)6 Fe(CN)6 Fe(CN)x or -CN Fe(CN)x or -CN 3+ Al Al Matrix dissolution Dealloying/ Matrix dissolution Al-Fe-Cu IMC Mg2Si IMC Al matrix Al2CuMg IMC
(a) CCC formation on IMCs at early stage (5 seconds)
Continous coating formation (3 minutes) Early coating formation (5 seconds)
Al matrix Al2CuMg IMC
Al-Fe-Cu IMC Mg2Si IMC
(b) CCC formation on IMCs at 3-minute coating time
Figure 3.18. Schematic of CCC formation on different IMCs in 7xxx aluminum alloy. (a) CCC formation on IMCs at early stage. Absorbed ferricyanide inhibited chromate reduction reaction on Al2CuMg and Al-Fe-Cu IMCs. Al2CuMg and Mg2Si IMCs showed anodic dissolution of Mg around IMCs. (b) CCC formation on IMCs during commercial coating time (3-minute). The surface height difference between IMCs and matrix was not significantly changed after early coating. Coating thicknesses are exaggerated about 20x compared to IMCs for comparison.
118
CHAPTER 4
CORROSION BEHAVIOR AND CHROMATE CONVERSION COATING
FORMATION ON THIN FILM ANALOGS OF η (MgZn2) AND S (Al2CuMg)
PHASES
4.1 INTRODUCTION
The electrochemical activities of intermetallic compounds (IMCs) are very important for understanding the localized corrosion, such as pitting corrosion, crevice corrosion, galvanic corrosion, intergranular corrosion (IGC), and exfoliation corrosion of high strength Al alloys. It is also important for understanding formation and breakdown of chromate conversion coatings. IMCs are always present in the microstructure of 2xxx and 7xxx alloys due to impurities and alloying element additions. IMCs can be divided into coarse constituent particles formed during solidification processing and fine precipitate hardening particles formed during natural or artificial aging processes.
The coarse constituent particles can be cathodic or anodic relative to the matrix.
Buchheit compiled the corrosion potentials of various IMCs in Al alloys [1]. The galvanic relations between IMCs and the matrix can contribute to the different
119
electrochemical properties observed in alloys. Among these IMCs, Cu-rich and Fe-rich
IMCs typically act as cathodes supporting oxygen reduction. This can lead to the development of local alkalinity that enhances corrosion of adjacent aluminum alloy matrix by cathodic corrosion [2-4]. The increased local pH also results in dealloying of
IMCs [5-8]. Preferential dissolution of Al from Al3Fe under open circuit conditions in
NaOH solution has been observed to cause an Fe rich surface to develop on Al3Fe, which reduced the cathodic behavior of Al3Fe due to the formation of protective Fe oxide [5].
Copper-rich IMCs are often thought to be local cathodes because of their high Cu content. However, they can also be active and sustain high anodic dissolution rate as more active elements in them preferentially dissolve. This dealloying leads to Cu enrichment and a reversal of the galvanic relationship with the surrounding matrix. The
Al2CuMg intermetallic compound, sometimes referred to as the S phase, is present as a coarse constituent particle with diameters of 0.5 to 5µm and as a fine lath-shaped precipitate in several important commercial Al-Cu-Mg and Al-Zn-Mg-Cu alloys. The compound is electrochemically active, and coarse particles significantly affect localized corrosion and surface finishing of alloys that contain them.
Direct measurements of the Al2CuMg compound synthesized in bulk for show that the phase exhibits corrosion potentials in the range of –0.88 to –0.93Vsce in neutral chloride solutions [9]. This makes the phase active with respect to the surrounding matrix phase under these conditions. Under free corrosion conditions, the phase dissolves by dealloying at an appreciable rate. This dealloying process can significantly ennoble the phase as the dissolving particle becomes enriched with Cu. Dealloyed particles can
120
catalyze oxygen reduction leading to the development of local alkalinity and cathodic corrosion of the surrounding matrix phase [10].
Interestingly, in some cases the dealloyed particle remnant can release Cu ions at potentials well below the reversible potential for Cu that might be estimated from the
Nernst equation [11]. High resolution electron microscopy has shown that the dealloyed particle structure can coarsen and in the act of coarsening can form nanometer diameter metallic clusters that are detached from the particle remnant [4]. Once detached, these Cu clusters are electrically isolated from the particle and can move to more positive potentials where they are presumably oxidized to form Cu ions that can be deposited elsewhere on the alloy surface in a form of deposition corrosion [8].
In Al-Cu-Mg alloy 2024-T3, Al2CuMg particles have been observed to interfere with the formation of chromate conversion coatings. Raman scattering at 860 cm-1 is indicative of Cr3+-O-Cr6+ bonding in CCCs and the intensity of the band is directly proportional to CCC thickness. Mircofocal Raman microscopy at 860cm-1 has shown that
CCCs are thinner on bulk synthesized Al2CuMg and on Al2CuMg particles in 2024-T3.
Al2CuMg particles are also present in Al-Zn-Mg-Cu alloy 7075. Localized corrosion on chromate conversion coated 7075 substrates after exposure to aerated chloride solutions is commonly associated with Al2CuMg particles present in the alloy in chapter 3.
The fine precipitate hardening particles MgZn2 is an important strengthening phase in Al-Zn-Mg-Cu alloys. This phase forms by nucleation and growth processes during artificial aging on grain boundaries and within grains resulting in a uniform, high number density dispersion of fine precipitates. The precipitation sequence is [12-22]:
121
GP → η’’→ η’ → η (MgZn2)
This precipitation process contributes greatly to alloy strengthening. Larger precipitates form on grain boundaries than in the matrix and attempts to measure grain boundary precipitate chemistry show that the phase contains Cu and perhaps Al as aging progresses from peak to overaged tempers. While precise chemistries have been difficult to obtain, the solubility for Cu and Al is so extensive, that MgZn2 and AlCuMg are isostructural [13,
23].
It is observed that localized corrosion susceptibility of Al-Zn-Mg-Cu alloys changes significantly from peakaged to overaged tempers. The overaged temper exhibits lower pitting potentials, but is less susceptible to intergranular corrosion. Distinctive differences in localized corrosion morphologies mainly related to the occurrence of intergranular corrosion have also been noted [24-27].
The electrochemistry of compositional analogs of compounds ranging in chemistry from MgZn2 to AlCuMg have been characterized using polarization techniques.
Results show that dissolution kinetics of the class of compounds are ennobled as Cu content increases. Ennoblement is significant when the Cu content exceeds 20 at. %.
While the change in intergranular corrosion susceptibility of Al-Zn-Mg-Cu alloys and ennoblement of η phase dissolution kinetics with increase aging has been noted, there is not broad agreement on a causal relationship [24-26, 28].
Alloy temper also affects the formation and protectiveness of chromate conversion coatings (CCCs) on Al-Zn-Mg-Cu alloys. For alloy 7475, conversion coatings were found to be slightly more protective when applied to overaged (T7) tempered material than to peakaged (T6) tempered material in chapter 6.
122
The apparent link between the η phase chemistry, η phase electrochemical behavior and alloy corrosion behavior suggests the possibility of a relation between the dependence of chromate conversion coating formation and breakdown behavior and η phase chemistry.
In this study, thin film compositional analogs of η and S phases were fabricated using the flash evaporation techniques and sputtering. The thickness of 3-min CCCs on η phase with different Cu contents was determined by Raman spectroscopy. Additionally, the effect of ferricyanide with Cu-bearing η phase was examined. Results show that the effects of Cu on electrochemical behavior and clarify the nature of the interaction of
3- Fe(CN)6 as Cu content in η increases. The reactivity of Al2CuMg before, during and after chromate conversion coating was measured by using the experimental approach based on a non-electrochemical quartz crystal microbalance measurement of thin film.
4.2 EXPERIMENTAL
4.2.1 Thin film analogs of IMCs
Thin film analogs of pure Al and η (Mg(Zn,xCu)2) were produced using a flash evaporation technique following the procedure of Ramgopal et al. using alloy powder feedstock produced by grinding cast ingots [26]. Samples with four different Cu contents ranging from 0 to 35 at. % were prepared. These ingots were fabricated using inductive arc melting. The starting materials were prepared by placing Mg, Zn, and Cu in an MgO crucible, which was contained in a quartz tube and was evacuated and Argon-backfilled to minimize the effect of oxygen. The encapsulated quartz tube was heated in an induction furnace to around 1200 °C and held several minutes to stir and mix the melt 123
homogeneously. After mixing, the power was turned off and the charge was allowed to cool down to room temperature naturally. The ingot was crushed into powder, which was used as the starting material for deposition of thin film analogs. A feeder in evaporation vacuum chamber was used to carefully drop the powder onto a heated tungsten boat where it evaporated quickly (Figure 4.1). The operating pressure of flash evaporation chamber was controlled below 2 × 10-7 Torr in order to minimize oxygen contamination
[26]. The thin films were deposited on Si wafers for electrochemical testing. The nominal thickness of the analog films was up to 2 µm for Si wafer substrates. After deposition, the compositions of thin film analogs were checked using energy dispersive spectroscopy
(EDS) and were found to have compositions corresponding closely to the composition of the powder feedstock. The samples prepared had the following compositions expressed on an atomic percentage basis and are referred to using their approximate stoichiometric ratios: 33Mg-67Zn (MgZn2), 33Mg-50Zn-17Cu (Mg3Zn5Cu2), 33Mg-42Zn-25Cu
(Mg3Zn4Cu3), 33Mg-32Zn-35Zn (MgZnCu). Whether any of these specific phases are present in Al-Zn-Mg-Cu alloys is unknown, however the trends as a function of Cu content in the electrochemical behavior and formation of conversion coatings on the η phase should be reflected in experiments conducted with these samples.
Sputter deposition was used for the preparation of thin film analogs of Al2CuMg from an alloy target with an S-phase composition (Plasmaterials, INC.) The base operating pressure was also about 2 × 10-7 Torr. The thickness of thin film after 12- minute sputtering was about 500 nm as measured by surface profilometry (Vecco:
Dektak3ST).
124
The thin films were deposited on Si wafers for electrochemical testing and on 10
MHz AT-cut quartz crystal for quartz crystal microbalance mass loss measurements. The nominal thickness of the thin films was up to 2 µm for Si wafer substrates. After deposition, the compositions of thin film analogs were checked using energy dispersive spectroscopy (EDS). The nominal compositions of the Al2CuMg analog films ranged from 55 at.%Al, 21 at.%Cu and 24 at.%Mg to 56 at.%Al, 22 at.%Cu and 22 at.%Mg.
4.2.2 Quartz substrate working electrode
Thin films were deposited on 10 MHz AT-cut quartz crystal substrate
(International Crystal Manufacturing Co, Inc.). These crystals demonstrate stability over a wide range of temperatures and resonant frequencies. Prior to thin film deposition, the crystals were sonicated sequentially in methanol and acetone for 3 minutes, then rinsed with distilled water and carefully dried with Ar gas to remove dust. After deposition, the crystal was mounted on a holder with the working electrode facing the solution. A schematic illustration of quartz working electrode is shown in Figure 4.2. The electrode consisted of the working electrode area and current lead. The quartz crystal holder was placed vertically into electrochemical cell. Teflon tape and silicone resin were used to prevent the holder from leaking. Three-electrode system was used in which Pt was used as a counter electrode, and a saturated calomel electrode (SCE) was used as a reference electrode.
125
4.2.3 Electrochemical quartz crystal microbalance measurement
The electrochemical quartz crystal microbalance (EQCM) is based on the resonant response of a piezoelectric quartz crystal on a thin film that enables in-situ measurement of mass changes on a thin film electrode with extremely high sensitivity.
The EQCM system consists of an electrochemical microbalance (EQCN-900, Elchema), a frequency scanner (EQCN 906, Elchema) and a potentiostat (PS-605, Elchema) as shown in Figure 4.3. EQCM measurements were carried out in a Faraday cage to minimize electromagnetic interference. The frequency difference (∆f) on the working quartz crystal facing the solution and the air was measured with microbalance composed of oscillator and frequency counter and then converted to a mass change (∆m) using the
Sauerbrey Eqn. [29]. At the same time, the current and potential was monitored using potentiostat. Over a wide mass range, the relationship between change in resonant frequency and mass on the surface is linear. The Sauerbrey Eqn. is as follows [29]:
Eqn. 4-1 ∆f = − K f ∆m, 2 2 f Eqn. 4-2 K f = A⋅ µq ρq
Eqn. 6-1 shows a linear relation between the resonant frequency change ∆f and the mass change ∆m. The value of the linearity coefficient Kf is proportional to the square of the frequency f, and depends on the density ρq, shear modulus of the quartz µq, and the surface area of the electrode A. The EQCM converts the change in the frequency into the mass change with assumption of constant Kf.
126
Anodic polarization experiments were carried out in a flat cell using a three- electrode configuration using a Pt counter electrode and a saturated calomel reference electrode. Scans were initiated below the steady-state corrosion potential and were continued past the apparent breakdown or pitting potential. All scans were carried out in deaerated 0.1M NaCl solution at a scan rate of 0.2 mV/s.
Chromate conversion coatings were applied to thin film analogs by simple immersion of the samples at their respective open circuit potentials. Coatings were applied using a ferricyanide-accelerated Alodine 1200S conversion coating bath formulation. The chemistry of this bath was approximately 3.95-4.74 g/L CrO3, 0.79-2.37 g/L K3Fe(CN)6, 0.79-2.37 g/L KBF4, 0.08-0.79 g/L NaF, and 0.08-0.79 g/L K2ZrF6 [30].
Selected conversion coatings were applied from conversion coating baths from which ferricyanide was withheld. The chemistry for these baths was 5.4 g/L CrO3 and 0.9 g/L
NaF.
4.2.4 Raman spectroscopy
The thickness of CCCs formed on the compositional analogs was estimated from the peak intensity of the 860cm-1 Raman scattering band. This peak is due to scattering by
Cr6+-O-Cr3+ bonds in the CCC structure [31, 32]. Samples were interrogated using a
514.5 nm laser, a 180° backscattered sampling geometry, and a laser spot size of about 50
µm diameter. The peak height at 860 cm-1 was analyzed after baseline correction.
127
4.3 RESULTS AND DISCUSSION
4.3.1 Effect of Cu on the anodic polarization behavior of the compositional analogs
Figure 4.4 shows anodic polarization curves for the compositional analogs in deaerated 0.l M NaCl. The polarization curves for MgZn2 and Mg3Zn5Cu2 are similar with corrosion potentials between –1200 and –1300 mV, and passive regions where the dissolution rates are several µA/cm2. Both curves exhibit poorly articulated breakdown potentials at about –1100 mV, above which the dissolution rate gradually increases from tens to hundreds of µA/cm2. This breakdown has been noted previously and has been attributed to the onset of Zn oxidation from the phase [33]. In these experiments, breakdown occurs at a potential that is about 100 mV more positive than the reversible potential of Zn, which was estimated to be –1180 mV assuming a Zn2+concentration of
10-6 M.
The polarization response of the MgZnCu analog exhibited significantly increased corrosion and pitting potentials and a reduced passive current density compared to the other two analogs. The pitting potential is very clearly articulated and occurs well positive of the expected reversible potential for Zn. Whether breakdown is associated with the onset of Zn oxidation or the breakdown of a protective Cu-rich surface film that forms during polarization is not obvious.
These data illustrate the ennoblement in the dissolution kinetics of η as it dissolved Cu. However, these data suggest that even with very high Cu concentrations η is still expected to be a very anodic phase in Al-Zn-Mg-Cu alloys where the free corrosion potential will range from –0.700 to –0.750 mV in aerated dilute chloride solutions.
128
4.3.2 Effect of Cu content in compositional analogs on CCC formation
CCC formation on Al-Cu-Mg intermetallic compounds has been observed to be inhibited as the Cu content of the phase increases [34]. The same appears to be true for η as it dissolves Cu. Figure 4.5 shows the 860 cm-1 Raman band for CCCs formed by a 3- minute immersion in the coating bath. The band is intense for MgZn2 suggesting robust
CCC formation. However as Cu content increases band intensity diminishes markedly. A weak band is observed for Mg3Zn5Cu2, but evidence of CCC formation on Mg3Zn4Cu3 and MgZnCu is completely absent.
In the case of Al-Cu-Mg intermetallics CCC formation was suppressed by surface adsorption of ferricyanide on Cu sites [34]. This adsorbed layer appeared to passivate the compounds suppressing normal CCC formation, but this layer itself did not appear to confer the level of corrosion protection provided by CCCs.
4.3.3 Effect of ferricyanide on CCC formation on η
The presence of ferricyanide in a CCC bath had a significant affect on CCC formation on the compositional analog of η. Figure 4.6 shows Raman spectra in the vicinity of the 860 cm-1 peak. There is evidence of strong CCC formation when coating is carried out in an Alodine 1200S bath. However, when coating is carried out in a bath where ferricyanide is withheld, coating formation is much weaker. This result shows that ferricyanide is an accelerator of CCC formation on η in much the same way as it is for pure aluminum.
Figure 4.6 also shows Raman spectra collected from Mg3Zn4Cu3 and MgZnCu compositional analogs that were coated in a bath from which ferricyanide was withheld. 129
Considering the Raman spectra from these analogs presented in Figure 4.5, it is clear that withholding ferricyanide from the coating bath does nothing to improve CCC formation on the analogs with high levels of Cu.
Several authors have suggested that the passivation of Cu-containing IMCs is caused by reductive adsorption of chromate and not by the cyanide adsorption. Halada et al. observed that F and CN ions were not found on Cu-containing IMCs after early coating formation [35]. Hurley et al. observed monolayers of chromium(III) film on Cu surface, which is very stable and inhibits electron transfer and further chromate reduction
[36]. They suggested a mechanism for chromium(III) film formation in which chromate is reductively adsorbed on Cu surface initially and then reduced to chromium (III) film very rapidly to form Cr3+/Cr6+ mixed oxide. After this mixed oxide formed, however, the chromate adsorption and reduction reactions were inhibited by this stable oxide on Cu surface. Campestrini et al. suggested that chromate reduction is intense on Cu-rich IMCs and leads to form CCC formation, which is not enough time to adsorb ferricyanide on
IMCs [37]. It is appropriate that the passivation on η phase might be associated with the fast chromate reduction on this Cu-rich η phase.
4.3.4 Reactivity of Al2CuMg in 0.5 M NaCl
Figure 4.7 shows the evolution of the open circuit potential (OCP) of several
Al2CuMg thin film analogs over 1 hour during immersion in aerated 0.5 M NaCl. The
OCP variation was reproducible and was characterized by a rapid increase within about
1200 s followed by stabilization. The OCP in the stable region ranged from -860 to -760 mVSCE, which is in the OCP range reported for this phase in the literature [1].
130
Figure 4.8 shows the variation in OCP and mass loss during an EQCM measurement. In this particular experiment the initial increase in OCP is accompanied by a small decrease in mass. An episode of stable potential and no mass change is then observed. After about 1000 s of exposure, there is a decrease in OCP and the onset of steady mass loss at a rate of about 0.08 µg/cm2⋅s. It is noted that this increase in OCP and the OCP value in the plateau region are different than that shown in Figure 4.7, however it is believed that the processes occurring are essentially the same in both experiments.
Figure 4.9b shows EDS spectra collected from thin film analogs withdrawn from solution at the times indicated in the OCP versus time plot in Figure 4.9. The EDS spectra indicate a progressive increase in Cu consistent with dealloying [11]. Overall, these data are taken to indicate an initial phase of dealloying that ennobles the OCP and produces a protective, Cu-rich film on the electrode surface. With time, this film breaks down leading to a decrease in OCP and the onset of an episode of stable mass loss that does not lead to the formation of a protective film.
4.3.5 Reactivity of Al2CuMg during chromate conversion coating formation
Conversion coatings were formed in a bath comprised of CrO3, K3Fe(CN)6, NaF,
KBF4 and K2ZrF6. During 3 minutes of coating formation on Al2CuMg, the OCP increased by about 50 mV while sustaining net mass loss of 0.007 µg/cm2⋅s (Figure 4.12).
During the coating, the rate of net mass loss decreased with time. Under identical conditions, a pure Al thin film gains mass due to coating formation at a rate of about
2. 0.003 µg/cm s (Figure 4.10). CCC formation on the Al2CuMg analog in this study was not carried out at a potential characteristic of coating formation on Al alloys, but these
131
results are consistent with other observations showing that conversion coatings are thicker on Al-rich matrix phase of Al alloys than on Al2CuMg intermetallic particles [34,
38].
4.3.6 Reactivity of Al2CuMg after chromate conversion coating formation
5 × 5 µm surface height and Volta potential maps of Al2CuMg thin film analogs before and after application of a 3-minute CCC are shown in Figure 4.12. Before CCC formation, the roughness of thin film analog was about 2.63 nm RMS; though small nodules about 100 nm in diameter are evident. The Volta potential map showed no contrast; even at nodule locations suggesting that the surface was compositionally homogeneous.
After application of the CCC, the surface roughness was increased slightly to 3.46 nm RMS. Contrast variations in the surface height map showed elevated domains several micrometers in diameter separated by narrow valleys in morphology reminiscent of shrinkage cracks commonly observed on chromate conversion coatings [32]. The Volta potential map showed that surface potential was slightly elevated in the depressed crack- like regions on the surface.
OCP/EQCM measurements of mass loss of the conversion coated Al2CuMg analog in aerated 0.5 M NaCl solutions (dried for 24 hours before immersion) are shown in Figure 4.13. The OCP rapidly decreases to a value characteristic of Al2CuMg without a conversion coating, however the nominal mass loss rate is reduced by the presence of the coating to a value of about 0.001µg/ cm2⋅s. This mass loss rate is more than two orders of magnitude lower than that of the uncoated analog. This comparison suggests that while
132
Al2CuMg forms a thinner and presumably inferior chromate conversion coating, the coating that does form inhibits dissolution of Al2CuMg to a significant degree.
Generally, Table 4.1 shows mass loss rates for the thin film analogs in aerated
0.5M NaCl, in a chromate conversion coating bath and in aerated 0.5M NaCl after chromate conversion coating. Two values for the mass loss are reported for the uncoated analogs reflecting the active and passive behaviors observed in these experiments. Mass loss rates are lower for the analog in the conversion coating bath, but stand out in comparison to pure Al, which gains mass during conversion coating. The absolute dissolution rate of Al2CuMg in a CCC bath is likely to be greater than that indicated in the table because this rate reflects the combination of mass gain from coating formation and mass loss from dissolution. Conversion coating slows, but does not stop Al2CuMg dissolution. How the dissolution rate of Al2CuMg compares with that of the conversion coated matrix phase in Al-Cu-Mg and Al-Zn-Mg-Cu alloys has not been established with certainty, however it has been observed that breakdown of conversion coated 7075 and
7475 alloy surfaces is usually associated with Al2CuMg particles in chapter 3.
EDS measurements shown in Figure 4.14 indicate that Cu is not strongly enriched in Al2CuMg by chromate conversion coatings of thin film analogs. This is probably due to the low dissolution rate of the phase and the comparatively short duration of the immersion. Cu enrichment of the conversion coated analog during exposure to aerated
0.5M NaCl solution was observed to be appreciable. This suggests that CCCs may not completely suppress Cu redistribution phenomena or the attendant deposition corrosion susceptibility in Al2CuMg-forming Al alloys
133
4.4 SUMMARY
Electrochemical characterization and chromate conversion coating formation experiments with analogs of η phase with various amounts of dissolved Cu and EQCM measurements of Al2CuMg thin film analogs suggest the following:
1. Cu dissolved in η over the range of 0 to 35 at. %, ennobles dissolution kinetics as
has been shown previously [33].
2. Cu in η tends to suppress the formation of chromate conversion coatings from
ferricyanide-bearing accelerated conversion coating baths. Between 15 and 25
at% Cu evidence of CCC formation is completely absent in Raman spectra.
3. Withholding ferricyanide from a coating bath strongly suppresses conversion
coating formation on η. However, there is no improvement in conversion coating
formation on Cu-rich compositional analog.
4. Al2CuMg particles in Al alloys exhibit active and passive behavior. Passivation of
the phase appears to be due to the formation of a Cu-rich surface film. When
passivated, mass loss is negligible. In these experiments, mass loss rates of 0.08
µg/cm2.s characterized active dissolution in aerated 0.5M NaCl solution. Net mass
losses were observed during immersion in an accelerated CCC bath. This stands
in contrast to the behavior of pure Al, which gains mass during coating formation.
5. Measurements of mass loss rates of conversion coated analog of Al2CuMg show
that the coating that does form reduces the dissolution rate by more than two
orders of magnitude during exposure to aerated chloride solutions, but does not
suppress Cu enrichment of the phase and may not suppress deposition corrosion
involving redistributed Cu. 134
REFERENCES
1 R.G. Buchheit, J. Electrochem. Soc., 142, 3994 (1995).
2 N. Dimitrov, J.A. Mann, and K. Sieradzki, J. Electrochem. Soc., 146, 98 (1999).
3 N. Dimitrov, J.A. Mann, M. Vukirovic, and K. Sieradzki, J. Electrochem. Soc., 147, 3283 (2000).
4 R.G. Buchheit, R.P. Grant, P.F. Hlava, B. Mckenzie, and G.L. Zender, J. Electrochem. Soc., 144, 2621 (1997).
5 K. Nisancioglu, J. Electrochem. Soc., 137, 69 (1990).
6 B. Mazurkiewicz, Corros. Sci., 23, 687 (1983).
7 B. Mazurkiewicz and A. Piotrowski, Corros. Sci., 23, 697 (1983).
8 R.G. Buchheit, R.K. Boger, M.C. Carroll, R.M. Leard, C. Paglia, and J.L. Searles, JOM, 53, 29 (2001).
9 R.G. Buchheit, L.P. Montes, M.A. Martinez, J. Micheal, and P.F. Hlava, J. Electrochem. Soc., 146, 4424 (1999).
10 T.J.R. Leclere and R.C. Newman, J. Electrochem. Soc., 149, B52 (2002).
11 R.G. Buchheit, M.A. Martinez, and L.P. Montes, J. Electrochem. Soc., 147, 119 (2000).
12 L.F. Mondolfo, Metallurgical Reviews, 153, 95 (1971).
13 L.F. Mondolfo, Aluminium alloys: Structure and properties, Buttenworths, Boston (1976).
14 H. Schmalzried and V. Gerold, Z. Metallkd., 49, 291 (1958).
15 L.F. Mondolfo, N.A. Gjostein, and D.W. Levinson, Trans. Amer. Inst. Min. Met. Eng., 206, 1378 (1956).
16 G. Thomas and J. Nutting, J. Inst. Metals, 88, 81 (1959-60).
17 R.B. Nicholson, G. Thomas, and J. Nutting, Brit. J. Appl. Phys., 9, 25 (1958).
18 R.B. Nicholson, G. Thomas, and J. Nutting, J. Inst. Metals, 87, 429 (1958-59).
19 J. Gjonnes and C.J. Simensen, Acta Met., 18, 881 (1970).
135
20 J.D. Embury and R.B. Nicholson, Acta Met., 13, 403 (1965).
21 G.W. Lorimer and R.B. Nicholson, Acta Met., 14, 1009 (1966).
22 G.W. Lorimer and R.B. Nicholson, The Mechanism of Phase Transformations in Crystalline Solids, Inst. Metals, London (1968).
23 J.E. Hatch, Aluminum propeties and physical metallurgy, American Society for Metals, Metals Park, Ohio (1984).
24 S. Maitra and G.C. English, Metallurgical Transaction A, 12A, 535 (1981).
25 S. Maitra and G.C. English, Metallurgical Transaction A, 13A, 161 (1982).
26 T. Ramgopal, Role of Grain boundary precipitates and Solute depleted zone in the Intergranular corrosion of Aluminum Alloy AA7150, The Ohio State University (2001).
27 Q. Meng and G.S. Frankel, J. Electrochem. Soc., 151, B271 (2004).
28 J.K. Park and A.J. Ardell, Acta Metall. Mater., 39, 591 (1991).
29 G. Sauerbrey, Z. Physik, 155, 206 (1959).
30 Henkel, Materials Safety Data Sheet for Alodine 1200S, Henkel Corporation.,
31 L. Xia and R.L. McCreery, J. Electrochem. Soc., 146, 3696 (1999).
32 W. Zhang, B.L. Hurley, and R.G. Buchheit, J. Electrochem. Soc., 149, B357 (2002).
33 T. Ramgopal, P. Schmutz, and G.S. Frankel, J. Electrochem. Soc., 148, B348 (2001).
34 W.R. McGovern, P. Schmutz, R.G. Buchheit, and R.L. McCreery, J. Electrochem. Soc., 147, 4494 (2000).
35 G.P. Halada, C.R. Clayton, M.J. Vasquez, J.R. Kearns, M.W. Kendig, S.L. Jeanjaquet, G.G. Peterson, G.S. McCarthy, and G.L. Carr., PV 99-27, The Electochemical Society Proceedings Series, Pennington, NY (1999).
36 B.L. Hurley and R.L. McCreery, J. Electrochem. Soc., 150, B367 (2003).
37 P. Campestrini, H. Terryn, J. Vereecken, and J.H.W. de Wit, J. Electrochem. Soc., 151, B359 (2004).
38 Q.J. Meng and G.S. Frankel, Surf. Int. Anal., 36, 30 (2004).
136
TABLE
Dissolution rate (µg/cm2⋅sec)
Bare Al2CuMg under OCP in 0.5 M NaCl 20/3600
CCC formation on Al2CuMg in Alodine 1200S 1.2/180
3-minute CCCs on Al2CuMg under OCP in 0.5 M NaCl 4/3600
Table 4.1: Dissolution rates of thin film analog of S phase in different conditions.
137
FIGURES
Evaporation Chamber
Sample Holder Si wafer QCM
Shutter
Feeder
W-Boat
Diffusion Pump
Figure 4.1. Schematic description of flash evaporation technique. Grounded cast bulk alloys are fed onto W boat and then evaporated from the W boat to Si wafer homogeneously.
138
Circular part Thin film deposition
Quartz crystal
Current lead part
(a) Top view (b) Side view
Figure 4.2. Schematic description of quartz crystal as working electrode, showing (a) front view and (b) side view.
139
Counter E
Working E Potentiostat Reference E
Oscillator
Frequency Counter
∆ƒ ∆m E i
Computer
Figure 4.3. Schematic representation of typical EQCM system, showing electrochemical parts of potentiostat and mass change measurement such as oscillator and frequency counter.
140
-0.6
MgZn2 -0.7 Mg(Zn17%Cu) 2 Mg(Zn35%Cu)2 -0.8
-0.9
-1.0
-1.1
Potential (V vs. SCE) -1.2
-1.3
-1.4 1e-9 1e-8 1e-7 1e-6 1e-5 1e-4
2 Current Density (A/cm )
Figure 4.4. Polarization curves of MgZn2, Mg3Zn5Cu2, and MgZnCu at a scan rate of 0.2 mV/s in deaerated 0.1 M NaCl.
141
10000
MgZn 2 Mg(Zn,17%Cu) 2 Mg(Zn,25%Cu) 8000 2 Mg(Zn,35%Cu)2
6000
4000
(a.u.) Intensity
2000
0
750 800 850 900 950
-1 Raman Shift (cm )
Figure 4.5. Raman spectra of 3-minute CCCs on thin film Mg(Zn,xCu)2, x=0, 17, 25 and 35 atom %.
142
10000 Cr+F_MgZn 2 Cr+F_Mg(Zn,25%Cu) 2 Cr+F_Mg(Zn,35%Cu) 8000 2 Alodine 1200s_ MgZn 2
6000
4000
Intensity (a.u.) 2000
0
-2000 750 800 850 900 950
-1 Raman Shift (cm )
Figure 4.6. Raman spectra of 3-minute CCCs on thin film Mg(Zn,xCu)2, x=0, 25 and 35 atom % in Cr+F synthesized bath and 3-minute CCCs on thin film MgZn2 in Alodine 1200S bath. Cr and F represented CrO3 and NaF respectively.
143
-0.7
-0.8
-0.9
-1.0
-1.1
Potential (V vs.Potential SCE)
-1.2
-1.3 0 1000 2000 3000
Time (sec)
Figure 4.7. OCPs of S phase in aerated 0.5 M NaCl during 1 hr immersion.
144
-600 5
-700 0 Potential ) -800 -5 2
g/cm
µ -900 -10
-1000 -15 Mass Change ( Potential (V vs.SCE)
Potential (mV vs. SCE) Mass change
-1100 -20
-1200 -25 0 1000 2000 3000 Time (sec)
Figure 4.8. OCP and mass change of thin film analog of Al2CuMg in aerated 0.5 M NaCl.
145
-0.7 (a)
-0.8
-0.9 (ii) Near peak: 49%Al-32%Cu-19%Mg -1.0 (iii) After 1 hr immersion: 42%Al-37%Cu-21%Mg
-1.1
(V vs. SCE) Potential
-1.2 (i) Before immersion: 55%Al-21%Cu-24%Mg -1.3 0 1000 2000 3000
Time (sec)
10000 (b) Si
8000 Cu Al Mg O (iii) 42%Al-37%Cu-21%Mg Cu 6000
(ii) 49%Al-32%Cu-19%Mg 4000
Intensity (counts) (i) 55%Al-21%Cu-24%Mg 2000
0 0246810 Energy (kV)
Figure 4.9 OCP and EDS of thin film analogs of S phase. (a) Open circuit potential of S phase in aerated 0.5 M NaCl. (b) EDS spectra of S phase in different immersion time. The intensities are offset vertically by 2000.
146
-600
0.0 Potential
-605 -0.2 )
2
-0.4 g/cm
µ
-610 -0.6
-0.8 Mass Change Potential (V vs. SCE) -615 -1.0 Mass Change (
-1.2
Potential (mV vs. SCE) -620 -1.4 0 20 40 60 80 100 120 140 160 180 200
Coating Time (sec)
Figure 4.10. CCC formation on S phase during 3-minute in Alodine 1200S.
147
420 1.0
Mass Change 410 0.8
) 400 Potential 0.6 2
g/cm
µ 390 0.4
380 0.2
Potential (V vs. SCE) (V vs. Potential 370 0.0 Mass Change ( Mass Change Potential (mV vs. SCE)
360 -0.2
350 -0.4 0 20 40 60 80 100 120 140 160 180 200
Coating Time (sec)
Figure 4.11. CCC formation on thin film of pure Al during 3-minute in Alodine 1200S.
148
Surface Height Volta Potential 0 5 µm 0 5 µm
(a) Before CCC formation
Surface Height Volta Potential 0 5 µm 0 5 µm
(b) After 3-minute CCC formation
Figure 4.12. Surface height and Volta potential of thin film analog of (a) before chromate coating and (b) after chromate coating on S phases.
149
-740 1
-760 0
-780 ) 2 -1
-800 g/cm µ Mass Change -820 -2
-840
Potential -3 vs. SCE) Potential (V Mass Change ( Mass Change Potential (mV vs. SCE) -860 -4 -880
-900 -5 0 1000 2000 3000 Time (sec)
Figure 4.13. 3-minute chromate coated S phase immersion to aerated 0.5 M NaCl for 1 hr.
150
Si 6000
Cu Al O Mg Cl Cu (d) After immersion: 37%Al-46%Cu-17%Mg 4000
(c) After immersion: 44.3%Al-33.3%Cu-22.4%Mg
2000
Intensity (counts) (b) After CCC: 55%Al-21%Cu-24%Mg
0 (a) Before CCC: 56%Al-22%Cu-22%Mg
0246810 Energy (kV)
Figure 4.14 EDS of thin film analogs of S phase. (a) and (b) are EDS spectra of S phase before and after CCC. (c) and (d) are EDS spectra of 3-minute CCC on S phase under OCP for 1 hr in 0.5 M NaCl. The intensities are offset vertically by 1800.
151
CHAPTER 5
CHROMATE CONVERSION COATING FORMATION AND BREAKDOWN ON
7X75 ALUMINUM ALLOYS
5.1 INTRODUCTION
Chromate conversion coatings are effectively and extensively used to enhance the localized corrosion resistance of high strength aluminum alloys, to improve the adhesion of paint, and to provide the metallic surface with decorative finish [1-3]. A great deal of research has been devoted to understanding the structure and chemical composition of
CCCs. It is generally accepted that CCCs are formed on metal surfaces as a result of a chemical and electrochemical processes during contact with an aqueous solution of chromic acid as a coating formation agent, hydrofluoric acid as an activator, and potassium ferricyanide as an accelerator [3, 4]. Through the process, an amorphous protective coating form which is composed of a complex chromium oxide/hydroxide and other components of the bath.
The formation of CCCs is complicated by the heterogeneous microstructure of high strength aluminum alloys. Aluminum-based precipitation age-hardened alloys, such
152
as 2xxx and 7xxx, contain intermetallic compounds (IMCs), which affect CCC formation and localized corrosion behavior. Among the intermetallic compounds, copper-containing
IMCs have been thoroughly studied [5-8]. Several physical models for CCC formation and growth have been proposed, but do not explicitly account for microstructural heterogeneity. These include a uniform coating growth model [2], an electron tunneling model [9-12], a layer structure model of spherical particles [13], and a hydrolysis- condensation polymerization model [14-17].
Using various surface-sensitive characterization techniques, protection mechanisms by CCCs have been proposed. Sato suggested a bipolar membrane model for the passivity of metals [18, 19]. An important component of protection mechanism of
CCCs is self-healing. When damage occurs, the soluble hexavalent chromate preserved in
CCCs releases into solution and moves to the active corrosion sites by diffusion and is reduced to form insoluble trivalent chrominum species, which suffocates the cathodic reactions. Corrosion at the damaged area, therefore, is inhibited [2, 16, 20, 21].
Chidambaram et al. proposed a new duplex model, which incorporated both the Sato- bipolar model and McCreery-pit repassivation model, for the protection mechanism of
CCCs [22].
Laget et al. observed the loss of corrosion protection and degradation of CCCs due to dehydration, which was associated with the structural changes in the coating [23].
They suggested that the loss of corrosion protection was directly associated with the formation of shrinkage cracks and the loss of Cr6+ leachability [23]. Most of investigations have been focused on CCC formation and growth, CCC protection
153
mechanisms, and the structure and chemistry of CCCs. There has been little attention given to relating formation and breakdown behavior on aluminum alloys.
In the current study, the electrochemical behavior of AA7075 and AA7475 during conversion coating formation and breakdown was studied using a multichannel microelectrode analyzer (MMA). In the case of AA7x75, one of the important factors in the coating formation and breakdown is the heterogeneous microstructure of the alloy substrate. For studies of coating formation and breakdown on these alloys, the MMA is useful. Using this tool, it is possible to simultaneously collect electrochemical responses from numerous identical electrodes. The MMA also allows measurement of coating formation current transients and breakdown potentials for each electrode individually, hence the distribution of breakdown potentials for an entire array. Zhang et al., used this approach to study the chromate conversion coating formation and breakdown on pure Al
[24]. Based on their results, it was suggested that there were two stages of CCC formation. The first stage consisted of a 30-second transient period characterized by measurable electrochemical activity. The second stage occurred with little measurable net activity but continuous thickening of the coating [24]. Although pure Al has been studied extensively, its practical use in structural applications is limited due to its low strength.
Therefore, it is necessary to understand the formation and breakdown of CCCs on the commercial high strength aluminum alloys such as AA7475 and AA7075, which are used in aircraft structural applications.
During aging of Al-Zn-Mg-(Cu) alloys, the main strengthening reaction is:
SSSS → GP → η’ → η (MgZn2)
154
where SSSS is the supersaturated solid solution, GP is Guinier-Preston zone, and η’ is intermediate precipitate phase. Using hot stage TEM with differential scanning calorimetry (DSC), Adler et al. found that GP zones are the predominant precipitates in the T651 temper, and η’ phases are dominant in T7 temper of AA7075 [25-27]. Park and
Ardell suggested that η predominated in AA7075-T7 and exhibited a bimodal size distribution, i.e., plate-shaped and coarse η phases and a finer η phase [28, 29]. The microstructure of T6 temper, on the other hand, contained predominantly the η’ transition phase with smaller amounts of the η phase. This fine coherent dispersion of small η’ transition particle causes maximum hardness in the T6 temper.
In this study, the effect of alloy temper on the formation and breakdown of CCC on AA7x75 has been characterized. Tempering is used to improve mechanical properties, fracture toughness, and stress corrosion resistance. In 7xxx high strength aluminum alloys, the T6 temper exhibits higher tensile strength than that of the T7 temper, while
SCC resistance of the T7 is higher than that of T6. For the study of the formation and breakdown of CCCs on AA7x75 was investigated using x-ray adsorption near-edge spectroscopy (XANES), Raman spectroscopy, a multichannel microelectrode analzer
(MMA). This MMA approach gave simple indications of coating formation and coating breakdown through evolution of electrode current. The coating thicknesses with different alloy tempers were carried out using Raman spectroscopy after chromate conversion coating formation. The Cr6+ content on coating was carried out using XANES. The effect of alloy purity and ferricyanide as ingredients in chromate bath on coating formation and breakdown was also investigated. These findings provide a better understanding of the
155
relationship between coating formation characteristics and coating breakdown during exposure to aggressive environments.
5.2 EXPERIMENTAL
5.2.1 Materials and heat-treatment
The materials used for this study were commercial 40 mm thick AA7475 and 8 mm thick AA7075. The specimens of AA7475 were cut to 5 mm thickness in order to reduce any possible effects due to quench sensitivity. The compositions of the AA7475 and AA7075 alloys were determined by inductively coupled plasma- mass spectrometry
(ICP-MS) using Perkin-Elmer Sciex ELAN 6000 (Table 5.1). The AA7475 alloy has higher purity level than AA7075. Specifically, Fe and Mn contents of AA7475 were 0.05 and 0.001 wt % while those of AA7075 were 0.28 and 0.029 wt %.
Standard heat-treatment schedules were used to achieve T6 and T7 tempers [30,
31]. Tempering treatments were identical for AA7075 and AA7475. Solution treatments of AA7475 and AA7075 were obtained by heat-treating at 515 °C and 490 °C, respectively for one hour and followed water quenching. The T6 temper was achieved by solution treatment and artificial aging for 24 hours at 120 °C. The T7 temper included over-aging at 107 °C for 6 hours and 163 °C for 24 hours.
In preparation for use in MMA experiments, samples were mechanically ground using successively finer SiC abrasive papers through 1200 grit. They were then polished using 3 and then 1 µm diamond paste on a nylon cloth. All samples were degreased ultrasonically with ethyl alcohol. During mechanical polishing, non-aqueous lubricating slurry (Blue Lube by Struers) was used to minimize any corrosion. All chemicals were 156
reagent grade from commercial vendors. All solutions were prepared with distilled water of 18 MΩ⋅cm resistivity. There was no acid and alkaline pretreatment before conversion coating. Chromate conversion coating was carried out using the Alodine 1200S product, which generally contains chromic acid, complex fluoro salts, and potassium ferricyanide.
5.4 g/L CrO3, 0.9 g/L K3Fe(CN)6, and 0.9 g/L NaF were used to prepare "in-house" chromate conversion coatings [24]. After chromate conversion coating, specimens were thoroughly rinsed with distilled water and dried for 24 hours.
5.2.2 Microelectrode arrays
A MMA (MMA 900 by Scribner Associates Inc.) was used to investigate the formation and breakdown of CCCs with a large number of samples. The small electrodes for the MMA were cut by diamond saw (ISOMETTM Low Speed Saw by Buehler). Each electrode had square cross section and was about 300 µm per side. The orientation of electrodes was such that the short-transverse orientation was the test surface. The electrode contained several grains which were 60~100 µm long and about 50 µm wide.
The electrode array contained 25 electrodes in a 5 × 5 configuration (Figure 5.1). The entire microelectrode array was mounted in a low-viscosity epoxy in order to fully wet the electrode edges. A simple diffusion path length calculation shows the chemical interaction between electrode elements was unlikely:
L = 2 Dt Eqn. 5-1 [32]
157
where L is diffusion length, D is a representative diffusion coefficient assumed to be
5×10-6 cm2/s, and t was coating time, which was 180 s in coating bath [24, 32]. The calculated diffusion length was 0.59 mm, which was much less than the 2.5 mm distance between electrodes. Therefore, chemical interaction between electrodes can be ruled out during coating formation. During potentiodynamic polarization, the duration of the experiment was 1500 s and the calculated diffusion length was 1.72 mm, which was still less than the distance between the electrodes [24].
For CCC formation, the electrode array was connected to the MMA and then immersed in the coating bath for different lengths of time. The sample rate on each microelectrode was 10 Hz. After coating formation, the electrode array was rinsed and dried in air for 24 hours. To characterize CCC breakdown, the electrode array was immersed in the aerated 0.5 M NaCl solution and then potentiodynamically polarized from the open circuit potential (OCP) of the array at a scan rate of 0.2 mV/s. Breakdown potentials on electrodes were indicated by a sharp increase in current.
5.2.3 Raman spectroscopy
Raman spectroscopy (Kaiser Optical Systems Inc.) was used to measure the 860 cm-1 peak due to Cr6+-O-Cr3+ bonding in CCCs [8, 14, 24]. The baseline corrected 860 cm-1 peak intensity gives a qualitative indication of coating thickness in CCCs [8].
Raman spectra were collected using a 514.5 nm laser and a 180° backscattered sampling geometry with a laser spot size of about 50 µm. The spectrometer was calibrated using the spectrum of naphthalene. The data acquisition time was 10 s and single spectra were collected from three different positions. A video charge-coupled device (CCD) camera
158
was used to focus the laser on the specimens. The temperature of instrument was cooled down –110 °C with liquid nitrogen.
5.2.4 X-ray absorption near edge spectroscopy
X-ray absorption near edge spectroscopy (XANES) was performed at the
Brookhaven National Laboratory of the National Synchrotron Light Source on beam line
X-19A. Spectra were obtained at the Cr K edge (E° = 5989 eV) on samples tilted at 45° to the incident beam. XANES for Cr was distinguished by a pre-edge peak for Cr(VI) at
5994 eV and an absorption edge. The edge height was total Cr content in the coating at
6250 eV. The ratio of the pre-edge peak height to the absorption edge height had a linear relationship with relative Cr(VI) content in coating [5, 23].
5.3 RESULTS AND DISCUSSION
5.3.1 Coating thickness determined by Raman spectroscopy
Raman spectroscopy was used to study chromate conversion coating formation [8,
14, 16, 17, 33]. Figure 5.2 shows Raman spectra for 3-minute CCC on AA7475-T7. The spectra were obtained of three different positions of same AA7475-T7 sample. The three spectra were identical to each other and show clearly the 860 cm-1 scattering band due to
Cr6+-O-Cr3+ bonding in the CCC [8, 14].
Figure 5.3 shows the variation in the 860 cm-1 band intensity as a function of immersion time in the coating bath for AA7075. As coating time increased from 15 seconds to 3 minutes, the band intensity in the coating increased steadily, which suggests that the relative coating thickness also increased with increasing coating time [8]. Figure
159
5.4 compares the intensity of the 860 cm-1 band on AA7475-T6 and –T7 coated for 3 minutes. This figure suggests that thicker coating forms on the alloy in the T7 temper.
Figure 5.5 shows the variation in 860 cm-1 intensity as a function of coating time for
AA7475 and AA7075 in the T6 and T7 tempers. As coating time increased, 860 cm-1 intensities generally increased. In all cases the 860 cm-1 intensity increases sharply up to about 30 seconds of coating time. Thereafter the intensity increases at a slower rate. The fact that all samples exhibited this two-stage behavior suggests that it is characteristic of
CCC growth. The two-stages of CCC growth have been discussed previously [24]. The average 860 cm-1 intensities of three samples for 3-minute CCC were as follows:
AA7475-T7 > AA7075-T7 ≈ AA7075-T6 > AA7475-T6
Considering the data from 1- and 3-minitue coating time, the plot in Figure 5.5 shows that there was no significant difference of 860 cm-1 intensities on AA7075-T6 and T7.
However, 860 cm-1 intensity of AA7475-T7 was much higher than that of AA7475-T6.
These findings suggest that alloy temper has an effect on the development of CCCs at both shorter and longer coating times, but that this effect is probably mediated by alloy impurity, i.e. constituent particles [5-8]. Alloy temper is more significant effect on 860 cm-1 intensity of AA7475, which has higher purity level than AA7075.
Figure 5.6 shows the morphologies of 3-minute CCCs on AA7075 and AA7475.
AA7075 is characterized by a higher population of Al-Fe-Cu, Mg2Si, and Al2CuMg constituent particles, which clearly affect coating formation. AA7075 contains many
160
IMCs, while AA7475 contains few particles. During heat-treatment of T6 and T7 tempers, the chemical compositions of coarse constituent IMCs are not changed.
5.3.2 Cr6+ content in the coating determined by XANES
The normalized XANES spectra in Figure 5.7 show characteristic absorption spectra for standard Cr metal, Cr(III)2O3 and Na2Cr(VI)O4 relative to Cr K-edge (E° =
5989 eV). There are two characteristic features associated with these spectra. The first is an absorption edge of the valence state of chromium, i.e. from Cr°, Cr3+ to Cr6+. As can be seen in plot, the energy of absorption edge increased with the increasing valence state of chromium. The second is a pre-edge peak due to tetrahedrally coordinated Cr6+ at 5994 eV [34]. XANES treated as a linear combination of spectra for Cr3+ and Cr6+ components allows quantitative evaluation of the Cr6+ content using ratio of absorption of Cr6+ pre- edge peak to total Cr above the edge [23, 35]. In the normalized spectra of Figure 5.8,
6+ there are three standard samples. The Cr2O3 contained almost no Cr while Na2CrO4 is
6+ almost pure Cr (96%). The pre-edge peak height of Na2CrO4 equals to the absorption
6+ edge itself. The mixture of 39% Na2Cr(IV)O4 and 61% Cr(III)2O3 has 33% Cr content in the sample. Therefore this pre-edge peak height indicates the fraction of Cr6+ in the sample. Figure 5.9 shows XANES spectra for 3-minute CCC on AA7475-T6 and –T7.
The Cr6+ contents of T6 and T7 tempers were not significantly different.
5.3.3 Formation of CCCs on electrode array
During the formation of chromate conversion coatings on electrode arrays of
AA7x75, three distinct current transient behaviors were demonstrated. These were
161
classified as net anodic behavior, net cathodic behavior, and mixed behavior. Similar transient behavior has been shown for pure Al [24]. Figure 5.10 shows net anodic behavior on an electrode during coating formation. The transient is sustained for about 30 seconds immediately after immersion to chromate bath and indicates alloy dissolution dominates the response. Net cathodic behavior was observed on many electrodes and indicates that cathodic reactions dominate the response (Figure 5.11). Other electrodes oscillated from anodic to cathodic behavior during early-stage coating formation (Figure
5.12). The different transient behaviors indicate that coating formation is not uniform on the array initially.
The length of the transients was typically about 30 s until the current decayed to background levels. This length of time is about the same as that for the rapid intensification of the 860 cm-1 band in Figure 5.5 suggesting a possible relationship between electrochemical activity and rapid coating growth in early stage coating formation.
In terms of occurrence, mixed behavior was observed more frequently than net anodic and cathodic behaviors. It was also noted that the peak current in mixed behavior electrodes, which was about 3×10-7 A, was typically smaller than that of the net anodes and cathodes. The duration of the electrochemical transients was about 30 seconds. These transients were similar to those observed on pure Al [24]. No aspect of the transient behavior appeared to depend on alloy purity or alloy temper. Characteristics considered in this regard were transient intensity, transient duration and polarity of electrode behaviors (e.g. net anodic, cathodic, or mixed).
162
5.3.4 Breakdown of CCCs on electrode array
Figure 5.13 shows representative breakdown behavior of bare and 3-minute chromate coated electrode arrays of AA7075-T7 in 0.5 M NaCl solution during anodic polarization. The breakdown behavior of the bare alloy was clearly seen during potentiodynamic polarization. The sharp current increase was observed at -735 mVSCE due to pitting. The breakdown behavior of the chromate coated alloy showed a gradual increase of current starting at about -700 mVSCE. Overall, the breakdown potential of the chromate coated alloy was higher than that of bare alloy, as might be expected.
Metastable pitting was sometimes observed with the bare alloy. No metastable pitting was detected in chromate coated alloy. This result possibly indicates that chromate conversion coatings provide anodic inhibition.
Ensembles of breakdown potential measurements were presented graphically as cumulative probability plots. Figure 5.14 shows a cumulative probability plots with breakdown potential distributions (BPDs) for bare AA7075 and AA7475 in the T6 and
T7 tempers during exposure to 0.5 M NaCl. The breakdown potentials of T7 tempers of both alloys were higher than those of T6 tempers at low cumulative probability, but the relationship is reversed at high cumulative probability.
Breakdown potentials at the low end of the distribution are likely to be from samples with local sites prone to breakdown. The breakdown potentials at low cumulative probability are believed to be representative of breakdown potentials on large area electrodes (e.g. > 0.1 cm2) that are likely to include such sites. In fact, the breakdown potentials at low cumulative probability match well with breakdown potentials measured on 0.15~0.2 cm2 electrodes. They also reflect the relationship
163
between breakdown potentials in T6 and T7 tempers (Figure A.7). It should be noted that
Ramgopal showed that the breakdown potential of AA7150-T7 was higher than that of
AA7150-T6 [37].
The slope of a cumulative distribution plot is an indication of variability in the quantity being analyzed. In the plot format used here, lower slope indicates increased variability. In Figure 5.14, the BPD for both alloys in the T6 tempers is much more variable than the BPD for the alloys in the T7 tempers. The lower variability in the T7 temper suggests that a single process or a single type of site governs breakdown. In the
T6 temper, breakdown may not be governed by a single site or is governed by a site whose resistance to breakdown is variable. BPDs were independent of alloy purity. Alloy temper had a more significant effect on breakdown potential of bare AA7075 and
AA7475 than alloy purity (Figure 5.14). These results are consistent with those of Meng et al., who suggested that the first breakdown potential of 7xxx is associated with matrix attack rather than breakdown at particles [38].
In general, AA7x75-T6 has two breakdown potentials. However, in the electrode array measurements at T6 electrodes, just one-breakdown potential was observed due to current limitation of the MMA. The MMA system used in this study was capable of measuring a maximum current at 10-6 A. For the nominal electrode areas used in these arrays, the maximum area was almost 10-3 cm2. The current density of breakdown potential for individual electrode was 10-3 A/cm2, which was slightly lower than the current density needed to resolve the second breakdown potential of T6 temper.
Figure 5.15 shows the BPD for a 3-minute chromate conversion coating on
AA7075 in the T6 and T7 tempers. Chromate coatings increased the breakdown potential
164
by about 50 mV in the noble direction compared to the bare alloy. The slope of BPD of chromate coated AA7075 was not changed. The chromate coated AA7075-T7 was steeper than that of T6; like the bare metal. The breakdown potentials of T6 temper were generally higher than those of T7, while at low cumulative probability the breakdown potentials of T7 temper were higher than those of T6. Figure 5.16 shows the BPDs of 3- minute CCCs on AA7475-T6 and -T7. Chromate coated AA7475 had higher breakdown potential than the bare alloy. Generally, chromate coated AA7475-T7 had higher breakdown potential than AA7475-T6. There was no significant difference of breakdown potential between chromate coated AA7475-T6 and bare metal at high cumulative probability.
Figure 5.17 shows that the effect of current transient polarity during coating formation on breakdown behavior of 3-minute CCCs. Net cathodes are perhaps slightly more resistant to breakdown than net anodes. BPDs between net anode and net cathode differed by about 20 mV. In similar experiments carried out with pure Al, electrodes that showed net cathodic behavior during early coating formation had breakdown potentials that were about 100 mV more positive than electrodes with net anodic behavior [24].
5.3.5 The effect of coating bath chemistry on CCC formation and breakdown
Common commercial chromate conversion coating formulations contain 3 main ingredients: chromate as a coating forming agent, fluoride as an activator, and ferricyanide as an accelerator. To understand the effect of each component of the bath on the coating formation, conversion coating baths with one or more of these components
165
withheld were synthesized. Formation and breakdown characteristics of coatings formed in these baths were compared to CCCs formed in an Alodine 1200S bath.
Figure 5.18 compares net anodic transients measured an AA7475-T6 in an
Alodine 1200S bath (ferricyanide-accelerated), a bath consisting of (Cr+F+Fe), and a bath consisting of (Cr+F). The bath consisting of (Cr+F+Fe) has the main ingredients of
Alodine 1200S, but lacks minor ingredients including tetrafluoroborate (KBF4) and potassium hexafluorozirconate (K2ZrF6) [39]. The bath consisting of (Cr+F) lacks ferricyanide and minor ingredients. In all three baths, the peak currents achieved and the transient durations are similar showing that these variations in bath chemistry do not have a significant effect on the evolution of anodic behavior during coating formation.
This is not the case for net cathodic behavior (Figure 5.19). For the current evolution associated with net cathodic behavior, there were significant changes between solutions during coating formation. In Alodine 1200S solution, there was a sharp increase in the current transient. In (Cr+F+Fe) and (Cr+F) solutions, the current transients gradually increased showing that the ingredients in Alodine 1200S strongly influence the current evolution of net cathodes. Chidambaram et al. found that hexafluorozirconate had significant effect on the surface activation of pure Al compared to tetrafluoroborate and fluoride [39]. These complex ions containing fluoride may effectively activate the surface of alloy matrix to support chromate reduction at high rates.
Figure 5.20 shows the effect of ferricyanide on breakdown potential of 3-minute
CCC on AA7075-T7. It also shows coating thickness as a function coating time as indicated by the 860 cm-1 Raman intensity. When ferricyanide is added to the coating bath, the increase of breakdown potential was more than 50 mV at the BPD median
166
compared to when ferricyanide was absent. The 860 cm-1 intensity without ferricyanide was lower than that with ferricyanide, which highlights the role of ferricyanide as a promoter of coating formation.
Figure 5.21 shows the effect of ferricyanide, alloy temper and alloy purity on 860 cm-1 intensity. Generally, the thickness in (Cr+F+Fe) bath was higher than (Cr+F) bath showing action of ferricyanide as an accelerator. It is well known that the coating thickness, coating weight, coating formation, and coating resistance are increased in ferricyanide containing solution [1, 4, 20, 40]. In particular, thickness of AA7475-T7 was always higher than AA7475-T6 in both (Cr+F) and (Cr+F+Fe) solutions. The thickness of AA7475 in (Cr+F+Fe) solutions was always greater than (Cr+F) solutions in both T6 and T7 tempers. However, the thickness of AA7075-T7 was greater than that of AA7075-
T6 in (Cr+F) solution while the thicknesses of AA7075-T7 and AA7075-T6 were similar in (Cr+F+Fe) solutions. Similar behavior is observed with Alodine 1200S for AA7075
(Figure 5.5). These results show that coating ingredients have a significant effect on coating formation and breakdown compared to alloy temper and alloy purity.
5.4 SUMMARY
The chromate conversion coating formation and breakdown of 7xxx aluminum alloys was investigated by using Raman spectroscopy, XANES and MMA. The core results are as follows:
1. As coating time increases during coating formation, coating thickness increases.
The coating thickness is dependent on alloy temper and alloy purity. The coating
167
thickness on 3-minute CCCs from Alodine 1200S solution is as follows: AA7475-
T7 > AA7075-T7 ≥ AA7075-T6 > AA7475-T6.
2. Results from coating formation experiments suggest that coating formation on
AA7x75 is not uniform in terms of coating thickness and perhaps chemistry.
3. Net cathodes have slightly higher breakdown potentials than net anodes. There is
no significant difference of current evolution during coating formation with alloy
temper and alloy purity. Current transients during coating formation are sustained
for a 30-second period and then drop to the background levels, though coating
thickness continues to increase up to 3 minutes of immersion.
4. Alloy temper has a more significant effect on breakdown potential of bare
AA7x75 than alloy purity. Chromate conversion coatings increase the breakdown
potentials of alloys in the noble direction. After CCC formation on AA7475, the
T7 temper has higher breakdown potential distribution than the T6 temper due to
higher coating thickness in the T7 temper, though there were similar Cr6+ contents
between T6 and T7 tempers. For the AA7075, T7 and T6 tempers have similar
breakdown potential distribution perhaps due to similar coating thicknesses.
5. The presence of ferricyanide increases the coating thickness and corrosion
resistance of coating in a synthesized chromate bath. AA7475 is more
significantly influenced by alloy temper than AA7075.
168
REFERENCES
1. A.E. Hughes, R.J. Taylor, and B.R.W. Hinton, Surface and interface analysis, 25, 223 (1997).
2. H.A. Katzman, G.M. Malouf, R. Bauer, and G.W. Stupian, Applications of Surface Science, 2, 416 (1979).
3. K.A. Korinek, Chromate Conversion Coatings, Vol. Metals Park, OH: ASM (1987).
4. P.L. Hagans and C.M. Haas, Chromate Conversion Coatings, Vol. Metals Park, OH: ASM International (1994).
5. J.R. Waldrop and M.W. Kendig, J. Electrochem. Soc., 145, L11 (1998).
6. G.M. Brown and K. Kobayashi, J. Electrochem. Soc., 148, B457 (2001).
7. Y. Baek and G.S. Frankel, J. Electrochem. Soc., 150, B1 (2003).
8. W.R. McGovern, P. Schmutz, R.G. Buchheit, and R.L. McCreery, J. Electrochem. Soc., 147, 4494 (2000).
9. G.M. Brown, K. Shimizu, K. Kobayashi, G.E. Thompson, and G.C. Wood, Corrosion Science, 33, 1371 (1992).
10. G.M. Brown, K. Shimizu, K. Kobayashi, G.E. Thompson, and G.C. Wood, Corrosion Science, 34, 1045 (1993).
11. G.M. Brown, K. Shimizu, K. Kobayashi, G.E. Thompson, and G.C. Wood, Corrosion Science, 34, 2099 (1993).
12. G.M. Brown, K. Shimizu, K. Kobayashi, G.E. Thompson, and G.C. Wood, Corrosion Science, 35, 253 (1993).
13. D.J. Arrowsmith, J.K. Dennis, and P.R. Sliwinski, Transactions of the Institute of Metal Finishing, 62, 117 (1984).
14. L. Xia and R.C. McCreery, J. Electrchem. Soc., 145, 3083 (1998).
15. G.S. Frankel, Mechanism of Al Alloy Corrosion and the role of Chromate Inhibitors, Air Force Office of Scientific Research, Contract No. F49620-96-1- 0479, Final Report (2001).
16. L. Xia and R.L. McCreery, J. Electrochem. Soc., 146, 3696 (1999).
17. L. Xia, E. Akiyama, G. Frankel, and R. McCreery, J. Electrochem. Soc., 147, 2556 (2000). 169
18. N. Sato, Corrosion, 45, 354 (1989).
19. N. Sato, Corrosion Science, 31, 1 (1990).
20. P.L. Hagans and C.M. Haas, Surface and Interface Analysis, 21, 65 (1994).
21. J. Zhao, G.S. Frankel, and R.L. McCreery, J. Electrochem. Soc., 145, 2258 (1998).
22. D. Chidambaram, C.R. Clayton, and G.P. Halada, J. Electrochem. Soc., 150, B224 (2003).
23. V. Laget, C.S. Jeffcoate, H.S. Isaacs, and R.G. Buchheit, J. Electrochem. Soc., 150, B425 (2003).
24. W. Zhang, B.L. Hurley, and R.G. Buchheit, J. Electrochem. Soc., 149, B357 (2002).
25. P.N. Adler, R. DeIasi, and G. Geschwind, Metall. Trans., 3, 3191 (1972).
26. P.N. Adler and R. DeIasi, Metall. Trans. A, 8A, 1185 (1977).
27. R. DeIasi and P.N. Adler, Metall. Trans. A, 8A, 1177 (1977).
28. J.K. Park and A.J. Ardell, Metall. Trans. A, 14A, 1957 (1983).
29. J.K. Park and A.J. Ardell, Mater. Sci. Engng., A114, 197 (1989).
30. ASM International, ASM handbook: Vol. 4 Heat Treating, Metal Parks: American Society for Metals (1991).
31. H.E. Boyer and T.L. Gall, Metals Handbook, Metal Parks: American Society For Metals (1985).
32. A.J. Bard and L.R. Faulkner, Electrochemical Methods: Fundamentals and Applications, New York: John Wiley & Sons (2001).
33. J.D. Ramsey, L. Xia, M.W. Kendig, and R.L. McCreery, Corrosion Science, 43, 1557 (2001).
34. H.A. Hall, Corrosion, 23, 173 (1967).
35. M.W. Kendig, A.J. Davenport, and H.S. Isaacs, Corrosion Science, 34, 41 (1993).
36. T. Ramgopal, Role of Grain boundary precipitates and Solute depleted zone in the Intergranular corrosion of Aluminum Alloy AA7150, The Ohio State University, (2001).
170
37. Q. Meng and G.S. Frankel, J. Electrochem. Soc., 151, B271 (2004).
38. P.L. Hagans and C.M. Haas, Surface Engineering, 13, 5, (1996).
171
TABLES
Alloy Zn Mg Cu Cr Fe Si Mn Ti Al (wt%)
7475 5.64 2.27 1.35 0.22 0.05 0.021 0.001 0.044 Balanced
7075 5.52 2.44 1.41 0.17 0.28 0.019 0.029 0.033 Balanced *Average of 3 samples
Table 5.1: Chemical composition of AA7475 and AA7075 alloys using ICP-MS.
172
FIGURES
300×300 µm
2.5 mm
Figure 5.1. Photograph of 5×5 electrodes array configuration. Electrodes spacing was 2.5 mm spacing within low-viscosity epoxy.
173
0.012
0.010
0.008
0.006
860 cm-1 Cr6+-O-Cr3+ 0.004
Intensity (a.u.) Intensity
0.002
0.000
-0.002 500 1000 1500 2000 Raman Shift (cm-1)
Figure 5.2. Raman spectra of 3-minute CCC on AA7475-T7. The spectra were obtained three different positions of AA7475-T7. Three spectra are identical. The intensity of 860 cm-1 indicates the thickness in the coating. The units are arbitrary intensity to relative Raman shift.
174
1.4e-3 15 sec 30 sec 1.2e-3 1 min 3 min 1.0e-3
8.0e-4
intensity (a.u.) intensity (a.u.)
6.0e-4
-1 (a.u.) Intensity 4.0e-4
860 cm 2.0e-4
0.0 700 750 800 850 900 950 1000 -1 Raman Shift (cm )
Figure 5.3. Raman spectra for 860 cm-1 intensity of CCC on AA7075-T6 at different coating times.
175
2.5e-3
T6
T7 2.0e-3
1.5e-3
1.0e-3 intensity (a.u.) intensity (a.u.)
Intensity (a.u.) Intensity -1
5.0e-4 860 cm
0.0 700 750 800 850 900 950 1000
Raman Shift (cm-1)
Figure 5.4. Raman spectra showing 860 cm-1 intensity of 3-minute CCC on AA7475 in different tempers.
176
14000 1st stage
growth nd 12000 2 stage growth
10000
8000
intensity (a.u.) intensity (a.u.) 6000
-1
intensity (a.u.)
6+ 4000 Cr
860 cm AA7075-T6 2000 AA7075-T7 AA7475-T6 0 AA7475-T7
0 20 40 60 80 100 120 140 160 180 200
Coating Time (sec)
Figure 5.5. The 860 cm-1 intensity of Raman spectra on CCCs for both AA7075 and AA7475 in T6 and T7 tempers.
177
(a)
Al-Fe-Cu IMCs
Mg Si IMCs Al2CuMg IMC 2
(b)
Figure 5.6. Scanning electron micrographs of the morphology of 3-minute CCCs on (a) AA7075-T7 and (b) AA7475-T7.
178
1.4 Pre-edge peak of Cr6+ 1.2
1.0
0.8
0.6 Absorption 0.4 Edge Cr0 metal Normalized Counts Cr2O3 Normalized absorbance 0.2 Na2CrO4
0.0
-0.2 -20 0 20 40 60
E-E (eV) 0
Figure 5.7. XANES spectra for Cr metal, Cr(III)2O3 and Na2Cr(VI)O4 showing normalized absorbance against energy relative to Cr K edge (E0 = 5989 eV). The energy of absorption edge increased with the valence state of chromium, i.e. from Cr0, Cr3+ to Cr6+. The pre-edge peak of Cr6+ is at 5994 eV.
179
3.0
2.5
2.0
1.5
1.0
Normalized Counts Cr(III) O 0.5 2 3 Normalized absorbance 3 min CCC AA7475-T7 39%Cr(VI) + 61%Cr(III)
0.0 Na2Cr(VI)O4
-0.5 -20 0 20 40 60
E-E0 (eV)
Figure 5.8. XANES spectra for standard Cr2O3 and Na2CrO4 with mixture between 39% Na2CrO4 and 61% Cr2O3 showing the spectrum from a 3-minute CCC on AA7475-T7, and the spectra are offset vertically by 0.5.
180
8
6
4 2.0 1.8 1.6 2 1.4 1.2
Absorbance (a.u.) 1.0 0.8 0 AA7475-T6 0.6 0.4 AA7475-T7 4.0 4.5 5.0 5.5 6.0 6.5 7.0
-2 0 100 200 300 E-E (eV) 0
Figure 5.9. XANES spectra for 3-minute CCC on AA7475-T6 and -T7 showing absolute absorbance with with relative energy.
181
5e-7 1st stage 2nd stage growth 4e-7 growth
3e-7
2e-7
(A) Current 1e-7
0
Start of CCC formation -1e-7 0 20406080100120
Time (sec)
Figure 5.10. Typical current vs. time plots for anodic behavior of CCC formation on AA7475-T6. The peak current sustained about 30 sec after immersion.
182
2e-7
0 Start of CCC formation -2e-7
st -4e-7 1 stage growth nd (A) Current -6e-7 2 stage growth
-8e-7
-1e-6 0 20 40 60 80 100 120
Time (sec)
Figure 5.11. Typical current vs. time plot for cathodic behavior of CCC formation on AA7475-T6. The peak current sustained about 30 sec after immersion.
183
4e-7
2e-7
0 Start of CCC formation
-2e-7 st Current (A) Current 1 stage 2nd stage growth growth -4e-7
-6e-7 0 20 40 60 80 100 120
Time (sec)
Figure 5.12. Typical current vs. time plot for cathodic and anodic mixed-behavior of CCC formation on AA7475-T6. The peak current sustained about 30 sec after immersion.
184
1.2e-6
1.0e-6
8.0e-7
6.0e-7
4.0e-7
Current (A)
2.0e-7
0.0 3 min CCC bare Metal
-2.0e-7 -0.85 -0.80 -0.75 -0.70 -0.65 -0.60 -0.55
Breakdown Potential (V vs. SCE)
Figure 5.13. Typical breakdown behavior for AA7075-T7 during anodic polarization with and without 3-minute CCC in 0.5 M NaCl solution.
185
1.2
1.0
0.8
0.6
0.4
AA7075-T6 AA7075-T7 0.2 AA7475-T6 Cumulative Probability of Breakdown Probability Cumulative AA7475-T7
0.0 -0.80 -0.76 -0.72 -0.68 -0.64 -0.60 -0.56
Breakdown Potential (VSCE)
Figure 5.14. Cumulative probability with breakdown potential distribution for AA7075 and AA7475 in the T6 and T7 tempers in 0.5 M NaCl solution.
186
1.2
1.0
0.8
0.6
0.4 AA7075-T6
Cumulative Probability 0.2 AA7075-T7 AA7075-T6 CCC
AA7075-T7 CCC 0.0 -0.80 -0.75 -0.70 -0.65 -0.60 -0.55 -0.50
Breakdown Potential (V ) SCE
Figure 5.15. BPDs for 7075-T6 and T7 with and without 3-minute CCC.
187
1.2
1.0
0.8
0.6
0.4
AA7475-T6
Cumulative Probability 0.2 AA7475-T7 AA7475-T6 CCC AA7475-T7 CCC 0.0 -0.80 -0.75 -0.70 -0.65 -0.60 -0.55 -0.50
Breakdown Potential (VSCE)
Figure 5.16. BPDs for 7475-T6 and -T7 with and without 3-minute CCC.
188
1.2
1.0
0.8
0.6
0.4
Probability Cumulative
0.2 Net Anode Net Cathode Mixed Character 0.0 -0.72 -0.71 -0.70 -0.69 -0.68 -0.67
Breakdown Potential (V vs. SCE)
Figure 5.17. Effect of polarity on breakdown behavior for 3-minute CCC of AA7475-T7 as by anodic polarization determined in aerated 0.5 M NaCl solution.
189
1.5e-6
Alodine 1200s solution 1.0e-6
Cr + F + Fe coating solution 5.0e-7 (A) Current
Cr + F coating solution 0.0 Start of coating formation
0 1020304050 Time (sec)
Figure 5.18. Current evolution of net anodic behavior for CCC formation on AA7475-T6 Cr+F, Cr+F+Fe, Alodine 1200S solutions. The currents are offseted vertically by 5×10-8 A between plots.
190
1.5e-6
Alodine 1200s solution
1.0e-6
Cr + F + Fe coating solution
5.0e-7
(A) Current Cr + F coating solution 0.0
Start of coating formation
-5.0e-7 0 1020304050
Time (sec)
Figure 5.19. Current evolution of net cathodic behavior for CCC formation on AA7475- T6 in Cr+F, Cr+F+Fe, Alodine 1200S solutions. The currents are offseted vertically by 5×10-8 A between plots.
191
1.2 (a) 1.0
0.8
0.6
0.4 AA7075-T7
Cumulative Probability Cumulative 0.2 CCC Cr+F Cr+F+Fe 0.0 -0.80 -0.75 -0.70 -0.65 -0.60 -0.55 -0.50
Breakdown Potential (VSCE)
15000 (b) Cr+F+Fe solution
Cr+F solution 10000 Alodine 1200s
intensity (a.u.)
1 - intensity (a.u.) intensity 5000 6+ Cr 860 cm
0 0 50 100 150 200 250
Coating time (sec) 3- Figure 5.20. Effect of Fe(CN)6 on (a) coating breakdown potential distribution and (b) 860 cm-1 intensity of AA7075-T7.
192
Empty 15000 AA7475-T7_(Cr+F+Fe) AA7075-T6_(Cr+F+Fe) symbols: AA7075-T7_(Cr+F+Fe) Cr+F+Fe AA7475-T7_(Cr+F) bath AA7475-T6_(Cr+F+Fe) AA7075-T7_(Cr+F) Solid 10000 AA7475-T6_(Cr+F) symbols: AA7075-T6_(Cr+F) Cr+F bath
intensity (a.u.) intensity (a.u.)
intensity (a.u.) intensity
6+ -1 5000 Cr
860 cm 0
0 50 100 150 200 250
Coating time (sec)
Figure 5.21. Effect of ferricyanide, alloy temper and alloy purity on the 860 cm-1 intensity.
193
CHAPTER 6
EFFECT OF ALLOY TEMPER ON CHROMATE CONVERSION COATING
PERFORMANCE OF Al-Zn-Mg-Cu ALLOYS
6.1 INTRODUCTION
It is known that subtle changes in the heat treatment can influence localized corrosion in Al-Zn-Mg-Cu high strength aluminum alloys. Artificial aging to a T6 temper results in maximum mechanical strength. However, the stress corrosion cracking (SCC) resistance of this temper is relatively low. Overaging to a T7 temper improves the SCC resistance at the expense of a 10 to 15% reduction in mechanical strength [1]. Many studies tried to explain the susceptibility to intergranular corrosion (IGC) in Al-Mg-Zn-
Cu alloy. The alloys in T7 and T6 tempers have one breakdown potential and two breakdown potentials, respectively [2-4]. Maitra and English suggested that the alloy was susceptible to IGC between the active and noble breakdown potentials in T6 temper [2].
Ramgopal et al. suggested that high Cu enriched grain boundary precipitates raised breakdown potential to more noble potential, which caused the IGC and SCC resistance in T7 temper [3]. Meng et al. also suggested that the first breakdown potential was
194
associated with the Cu content in precipitate hardening particles and the second breakdown potential was associated with IGC and selective grain attack in T6 temper [4].
More recently, retrogression and reaging tempers (RRAs) have been developed, which enable alloys to exhibit the mechanical properties of T6 temper with SCC resistance of the T7 temper [5].
Microstructural heterogeneity, including precipitate hardening particles and coarse constituent particles makes Al-Zn-Mg-Cu alloys susceptible to localized corrosion, such as pitting, crevice corrosion, IGC and exfoliation corrosion. In order to prevent localized corrosion, chromate conversion coatings (CCCs) are used alone or in conjunction with organic top coats. Because the structure and composition of CCCs is affected by the microstructure of the alloy on which it forms, CCC performance is expected to depend to some extent on alloy temper.
Many recent studies have focused on the formation of CCCs, mainly on 2xxx Al-
Cu-Mg alloys. On the basis of this work, the formation of CCCs is understood to involve a redox reaction between chromate in solution and the aluminum [6-8]. The chromate oxidizes the aluminum in the presence of fluoride and Cr6+ in solution is reduced to Cr3+.
These Cr6+/Cr3+ redox reactions are accompanied by an increase of the local pH. The pH increase results in the precipitation of an amorphous mixture of hydrated aluminum and mixed Cr3+ and Cr6+ oxides [9, 10].
Both Cr3+ and Cr6+ in CCCs contribute to coating corrosion resistance. The
3+ insoluble Cr species is present as a Cr(OH)3 polymer network that acts as a barrier to aggressive environment, while soluble Cr6+ species is mobile and releasable providing active corrosion protection in the form of self-healing [10-15].
195
In general, corrosion protection by CCCs increases with coating thickness and
Cr6+ content in coating. For this reason, the ability to accurately measure Cr6+ in CCCs is needed. X-ray photoelectron spectroscopy (XPS) measurements have been used in the determination of Cr6+ contents, however, reduction of Cr6+ in the ultrahigh vacuum environments during XPS measurement prevent accurate determination of Cr6+ species [8,
16]. X-ray absorption spectroscopy (XAS) is an attractive alternative to XPS because this technique can be performed under ambient conditions. X-ray absorption near edge structure (XANES), which covers the range near the threshold for the absorption, is one form of XAS. Kendig et al. used XANES to show that Cr6+ content was about 20 % of the total chromium content on a 5-minute CCCs of AA2024-T3. It was also observed that the Cr6+ ratio increased with coating time up to 5 minutes [6]. Lytle et al. found that the
Cr6+ content was about 23% on AA2024-T3 in Alodine 1200S solution for 1-minute and
3-minute coating [17]. Wan and Thompson observed that Cr6+ content was 30% on pure
Al [18].
In this study, the effect of alloy temper on CCC performance in AA7x75 was examined using several methods including electrochemical impedance spectroscopy
(EIS), XANES, and salt spray fog exposure. The objective of this study was to characterize the magnitude of any temper effect in terms of the resulting coating chemistry and corrosion protection afforded by the coating. An additional objective was to compare the effect of temper on CCC protection against other factors that influence coating protectiveness including alloy purity, surface pretreatment and coating time.
196
6.2 EXPERIMENTAL
6.2.1 Materials and Heat-treatment.
The materials used for this study were commercial 40 mm thick AA7475 (Al-
5.64Zn-2.27Mg-1.35Cu) and 8 mm thick AA7075 (Al-5.52Zn-2.44Mg-1.41Cu). Alloy
AA7475 has higher purity than AA7075. The minor elements in AA7475 were 0.22 Cr,
0.05 Fe, 0.021 Si, 0.001 Mn, and 0.044 Ti 0.044. In AA7075, low level additions were
0.17 Cr, 0.28 Fe, 0.019 Si, 0.029 Mn, and 0.033 Ti with balance Al (wt%). The compositions of the AA7475 and AA7075 alloys were determined by inductively coupled plasma mass spectrometry (ICP-MS) using Perkin-Elmer Sciex ELAN 6000 (Table 6.1).
The specimens for electrochemical impedance experiments of AA7475 were cut to 5 mm thickness in order to reduce the effect of thickness during heat treatment. The specimens were solution-heat treated, quenched and then aged to W, T6, T7, and RRA tempers [19, 20]. Figure 6.1 shows the heat-treatment schedules used. Samples were mechanically ground using successively finer SiC abrasive papers through 1200 grit.
They were then polished using 3 and 1 µm diamond paste on a nylon cloth. All samples were degreased ultrasonically with ethyl alcohol. During preparation of samples, a non- aqueous lubricating slurry (Blue Lube by Struers) was used to minimize localized corrosion.
6.2.2 Chemicals and chromate conversion coatings.
All chemicals used were reagent grade and were obtained from commercial vendors. All solutions were prepared with deionized water with 18 MΩ⋅cm resistivity.
Chromate conversion coating (CCC) formation was carried out using 7.55 g of Alodine
197
1200S powder in 1000 ml deionized water and adjusted pH to 1.5 ~ 1.7 by HNO3 at room temperature. The samples were immersed in Alodine solution for 1-minute and 3-minute intervals, then rinsed thoroughly with deionized water and dried in air for 24 hours.
Before chromate coating, some samples were cleaned in an alkaline bath and deoxidized in an acid bath as pretreatment (Table 6.2).
6.2.3 Electrochemical measurements.
Electrochemical impedance spectroscopy (EIS) was used to evaluate the corrosion behavior of chromate-coated surfaces in aerated 0.5 M NaCl solution. A Princeton
Applied Research (PAR) 273A potentiostat and Solartron 1255 frequency response analyzer with ZplotTM software were used to perform the measurements. A single sinusoidal potential of 10 mV amplitude was superimposed on the open-circuit potential
(OCP). Measurements were made between 10 mHz and 10 kHz from high to low values.
The sample area was 1.0 cm2.
EIS data were evaluated by equivalent circuit modeling. The physical model and corresponding equivalent circuit model that formed the basis of this analysis are shown in
Figure 6.2. The response of this intact conversion coating was represented by oxide capacitance (Coxide) and oxide resistance (Roxide). Cracks in the coating were treated defects that could contribute to measured impedance. Crack defects were represented with a defect capacitance (Cd), a defect resistance (Rd) and a pore resistance (Rp). Rs in
Figure 6.2 refers to bulk solution resistance. A simplified single time constant model derives from the model in Figure 6.2 if it is assumed that Roxide is very large it behaves an open while Rp behaves as a short. The oxide capacitance, Coxide and defect capacitance,
198
Cd add in parallel to give an imperfect capacitive response introduced by a constant phase element (CPE). Figure 6.3 shows the simplified equivalent circuit model of chromate conversion coating for this data analysis. The defect resistance is termed Rcoat and is an overall indication of coating protection that likely indicates the resistance at these types of defects:
1 Z CPE = Eqn. 6-1 C( jw) P
where C is an amplitude [Ω⋅s1-P] and P is an exponent that varies from -1 to 1. For these experiments, the values of p ranged from 0.85 to 1.0. A P value of 1.0 corresponds to perfect capacitive behavior. Through EIS experiments, the solution resistance, Rs was found to be about 20 Ω. Typical Bode and Nyquist plots representing the EIS response for chromate coated AA7475-T7 in 0.5 M NaCl solution are shown in Figure 6.4.
Complex non-linear least-squares (CNLS) fits are also shown. Overall, the simplified model in Figure 6.3 appears to match the experimental data well.
6.2.4 X-ray absorption near edge spectroscopy
X-ray absorption near edge spectroscopy (XANES) was obtained at the
Brookhaven National Laboratory of the National Synchrotron Light Source on beam line
X-19A. Spectra were obtained at the Cr K edge (E° = 5989 eV) on samples. XANES for
Cr was distinguished by a pre-edge peak for Cr(VI) at 5994 eV and total Cr was determined at 6250 eV. The total Cr content and Cr(VI) content were measured at edge height and pre-edge peak height, respectively [6, 17, 21].
199
6.2.5 Salt spray exposure.
The salt spray exposure testing was conducted using Q-Fog (Q-Panel Lab
Products) according to ASTM B117 [22]. The fog was generated from a 5 wt % sodium chloride solution whose pH ranged from 6.5 to 7.2 pH. The fog deposition rate ranged from 1 to 2 ml/h and the temperature of chamber was 35 °C. The sample sizes of
AA7075 and AA7475 were 4×5” and 1×1.5” (limited amount of material) respectively.
The front-to-back spacing of samples in the panel mounting racks was 0.5 inch. And all samples faced the same way at the angle of 60◦ from horizontal. After exposure and rinsing samples with deionized water thoroughly, corrosion damages were inspected visually.
6.3 RESULTS
6.3.1 EIS of conversion-coated samples
The EIS response of chromate coated AA7x75 was evaluated to characterize the effects of alloy temper, alloy purity, immersion time, and pretreatment on corrosion behavior. Figure 6.5 shows typical impedance spectra of 3-minute CCCs for AA7475-T7 temper collected during immersion in 0.5 M NaCl. The total impedance increased up to 7 days immersion. After 7 days immersion, the impedance tended to decrease. Figure 6.6 shows the effect of coating time on the impedance response of 1-minute and 3-minute
CCCs on AA7475-T7 substrates exposed to aerated 0.5 M NaCl. The maximum corrosion resistance of 3-minute coating was 7.0 MΩcm2 at 72 hr immersion while that of 1-minute coating was 3.17 MΩcm2 at 104 hr immersion. As might be expected, 3- minute coatings consistently demonstrated greater impedance than 1-minute coatings. 200
Figure 6.7 and Figure 6.8 show the effect of temper on the impedance for 1 and
3-minute CCCs on AA7475. As can be seen in Figure 6.7, Rcoat for a 3-minute coating on
AA7475-T6 was far below 1 MΩcm2 while that for other tempers were above 1 MΩcm2.
Figure 6.8 shows EIS response of 1-minute CCCs at 68 hr immersions in 0.5 M NaCl.
The coating resistance of T7 and RRA tempers are 2.94 and 1.06 MΩcm2 while those of
T6 and W tempers are 0.57 and 0.45 MΩcm2. The corrosion resistances with various tempers of 1 and 3-minute CCCs on AA7475 are shown in Figure 6.9. The impedance parameters of 1 and 3-minute CCCs on AA7475 are summarized in Table 6.3 and Table
6.4. The corrosion resistances of AA7475-T7 with 1-minute and 3-minute CCCs were always higher than those of AA7475-T6. The retrogression and reaged (RRA) temper and solution treated (W) temper behaved like T7-tempered alloys with 3-minute CCC but behaved like T6-tempered alloys in 1-minute CCC on AA7475. The exponent value of
CPE, P, in T6 temper on 3-minute CCCs was much lower than those of other tempers indicating greater dispersion in the capacitive response. In addition, the amplitude of CPE,
C, in T6 temper was also higher than those of other tempers; consistent with a greater contribution from the defect capacitance.
Figure 6.10 and Figure 6.11 show the impedance spectra on 3-minute chromate coated AA7075 during 7 days immersion in 0.5 M NaCl. Figure 6.12 shows the effect of alloy temper on corrosion resistance of AA7075-T6 and -T7 as a function of immersion time up to 7 days. These data show that the corrosion resistance between T6 and T7 tempers is much less than that in AA7475 (as in Figure 6.9 for example). It is possible that the presence of an increased concentration of IMCs in AA7075 diminishes the effect of temper on the EIS response. There is a significant spectra difference between T6 and
201
T7 tempers at low frequencies (Figure 6.10 and Figure 6.11). T6 tempered alloy exhibited a distinct DC limit while T7 tempered alloys exhibited a low frequency response that was suggestive of a diffusional impedance. The different behaviors are presently not understood.
Pretreating alloy surfaces prior to corrosion coating did not significantly affect corrosion resistances on chromate coated AA7475. Figure 6.13 and Figure 6.14 show the corrosion resistance on 1-minute and 3-minute chromate coated AA7475 during 168 hr immersion in 0.5 NaCl without pretreatment. These surfaces were degreased and deoxidized according to the procedures shown in Table 2 immediately prior to conversion coating. T7 temper of 1-minute and 3-minute coated AA7475 have higher corrosion resistance than T6 temper. However, the RRA and W tempers behaved like T7-tempered alloys with 1-minute CCC but behaved like T6-tempered alloys in 3-minute CCC on
AA7475. No general conclusions could be drawn for effects of RRA and W tempers on corrosion resistance. Further study is necessary to correlate electrochemical impedance data and microstructure.
6.3.2 Cr6+ contents in the coating
In X-ray absorption spectra of Cr, the intensity of the pre-edge peak gives an indication of Cr6+ content in the coating, and the edge height indicates the total quantity of Cr in the coating [6, 21]. In these experiments, the total amount of chromium increased with the increasing coating time. Figure 6.15 shows that the edge height on 3-minute
CCC of AA7475-T7 was almost twice that of 1-minute CCC of AA7475-T7 and the pre- edge peaks of Cr6+ between two coatings were different by a factor of two. The ratio of
202
the pre-edge peak height to the absorption edge height had a linear relationship with Cr6+ content in coating [6, 21]. Therefore, the relative Cr6+ content of 1-minute CCCs was
29.6 % and that of 3-minute CCCs was 27.6 %. These results indicate that coating thickness is increased with coating time, while the Cr6+ content in the coating remained essentially constant up to 3-minute immersion time.
Figure 6.16 shows XANES spectra of 3-minute CCCs on different tempers of
AA7475. The total chromium and Cr6+ content in terms of background corrected and non-normalized absorbance was as follows (Figure 6.16a):
Total Cr: T7 (6.0994) > T6 (5.7461) > W (5.3127) > RRA (5.2886)
Cr6+: T7 (1.6843) > W (1.5049) > RRA (1.4171) > T6 (1.3562)
The relative Cr6+ contents in the coating were as follows (Figure 6.16b):
W (28.3 %) ≈ T7 (27.6 %) ≈ RRA (26.8 %) > T6 (23.6 %)
The Cr6+ contents are not significantly different from temper to temper and are generally in agreement with Cr6+ contents reported in the literature [7, 17, 18]. These XANES spectra were collected from the exact same samples used for the EIS measurements presented in Figure 6.9a. Interestingly, the relative Cr6+ contents of W, T7 and RRA tempers are higher than that of T6 temper, which is similar to the trend of impedance spectra. The total impedance values of W, T7 and RRA tempers are much higher than that of T6 temper on 3-minute CCC on AA7475 (Figure 6.9a).
203
Figure 6.17 and Table 6.5 give information illustrating the differences in the Cr6+ leached from 1-minute and 3-minute CCCs on AA7475-T7 and RRA. Cr6+ enables the coating to self-heal [13]. By comparing absorption spectra before and after exposure to
0.5 M NaCl for 7 days, Cr6+ contents were calculated from the spectra in Fig. 17. Results showed that more than three times as much Cr6+ is released from a 3-minute CCC than from a 1-minute coating. The corrosion resistance of a 3-minute coating is nearly twice that of a 1-minute coating, due in part from the large inhibitor reservoir in the 3-minute coating [13].
6.3.3 Salt spray testing
Salt spray testing (SST) was carried out to assess the effect of alloy pretreatment on coating corrosion resistance and to characterize the effect of temper and alloy purity in a way that would enable comparison to literature reports on these effects [23]. This SST was performed in 5 wt% NaCl neultral salt spray [22]. The conversion-coated samples were inspected regularly and photographs were recorded after 651 and 1310 hours of exposure to assess corrosion damage.
Figure 6.18 shows photographs illustrating the type of corrosion damage observed.
The samples were ranked from 0 to 5 according to the number of pits and the corrosion product accumulated. A rank of 0 indicated almost no corrosion and rank of 5 indicated severe corrosion (Table 6.6).
Figure 6.19 and Figure 6.20 show the effect of alloy temper, alloy purity, and pretreatment on corrosion damage. Notations A, B, and C indicate alkaline pretreatment, acid pretreatment, and chromate conversion coating respectively. After 651 and 1310
204
hours of exposure, a significant effect of pretreatment on AA7075 was noted, while there was essentially no effect of pretreatment on AA7475. Specifically, AA7075_(A+C) was severely corroded, whereas AA7475_(A+C) exhibited good corrosion performance; similar to that of AA7475_(A+B+C). Figure 6.21 shows the effect of pretreatment on
CCC performance. AA7075-T6 with (A+B+C) had best performance among all samples tested. AA7475 showed little dependence on pretreatment due to the low population of constituent particles initially.
After 651 hr exposure, AA7075 and AA7475 in both the T6 and T7 tempers performed well when fully pretreated. Representative photographs at 651 hr and 1310 hr immersion are shown in Figure 6.22 and Figure 6.23. illustrates the effect of temper on
AA7075 after 651 and 1310 hours of exposure. At 651 hr immersion time, there was no significant sign of corrosion on either T6 or T7 surfaces. However, after 1310 hr of exposure, pitting and corrosion product were evident. The T7 samples demonstrated greater incidence of pitting. The T7 samples had less pitting, but appeared to generate more corrosion product than T7. AA7475 exhibited similar behavior (Figure 6.23). After
651 hr immersion, there was no significant difference in the corrosion behavior of T6 and
T7 samples on AA7475. After 1310 hr exposure, the T6 samples had produced more corrosion product and pits than T7 samples. From the above results, it can be concluded that pretreatment has more significant effect on the performance of chromate conversion coating on AA7x75 than alloy temper.
205
6.4 DISCUSSION
The temper effect is pronounced in 7475. Corrosion resistance is consistently greater in the T7 temper than in the T6 temper. The effect is detectable in EIS measurements across a range of exposure times and in pitting potential measurements.
The reason that this effect is so pronounced may be related to the relative lack of IMC particles in the alloy. Without these particles, the number of sites where thin, less protective coatings form is reduced and the effect of increased coating thickness on 7475-
T7 on corrosion resistance is more evident.
Based on the same arguments, the difference in corrosion resistance of conversion coated 7075 in the T6 and T7 tempers is less because coating breakdown is more strongly affected by intermetallic particles. While a thicker coating may form on the matrix of
7075, a significantly higher corrosion resistance is not detected because coating breakdown initiates at IMC particles, which are plentiful in this alloy. In addition to the coating thickness difference that appears to exist between the T6 and T7 tempers, there is the additional effect of an increase in the intrinsic resistance to localized corrosion by
AA7475 in the T7 temper in chapter 5. This may also be contributing to the apparent increase in corrosion resistance of conversion coated alloys in the T7 temper. The RRA and W tempers did not exhibit regular enough behavior to help contribute to the interpretation of the temper effect.
XANES shows that Cr6+ content is more or less the same across all alloy tempers except T6 temper. These data help in the interpretation of the 860 cm-1 Raman intensity variations observed for alloys in the T6 and T7 tempers in chapter 5. The 860 cm-1 intensity of coatings in T7 temper was higher than that of T6 temper and variations in the
206
860 cm-1 intensity are associated with variations in coating thickness. The fact that the
860 cm-1 intensity was greater on T7 alloys than T6 suggests that CCCs on T7 had thicker coating. XANES shows that the Cr6+ to total Cr ratio is about the same for 1-min and 3- min CCCs although the absolute Cr6+ content of 3-minute CCCs is higher than that of 1- minute CCCs. Because the 3-minute coating is thicker, it is a better barrier and the total
Cr6+ reservoir is presumably larger.
Salt spray exposure testing does not reveal a strong temper effect suggesting that
EIS and pitting potential measurements are in fact more sensitive methods for characterizing coating corrosion protection. SST did not replicate results reported in a similar prior study, which suggested that 7075-T6 was better protected than 7075-T7 [23].
Salt spray testing supports the widely understood fact that degreasing and deoxidizing alloy substrates prior to conversion coating has a significant positive effect on subsequent corrosion protection. Perhaps not so surprisingly, cleaning has a larger effect on AA7075 than on AA7475. This seems to be in line with the notion that cleaning removes constituent particles that are difficult to coat and are sites for coating breakdown during exposure to aggressive environments.
While the temper effect is not well differentiated by SST, it is possible to see differences in corrosion resistance due to alloy purity. AA7075 performs worse than
AA7475 at equivalent exposure times. Overall, these results show that there is distinct temper dependence in the corrosion resistance of conversion coated 7x75 alloys. This effect is readily detected in electrochemical testing, but is not readily detectable in exposure testing. The origin of the effect appears to lie with an increase in the intrinsic
207
localized corrosion resistance of the alloy and the ability of the matrix phase in the T7 alloy to grow a thicker conversion coating.
6.5 SUMMARY
The effect of alloy temper on chromate conversion coating performance of
AA7075 and AA7475 were investigated by EIS, XANES, and salt spray exposure testing.
The performance of chromate conversion coating is influenced by not only alloy temper but also alloy impurity and surface pretreatment. The details of the results are as follows:
1. The corrosion resistance of AA7475-T7 was better than that of AA7475-T6 in 1-
minute and 3-minute CCCs with/without acid and alkaline pretreatment. The
corrosion resistance of AA7075 was, however, almost independent of alloy
temper in 3-minute CCCs.
2. XANES shows that the Cr6+ content of CCCs are nearly independent of alloy
temper though T6 tempered alloy has lowest Cr6+ content. The more released Cr6+
compared absorption spectra before and after exposure indicates higher corrosion
resistance.
3. Salt spray exposure testing reveals a stronger surface pretreatment effect on CCC
performance than alloy temper. 1310 hr salt spray test revealed that AA7475-T7
had less corrosion products and pits than AA7475-T6, whereas AA7075-T6 had
less corrosion products and pits than AA7075-T7.
208
REFERENCES
1. J.K. Park and A.J. Ardell, Scripta Metall., 22, 1115 (1988).
2. S. Maitra and G.C. English, Metallurgical Transaction A, 12A, 535 (1981).
3. T. Ramgopal, Role of Grain boundary precipitates and Solute depleted zone in the Intergranular corrosion of Aluminum Alloy AA7150, The Ohio State University (2001).
4. Q. Meng and G.S. Frankel, J. Electrochem. Soc., 151, B271 (2004).
5. B. Cina, U.S. Patent 3,856,584 (1974).
6. M.W. Kendig, A.J. Davenport, and H.S. Isaacs, Corrosion Science, 34, 41 (1993).
7. J.R. Waldrop and M.W. Kendig, J. Electrochem. Soc., 145, L11 (1998).
8. K. Asami, M. Oki, G.E. Thompson, G.C. Wood, and V. Ashworth, Electrochemica Acta, 32, 337 (1987).
9. L. Xia and R.C. McCreery, J. Electrochem. Soc., 145, 3083 (1998).
10. L. Xia and R.L. McCreery, J. Electrochem. Soc., 146, 3696 (1999).
11. H.A. Katzman, G.M. Malouf, R. Bauer, and G.W. Stupian, Applications of Surface Science, 2, 416 (1979).
12. P.L. Hagans and C.M. Haas, Surface and Interface Analysis, 21, 65 (1994).
13. J. Zhao, G.S. Frankel, and R.L. McCreery, J. Electrochem. Soc., 145, 2258 (1998).
14. B.L. Hurley and R.L. McCreery, J. Electrochem. Soc., 150, B367 (2003).
15. L. Xia, E. Akiyama, G. Frankel, and R. McCreery, J. Electrochem. Soc., 147, 2556 (2000).
16. G.P. Halada and C.R. Clayton, J. Electrochem. Soc., 138, 2921 (1991).
17. F.W. Lytle, R.B. Greegor, G.L. Bibbins, K.Y. Blohowiak, R.E. Smith, and G.D. Tuss, Corrosion Science, 37, 349 (1995).
18. J. Wan, G.E. Thompson, K. Lu, and C.J.E. Smith, Physica B, 208&209, 511 (1995).
209
19. ASM International, ASM handbook: Vol. 4 Heat Treating, American Society for Metals, Metal Parks (1991).
20. H.E. Boyer and T.L.E. Gall, Metals Handbook, American Society For Metals, Metal Parks (1985).
21. V. Laget, C.S. Jeffcoate, H.S. Isaacs, and R.G. Buchheit, J. Electrochem. Soc., 150, B425 (2003).
22. ASTM B117, Standard Practice for Operation Salt Spray (Fog) Apparatus, ASTM, West Conshohochen, PA (1997).
23. L.J. Bailin, P. Fitzpatrick, and M.J. Joyce, Evaluation of Unpainted Alodine Chromate Conversion Coatings for Corrosion Resistance and Electrical Conductivity, Lockheed Missiles and Space Co., Report F035575 (1985).
210
TABLES
Alloy Zn Mg Cu Cr Fe Si Mn Ti Al(wt%)
7475 5.64 2.27 1.35 0.22 0.05 0.021 0.001 0.044 Balanced
7075 5.52 2.44 1.41 0.17 0.28 0.019 0.029 0.033 Balanced *Average of 3 samples
Table 6.1: Composition of the AA7475 and AA7075 alloys by ICP-MS.
Steps Process
32 g/L Na2SiO3 Cleaning & degreasing 48 g/L Na2CO3 at 65 °C for 2 minutes Rinse 72 ml/L HNO3 Deoxidizing 30 g/L Sanchem1000 at 55 °C for 3 minutes Rinse 7.55 g/L Alodine® 1200S Ajust pH to 1.5 ~ 1.7 by HNO3 at room temp. Chromate conversion coating for 1 and 3 minutes Rinse Drying for 24 hours
Table 6.2: Chromate conversion coating process
211
2 t (hr) Rs (Ω) CCPE PCPE Rcoat (Ωcm ) 24 19.31 6.5016×10-6 0.94922 3.4445×105 (1.22%) 68 20.2 6.5008×10-6 0.93914 4.5796×105 (1.08%) W 104 21.09 6.2004×10-6 0.94113 2.4819×105 (1.58%) 120 21.47 5.9199×10-6 0.94589 6.2975×105 (1.19%) 144 20.82 5.7153×10-6 0.94962 6.2975×105 (1.19%) 24 18.89 1.4099×10-5 0.93681 7.1295×105 (1.06%) 68 19.28 1.3539×10-5 0.93069 5.7341×105 (0.75%) T6 104 19.96 1.3001×10-5 0.93177 6.9979×105 (0.88%) 120 19.48 1.2858×10-5 0.93353 4.9978×105 (0.94%) 144 19.07 1.2859×10-5 0.93272 4.3498×105 (0.88%) 24 19.59 6.3778×10-6 0.94992 1.8642×106(1.21%) 68 20.5 6.1096×10-6 0.95175 2.9473×106(1.19%) T7 104 20.73 6.0365×10-6 0.94806 3.1722×106(1.47%) 120 20.22 6.0472×10-6 0.94676 2.5755×106(1.36%) 144 19.9 5.7939×10-6 0.94658 1.9039×106(2.16%) 24 19.36 9.2133×10-6 0.94812 9.3267×105 (0.95%) 68 19.94 8.9222×10-6 0.9454 1.0669×106 (1.15%) RRA 104 19.9 8.8915×10-6 0.94097 1.0190×106(0.75%) 120 19.67 8.8424×10-6 0.9412 8.3286×105 (1.38%) 144 19.31 8.7096×10-6 0.94105 1.2565×106(1.89%) CPE: Constant Phase Element 1 Z CPE = C( jw) P C: amplitude, P: exponent
Table 6.3: Impedance parameter values of 1-minute chromate coated AA7475 after exposure to 0.5 M NaCl solution.
212
2 t (hr) Rs (Ω) CCPE PCPE Rcoat (Ωcm ) 6 23.42 2.8905×10-6 0.97929 4.7174×106(5.62%) 36 20.91 2.9113×10-6 1.006 6.6367×106(9.82%) W 48 20.6 2.9299×10-6 1.012 7.0193×106(11.4%) 72 20.14 3.1661×10-6 1.007 5.0919×106(11.1%) 170 18.57 4.1227×10-6 0.96951 1.9883×106(1.88%) 6 20.99 8.5374×10-6 0.92962 3.8763×105 (1.67%) 36 20.87 8.3546×10-6 0.93756 2.9608×105 (1.07%) T6 48 19.94 8.3595×10-6 0.93831 3.5299×105 (0.88%) 72 19.48 8.2537×10-6 0.94078 3.4687×105 (0.90%) 170 20.77 8.7712×10-6 0.92972 2.9276×105 (3.81%) 6 23.24 4.2964×10-6 0.97941 3.4157×106(9.42%) 36 21.51 4.3467×10-6 0.99802 4.4243×106(12.4%) T7 48 21.15 4.2235×10-6 1.005 6.3696×106(17.2%) 72 21.22 4.3595×10-6 1.003 7.0305×106(19.1%) 170 19.38 8.7855×10-6 0.9536 1.9669×106(2.35%) 6 22.05 6.2015×10-6 0.99497 3.2895×106(17.7%) 36 21.54 6.0508×10-6 1.013 4.0686×106 (20.3%) RRA 48 21.32 4.9782×10-6 1.05409 5.1393×106(27.6%) 72 20.97 5.9698×10-6 1.018 4.4689×106 (22.6%) 170 17.94 4.7883×10-6 0.96018 4.3865×106(2.44%)
Table 6.4: Impedance parameter values of 3-minute chromate coated AA7475 after exposure to 0.5 M NaCl solution
213
Unexposed Exposed ***Released EIS Cr6+ *Cr **% Cr6+ Cr % Cr6+ (%) (MΩcm2) 3-min CCC T7 1.6843 6.0994 27.6 0.76609 5.3443 14.3 0.92 (13.3) 3.6226
1-min T7 0.94121 3.1939 29.6 0.66806 2.8218 23.7 0.27(5.9) 1.9039 CCC RRA 0.80989 3.1503 26.1 0.64355 3.0541 20.7 0.17(5.4) 1.2565 *Cr: total chromium (Cr3+ + Cr6+ + Cr metal), Cr metal = 0.87504 **%: Cr6+/Cr in coating ***Released Cr6+: Non-exposure Cr6+ – Exposure Cr6+
Table 6.5: Comparison of Cr6+ content of XANES for 3-minute and 1-minute CCCs of AA7475-T7 and AA7475-RRA in 0.5 M NaCl after 7 days.
214
Specimen (#) 651 hr visual inspection 1310 hr visual inspection AA7075_W_(*A+B+C) (1) **2 2 AA7075_W_(A+B+C) (2) 2 3 AA7075_W_(A+B+C) (3) 2 2 AA7075_W_(A+B+C) (4) 2 3 AA7075_W_(A+B+C) (5) 2 2
AA7075_T6_(A+B+C) (1) 0 2 AA7075_T6_(A+B+C) (2) 0 1 AA7075_T6_(A+B+C) (3) 0 0 AA7075_T6_(A+B+C) (4) 0 2 AA7075_T6_(A+B+C) (5) 1 1
AA7075_T7_(A+B+C) (1) 0 2 AA7075_T7_(A+B+C) (2) 0 2 AA7075_T7_(A+B+C) (3) 1 2 AA7075_T7_(A+B+C) (4) 0 3 AA7075_T7_(A+B+C) (5) 0 3
AA7075_RRA_(A+B+C) (1) 1 3 AA7075_RRA_(A+B+C) (2) 0 3 AA7075_RRA_(A+B+C) (3) 2 4 AA7075_RRA_(A+B+C) (4) 1 3 AA7075_RRA_(A+B+C) (5) 1 3
AA7475_W_(A+B+C) (1) 1 3 AA7475_W_(A+B+C) (2) 2 4 AA7475_W_(A+B+C) (3) 2 4
AA7475_T6_(A+B+C) (1) 1 3 AA7475_T6_(A+B+C) (2) 1 3 AA7475_T6_(A+B+C) (3) 1 4
AA7475_T7_(A+B+C) (1) 1 3 AA7475_T7_(A+B+C) (2) 1 1 AA7475_T7_(A+B+C) (3) 1 2
Table 6.6: Rankings of the degree of corrosion in salt spray test at 651 hr and 1310 hr visual inspection.
(Continued)
215
Table 6.6: (Continued) AA7475_RRA_(A+B+C) (1) 1 3 AA7475_RRA_(A+B+C) (2) 1 2 AA7475_RRA_(A+B+C) (3) 2 3
AA7075_W_(A+C) (1) 5 5 AA7075_W_(A+C) (2) 5 5
AA7075_T6_(A+C) (1) 3 5 AA7075_T6_(A+C) (2) 4 5
AA7075_T7_(A+C) (1) 3 5 AA7075_T7_(A+C) (2) 3 5
AA7075_RRA_(A+C) (1) 5 5 AA7075_RRA_(A+C) (2) 5 5
AA7475_W_(A+C) (1) 2 3 AA7475_T6_(A+C) (1) 1 3 AA7475_T7_(A+C) (1) 1 2 AA7475_RRA_(A+C) (1) 1 1
AA7075_W_(C) (1) 3 4 AA7075_T6_(C) (1) 3 5 AA7075_T6_(C) (2) 5 5 AA7075_T7_(C) (1) 2 4 AA7075_T7_(C) (2) 1 3 AA7075_RRA_(C) (1) 4 5 AA7075_RRA_(C) (2) 3 4
AA7475_W_(C) (1) 2 3 AA7475_T6_(C) (1) 1 3 AA7475_T7_(C) (1) 4 5 AA7475_RRA_(C) (1) 1 4
AA7075_W_(B+C) (1) 3 4 AA7075_T6_(B+C) (1) 3 5 AA7075_T7_(B+C) (1) 1 2 AA7075_RRA_(B+C) (1) 3 4
*A: Alkaline pretreatment, B: Acid pretreatment, C: Chromate conversion coating **0 to 5 indicates no corrosion to severe corrosion.
216
FIGURES
490 or 515oC/1hr o 204 C/1hr 120 oC/24hr 120 oC/24hr
WQ AC W WQ Retrogression T6 temper Re-aging RRA temper 490 or 515oC/1hr 163 oC/24hr 107 oC/6hr T7 temper
WQ AC AC
Figure 6.1. Schematic illustration of heat treatment on AA7075 and AA7475. W temper of AA7075 is 490 °C and W temper of AA7475 is 515 °C. WQ and AC indicate water quenching and air-cooling respectively.
217
Mud-crack
5 µm
(a) 3-minute chromate conversion coating
Al Alloy Chromate Conversion Substrate Coating
Coxide
Rs R Cd oxide
Rp Rd Mud-crack
(b) Equivalent circuit models
Figure 6.2. Physical model and equivalent circuit model of chromate conversion coating.
218
C oxide C Coxide oxide
Rs Rs Rs Roxide Cd Cd Cd
R R p R p Rd Rd d (R ↑↑:open) (Rp ↓↓:short) oxide CPE
Ccoat
R Rs s
Rcoat
(Coxide ≈ Cdl) Rcoat
Figure 6.3. Simplified equivalent circuit model of chromate conversion coating analysis. Roxide is an electrical open due to high resistance and Rp is a short due to low resistance. Ccoat represents the summation of Coxide and Cd. CPE stand for a constant phase element for Ccoat.
219
-2e+6 (a)
-2e+6
]
2
cm Ω -1e+6
Im(Z) [
-5e+5
0 0.0 5.0e+5 1.0e+6 1.5e+6 2.0e+6 2.5e+6 Re(Z) [Ωcm2]
107 -100 (b)
106 -80
105
) -60 2
cm 104 Ω
(degrees) θ |Z| ( -40 3 10
-20 102 Real data Fitting data
101 0 10-2 10-1 100 101 102 103 104 Frequency (Hz)
Figure 6.4. Typical Nyquist and Bode plots for 3-minute CCC on AA7475-T7 with complex nonlinear least-square fitting in 0.5 M NaCl solution for 6 hours.
220
-4e+6 (a)
-3e+6
] 2
cm Ω -2e+6
[ Im(Z) 6hr
-1e+6 36hr 48hr 72hr 170hr 0 0 1e+6 2e+6 3e+6 4e+6 5e+6
Re(Z) [Ωcm2]
108 -100 (b)
7 10 -80 106
) 5
2 10 -60 cm 104 Ω
degrees) 3 -40
|Z| ( 10
θ ( 6hr 2 10 36hr -20 48hr 101 72hr 170hr 100 0 10-2 10-1 100 101 102 103 104
Frequency (Hz)
Figure 6.5. (a) Nyquist and (b) Bode plots of impedance for 3-minute CCCs on AA7475- T7 of different immersion time in 0.5 M NaCl solution.
221
-4e+6
-3e+6
] 2
cm Ω -2e+6 1 min CCC-24hr 1 min CCC-68hr 1 min CCC-104hr Im(Z) [ 1 min CCC-120hr 1 min CCC-144hr -1e+6 3 min CCC-6hr 3 min CCC-36hr 3 min CCC-48hr 3 min CCC-72hr 3 min CCC-170hr 0 0 1e+6 2e+6 3e+6 4e+6 5e+6
Re(Z) [Ωcm2]
Figure 6.6. Nyquist plots of impedance for 1 and 3-minute CCCs on AA7475-T7 in different immersion time of 0.5 M NaCl solution.
222
-4e+6 (a)
-3e+6
] 2
cm
Ω -2e+6
[ Im(Z) -1e+6 W T6 T7 RRA
0 0 1e+62e+63e+64e+6
2 Re(Z) [Ωcm ]
108 -100 (b)
107
-80 106
) 5
2 10 -60 cm 104 Ω
3 -40 (degrees)
|Z| ( 10 θ
2 10 W -20 T6 101 T7 RRA 100 0 10-2 10-1 100 101 102 103 104
Frequency (Hz)
Figure 6.7. (a) Nyquist and (b) Bode plots of impedance for 3-minute CCC on AA7475- W, T6, T7, and RRA tempers in 0.5 M NaCl solution for 72 hours.
223
-1e+6 (a)
-8e+5
] 2 -6e+5
cm Ω
-4e+5 Im(Z) [
W
-2e+5 T6 T7 RRA 0 0.0 2.0e+5 4.0e+5 6.0e+5 8.0e+5 1.0e+6 1.2e+6
2 Re(Z) [Ωcm ]
108 -100 (b)
107 -80 6 10
5
) 10 2 -60
cm 4
Ω 10
(degrees) |Z| ( -40 103 θ
102 W T6 -20 101 T7 RRA
100 0 10-2 10-1 100 101 102 103 104
Frequency (Hz)
Figure 6.8. (a) Nyquist and (b) Bode plots of impedance for 1-minute CCC on AA7475- W, T6, T7, and RRA tempers in 0.5 M NaCl solution for 68 hours.
224
(a) W T6 T7 1e+7 RRA
) 2
cm Ω 1e+6
|Z| (
1e+5
0 20 40 60 80 100 120 140 160 180 Immersion time (hours)
(b) W T6 1e+7 T7 RRA )
2
cm
Ω 1e+6
|Z| (
1e+5
0 20406080100120140160
Immersion time (hours)
Figure 6.9. Corrosion resistance variation with tempers for (a) 3-minute CCC and (b) 1- minute CCCs of AA7475 in 0.5 M NaCl solution.
225
-2e+5 (a) 24hr 48hr 72hr 96hr -2e+5 120hr 144hr
] 168hr
2
cm Ω -1e+5
[ Im(Z)
-5e+4
0 0.05.0e+41.0e+51.5e+52.0e+52.5e+5 Re(Z) [Ωcm2]
106 -100 (b) 24hr 48hr 72hr 105 96hr -80 120hr 144hr 104 168hr
) -60 2 cm 103 Ω
(degrees) |Z| ( |Z|
-40 θ 102
-20 101
100 0 10-2 10-1 100 101 102 103 104 105 Frequency (Hz)
Figure 6.10. (a) Nyquist and (b) Bode plots of impedance for 3-minute CCCs on AA7075-T6 of different immersion time in 0.5 M NaCl solution.
226
(a) 24hr -3e+5 48hr 72hr 96hr 120hr 144hr
]
2 168hr -2e+5 cm
Ω
[ Im(Z)
-1e+5
0 0 1e+5 2e+5 3e+5 4e+5 2 Re(Z) [Ωcm ]
106 -100 (b) 24hr 48hr 72hr 105 96hr -80 120hr 144hr 104 168hr ) -60 2 cm 103 Ω
(degrees) |Z| ( |Z|
-40 θ 102
-20 101
100 0 10-2 10-1 100 101 102 103 104 105 Frequency (Hz)
Figure 6.11. (a) Nyquist and (b) Bode plots of impedance for 3-minute CCCs on AA7075-T7 of different immersion time in 0.5 M NaCl solution.
227
T6 T7
1e+6
)
2
cm
Ω
|Z| ( 1e+5
1e+4 0 20 40 60 80 100 120 140 160 180 Immersion Time (hours)
Figure 6.12. Corrosion resistance variation of 3-minute CCCs on AA7075 with T6 and T7 tempers in 0.5 M NaCl solution.
228
7 10
6 10 )
2
Cm
Ω
|z| ( 5 10
W T6 T7 RRA 10 4 24 48 72 96 120 144 168 Immersion time (hours)
Figure 6.13. Corrosion resistance variation of 1-minute CCCs on AA7475 with all tempers in 0.5 M NaCl solution. Only 1 µm mechanical polishing without pretreatment.
229
7 10
) 2 6
Cm 10 Ω
|z| (
W T6 T7 5 RRA 10 24 48 72 96 120 144 168
Immersion time (hours)
Figure 6.14. Corrosion resistance variation of 3-minute CCCs on AA7475 with all tempers in 0.5 M NaCl solution. Only 1 µm mechanical polishing without pretreatment.
230
8
6
)
a.u. ( 4
2
Absorbance
units) (arbitrary Fluorescence α 3 min CCC Cr K 0 1 min CCC
0 50 100 150 200 250 300
E-E0 (eV)
Figure 6.15. XANES spectra for AA7475-T7 with 3-minute CCC and 1-minute CCC. XANES spectra showing absolute absorbance against energy relative to Cr K edge (E0 = 5989 eV). The pre-edge peak of Cr6+ is at 5994 eV.
231
8 (a)
6
)
a.u. 4 W ( 2.0 T6 1.8 T7 2 1.6 RRA 1.4 Fluorescence (a.u.) Fluorescence 1.2 α 1.0 0 Absorbance 0.8 Cr K 0.6 -2 4.04.55.05.56.06.57.0 0 50 100 150 200 250 300
E-E (eV) 0
1.4 (b) 1.2
1.0
0.8 0.30 W 0.6 T6 T7 0.25 0.4 RRA 0.20 0.2 Normalized Counts Normalized absorbance 0.15 0.0 0.10 -0.2 4.0 4.5 5.0 5.5 6.0 6.5 7.0 0 50 100 150 200 250 300
E-E0 (eV)
Figure 6.16. XANES spectra for AA7475 of 3-minute CCC showing (a) absolute absorbance (b) normalized absorbance with W, T6, T7, and RRA tempers.
232
1.4 (a) 1.2
1.0
0.8 Cr6+ content 0.6 0.4
T7-3min CCC 0.2 Normalized Counts T7-3min CCC (exposed) RRA-3min CCC 0.0 RRA-3min CCC (exposed) Normalized absorbance -0.2 0204060
E-E0 (eV)
1.4 (b) 1.2
1.0
0.8 Cr6+ content 0.6
0.4 0.2 T7-1min CCC Counts Normalized T7-1min CCC (exposed) Normalized absorbance RRA-1min CCC 0.0 RRA-1min CCC (exposed)
-0.2 0204060
E-E0 (eV)
Figure 6.17. XANES spectra for AA7475-T7 and AA7475-RRA with (a) 3-minute CCC and (b) 1-minute CCC exposed to 0.5 M NaCl solution after 7 days.
233
Rank 0 Rank 1 Rank 2
No corrosion # pit < 10 # pit < 50
Rank 3 Rank 4 Rank 5
# pit > 50 # pit > 50 with # pit >> 50 with corrosion smut corrosion smut
Figure 6.18. Rankings of the degree in salt spray test according to surface morphology.
234
5 W T6 T7 RRA 4
3
Rank
2
1
0
AA7075_(C) AA7475_(C) AA7075_(A+C) AA7075_(A+C) AA7475_(A+C) AA7075_(B+C)
AA7075_(A+B+C) AA7075_(A+B+C) AA7475_(A+B+C)
Figure 6.19. Rankings of the degree of corrosion at 651 hr visual inspection for chromate coated 7xxx with / without pretreatment. (A: alkaline pretreatment, B: acid pretreatment, C: chromate conversion coating).
235
W 5 T6 T7 RRA
4
3
Rank
2
1
0
AA7075_(C) AA7075_(C) AA7475_(C)
AA7075_(A+C) AA7075_(A+C) AA7475_(A+C) AA7075_(B+C)
AA7075_(A+B+C) AA7475_(A+B+C)
Figure 6.20. Rankings of the degree of corrosion at 1310 hr visual inspection for chromate coated 7xxx with / without pretreatment. (A: alkaline pretreatment, B: acid pretreatment, C: chromate conversion coating).
236
(a) AA7075-T6 with (A+B+C) (b) AA7075-T6 with (A+C)
(c) AA7075-T6 with (C) (d) AA7075-T6 with (B+C)
Figure 6.21. Optical photographs of salt spray test for pretreatment of AA7075-T6 samples at 1310 hr exposure. (a) AA7075-T6 with (A+B+C), (b) AA7075-T6 with (A+C), (c)AA7075-T6 with (C), and (d)AA7075-T6 with (B+C). A: Alkaline pretreatment, B: Acidic pretreatment, and C: Chromate conversion coating.
237
(a) AA7075-T6 @651 hr (b) AA7075-T7 @651 hr
(c) AA7075-T6 @1310 hr (d) AA7075-T7 @1310 hr
Figure 6.22. Optical photographs of salt spray test for AA7075 at comparison 651 hr and 1310 hr exposure. (a) AA7075-T6 @651 hr, (b) AA7075-T7 @651 hr, (c)AA7075-T6 @1310 hr, and (d)AA7075-T7 @1310 hr.
238
(a) AA7475-T6 @651 hr (b) AA7475-T7 @651 hr
(c) AA7475-T6 @1310 hr (d) AA7475-T7 @1310 hr
Figure 6.23. Optical photographs of salt spray test for AA7475 at 651 hr and 1310 hr exposure. (a) AA7475-T6 @651 hr, (b) AA7475-T7 @651 hr, (c) AA7475-T6 @1310 hr, and (d)AA7475-T7 @1310 hr.
239
CHAPTER 7
CONCLUSIONS AND FUTURE WORK
7.1 CONCLUSIONS
In this study, chromate conversion coating formation and breakdown on intermetallic compounds of heterogeneous 7xxx aluminum alloys was studied. The effect of alloy temper on corrosion behavior and chromate conversion coating performance of
7xxx aluminum alloys was characterized. Thin film analogs of η and S phases were prepared for studies of corrosion behavior and coating formation. The primary aim of the present study provides a better understanding of the formation and breakdown of CCCs on 7xxx aluminum alloys. The following are the main findings:
1. The chromate conversion coating formation on 7x75 aluminum alloys was
strongly influenced by microstructural heterogeneity. Conversion coatings on
constituent particles were thinner than on the matrix. Coatings appeared to be
thinnest on Al2CuMg particles. The poorly formed coating on Al2CuMg does not
provide much protection to the phase as demonstrated in thin film analog
experiments. In subsequent exposure testing of alloy electrodes, breakdown of the
240
coating appeared to be initiated at these sites, indicating that these particles are the
weakest points in the coated surface.
2. After 5-second coating times at room temperature, the surface height and Volta
potential differences between IMCs and matrix were almost identical with those
of 3-minute coating. CCCs were continous across the surface, but were non-
uniform in thickness. After initial coating formation, the coating appeared to grow
uniformly, preserving the differences in apparent coating thickness. Trenches are
observed at the particle-matrix interface. This trench is associated peripheries of
cathodic particles due to particle dissolution in the case of Al-Fe-Cu IMC and
metallic dissolution in the case of Mg2Si and Al2CuMg. There is no significant
effect of alloy temper on the difference in coating thickness or Volta potential
between constituent particles and the surrounding matrix.
3. The η phase receives a conversion coating well, but its receptiveness decreases as
Cu content in the phase increases. However, as Cu content goes up the intrinsic
corrosion resistance of the phase also seems to increase. How these seemingly
opposing factors contribute to the larger issue of the effect of temper on corrosion
resistance was not resolved here.
4. Evidence indicates that thicker coatings form on the matrix of 7X75 alloys when
they are over-aged. This has the effect of increasing corrosion resistance on more
pure AA7475. Smaller increases in corrosion resistance are observed on AA7075-
T7, perhaps due to the effect of constituent particles, which initiate corrosion
more easily and lower corrosion resistance.
241
5. The performance of chromate conversion coating is dependent on alloy purity,
coating ingredients, coating time, pretreatment as well as alloy temper. Bath
chemistry changes and coating process times induce much more significant
changes in overall corrosion protection.
7.2 FUTURE WORK
This study has opened up several issues for the more studies of chromate conversion coating formation and breakdown as future work.
1. This study confirmed that the formation of chromate conversion coating was not
uniform. The matrix developed thicker coatings than IMCs. More detailed
characterizations such as focused ion beam (FIB) cross section with TEM and
AFM are necessary to explain the different coating formation thicknesses on these
IMCs. In addition, each ingredient of CCCs on formation and breakdown of these
IMCs is also needed to investigate for understanding CCC mechanism.
2. In Raman spectra, AA7475-T7 having higher 860 cm-1 intensity band than
AA7475-T6 had better corrosion resistance, higher breakdown potential, and
better performance of SST than AA7475-T6. However, AA7075-T6 was better
than AA7075-T7 though AA7075-T6 and AA7075-T7 had similar 860 cm-1
intensity band in the coating using Raman spectra. In XANES, absolute Cr6+,
relative Cr6+, and released Cr6+ contents in the coating are needed to clarify the
coating performance quantitatively, which could be possible using XANES and
Raman spectra combined with surface techniques such as AES, AFM and SIMS.
The structural and compositional changes of CCC during coating formation are
242
also needed to be characterized using AFM, EXAFS, Raman spectroscopy,
especially to clarify the early coating formation and growth.
3. This work showed the breakdown potentials of η phase increased with increasing
Cu contents while the Cr6+ contents decreased with increasing Cu contents.
However, there was considerable difference between the breakdown potentials of
thin film analogs of η phase and 7xxx. It is necessary to understand the role of η
phase in the breakdown behavior of 7xxx.
4. This work also confirmed that AA7475-T6 and AA7075-T6 had two breakdown
potentials but the other tempers demonstrated just one breakdown potential. Each
characteristic potential such as Ecorr, Ebr, and Erepass was dependent on the temper
though these potentials were independent of crystal orientation. However, it is not
clear why only T6 temper has two breakdown potentials. RRA temper is an
intermediate microstructure between T6 and T7 temper and needed to identify for
its coating performance. Careful characterization is necessary to understand the
breakdown behavior and CCC performance of 7xxx according to temper. The
chemical composition of precipitate on grain boundary and matrix, solute deplete
zone, precipitate free zone, and coarse IMCs are to be examined to understand the
effect of alloy temper using TEM, FIB, AES, and nano-EDS analysis
systemically.
5. Thin film analogs of η and S phases showed the corrosion behaviors and
chromate coating formation behaviors. However, it is still necessary to confirm
the difference between real IMCs of 7xxx aluminum alloys and these thin film
analogs using XRD, heat-treatment combined with electrochemical approaches.
243
And the other IMCs, such as Mg2Si and Al-Fe-Cu IMCs, must also be fabricated and investigated to understand their corrosion behavior and chromate coating formation and breakdown behaviors.
244
APPENDIX A
EFFECT OF ALLOY TEMPER ON CORROSION BEHAVIOR
OF Al-Zn-Mg-Cu ALLOY 7475
A.1 INTRODUCTION
There has been significant interest in the corrosion behavior of high strength 7xxx aluminum alloys, which are of importance for aircraft structural applications due to high strength and low density. The 7xxx aluminum alloys, however, are susceptible to pitting, crevice corrosion, intergranular corrosion (IGC), exfoliation corrosion, hydrogen embrittlement and stress corrosion cracking (SCC). The resistance to exfoliation and SCC can be improved by overaging to the T7 temper, but only at the expense of a 10 to 15% reduction in strength. More recently, retrogression and reaging (RRA) tempers have been developed, which have the high level of strength of the T6 temper combined with SCC resistance of the T7 temper [1].
Microstructural heterogeneity, specifically precipitate hardening particles and coarse intermetallic compounds, makes these alloys highly susceptible to localized corrosion. In order to retard corrosion, chromate conversion coatings (CCCs) are used.
245
The coatings inhibit corrosion and promote adhesion of protective organic coatings. In order to replace CCCs, a performance equivalent coating must be developed. A great deal of research into the formation and protection mechanisms of CCCs has been undertaken so that these mechanisms might be identified and duplicated in Cr-free coating. It has also been necessary to understand localized corrosion behavior of high strength Al alloys or address issues that are relavent to chromate conversion protection but were never thoroughly studied. A great deal of research has been aimed at the role of grain boundary precipitates [2-6], precipitate-free zones (PFZs) [7, 8], matrix precipitates [9, 10], solute depleted zones (SDZs) [11, 12], and intermetallic particles in the matrix [13-16].
In addition, most of the research reported in the literature has focused on
AA7075. Little attention has been paid to the effect of alloy temper on corrosion behavior of AA7475, which has higher purity level than AA7075. The aim of this study is to examine the effect of alloy temper on corrosion behavior of AA7475 using cyclic potentiodynamic polarization method with SEM/EDS.
A.2 EXPERIMENTAL
A.2.1 Materials and Heat treatment.
The material used for this study was commercial 40 mm thick AA7475, which was cut to 5 mm thickness to maximize quench rates after heat treatment. Heat treatments were carried out according to ASM standards to achieve W, T6, T7 and RRA tempers [17,
18].
Samples were prepared to study the dependence of orientation to the microstructure on the corrosion behavior. There are three different orientation planes.
246
The longitudinal plane or rolling plane was composed of longitudinal direction and transverse direction, the transverse plane was composed of longitudinal direction and short transverse direction, and the short transverse plane was composed of transverse direction and short transverse direction (Figure A.1).
A.2.2 Electrochemical measurements.
The cyclic polarization experiments were carried out as follows. The experiment was based on a linear sweep of the potential at 0.2 mV/sec. The initial sweep was in the positive direction starting 30 mV below corrosion potential. When the measured current
-3 2 reached an apex potential i.e. 0.5 VSCE or apex current density of 10 A/cm , the sweep direction was reversed. The experiments were performed in a three-electrode cell. A saturated calomel electrode (SCE) and a platinum mesh were used as reference and auxiliary electrodes, respectively. The electrolyte was a 0.5 M NaCl solution, which was deaerated by bubbling argon through the solution for at least 1 day before immersing the specimen. Each specimen was masked along its edges with black wax prior to testing exposing surface areas of about 0.15~0.2 cm2.
A.3 RESULTS AND DISCUSSION
A.3.1 Surface structures
The microstructure of polished samples was revealed using Keller’s reagents.
Figure A.1 is a typical optical micrograph showing the grain morphology for AA7475-T7 alloy. Grains were pancake shaped characteristic of wrought product.
247
Representative micrographs of AA7475 are illustrated in the scanning electron micrographs of Figure A.2. The microstructures consist of coarse intermetallic compounds (IMCs). EDS analysis indicated that there were three main types of IMCs: blocky and irregularly shaped Al-Fe-Cu particles, smaller spherically shaped Al2CuMg
(S phase) particles, and another blocky irregularly shaped MgSi2 particles. Thoughout visible in these micrographs, dispersoids, which were formed during solidification, and fine hardening precipitates, which were formed during artificial aging, were present in the microstructure. Coarse IMC particles are sites for pit initiation, but are not expected to be affected by low temperature artificial aging heat treatments [2, 3, 16, 19].
A.3.2 Potentiodynamic polarization curves
Anodic potentiodynamic polarization experiments were performed on all tempers of AA7475 in deaerated 0.1 M and 0.5 M NaCl (Figure A.3). The anodic polarization curves showed corrosion potentials and breakdown potentials indicated by a sharp increase in current density. Of all the tempers examined, only the T6 tempers exhibited two breakdown potentials. The first breakdown potential was characterized by a sharp increase in current density at the end of the passive range. As the potential increased further, the current density increased up to 1 mA/cm2 and then immediately decreased by as much as a factor of 5. Above second breakdown potential, the current density increased until the reaction became transport-limited. The effect of chloride activity was clearly seen on breakdown potential and passivation current density in 0.1 M and 0.5 M
NaCl [20]. As the chloride concentration increased, the breakdown potential decreased about 100 mV while the passive current density increased.
248
To accurately characterize the effect of temper on polarization response one series of experiment was carried out according to a strict protocol. Before measuring polarization curves, the open circuit potential (OCP) was monitored for 2 hours in deaerated 0.5 M NaCl solution. The OCP evolution of AA7475 for all temper was shown in Figure A.4. Several sharp cathodic drops in the OCP curves were observed immediately after immersion in solution and increases to more positive potentials were followed. The OCP values reached relatively stationary values after one hour. The 200 mV OCP transient may be associated with dissolution of active materials, such as Mg by dealloying of Al2CuMg, dissolution of MgZn2, or dissolution adjacent to Al-Fe-Cu IMCs due to galvanic interaction during exposure.
The polarization response was characterized by the characteristic potentials; corrosion potential (Ecorr), breakdown potential (Ebr), and repassivation potential (Erepass)
(Figure A.5). The repassivation potential was taken to be the sharp decrease in current density during cathodic reverse scans even though the current density at this point was typically higher than that of the passive current density. The 2nd zero current potential
(Ecorr’ on reverse scan) was taken to another corrosion potential during backward scan.
Measurements were carried out at least three times on different samples. Figure A.6 shows the typical cyclic polarization curves for all tempers on rolling planes. The solution heat-treated samples had slightly lower breakdown potential of Ebr values of about –0.775 VSCE compared to the others. The T6 temper had a Ebr of –0.755 VSCE and both T7 and RRA tempers had Ebr values of about –0.725 VSCE. The passive current densities of W and RRA tempers were lower than those of T6 and T7 tempers. The passivation current densities were about 1 µA/cm2. The repassive current density of
249
solution heat-treated samples was about 0.1 mA/cm2, which was almost one order of magnitude higher than other tempers. The repassive current density decreased in the following order: W > T6 > T7 > RRA. The scatter in the repassivation potential is very small, within 5 mV compared to the other potentials. The repassivation potentials of T7 and RRA tempers were higher than those of T6 and W tempers (Table A.1).
The results of cyclic polarization curves are summarized in Figure A.7. The polarization behavior was independent of the orientations of planes, even though there was significant difference of aspect ratio of grains. The breakdown potential (Ebr) of T7 temper was higher than that of T6 temper on all three planes, which was consistent with
MMA polarization described in chapter 5. The Ebr of RRA temper was similar to that of
T7 temper. The Ebr of W temper was lowest among all tempers. Because the corrosion potential (Ecorr) of T6 temper was higher than that of T7 temper, the difference between
Ebr and Ecorr in T7 temper was larger than T6 temper. It is interesting to consider that this difference of potentials may be an indication of localized corrosion resistance.
A.4 SUMMARY
The corrosion behavior on AA7475 with different alloy tempers has been investigated by potentiostatic and potentiodynamic polarization and SEM/EDS. The results in detail are as follows:
1. The characteristic potentials of AA7475 were independent of the microstructural
orientation. AA7475-T7 had higher breakdown potential than AA7475-T6.
2. The current densities of passivation for T6 and T7 tempers were almost identical,
whereas the current density of repassivation of T7 temper was slightly lower than
250
that of T6 temper. The breakdown potentials for all planes were in this order: T7 temper > T6 temper. The corrosion potentials for all planes were in the order of
T6 temper > T7 temper.
251
REFERENCES
1. B. Cina, U.S. Patent 3,856,584 (1974).
2. T. Ramgopal, P. Schmutz, and G.S. Frankel, J. Electrochem. Soc., 148, B348 (2001).
3. S. Maitra and G.C. English, Metallurgical Transaction A, 12A, 535 (1981).
4. K.G. Kent, J. Inst. Metals, 97, 127 (1969).
5. P.N. Adler, R. DeIasi, and G. Geschwind, Metall. Trans., 3, 3191 (1972).
6. P.K. Poulose, J.E. Morall, and A.J. McEvily, Met. Trans., 3A, 3191 (1974).
7. A.J. Sedriks, P.W. Slattery, and E.N. Pugh, Trans. ASM, 62, 238 (1969).
8. H.A. Holl, Corrosion, 23, 173 (1967).
9. A.J. DeArdo and R.D. Townsend, Met. Trans., 1A, 2573 (1970).
10. N. Ryum, Acta Met., 17, 821 (1969).
11. C.R. Shastry, M. Levy, and A. Joshi, Corros. Sci., 21, 673 (1981).
12. R.P. Wei, M. Gao, and P.S. Pao, Scripta Met., 18, 1195 (1984).
13. R.P. Wei, C.M. Liao, and M. Gao, Met. Trans. A, 29A, 1153 (1998).
14. P.S. Pao, C.R. Feng, and S.J. Gill, Corrosion, 10, 1022 (2000).
15. G.S. Chen, M. Gao, and R.P. Wei, Corrosion, 52, 8 (1996).
16. R.G. Buchheit, R.P. Grant, P.F. Hlava, B. Mckenzie, and G.L. Zender, J. Electrochem. Soc., 144, 2621 (1997).
17. ASM International, ASM handbook: Vol. 4 Heat Treating, Metal Parks: American Society for Metals (1991).
18. H.E. Boyer and T.L. Gall, Metals Handbook, Metal Parks: American Society For Metals (1985).
19. S. Maitra and G.C. English, Metallurgical Transaction A, 13A, 161 (1982).
20. H.P. Leckie and H.H. Uhlig, J. Electrochem. Soc., 113, 1262 (1966).
252
TABLES
VSCE (mVSCE) Longitudinal Plane Transverse Plane Short Transverse Plane
Ecorr -1.008 (±51.0) -1.020 (±5.5) -1.014 (±34.5)
W Ebr -0.775 (±10.5) -0.771 (±4.0) -0.769 (±7.5)
Erepass -0.799 (±10.0) -0.787 (±21.0) -0.787 (±1.5)
Ecorr -1.031 (±1.0) -1.008 (±17.0) -0.974 (±9.5)
T6 Ebr -0.754 (±26.5) -0.765 (±29.5) -0.752 (±14.5)
Erepass -0.807 (±1.5) -0.809 (±0.5) -0.808 (±2.5)
Ecorr -1.086 (±14.5) -1.026 (±12.0) -1.016 (±48.0)
T7 Ebr -0.728 (±21.0) -0.725 (±10.0) -0.730 (±21.0)
Erepass -0.750 (±4.0) -0.755 (±4.5) -0.755 (±6.5)
Ecorr -1.032 (±14.5) -1.041 (±25.5) -1.031 (±10.5)
RRA Ebr -0.727 (±1.5) -0.725 (±2.5) -0.724 (±19.0)
Erepass -0.739 (±1.0) -0.745 (±2.5) -0.748 (±1.0)
Ecorr: corrosion potential; Ebr: breakdown potential; Erepass: repassivation potential
Table A.1: Corrosion potentials, breakdown potentials, and repassivation potentials for AA7475 of all tempers and three different planes.
253
FIGURES
Rolling direction
ST T
60 µm L
Figure A.1. Three-dimensional representation of polished and etched grain morphology for AA7475-T7 alloy. L, T, and ST indicate longitudinal, transverse and short transverse directions respectively. The rolling plane consists of L and T directions.
254
(a)
Al-Fe-Cu IMCs
Mg2Si
Al2CuMg
(b)
Figure A.2. Scanning electron micrograph of AA7475-T7 showing the Al-Fe-Cu and Al2CuMg intermetallic compounds (white) and Mg2Si (black) in (a) SE and (b) BSE images.
255
-0.4 -0.4 (a) W temper (b) T6 temper -0.5 -0.5
-0.6 -0.6
-0.7 -0.7
-0.8
-0.8 -0.9 Potential (V SCE) vs. Potential Potential (V SCE) vs. Potential -0.9 -1.0 0.1 M NaCl 0.5 M NaCl -1.0 -1.1 1e-11 1e-10 1e-9 1e-8 1e-7 1e-6 1e-5 1e-4 1e-3 1e-2 1e-111e-101e-91e-81e-71e-61e-51e-41e-31e-2 Current Density (A/cm2) Current Density (A/cm2)
-0.4 -0.3 (c) T7 temper (d) RRA temper -0.4 -0.6 -0.5
-0.6 -0.8
-0.7
-1.0 -0.8 Potential (V SCE) vs. Potential (V SCE) vs. Potential -0.9 -1.2
-1.0 1e-12 1e-11 1e-10 1e-9 1e-8 1e-7 1e-6 1e-5 1e-4 1e-3 1e-2 1e-11 1e-10 1e-9 1e-8 1e-7 1e-6 1e-5 1e-4 1e-3 1e-2 Current Density (A/cm2) Current Density (A/cm2)
Figure A.3. Anodic potentiodynamic polarization curves for all tempers of AA7475 in deaerated 0.1 M NaCl (⎯) and 0.5 M NaCl (---) at a scan rate of 0.2 mV/s. (a) W, solution heat-treatment, (b) T6, artificial peak-aging treatment (c) T7, over-aging treatment, and (d) RRA, retrogression and reaging temper.
256
-1.00
-1.05
-1.10
-1.15
Potential (V vs. SCE) (V Potential W T6 -1.20 T7 RRA
-1.25 0 2000 4000 6000 8000
Time (sec)
Figure A.4. Open circuit potential (OCP) against time curves in deaerated 0.5 M NaCl for different heat treatments on rolling planes (●: W temper, ▼: T6 temper, ■: T7 temper, and ¡: RRA temper).
257
-0.6
Breakdown Potential, -0.7 Ebr
-0.8
2nd zero current -0.9 Potential, Ecorr’ Repassivation Potential, Erepass
Potential (V vs. SCE) -1.0
-1.1 Corrosion Potential, E corr -1.2 10 -10 10-9 10-8 10-7 10-6 10-5 10-4 10 -3 10 -2
Current Density (A/cm2)
Figure A.5. Characteristic potentials of AA7475-T7: Corrosion potential: Ecorr, nd Breakdown potential: Ebr, Repassivation potential: Erepass, and 2 zero current potential: Ecorr’ on reverse scan.
258
-0.6
-0.7
-0.8
-0.9
-1.0
Potential (V vs. SCE)
W -1.1 T6 T7 RRA
-1.2 1e-10 1e-9 1e-8 1e-7 1e-6 1e-5 1e-4 1e-3 1e-2
Current Density (A/cm2)
Figure A.6. Anodic polarization curves of AA7475 after 2hr OCP in deaerated 0.5 M NaCl for different heat treatments on rolling planes.
259
-0.6 (a) -0.7
-0.8
-0.9
-1.0
E corr -1.1 E Potential (V vs. SCE) br Erepass -1.2 Ecorr'
WT6T7RRA
Temper
-0.6 (b) -0.7
-0.8
-0.9
-1.0
Ecorr -1.1 E
Potential (V vs. SCE) br
Erepass -1.2 E corr' W T6 T7 RRA
Temper
Figure A.7. Corrosion potential (Ecorr), breakdown potential (Ebr), repassivation potential nd (Erepass), and 2 zero current potential (Ecorr’ on reverse scan) in deaerated 0.5 M NaCl as a function of the heat-treatment for (a) longitudinal plane, (b) transverse plane, and (c) short transverse plane. (Continued)
260
Figure A.7: (Continued)
-0.6 (c)
-0.7
-0.8
-0.9
-1.0 E -1.1 corr
vs. (V SCE) Potential Ebr E -1.2 repass Erecorr'
WT6T7RRA Temper
261
BIBLIOGRAPHY
Abd Pabbo, M.F., J.A. Richardson, and G.C. Wood, Corrosion Science, 18, 117h (1978).
Adler, P.N., R. DeIasi, and G. Geschwind, Metall. Trans., 3, 3191 (1972).
Adler, P.N. and R. DeIasi, Metall. Trans. A, 8A, 1185 (1977).
Akiyama, E. and G.S. Frankel, J. Electrochem. Soc., 146, 4095 (1999).
Andreatta, F., H. Terryn, and J.H.W. de Wit, Corros. Sci., 45, 1733 (2003).
Argon, A.S., "Physics of Strength and Plasticity", MIT Press:Cambridge (1969).
Arrowsmith, D.J., J.K. Dennis, and P.R. Sliwinski, Transactions of the Institute of Metal Finishing, 62, 117 (1984).
Asami, K., M. Oki, G.E. Thompson, G.C. Wood, and V. Ashworth, Electrochemica Acta, 32, 337 (1987).
ASM International, "ASM handbook: Vol. 4 Heat Treating", American Society for Metals:Metal Parks (1991).
ASTM B117, "Standard Practice for Operation Salt Spray (Fog) Apparatus", ASTM:West Conshohochen, PA (1997).
Auld, J.H. and S. Mck. Cousland, Scripta Metall., 5, 765 (1971).
Ayer, R., J.Y. Koo, J.W. Steeds, and B.K. Park, Metallurgical Transaction A, 16A, 1925 (1985).
Babic, R., M. Metikos-Hukovic, and A. Jukic, J. Electrochem. Soc., 148, B146 (2001).
Badawy, W.A., F.M. Al-Kharafi, and A.S. El-Azab, Corrosion Science, 41, 709 (1999).
Baek, Y. and G.S. Frankel, J. Electrochem. Soc., 150, B1 (2003).
Bailin, L.J., P. Fitzpatrick, and M.J. Joyce, "Evaluation of Unpainted Alodine Chromate Conversion Coatings for Corrosion Resistance and Electrical Conductivity", Lockheed Missiles and Space Co., Report F035575 (1985). 262
Baldantoni, A., Mater. Sci. Engr., L5, (1985).
Bard, A.J. and L.R. Faulkner, "Electrochemical Methods: Fundamentals and Applications", John Wiley & Sons:New York (2001).
Bardwell, J.A., B. MacDougall, and M.J. Graham, J. Electrochem. Soc., 135, 413 (1988).
Bardwell, J.A., B. MacDougall, and G.I. Sproule, J. Electrochem. Soc., 136, 1331 (1989).
Beavers, J.A., C.L. Durr, and N.G. Thompson, "Unique interpretations of potentiodynamic polarization technique", Corrosion'98 NACE, (1998).
Beck, T.R. and S.G. Chan, Corrosion, 37, 665 (1981).
Beck, T.R., J. Electrochem. Soc., 129, 2412 (1982).
Beck, T.R., Electrochimica Acta, 29, 485 (1984).
Bergmann, G., L.T. Waugh, and L. Paulng, Acta Cryst., 10, 254 (1957).
Biestek, T. and J. Weber, "Electrolytic and chemical conversion coatings: a concise survey of their production, properties and testing", Portcullis press limited-redhill (1976).
Blanc, C., B. Lavelle, and G. Mankowski, Corrosion Science, 39, 495 (1997).
Blanc, C. and G. Mankowski, Electrochemical Society Proceedings, 26, 677 (1997).
Bohni, H. and H.H. Uhlig, J. Electrochem. Soc., 116, 906 (1969).
Bohni, H., T. Suter, and A. Schreyer, Electrochimica Acta, 40, 1361 (1995).
Bonnel, K., C.L. Pen, and N. Pebere, Electrochinica Acta, 44, 4259 (1999).
Boyer, H.E. and T.L. Gall, "Metals Handbook", American Society For Metals:Metal Parks (1985).
Brossia, C.S. and R.G.y. Kelly, Corrosion Science, 40, 1851 (1998).
Brown, G.M., K. Shimizu, K. Kobayashi, G.E. Thompson, and G.C. Wood, Corrosion Science, 33, 1371 (1992).
Brown, G.M., K. Shimizu, K. Kobayashi, G.E. Thompson, and G.C. Wood, Corrosion Science, 34, 1045 (1993).
263
Brown, G.M., K. Shimizu, K. Kobayashi, G.E. Thompson, and G.C. Wood, Corrosion Science, 34, 2099 (1993).
Brown, G.M., K. Shimizu, K. Kobayashi, G.E. Thompson, and G.C. Wood, Corrosion Science, 35, 253 (1993).
Brown, G.M. and K. Kobayashi, J. Electrochem. Soc., 148, B457 (2001).
Buchheit, R.G., J. Electrochem. Soc., 142, 3994 (1995).
Buchheit, R.G., R.P. Grant, P.F. Hlava, B. Mckenzie, and G.L. Zender, J. Electrochem. Soc., 144, 2621 (1997).
Buchheit, R.G., Journal of applied electrochemistry, 28, 503 (1998).
Buchheit, R.G., M. Cunningham, H. Jensen, M.W. Kendig, and M.A. Martinez, Corrosion, 54, 61 (1998).
Buchheit, R.G., "Critical Factors for the Transition from Chromate to Chromate-Free Corrosion", Strategic Environmental Research and Development Program, Contract No. DACA72-99-C-0002 (1999).
Buchheit, R.G., L.P. Montes, M.A. Martinez, J. Micheal, and P.F. Hlava, J. Electrochem. Soc., 146, 4424 (1999).
Buchheit, R.G., M.A. Martinez, and L.P. Montes, J. Electrochem. Soc., 147, 119 (2000).
Buchheit, R.G., R.K. Boger, M.C. Carroll, R.M. Leard, C. Paglia, and J.L. Searles, JOM, 53, 29 (2001).
Buchler, M., J. Kerimo, F. Guillaume, and W.H. Smyrl, J. Electrochem. Soc., 147, 3691 (2000).
Burstein, G.T. and A.J. Davenport, J. Electrochem. Soc., 136, 936 (1989).
Cabot, P.L., F. Centellas, J.A. Garrido, R.M. Rogriguez, E. Brillas, E. Perez, A.V. Benedetti, and P.T.A. Sumodjo, J. Appl. Electrochem, 22, 541 (1992).
Cabot, P.L., J.A. Garrido, E. Perez, A.H. Moreira, P.T.A. Sumodjo, and A.V. Benedetti, J. Appl. Electrochem, 25, 781 (1995).
Cabot, P.L., A.H. Moreira, J.A. Garrido, P.T.A. Sumodjo, E. Perez, and W. Proud, Electrochem. Acta., 40, 447 (1995).
Campestrini, P., H. Terryn, J. Vereecken, and J.H.W. de Wit, J. Electrochem. Soc., 151, B359 (2004). 264
Chao, C.Y., L.F. Lin, and D.D. Macdonald, J. Electrochem. Soc., 128, 1187 (1981).
Chen, G.S., M. Gao, and R.P. Wei, Corrosion, 52, 8 (1996).
Chen, J.M., T.S. Sun, R.K. Viswanadham, and J.A.S. Green, Metallurgical Transaction A, 8A, 1935 (1977).
Cheng, Y.F. and J.L. Luo, J. Electrochem. Soc., 146, 970 (1999).
Cheng, Y.F., M. Wilmott, and J.L. Luo, British Corrosion Journal, 34, 280 (1999).
Cheng, Y.F., J.L. Luo, and M. Wilmott, Electrochimica Acta, 45, 1736 (2000).
Chidambaram, D., C.R. Clayton, and G.P. Halada, J. Electrochem. Soc., 150, B224 (2003).
Chidambaram, D., C.R. Clayton, and G.P. Halada, J. Electrochem. Soc., 151, B151 (2004).
Chung, S.W.M., J. Robinson, G.E. Thompson, G.C. Wood, and H.S. Isaacs, Philosophical Magazine B, 63, 557 (1991).
Cina, B., U.S. Patent 3,856,584 (1974).
Cohen, S.M., Corrosion, 51, 71 (1995).
Colner, W.H. and H.T. Francis, J. Electrochem. Soc., 105, 377 (1958).
Conde, A. and J. De Damborenea, Corrosion Science, 39, 295 (1997).
Conserva, M., Alumino E. Nuova Metallurgia, 39, 515 (1970).
Cooper, K.R. and R.G. Kelly, J. Chromatography A, 739, 183 (1996).
Cordier, H., C. Dumont, and W. Gruhl, Aluminium, 55, 777 (1979).
Cordier, H., C. Dumont, W. Gruhl, and B. Grzemba, Metall., 36, 33 (1982).
Cuynen, E., P.V. Espen, G. Goeminne, and H. Terryn, Journal of analytical Atomic Spectrometry, 14, 483 (1999).
Danh, N., K. Rajan, and W. Wallace, Metall. Trans., 14A, 1843 (1983).
Danh, N., K. Rajan, and W. Wallace, Metall. Trans., 16A, 2068 (1985).
265
Davenport, A.J. and H.S. Isaacs, Corrosion Science, 31, 105 (1990).
Davenport, A.J., J.A. Bardwell, and C.M. Vitus, J. Electrochem. Soc., 142, 721 (1995).
Davenport, A.J. and M. Sansone, J. Electrochem. Soc., 142, 725 (1995).
Davenport, A.J., L.J. Oblonsky, M.P. Ryan, and M.F. Toney, J. Electrochem. Soc., 147, 2162 (2000).
Davies, D.E., J.P. Dennison, and M.L. Mehta, "Evidence of intergranular corrosion in a high purity lauminum-zinc-magnesium alloy in unstressed condition", U. R. Evans Conference on Localized Corrosion, Williamsburg, Va (1971).
Davis, J.R., "Corrosion of aluminum and aluminum alloys", ASM International:Materials Park, OH (1999).
Davis, J.R., "Alloying:Understanding the basics", ASM International:Materials Park, OH (2001).
DeArdo, A.J. and R.D. Townsend, Met. Trans., 1A, 2573 (1970).
DeIasi, R. and P.N. Adler, Metall. Trans. A, 8A, 1177 (1977). deWexler, S.B. and J.R. Galvele, J. Electrochem. Soc., 121, 1271 (1974).
Dimitrov, N., J.A. Mann, and K. Sieradzki, J. Electrochem. Soc., 146, 98 (1999).
Dimitrov, N., J.A. Mann, M. Vukirovic, and K. Sieradzki, J. Electrochem. Soc., 147, 3283 (2000).
Dix, J.E.H., AIME, 137, 11 (1940).
Dix, J.E.H., ASM, 42, 1057 (1950).
Doig, P. and J.W. Edington, British Corrosion Journal, 9, 22 (1974).
Doig, P. and J.W. Edington, British Corrosion Journal, 9, 220 (1974).
Doig, P. and J.W. Edington, British Corrosion Journal, 9, 88 (1974).
Doig, P. and J.W. Edington, Metall. Trans. A, 6A, 943 (1975).
Doig, P. and J.W. Edington, Corrosion, 31, 347 (1975).
Dunn, C.G. and R.B. Bolon, J. Electrochem. Soc., 116, 1050 (1969).
266
Echeverria, F., P. Skeldon, G.E. Thompson, H. Habazaki, and K. Shimizu, J. Electrochem. Soc., 146, 3711 (1999).
EL Shayeb, H.A., F.M.A. EL Wahab, and S.Z. EL Abedin, British Corrosion Journal, 34, 37 (1999).
Embury, J.D. and R.B. Nicholson, Acta Met., 13, 403 (1965).
Faita, G., F. Mazza, and G. Bianchi, "Role of Water and Ionic Solvation in Localized Corrosion Phenomena", R.W. Stahle et al eds., NACE, Houston (1974).
Falkenberg, F., V.S. Raja, and E. Ahlberg, J. Electrochem. Soc., 148, B132 (2001).
Ferragut, R., A. Somoza, and A. Tolley, Acta mater, 47, 4355 (1999).
Fin, N., H. Dodiuk, A.E. Yaniv, and L. Drori, Applied Surface Science, 28, 11 (1987).
Foley, R.T., "Localized corrosion of aluminum alloys - a review", Corrosion'86 NACE, (1986).
Frankel, G.S., L. Stockert, F. Hunkeler, and H. Boehni, Corrosion, 43, 429 (1987).
Frankel, G.S., M.A. Russak, C.V. Jahnes, M. Mirzamaani, and V.A. Brusic, J. Electrochem. Soc., 136, 1243 (1989).
Frankel, G.S., Corrosion Science, 30, 1203 (1990).
Frankel, G.S., J.O. Dukovic, V. Brusic, B.M. Rush, and C.V. Jahnes, J. Electrochem. Soc., 139, 2196 (1992).
Frankel, G.S., R.C. Newman, C.V. Jahnes, and M.A. Russak, J. Electrochem. Soc., 140, 2192 (1993).
Frankel, G.S., C.V. Jahnes, V. Brusic, and A.J. Davenport, J. Electrochem. Soc., 142, 2290 (1995).
Frankel, G.S., J.R. Scully, and C.V. Jahnes, J. Electrochem. Soc., 143, 1834 (1996).
Frankel, G.S., "Mechanism of Al Alloy corrosion and the role of Chromate Inhibitors", Air Force Office of Scientific Research, Contract No. F49620-96-1-0479, Second Annual Report (1998).
Frankel, G.S., J. Electrochem. Soc., 145, 2186 (1998).
267
Frankel, G.S., "Mechanism of Al Alloy Corrosion and the role of Chromate Inhibitors", Air Force Office of Scientific Research, Contract No. F49620-96-1-0479, Final Report (2001).
Frankel, G.S. and R.L. McCreery, "Inhibition of Al Alloy Corrosion by Chromates", Inferface, 10, 34 (2001).
Furneaux, R.C., G.E. Thompson, and G.C. Wood, Corrosion Science, 19, 63 (1979).
Galvele, J.R. and S.M. de De Micheli, Corrosion Science, 10, 795 (1970).
Galvele, J.R., S.M. de De Micheli, I.L. Muller, S.B. de Wexler, and I.L. Alanis, "Critical Potentials for Localized Corrosion of Aluminum Alloys", R.W. Stahle et al eds., NACE, Houston, Texas (1974).
Galvele, J.R., J. Electrochem. Soc., 123, 464 (1976).
Galvele, J.R., "Present State of Understanding of the Breakdown of Passivity and Repassivation", R.P. Frankkethal et al. eds., The Electrochemical Society-4th International Symposium on Passivity, Princeton, NJ (1978).
Galvele, J.R., Corrosion Science, 21, 551 (1981).
Gao, M., C.R. Feng, and R.P. Wei, Metallurgical and Materials Transactions A, 29A, 1145 (1998).
Gili, P. and P.A. Lorenzo-Luis, Coordination Chemistry Reviews, 193-195, 747 (1999).
Gjonnes, J. and C.J. Simensen, Acta Met., 18, 881 (1970).
Graf, R., Compt. Rend., 242, 1311 (1956).
Graf, R., Compt. Rend., 244, 337 (1957).
Gruhl, W., Aluminum, 38, 775 (1962).
Gruhl, W. and H. Cordier, Aluminum, 44, 403 (1968).
Gruhl, W., "International Congress on Aluminium Alloys in the Aircraft Industry", Turin (1976).
Guillaumin, V. and G. Mankowski, Corrosion Science, 41, 421 (1999).
Guillaumin, V., P. Schmutz, and G.S. Frankel, J. Electrochem. Soc., 148, B163 (2001).
268
Habazaki, H., K. Shimizu, P. Skeldon, G.E. Thompson, X. Zhou, J. De Laet, and G.C. Wood, Corrosion Science, 39, 719 (1997).
Hackerman, N. and R.A. Powers, Journal of Physical Chemistry, 57, 139 (1953).
Hagans, P.L. and C.M. Haas, Surface and Interface Analysis, 21, 65 (1994).
Hagans, P.L. and C.M. Haas, "Chromate Conversion Coatings", in ASM Handbook, Vol. 5, Surface Engineering, ASM International: Metals Park, OH (1994).
Hagans, P.L. and C.M. Haas, Surface Engineering, 13, 5 (1996).
Halada, G.P. and C.R. Clayton, J. Electrochem. Soc., 138, 2921 (1991).
Halada, G.P., C.R. Cayton, M.J. Vasquez, and J.R. Kearns, "Spatially-resolved microchemical analysis of chromate conversion coated aluminum alloys and constituent intermetallic particle analogs"(1998).
Halada, G.P., C.R. Clayton, M.J. Vasquez, J.R. Kearns, M.W. Kendig, S.L. Jeanjaquet, G.G. Peterson, G.S. McCarthy, and G.L. Carr, eds., PV 99-27, The Electochemical Society Proceedings Series, (1999).
Hall, H.A., Corrosion, 23, 173 (1967).
Hashimoto, M., S. Miyajima, and T. Murata, Corrosion Science, 33, 885 (1992).
Hatch, J.E., "Aluminum propeties and physical metallurgy", American Society for Metals:Metals Park, Ohio (1984).
Hawkins, J.K., H.S. Isaacs, S.M. Heald, J. Tranquada, G.E. Thompson, and G.C. Wood, Corrosion Science, 27, 391 (1987).
He, J., V.J. Gelling, D.E. Tallman, and G.P. Bierwagen, J. Electrochem. Soc., 147, 3661 (2000).
Hebert, K. and R. Alkire, J. Electrochem. Soc., 130, 1001 (1983).
Heimgartner, P. and H. Bohni, Corrosion-NACE, 41, 715 (1985).
Heine, M.A. and M.J. Pryor, J. Electrochem. Soc., 114, 1001 (1969).
Hevbare, G.O. and J.R. Scully, J. Electrochem. Soc., 148, B196 (2001).
Hinton, B.R., D.R. Arnott, and N.E. Ryan, Metals Forum, 7, 211 (1984).
Hinton, B.R., D.R. Arnott, and N.E. Ryan, Materials Forum, 9, 162 (1986). 269
Hinton, B.R., Metal Finishing, 89, 15 (1991).
Hisamatsu, Y., "Pitting Corrosion of Stainless Steels in Chloride Solutions", R.W. Stahle et al. eds., NACE, Houstion, Texas (1974).
Hisamatsu, Y., T. Yoshii, and Y. Matsumura, "Electrochemical and Microscopical Study of Pitting Corrrosion of Austenitic Stainless Steel", R.W. Stahle et al. eds., NACE, Houston, Texas (1974).
Hoar, T.P., D.C. Mears, and G.P. Rothwell, Corrosion Science, 5, 279 (1965).
Hoar, T.P., Corrosion Science, 7, 341 (1967).
Holl, H.A., Corrosion, 23, 173 (1967).
Hollingsworth, E.H. and H.Y. Hunsicker, "Corrosion and Corrosion Protection Handbook", Marcel Dekker:New York (1983).
Hughes, A.E., J.D. Gorman, and P.J.K. Paterson, Corrosion Science, 38, 1957 (1996).
Hughes, A.E., R.J. Taylor, and B.R.W. Hinton, Surface and interface analysis, 25, 223 (1997).
Hunkeler, F. and H. Bohni, Corrosion, 37, 645 (1981).
Hunsicker, H.Y., eds., Rosenhain Centenary Conference on the Contribution of Physical Metallurgy to Engineering Practice, London (1976).
Hurley, B.L. and R.L. McCreery, J. Electrochem. Soc., 150, B367 (2003).
Ilevbare, G.O. and J.R. Scully, J. Electrochem. Soc., 148, B196 (2001).
Inman, M.E., R.G. Kelly, S.A. Willard, and R.S. Piascik, Proceeding of the FAA-NASA Symposium on Continued Arworthiness of Aircraft Structures, (1996).
Isaacs, H.S., J. Electrochem. Soc., 120, 1456 (1973).
Isaacs, H.S., "Potential Scanning of Stainless Steel during Pitting Corrosion", R.W. Stahle et al. eds., NACE, Houston, TX (1974).
Isaacs, H.S. and R.C. Newman, "Local electrochemistry of pitting corrosion in stainless steels" (1987).
Isaacs, H.S., Corrosion Science, 34, 525 (1993).
270
Isaacs, H.S., J.H. Cho, M.L. Rivers, and S.R. Sutton, J. Electrochem. Soc., 142, 1111 (1995).
Ivanov, E.S. and N.G. Klyuchnikov, Protection of Metals, 5, 476 (1969).
Jeffcoae, C.S., H.S. Isaacs, A.J. Aldykiewicz, and M.P. Ryan, J. Electrochem. Soc., 147, 540 (2000).
Jeffcoate, C.S., H.S. Isaacs, J. Hawkins, and G.E. Thompson, "The effect of chromate concentration on the repassivation of corroding aluminum", Electrochemical Society Proceedings, (1998).
Kammer, C., "Aluminum Handbook", Aluminium-Zentrale:Oldenburg, Germany (1999).
Karlberg, G. and G. Wrangen, Corrosion Science, 11, 499 (1971).
Karlik, M. and B. Jouffrey, Acta mater, 45, 3251 (1997).
Katzman, H.A., G.M. Malouf, R. Bauer, and G.W. Stupian, Applications of Surface Science, 2, 416 (1979).
Kendig, M., E. Addison, and S. Jeanjaquet, J. Electrochem. Soc., 146, 4419 (1999).
Kendig, M. and R. Buchheit, "Corrosion Inhibition of Al and Al Alloys by Hexavalent Cr Compounds-A Mechanistic Overview", R.G. Buchheit et al. eds., NACE, Houston, TX (2000).
Kendig, M., S. Jeanjaquet, R. Addison, and J. Waldrop, Surface and Coatings Technology, 140, 58 (2001).
Kendig, M.W., A.J. Davenport, and H.S. Isaacs, Corrosion Science, 34, 41 (1993).
Kent, K.G., J. Inst. Metals, 97, 127 (1969).
Ketcham, S.J. and F.H. Haynie, Corrosion, 19, 242t (1963).
Kim, H., N. Akao, N. Hara, and K. Sugimoto, J. Electrochem. Soc., 145, 2818 (1998).
Kim, H., N. Hara, and K. Sugimoto, J. Electrochem. Soc., 146, 3679 (1999).
Kim, H.P., R.H. Song, and S.I. Pyun, Br. Corros. J., 23, (1998).
Kintrup, L., L. de Riese-Meyer, and H. KGaA, Inst. Phys. Cont. Ser, 119, 257 (1991).
Kobayashi, Y., S. Virtanen, and H. Bohni, J. Electrochem. Soc., 147, 155 (2000).
271
Konno, H., S. Kobayashi, H. Takahashi, and M. Nagayama, Corrosion Science, 22, 913 (1982).
Korinek, K.A., "Chromate Conversion Coatings", ASM: Metals Park, OH (1987).
Koudelkova, M., J. Augustynski, and H. Berthou, J. Electrochem. Soc., 124, 1165 (1977).
Kowal, K., J. Deluccia, J.Y. Josefowicz, C. Laird, and G.C. Farrington, J. Electrochem Soc, 143, 2471 (1996).
Kowal, K., J. Deluccia, J.Y. Josefowicz, C. Laird, and G.C. Farrington, J. Electrochem. Soc., 143, 2471 (1996).
Laget, V., C.S. Jeffcoate, H.S. Isaacs, and R.G. Buchheit, J. Electrochem. Soc., 150, B425 (2003).
Laycock, N.J., M.H. Moayed, and R.C. Newman, "Prediction of Pitting Potential and Critical Pitting Temperatures", P.M. Natishan et al. eds., The Electrochemical Society, 95-15, Pennington, NJ (1996).
Laycock, N.J., J. Stewart, and R.C. Newman, Corrosion Science, 39, 1791 (1997).
Laycock, N.J., M.H. Moayed, and R.C. Newman, J. Electrochem. Soc., 145, 2622 (1998).
Leblanc, P. and G.S. Frankel, J. Electrochem. Soc., 149, B239 (2002).
Leckie, H.P. and H.H. Uhlig, J. Electrochem. Soc., 113, 1262 (1966).
Leclere, T.J.R. and R.C. Newman, J. Electrochem. Soc., 149, B52 (2002).
Lectere, T.J.R., A.J. Davenport, J. Deakin, D. Raikes, and R.C. Newman, Electrochemical Society Proceedings, 17, 130 (1998).
Lee, T.S., R.M. Kain, and J.W. Oldfield, Materials Performance, 9 (1984).
Lekie, H.P. and H.H. Uhlig, J. Electrochem. Soc., 113, 1262 (1966).
Lin, L.F., C.Y. Chao, and D.D. Macdonald, J. Electrochem. Soc., 128, 1194 (1981).
Llevbare, G.O., J.R. Scully, J. Yuan, and R.G. Kelly, Corrosion, 56, 227 (2000).
Lloyd, D.J. and M.C. Chaturvedi, J. Mater. Sci., 17, 1819 (1982).
Lorimer, G.W. and R.B. Nicholson, Acta Met., 14, 1009 (1966).
272
Lorimer, G.W. and R.B. Nicholson, "The Mechanism of Phase Transformations in Crystalline Solids", Inst. Metals:London (1968).
Lott, S.E. and R.C. Alkire, Corrosion Science, 28, 479 (1988).
Lott, S.E. and R.C. Alkire, J. Electrochem. Soc., 136, 973 (1989).
Lunt, T.T., S.T. Pride, J.R. Scully, and J.L. Hudson, J. Electrochem. Soc., 144, 1620 (1997).
Luo, J.L. and M.B. Ives, "Factors Controlling Pit Development", G.S. Frankel et al. eds., The Electrochemical Society, PV92-9, Princeton, NJ (1992).
Lytle, F.W., R.B. Greegor, G.L. Bibbins, K.Y. Blohowiak, R.E. Smith, and G.D. Tuss, Corrosion Science, 37, 349 (1995).
Maitra, S. and G.C. English, Metallurgical Transaction A, 12A, 535 (1981).
Maitra, S. and G.C. English, Metallurgical Transaction A, 13A, 161 (1982).
Mansfeld, F., S. Lin, Y. Wang, and H. Shih, Corrosion, 45, 615 (1989).
Mansfeld, F., S. Lin, S. Kim, and H. Shih, J. Electrochem. Soc., 137, 78 (1990).
Mansfeld, F., Y. Wang, and H. Shih, J. Electrochem. Soc., 138, L74 (1991).
Mansfeld, F., Y. Wang, and H. Shih, Electrochem. Acta., 37, 2277 (1992).
Mansfeld, F. and J.C.S. Fernandes, Corrosion Science, 34, 2105 (1993).
Marcus, P. and J.M. Herbelin, Corrosion Science, 34, 1123 (1993).
Marcus, P. and J. Oudar, "Corrosion Mechanisms in theory and practice" (1995).
Matienzo, L.J. and K.J. Holub, Applications of Surface Science, 9, 47 (1981).
Matsuda, S. and H.H. Uhlig, J. Electrochem. Soc., 111, 156 (1964).
Mazurkiewicz, B., Corros. Sci., 23, 687 (1983).
Mazurkiewicz, B. and A. Piotrowski, Corros. Sci., 23, 697 (1983).
McCafferty, E., J. Electrochem. Soc., 137, 3731 (1990).
McCafferty, E., Corrosion Science, 39, 243 (1997).
273
McEvily, A.J., J.B. Clark, and A.P. Bond, Trans. ASM, 60, 661 (1967).
McGovern, W.R., P. Schmutz, R.G. Buchheit, and R.L. McCreery, J. Electrochem. Soc., 147, 4494 (2000).
Mclntyre, J.F. and T.S. Dow, Corrosion, 48, 309 (1992).
Meng, Q. and G.S. Frankel, J. Electrochem. Soc., 151, B271 (2004).
Meng, Q.J., "Effect of Cu content on Corrosion Behavior and Chromate Conversion Coating Protection of 7XXX series Al alloys", The Ohio State University (2003).
Meng, Q.J. and G.S. Frankel, Surf. Int. Anal., 36, 30 (2004).
Mestres, J., M. Duran, P. Matrin-Zarza, E. Medina de la Rosa, and P. Gili, Inorg. Chem., 32, 4708 (1993).
Moayed, M.H. and R.C. Newman, Corrosion Science, 40, 519 (1998).
Mondolfo, L.F., N.A. Gjostein, and D.W. Levinson, Trans. Amer. Inst. Min. Met. Eng., 206, 1378 (1956).
Mondolfo, L.F., Metallurgical Reviews, 153, 95 (1971).
Mondolfo, L.F., "Aluminium alloys: Structure and properties", Buttenworths:Boston (1976).
Moshier, W.C., G.D. Davis, J.S. Ahearn, and H.F. Hough, J. Electrochem. Soc., 133, 1063 (1986).
Moshier, W.C., G.D. Davis, J.S. Ahearn, and H.F. Hough, J. Electrochem. Soc., 134, 2677 (1987).
Moshier, W.C., G.D. Davis, and G.O. Cote, J. Electrochem. Soc., 136, 356 (1989).
Moshier, W.C. and G.D. Davis, Corrosion, 46, 43 (1990).
Mukhopadhyay, A.K., Q.B. Yang, and S.R. Singh, Acta Metall. Mater., 42, 3083 (1994).
Muller, I.L. and J.R. Galvele, Corrosion Science, 17, 179 (1977).
Muller, I.L. and J.R. Galvele, Corrosion Science, 17, 995 (1977).
Nazarov, A.P., M.A. Petrunin, and Y.N. Mikhailovskii, Protection of Metals, 28, 432 (1993).
274
Newhard, N.J., Met. Finish, 70, 49 (1972).
Newhard, N.J., Met. Finish, 7, 69 (1972).
Newman, J., "Mass Transport and Potential Distribution in the Geometries of Localized Corrosion Phenomena", R.W. Stahle et al eds., NACE, Houston (1974).
Newman, R.C. and H.S. Isaacs, J. Electrochem. Soc, 130, 1621 (1983).
Newman, R.C., Corrosion Science, 25, 341 (1985).
Newton, C.J. and N.J.H. Holroyd, "Time-Lapse Video Technique in the Corrosion Testing of Aluminum Alloys", V.S. Agarwala et al. eds., ASTM, Philadelphia (1992).
Nguyen, T.H. and R.T. Foley, J. Electrochem. Soc., 127, 1980 (1980).
Nicholson, R.B., G. Thomas, and J. Nutting, Brit. J. Appl. Phys., 9, 25 (1958).
Nicholson, R.B., G. Thomas, and J. Nutting, J. Inst. Metals, 87, 429 (1958-59).
Nisancioglu, K., J. Electrochem. Soc., 137, 69 (1990).
Oldfield, J.W. and W.H. Sutton, British Corrosion Journal, 13, 13 (1978).
Orazem, M.E., "A tutorial on impedance spectroscopy", Corrosion'98 NACE, (1998).
Ortner, S.R., C.R.M. Grovenor, and B.A. Shollock, Scripta Metall., 22, 839 (1988).
Pao, P.S., C.R. Feng, and S.J. Gill, Corrosion, 10, 1022 (2000).
Park, J.K. and A.J. Ardell, Metall. Trans. A, 14A, 1957 (1983).
Park, J.K. and A.J. Ardell, Metall. Trans. A, 15A, 1531 (1984).
Park, J.K., Mater. Sci. Engng., A103, 223 (1988).
Park, J.K. and A.J. Ardell, Scripta Metall., 22, 1115 (1988).
Park, J.K. and A.J. Ardell, Mater. Sci. Engng., A114, 197 (1989).
Park, J.K. and A.J. Ardell, Acta Metall. Mater., 39, 591 (1991).
Park, J.O., C.H. Paik, Y.H. Huang, and R.C. Alkire, J. Electrochem. Soc., 146, 517 (1999).
275
Pashley, D.W., M.H. Jacobs, and J.T. Veitz, Phil. Mag., 16, 51 (1967).
Pearson, "Handbook of Lattice Spacing and Structure of Metals and Alloys", Pergamon Press:Oxford (1958).
Pessall, N. and C. Liu, Electrochimica Acta, 16, 1987 (1971).
Petrunin, M.A., V.D. Gil'dengorn, and A.P. Nazarov, Protection of Metals, 30, 130 (1994).
Pistorius, P.C. and G.T. Burstein, Corrosion Science, 33, 1882 (1992).
Pistorius, P.C. and G.T. Burstein, Corrosion Science, 36, 525 (1994).
Polmear, I.J., "Light Alloys: Metallurgy of the Light Metals", J. Wiley & Sons:New York (1996).
Poulose, P.K., J.E. Morall, and A.J. McEvily, Met. Trans., 3A, 3191 (1974).
Pourbaix, M., "Atlas of Electrochemical Equilibrium in Aqueous Solutions", Pergamon Press:Oxford (1966).
Pride, S.T., J.R. Scully, and J.L. Hudson, J. Electrochem. Soc., 141, 3028 (1994).
Pyun, S.I., S.M. Moon, S.H. Ahn, and S.S. Kim, Corrosion Science, 41, 653 (1999).
Quarto, F.D., M. Santamaria, N. Mallandrino, V. Laget, R. Buchheit, and K. Shimizu, J. Electrochem. Soc., 150, B462 (2003).
Rajan, K., W. Wallace, and J.C. Beddoes, J. Mater. Sci., 17, 2817 (1982).
Ramgopal, T., "Role of Grain boundary precipitates and Solute depleted zone in the Intergranular corrosion of Aluminum Alloy AA7150", The Ohio State University (2001).
Ramgopal, T., P. Schmutz, and G.S. Frankel, J. Electrochem. Soc., 148, B348 (2001).
Ramgopal, T., P.I. Gouma, and G.S. Frankel, Corrosion, 58, 687 (2002).
Ramsey, J.D., L. Xia, M.W. Kendig, and R.L. McCreery, Corrosion Science, 43, 1557 (2001).
Reich, L., M. Murayama, and K. Hono, Acta mater, 46, 6053 (1998).
Revie, R.W., "Uhlig's corrosion handbook", Wiley:New York (2000).
276
Richardson, J.A. and G.C. Wood, J. Electrochem. Soc., 120, 193 (1973).
Ringer, S.P., T. Sakurai, and I.J. Polmear, Acta mater, 45, 3731 (1997).
Ryan, M.P., N.J. Laycock, R.C. Newman, and H.S. Isaacs, J. Electrochem. Soc., 145, 1566 (1998).
Ryan, M.P., N.J. Laycock, H.S. Isaacs, and R.C. Newman, J. Electrochem. Soc., 146, 91 (1999).
Rynders, R.M., C. Paik, R. Ke, and R.C. Alkire, J. Electrochem. Soc., 141, 1439 (1994).
Ryum, N., Acta Met., 17, 821 (1969).
Sakashita, M. and N. Sato, Corrosion, 35, 351 (1979).
Sarmaitis, R. and E. Juzeliunas, Plating and Surface Finishing, 72 (1991).
Sato, F. and R.C. Newman, Corrosion, 54, 955 (1998).
Sato, N. and M. Cohen, J. Electrochem. Soc., 111, 512 (1964).
Sato, N. and M. Cohen, J. Electrochem. Soc., 111, 519 (1964).
Sato, N., Electrochimica Acta, 16, 1683 (1971).
Sato, N., J. Electrochem. Soc., 129, 260 (1982).
Sato, N., Corrosion, 45, 354 (1989).
Sato, N., Corrosion Science, 31, 1 (1990).
Sauerbrey, G., Z. Physik, 155, 206 (1959).
Schmalzried, H. and V. Gerold, Z. Metallkd., 49, 291 (1958).
Schmuki, P., S. Virtanen, A.J. Davenport, and C.M. Vitus, J. Electrochem. Soc., 143, 574 (1996).
Schmuki, P., S. Virtanen, H.S. Isaacs, M.P. Ryan, A.J. Davenport, H. Bohni, and T. Stenberg, J. Electrochem. Soc., 145, 791 (1998).
Schmutz, P. and G.S. Frankel, J. Electrochem. Soc., 145, 2295 (1998).
Schmutz, P. and G.S. Frankel, J. Electrochem. Soc., 145, 2285 (1998).
277
Schram, T., J. De Laet, and H. Terryn, J. Electrochem. Soc., 145, 2733 (1998).
Scully, J.R., D.E. Peebles, A.D. Romg, D.R. Frear Jr, and C.R. Hills, Metallurgical Transaction A, 23A, 2641 (1992).
Scully, J.R., T.O. Knight, R.G. Buchheit, and D.E. Peebles, Corros. Sci., 35, 185 (1993).
Scully, J.R., "The polarization resistance method for determination of instanteous corrosion rates: A review", Corrosion'98 NACE, (1998).
Sedriks, A.J., P.W. Slattery, and E.N. Pugh, Trans. ASM, 62, 238 (1969).
Seegmiller, J.C. and D.A. Buttry, J. Electrochem. Soc., 150, B413 (2003).
Sehgal, A., D. Lu, and G.S. Frankel, J. Electrochem. Soc., 145, 2834 (1998).
Sehgal, A., G.S. Frankel, B. Zoofan, and S. Rokhlin, J. Electrochem. Soc., 147, 140 (2000).
Seiler, H.G., "Handbook on Toxicity of Inorganic Compounds", Marel Dekker:New York (1988).
Sharland, S.M., Corrosion Science, 33, 183 (1992).
Shastry, C.R., M. Levy, and A. Joshi, Corros. Sci., 21, 673 (1981).
Shaw, B.A., T.L. Fritz, G.D. Davis, and W.C. Moshier, J. Electrochem. Soc., 137, 1317 (1990).
Shaw, B.A., P.J. Moran, and P.O. Gartland, Corrosion Science, 32, 707 (1991).
Shibata, T., Corrosion, 52, 813 (1996).
Shimizu, K., G.M. Brown, K. Kobayashi, G.E. Thompson, and G.C. Wood, Corrosion Science, 34, 1853 (1993).
Shimizu, K., K. Kobayashi, P. Skeldon, G.E. Thompson, and G.C. Wood, Corrosion Science, 39, 701 (1997).
Shimizu, K., K. Kobayashi, G.E. Thompson, P. Skeldon, and G.C. Wood, Corrosion Science, 39, 281 (1997).
Sieradzki, K. and R.C. Newman, Journal of phisicis and chemistry of solids, 48, 1101 (1987).
278
Sieradzki, K., R.R. Corderman, K. Shukla, and R.C. Newman, Philosophical Magazine, 59, 713 (1989).
Siitari, D.W. and R.C. Alkire, J. Electrochem. Soc., 129, 481 (1982).
Smith, A.J., T. Tran, and M.S. Wainwright, Journal of Applied Electrochemistry, 29, 1085 (1999).
Son, M., N. Akao, N. Hara, and K. Sugimoto, J. Electrochem. Soc., 148, B43 (2001).
Speidel, M.O., Metall. Trans., 6A, 631 (1975).
Sprowls, D.O. and R.H. Brown, "Stress Corrosion Mechanisms for Aluminum Alloys", R.W. Stahle et al eds., NACE, Houston, Texas (1969).
Staley, J.T., H.Y. Hunsicker, and R.H. Brown, U.S. Patent 3,881,966 (1975).
Stockert, L. and H. Bohni, Materials Science Forum, 44&45, 313 (1989).
Stratmann, M., Corros. Sci., 27, 869 (1987).
Stratmann, M. and H. Streckel, Corros. Sci., 30, 681 (1990).
Stratmann, M. and H. Streckel, Corros. Sci., 30, 697 (1990).
Stratmann, M. and H. Streckel, Corros. Sci., 30, 715 (1990).
Strehblow, H.H., Werkstoffe und Korrosion, 27, 792 (1976).
Strehblow, H.H. and M.B. Ives, Corrosion Science, 16, 317 (1976).
Strehblow, H.H. and J. Wenners, Electrochem. Acta., 22, (1977).
Sugimoto, K., K. Hoshino, M. Kageyama, S. Kageyama, and Y. Sawada, Corrosion Science, 15, 709 (1975).
Suter, T. and R.C. Alkire, Electrochemical Society Proceedings, 17, 118 (1998).
Suter, T. and R.C. Alkire, J. Electrochem. Soc., 148, B36 (2001).
Suter, T., E.G. Webb, H. Bohni, and R.C. Alkire, J. Electrochem. Soc., 148, B174 (2001).
Talianker, M. and B. Cina, Metall. Trans. A, 20A, 2087 (1989).
Tasi, T.C. and T.H. Chuang, Metall. Trans., 27A, 2617 (1996).
279
Tester, J.W. and H.S. Isaacs, J. Electrochem. Soc., 122, 1438 (1975).
Thackery, P.A., J. Inst. Metals, 96, 228 (1968).
Thomas, G. and J. Nutting, J. Inst. Metals, 88, 81 (1959-60).
Thomas, G., J. Inst. Metals, 89, 287 (1960-61).
Townsend, H.E. and R.G. Hart, J. Electrochem. Soc., 131, 1345 (1984).
Trasatti, S.P. and F. Mazza, British Corrosion Journal, 30, 275 (1995).
Treverton, J.A. and N.C. Davies, Metals Technology, 10, 480 (1977).
Treverton, J.A. and N.C. Davies, Surface and Interface Analysis, 3, 194 (1981).
Treverton, J.A. and M.P. Amor, Transactions of the Institute of Metal Finishing, 60, 92 (1982).
Treverton, J.A. and M.P. Amor, Journal of Materials Science, 23, 1706 (1988).
Treverton, J.A., M.P. Amor, and A. Bosland, Corrosion Science, 33, 1411 (1992).
Twite, R.L. and G.P. Bierwagen, Progress in organic coatings, 33, 91 (1998).
Uchi, H., T. Kanno, and R.S. Alwitt, J. Electrochem. Soc., 148, B17 (2001).
Uhlig, H.H., J. Electrochem. Soc., 97, 215C (1950).
Uhlig, H.H. and P.F. King, J. Electrochem. Soc., 106, 1 (1959).
Uhlig, H.H., "Competitive Adsorption as a Mechanism for Breakdown of Passivity", R.W. Stahle et al eds., NACE, Houston, Texas (1976).
Unwin, P.N.T. and R.B. Nicholson, Acta Met., 17, (1969).
Urquidi, M. and D.D. Macdonald, J. Electrochem. Soc., 132, 555 (1985).
Vasquez, M.J., G.P. Halada, C.R. Clayton, and J.P. Longtin, Surf. Int. Anal., 25, 223 (1997).
Verhoff, M. and R. Alkire, J. Electrochem. Soc., 147, 1349 (2000).
Wainright, J.S., O.J. Murphy, and M.R. Antonio, Corrosion Science, 33, 281 (1992).
Waldrop, J.R. and M.W. Kendig, J. Electrochem. Soc., 145, L11 (1998). 280
Wall, F.D. and M.A. Martinez, J. Electrochem. Soc., 150, B146 (2003).
Wan, J., G.E. Thompson, K. Lu, and C.J.E. Smith, Physica B, 208&209, 511 (1995).
Webb, E.G., T. Suter, and R.C. Alkire, J. Electrochem. Soc., 148, B186 (2001).
Wei, R.P., M. Gao, and P.S. Pao, Scripta Met., 18, 1195 (1984).
Wei, R.P., C.M. Liao, and M. Gao, Met. Trans. A, 29A, 1153 (1998).
Westmacott, K.H., R.S. Barnes, D. Hull, and R.E. Smallman, Phil. Mag., 6A, 929 (1961).
Williams, D.E., C. Westcott, and M. Fleischmann, J. Electrochem. Soc., 132, 1796 (1985).
Williams, D.E., J. Stewart, and P. Balkwill, Corrosion Science, 36, 1213 (1994).
Williams, L.F.G., Surface Technology, 5, 105 (1977).
Wu, B., J.R. Scully, and J.L. Hudson, J. Electrochem. Soc., 144, 1614 (1997).
Xia, L. and R.C. McCreery, J. Electrchem. Soc., 145, 3083 (1998).
Xia, L. and R.L. McCreery, J. Electrochem. Soc., 146, 3696 (1999).
Xia, L., "Formation and function of chromate conversion coating on aircraft aluminum alloy" PhD dissertation, The Ohio State University (2000).
Xia, L., E. Akiyama, G. Frankel, and R. McCreery, J. Electrochem. Soc., 147, 2556 (2000).
Xu, N., G.E. Thompson, J.L. Dawson, and G.C. Wood, Corrosion Science, 34, 461 (1993).
Xu, N., G.E. Thompson, J.L. Dawson, and G.C. Wood, Corrosion Science, 34, 479 (1993).
Yang, H.S., Materials Characterization, 38, 165 (1997).
Yoon, Y. and R.G. Buchheit, "The Effect of Heat Treatment on the Corrosion behavior and Chromate Conversion Performance of Al-Zn-Mg-Cu Alloy 7475", NACE, Houston, TX (2001).
281
Yoon, Y., V. Laget, and R.G. Buchheit, "The Effect of Artificial Heat Treatment on the Chromate Conversion Performance of Al-Zn-Mg-Cu alloys", Tri-Service Corrosion, San Antonio, TX (2002).
Yoon, Y. and R.G. Buchheit, Corros. Sci., Submitted (2004).
Yoon, Y. and R.G. Buchheit, J. Electrochem. Soc., Submitted (2004).
Yu, S.Y., W.E. O'Grady, D.E. Ramaker, and P.M. Natishan, J. Electrochem. Soc., 147, 2952 (2000).
Yu, Z., H. Ni, G. Zhang, and Y. Wang, Applied Surface Science, 62, 217 (1992).
Zhang, W. and G.S. Frankel, J. Electrochem. Soc., 149, B510 (2002).
Zhang, W., B.L. Hurley, and R.G. Buchheit, J. Electrochem. Soc., 149, B357 (2002).
Zhao, J., G.S. Frankel, and R.L. McCreery, J. Electrochem. Soc., 145, 2258 (1998).
Zhu, Y. and D.E. Williams, J. Electrochem. Soc., 144, L43 (1997).
Zuo, Y. and S. Fu, Corrosion, 54, 313 (1998).
282