Hardness is not really related to corrosion resistance. Corrosion resistance is proportional to the anodic layer thickness, as well as the presence of a sealer. Typical seals include immersion of the anodized component in boiling water or an aqueous solution of nickel acetate. The sealer closes off the porosity that forms during the anodizing process. Properly anodized and sealed components should show several thousand hours of salt spray resistance (ASTM B 117).
The thicker anodizing will provide greater corrosion resistance because of its greater thickness. However, sealing is critical. You can also seal the anodizing with a good epoxy.
Thicker, hard anodizing will register greater hardness under a hardness test with deeper penetration into the substrate (e. g., Rockwell). In this sense you can correlate hardness to corrosion resistance.
Thick, full anodizing also ensures an outer portion which is fully stoichiometric Al2O3.
It isn’t so simple. Greater anodize thickness and hardness do not equate to longer corrosion resistance. Hard anodize is usually left unsealed for maximum hardness and wear resistance. Where corrosion resistance is required, it is given a dichromate seal (better than boiling DI or nickel acetate or fluoride) which lowers the hardness slightly. See MIL-A-8625F (get at
The corrosion resistance of hard anodize depends primarily upon freedom from physical defects (e.g., cracks) and freedom from chemical defects (resulting from intermetallics and Si particles in the Al alloy).
Cracking is more likely for thicker anodic coatings, which need a greater radius of curvature at edges (see MIL-A-8625F again), are less flexible if the substrate is flexed (one reason why thin chromic acid anodize is used on airplanes) or can crack from the CTE mismatch with the alloy. Hard anodize of standard 0.002” thickness will often show a stress cracking pattern know as ‘crazing’ after the hot sealing [note: hard anodizing is typically conducted at 32oF. Easily observed at 20x magnification. I did some research a few years back and found that lowering the dichromate sealing temperature to 160oF (from 202+10oF) minimized crazing.
Chemical inhomogeneities in the alloy result in inhomogeneities in the anodic coating. Part of the reason that hard anodizing per MIL-A-8625F is restricted to alloys of not more than 5% Cu or 8% Si. Alloying elements present as intermetallics in the Al matrix are a major problem if the size is anywhere near that of the anodic thickness. The other metal may anodize and be incorporated in the alumina as foreign cations, or resist oxidation and be incorporated as metal (case for Si). This causes local stresses as the downward growing anodic layer is stretched over the particle, plus galvanic corrosion can result, which eventually undermines the anodic coating. Hard anodize on highly alloyed Al 2024, 7075 and casting alloys has less corrosion resistance than that on Al 1100, 3003, 5xxx or 6xxx alloys. Some info on the effect of alloying elements upon anodic coatings is given in the AAC’s Reference Guide:
I've worked with anodizing. Maximum corrosion resistance comes with maximum thickness. You find this in a full hard (usually black) anodizing, which is generally 0.002" into the substrate and 0.002" grown over it. It forms a collumnar structure which is fully stoichiometric and minimally permeable at full thickness. If one seals with a stable epoxy, the salt fog resistance is exceptional.
JimMetalsCeramics,
It is clear that you have limited familiarity with hard anodized aluminum.
"It must not be assumed that the thicker coatings will necessarily give better protection, as in thick coatings the tendancy to crack undoubtedly decreases the corrosion resistance by acting as capillaries for the entry of moisture." -- The Surface Treatment and Finishing of Aluminum and its Alloys, 6th Edn.,, vol. 2, p.810 (2001).
The total thickness of 0.004" which you say is general cannot be achieved on cast alloys and many wrought alloys. Hard anodizing is generally 0.002" thick (0.001" into the substrate + 0.001" buildup). From MIL-A-8625F:
"3.7.2 Type III coatings.
3.7.2.1 Thickness of coating. Unless otherwise specified in the contract, purchase order, or applicable drawing (see 6.2), the nominal thickness of the coating shall be 0.002 inch (2 mils)"
Also, 'sealing' as used with respect to anodizing generally refers to hydration of the aluminum oxide pore walls. Alternatives to sealing by the use of materials which enter the pores, e.g., with wax, oil or silicone is referred to as 'impregnation.' Lockheed Martin has a process wherein methyl methacrylate monomer is impregnated into the pores and then thermally polymerized; it gives years of salt spray resistance. PTFE also gives exceptional performance if submicron particles are used to allow some pore impregnation in addition to surface coating.
Application of epoxy is known as 'painting.' This is usually performed over dichromate-sealed hard anodize for best corrosion resistance, e.g., in the Navy work for which I solved a crazing problem.
If one goes to a competent anodizing house (aluminum anodizing), there will not be fissures in the coating. The columnar structure can have areas between columns which are open or thin. In a set of exhaustive studies, it was clear, as can be expected, that the thicker the anodizing, the more protected was the surface. Dichromate is all well and good, but a thick anodizing will ensure that even with surface defects, the underlying oxide is thick and stoichiometric. The requirement for sealing was and is substantially less if the anodizing is at maximum thickness.
The full anodized coating has the opportunity to have the H2SO4 penetrate and build up the full oxide below and in between any columnar grains which are not packed to full areal density. What happens is the immediately available surface builds the initial oxide thick and stable. The rate of oxide growth in pits and grain boundary regions can be much slower.
Jim’misinformation’MetalsCeramics,
Such naïve hyperbole about ‘full, thick, dense, stoichiometric’ hard anodize suggests that you are a salesman with a superficial knowledge of the anodizing mechanism and little knowledge of chemistry or physics involved. To begin with, anodize growth is only possible because of its cellular porosity (reaching to a thin barrier oxide bottom layer), which amounts to some 20 vol. % for Al 1100 and 35% for 2000 series alloys. Not fully dense. Stoichiometry isn’t quite true either, as the anodize includes 10-15 wt% sulfate.
It follows from basic scientific principles that both conventional and hard anodize become softer and more porous beyond certain (alloy-dependent) thicknesses, and this is reflected in specifications. For instance, the nominal thickness for hard anodize per MIL-A-8625F, AMS 2468E and BS 5599:1993 is 2 mils (50 microns). The latter specification allows lower microhardness values for coatings of more than 60 microns (2.36 mils).
As previously mentioned, defects increase with increasing anodize thickness for both inside corners and outside edges. During growth at a corner, the 2 growth interfaces intersect at a right angle and the corner gets starved (and this is the area of fastest dissolution, too). At inside corners, the outwardly growing surfaces intersect, resulting in a line of defects. This is the reason that greater radii of curvature are required with increasing anodize thickness per MIL-A-8625F and ALCOA studies; see photos in The Surface Treatment and Finishing of Aluminum and its Alloys, 6th Edn., vol. 2, p. 787-788 (2001).
Also, ‘crazing,’ cracking on all surfaces due to CTE mismatch between the oxide and the Al alloy increases with increasing thickness; see photo ibid. p. 786. “With hard anodizing carried out at low temperatures in the region of 0-5oC, the increase to room temperature is sometimes sufficient to cause the film to fracture under tension, and a dip in hot water is almost certain to cause it.” -- ibid. p. 945.
The main reason for increasing anodize softness and porosity with anodize thickness is heat. This heat has 2 sources: 1) the heat of formation of the alumina, about 1670 kJ/mol and 2) ohmic or Joule heating due to the electrical resistance of the oxide. Hard anodize is typically conducted with a current density of about 36 Amps/ sq. ft, and the voltage is gradually raised from about 20 to 80 V as the oxide thickens to maintain the growth. To reach a 4 mil thickness would require about 100 V for some alloys. This 3600+ watts/sq. ft of heat must be removed from the anodize via conduction and transfer to the agitated solution. It is impossible for the growth interface to remain at 0oC, the usual solution temperature, and thus the porosity increases. If there is excessive temperature rise, the anodize can undergo structural changes, even crystallization, and may even become powdery.
The other reason for increasing softness and porosity with anodize thickness is a matter of chemistry and time. The oxide is continually dissolving into the electrolyte at its outer surface and within the cellular pores:
“the extent of pore widening depends upon the length of time in contact with the acid solution. Consequently, the first-formed anodic film material which is at the outer surface of the film, suffers most attack. Although film growth is coulombic, when the extent of chemical dissolution is such that the pore diameter at the outer surface equals the cell diameter, further film thickening ceases.” – ibid., vol. 1, p. 377.
This is more of a factor for Type II than for hard anodizing, though.
The decrease in abrasion wear (Taber testing) and in hardness with increasing thickness is considerable for Al 2024, less so for Al 6061. See figures in ibid., vol. 2, p. 799-800.
An interesting study simulating the effect of rainfall on the leading edges of aerofoils found that “With increasing [hard anodize] thickness the resistance to rainfall erosion is decreased. Failure in the case of 2024 (not clad) was shown by spalling of the coating in layers while on 6061 and 7075, failure followed the micro-crack structure.” – ibid., p. 810.
In view of the greater defect density of ‘thick, full’ hard anodize, I suggest that the need for (dichromate) sealing becomes more important for increasing anodize thickness.
There was also a comment that open areas can be healed by penetration of H2SO4 below and between alumina grains. While anodic thickness is normally very uniform since current flows more readily through areas of lesser resistance, bare or weak spots that occur after significant growth (such as can occur from intermetallics in 2024) are likely to burn (electrochemically erode at a rapid rate):
“With the high current densities and high voltages involved in hard anodizing, any weak point in the film may allow a very high proportion of the available current to flow at that point. Enormous local current densities are achieved and the high temperature produced means that the film is dissolved as fast as it is formed, and rapid dissolution of the base metal takes place.” – ibid., p. 801.
I might also point out that in the licensed Metalast anodizing process, target anodize thicknesses are 0.7 mild for Type II and 1.5 mils for Type III (hard anodize). E.g., see
Note: the corrosion resistance results shown are for 0.7 mil thick Type II anodize; Metalast hasn’t published any results that I could find on the corrosion resistance of their hard anodize.
Suggest that you read ‘Chapter 6. The Fundamentals of Anodizing’ and ‘Chapter 9. Hard Anodizing’ in The Surface Treatment and Finishing of Aluminum and its Alloys, 6th Edn., (2001).
Any references for the extensive studies you have mentioned?
stubby,
Hard anodize is usually considered to have corrosion resistance as good or better than Type II anodize of the same thickness, but sealing is a good idea if salt exposure is involved. The corrosion resistance does depend upon the alloy anodized, being quite good for Alclad and 6061, and relatively poor for 2024. So, what alloy did you use? Second, how did you get a Rockwell C reading of the anodize? Maybe file testing?
Thanks much, TVP.
The first link shows that Metalast found poorer corrosion resistance to salt spray for a sealed, 2 mil hard anodize coating than previously for sealed 0.7 mil Type II anodize. This lesser corrosion resistance was ascribed to thermal 'crazing' of the hard anodize, which I have described above.
The second link is an excellent article on the sealing (of the pores) of anodize coatings, with schematic diagrams of the hyration process. Metalast found that lowering the sealing temperaure reduces hard anodize crazing, as I had mentioned in my 1st Aug. 7 post. What was new to me was the big difference in performance between the dichromate and nickel acetate processes depending upon corrosive environment: Dichromate better in salt, Ni acetate in acids.
