What is anodizing and how does it protect aluminium parts?

The short answer

Why anodizing is a conversion, not a coating

Anodizing is often described as a coating, and that description hides the part that matters. A coating is applied on top of a surface and can be peeled, chipped or flaked away. Anodizing does something different: it passes current through the aluminium while the part sits as the anode in an acid electrolyte, and the oxygen released at the surface combines with the aluminium that is already there. The result is aluminium oxide, the same material as the natural passive film but grown to a controlled thickness. Because the layer is chemically continuous with the metal beneath it, there is no interface to delaminate.

That continuity is why anodized aluminium survives handling, cleaning and light abrasion that would lift a paint film. It also explains the one surprise buyers meet first: the oxide does not sit only on the surface. Roughly half of the layer grows inward and half grows outward from the original surface, so the part gets slightly larger and the underlying metal is consumed at the same time. A 20 micron Type II layer therefore adds about 10 microns per surface, and every tolerance, thread and bore the finish touches has to be planned around that growth.

What the oxide layer does to the part

Four properties change the moment the layer forms, and they explain almost every reason a designer specifies anodizing. Corrosion resistance comes from the dense oxide itself: a sealed Type II layer passes 336 hours of neutral salt spray as a baseline, and well-run lines reach 500 to 1,000 hours. Surface hardness rises from roughly 20 to 30 HV on bare aluminium to between 200 and 500 HV depending on the type, which turns a soft, easily marked surface into one that resists scuffs and light contact. The layer is also an electrical insulator, with a dielectric breakdown of roughly 400 volts per 25 microns on Type II and 800 to 1,000 volts per 25 microns on Type III, which is useful on a housing and a problem on a grounding contact. Finally, the porous structure accepts dye before it is sealed, so the part can be finished in a durable colour that is inside the oxide rather than on top of it.

How the anodizing line runs

The commercial process is a short sequence of tanks, and each stage has a job that shows up later as a defect or a cost. Cleaning and etching remove machining oils and give a uniform surface; a poor etch produces a blotchy finish that no amount of dye will hide. Anodizing is where the layer grows, and the thickness is set by the current density multiplied by the time the part spends in the bath, summarised by the well-known 720 rule: about 720 amp-minutes per square foot grows roughly one mil, or 25.4 microns, of oxide. Dyeing is optional and only works because the fresh oxide is porous. Sealing, usually in hot deionised water or a nickel acetate bath, closes the pores and locks the colour in.

Thickness by type and application

Anodizing is not one process but a family, and the type fixes the thickness band you can expect. Chromic acid anodizing, Type I, builds only 0.5 to 2.5 microns and is used where fatigue life and tight tolerances matter more than corrosion, because it removes almost nothing from the fatigue performance of the alloy. The thin sulfuric variant, Type IIB, sits at 1 to 5 microns for precision parts that need a conductive-free finish with minimal growth. Standard Type II sulfuric anodizing is the workhorse at 5 to 25 microns, with production usually targeting 10 to 18. Type III hard anodizing starts where Type II stops, at 25 microns, and runs up to about 125, and its chilled bath and higher current are what produce the dense, hard layer.

The 50/50 growth rule and why it scraps parts

The single most common production failure is a dimension that was correct before anodizing and wrong after it. Because the layer grows about half inward and half outward, a coated bore shrinks and a coated shaft grows, and a tight fit can be lost by the same micron figure that gives the part its protection. The practical answer is to state on the drawing whether the tolerance applies before or after finishing, and to leave a stock allowance on every face the oxide will touch. Threads deserve special care: an anodized thread changes pitch diameter, so it is normal to mask threads, or to specify that the thread be cut after finishing, or to allow a slightly larger pre-finish thread. Sealing faces, bearing seats, press-fit bores and any electrical contact should be listed as masked areas, and masking is hand labour, so a design that keeps the masked zones simple and few is cheaper to finish.

Where anodizing is the wrong choice

Anodizing suits aluminium and titanium and almost nothing else, so the first limit is the substrate: it cannot be used on steel or stainless, and on high-silicon or copper-rich casting alloys the oxide forms unevenly and often looks patchy. The second limit is wear. Type II anodizing is a corrosion and cosmetic finish, not an armour, and a surface that slides against another part will polish through a thin Type II layer in service. The third limit is colour on Type III: hard anodizing is inherently dark grey to black and does not dye to bright shades, so a colour-matched requirement belongs to Type II. There is a fatigue trade-off as well, because the thicker, more brittle hard layer can reduce fatigue life on parts under cyclic loading, which is why fatigue-critical work often specifies a thin chromic layer instead. Every figure here is a planning range, and the layer a shop actually achieves depends on the alloy, bath chemistry, part geometry and racking, so the reliable number is a thickness measurement and a salt-spray result on your own part.

See surface finishing for the finishes we run, CNC machining for the parts that most often go into an anodizing bath, and die casting for cast aluminium housings that anodize less evenly than wrought alloys.