Optimizing Color Development

When you develop a new color - or finish - there are always lots of considerations to be done. Let me share some of the learnings I have made along the way when developing new colors on anodized aluminum. The learnings are both around the evaluation of the color on the anodized aluminum part and the development of the color including the anodizing process.

Color appearance

The designer has an idea on color and might share a color target with you as a paint chip, a piece of leather or as an online reference. Then you set out to reproduce this color on an anodized aluminium surface - ideally with the product’s final shape, but this isn’t always possible either.

It is all about appearance! The anodized aluminum will always differ from the color targets mentioned here, where the coloring is solid and reflectivity comes from the top surface of the color target. The reflectivity of aluminum is a function of reflectivity of the base aluminum and the aluminum oxide. The color of the aluminum is also a function of the base aluminum alloy and the created aluminum oxide and the dyestuff added to create the intended color.

Left image shows appearance of color target. Light is reflected from top of surface. Right image shows appearance of aluminum sample. Light is reflected from both top of aluminum oxide and aluminum-aluminum oxide interface.

The team developing the color will need to rely on visual evaluation under agreed light conditions and ideally in combination with a color measurement to support parties in the work.

Why do you need to standardize light conditions? Many people use the window sill for evaluations as this is where they get most light. But as the time of day changes and the sun’s position on the sky changes the color will appear differently. Below pictures compare color of an anodized aluminium part under different light conditions. Industry standard is to evaluate color under the same light conditions every time, this could be in a light box using CIE standard illumination “D65”, also called “Daylight”. “D65” corresponds roughly to average midday light in Northern Europe and the light is composed of a mixture of direct sunlight and light diffused by a clear sky.

Left photo is taken under “natural light conditions” and right photo under “sunset conditions”. The right side photo shows a more reddish and slightly darker color, both evaluating the wooden background and the aluminum sample.

With standardized light conditions you will have more stable evaluation from sample 1 to 20 as you continue optimizing your process towards meeting the color target repeatably. Furthermore the feedback from the designer will also correspond to the visual appearance in your light box when conditions are standardized.

If available, adding a color measurement will help you speed up the process. This can help you understand if the difference is related to saturation, hue or chroma - or all of them? Whether you use hue and chroma or L*, a*, b* depends on preferences and maybe experience.

The L*a*b* Color Space system is widely used in different fields and is a mathematical presentation of color in a three dimensional space, very engineer friendly. L* represents lightness, a* and b* are chromaticity coordinates. +a* is the red direction, -a* is the green direction. +b* is the yellow direction, -b* is the blue direction. Color saturation increases as the absolute value of a* and b* increases.

The L*C*h Color Space system correlates well with how we perceive color and is often preferred by designers. L* represents lightness, C* chroma and h the hue angle. The value of chroma is the distance from center. Hue is the angle measured from +a* axis and is expressed in degrees.

Color development

Now we have an understanding of how to evaluate color and its time to start developing the anodizing process and the coloring step.

The anodizing process used for natural aluminium anodizing is a good starting point here. Any variation in film thickness and color of AAO, anodized aluminum oxide, will influence final color, therefore good electrical contact is essential and stable anodizing conditions. Having that under control we are ready to develop a custom color.

A custom color often consists of two or three organic dyestuffs that are mixed in an aqueous solution. Variation of ratio, total concentration of dyestuff will influence the final color. So will dye time, pH and potential additives.

In my experience it’s best to start out with creating approximately 10-30 samples with varying dye ratio and dye time using the two - three dyestuffs that I expect will work best with a given color target, and then compare samples to the color target under standardized conditions. Will this dye package be able to hit the color or shall one or two dyestuffs be replaced?

Which dyestuff to use? Selecting dyestuff is always tricky as there are multiple ways to reach a color. Here experience with the dyestuffs come in very handy, as some dyestuffs should be handled at specific pH values and others are more flexible; some dyestuffs mix well and others might have a fast dye uptake and therefor tend to adsorb deeper in the aluminum oxide.

When I have the dye package that I believe in for this project it is time to start tweaking - and this is where color measurements come in handy. Target is to 1) hit the color and 2) repeat the dye process and 3) optimize dye process to be a stable and repeatable process.

Ideally you want a dye process where a few extra seconds due to crane delay isn’t causing color reject and where the dye bath will last for a long time instead of color variation happening from flight bar to flight bar. These are all factors that need to be evaluated during color development.

Bath stability can be measured over three to five days, where you anodize and dye parts on eg three flight bars a day. Most often you have spare scrap material that can help fill up the flight bar reducing the number of “good parts” needed. Color measurements and overall monitoring of your process will help you evaluate the process stability and thereby gain trust in your new process.

Other considerations

It isn’t enough to develop a color that matches the color target in the light box, you also need to evaluate color inconstancy and maybe also metamerism if more materials are to match. From a quality point of view there are a lot of important color reliability metrics like heat and light stability as well as corrosion resistance and much more. More about all these parameters in another post!