Topics covered in this article: Substrate Problems; Metallurgical
Properties; Handling Steps; Pretreatment Line; Anodizing Concerns;
Sealing Process; Constant Current; Rinsing Stages; and Troubleshooting.
Everything from the condition of the aluminum to the application
of the seal can affect the quality of sulfuric acid anodizing.
Anodizing is one of many steps in a process that begins with the
alloying of pure metal in a cast house and doesnt end until the
product has been inspected and approved. In between exists a multitude
of opportunities for the emergence of problems that can affect
finish durability and appearance.
Such problems should be identified during the inspection of finished
loads when parts are unracked (if not before). Failure to pinpoint
the culprit could lead to reject material in the hands of the
customer. Proper evaluation at inspection is crucial to determine
if the problem is unique to the individual workload or if something
has developed in the system that will affect other loads as well.
That evaluation is dependent on a proper understanding of anodizing.
Anodizing is the generation of an aluminum oxide film under controlled
electrochemical conditions to uniformly cover and protect the
surface of bare aluminum metal. As oxidation takes place, the
quality of the film is influenced by the physical and chemical
characteristics of the aluminum basis metal from which it is formed.
The anodized film is transparent and does not have the capability
to mask over metal surface defects, even when colored. Therefore,
the final finish appearance depends largely on the condition of
the aluminum workpiece.
Most anodizing-related problems arise from deficiencies in the
quality of the basis metal or the anodic film, or from a combination
of the two. Hence, one of the most important skills in troubleshooting
is being able to visualize a problem as a result of the influence
these factors have on the chemical and physical uniformity of
the finish. To do that effectively, the troubleshooter must be
knowledgeable about metal production, handling procedures, racking
techniques, cleaning and other pretreatment operations, the anodizing
process itself, sealing, rinsing, and testing. Some of the major
considerations are addressed here.
Substrate Problems
The first source of trouble may be in the metal itself. One of
the most common occurrences is the inclusion of dross or bath
particles in the ingot or billet. These particles can spread through
the surface of the metal during rolling or extrusion. In the case
of rolled products, the oxides may not have originated within
the metal, but from oxides rolled into the surface. These "zones"
of pre-existing oxides do not take part in the anodizing process
and therefore become open areas in the anodized finish. If a post
anodizing color step is involved, quite often these zones will
remain uncolored and appear as white or gray streaks.
The most common alloy used for anodizing is 6063, and most anodizers
are familiar with its appearance after anodizing. This is not
something that just happened, but is the result of much experimentation
to produce a finish that is attractive, whether natural or colored.
The alloying constituents (elemental metals) are finely dispersed
throughout the aluminum and, as with inclusions, they do not oxidize
during the anodizing step. Instead, as the oxidation of the aluminum
takes place, the finely dispersed particles are lifted and become
a part of the transparent film or dissolve, leaving voids throughout
(Fig. 1). Either can lead to a dull or grayish appearance, a condition
that becomes more pronounced with the thickness of the film or
the concentration of the alloying constituents. During the pretreatment
step of etching, alloys that have a high concentration of alloying
constituents can be identified by the heavy gray smut that forms.
Film uniformity is not affected as adversely by alloying constituents
as by non-metallic inclusions. However, abrasion and corrosion
resistance of the anodized film may suffer from alloying constituents
because the particles and voids can reduce the integrity and strength
of the cellular structure of the aluminum oxide film. This condition
is best exemplified by the high-copper-containing alloys such
as 2024. Because of the dissolution of the copper particles during
anodizing, a dull gray or gold finish that has low abrasion resistance
is the result.
Practically all aluminum alloys can be anodized in some form or
another, but because of the effects of the alloying elements on
appearance and film quality, consideration has to be given to
the end use of the coating. Alloys to be used for extrusion and
anodizing purposes are generally formulated from primary metal
and controlled during casting and subsequent operations to produce
the needed metal quality. However, it is common to encounter secondary
metal that will result in a more mat or gray appearance due to
high concentrations of alloying ingredients, especially silicon
and iron. Therefore, if the final anodized appearance is not the
one expected, this could be an indication of improper alloy.
Metallurgical Properties
The function of an alloy is dependent on its metallurgical properties.
The most important is that of structural strength, which for architectural
applications must be coupled with a quality anodized finish. Both
are dependent on grain structure.
If the alloying ingredients are not finely and uniformly distributed
or chemically combined, the resulting surface pattern after etching
will be in one of two conditions. First, it could be coarse and
grainy because of the selective dissolution of aluminum from around
the large, non-dispersed particles. Or it could exhibit a frost-like
pattern as a result of slow crystalline growth of the aluminum
and its constituents. This can occur as a result of improper homogenization
of billet, poor quenching during extruding, or improper heat treatment
of extruded parts.
A common problem is the hot/cold spot, which may occur in a random
or regular pattern along an extrusion. This is usually due to
contact between the part and the run out table during extruding,
thus interfering with the cooling process. Cooling is generally
slower at that spot, and as a result, the grain structure is larger
and the particles less well dispersed. After etching, this shows
up with a darker coating of smut, and after anodizing the spot
will become grayer and more mat. This is due to the large localized
particulate concentrations and cannot be corrected or hidden by
the anodizing process.
Now a word about heat treatment. The designation that relates
the desired metallurgical qualities to the degree of post-extrusion
heat treatment required is called temper. Generally, the higher
the temper the fine will be the grain structure and the better
the anodizing response, producing a brighter, smoother, and less
gray finish (Fig. 2). Different structural requirements dictate
a wide variety of tempers, so it is important to be aware of what
is being anodized and what kind of response is expected.
Handling Steps
From the time parts are extruded or sheet is rolled, several handling
steps are employed. The material is cut to length, stacked on
carts, heat treated, stored for cool down, dismounted, possibly
stored again until ready for anodizing, and finally racked. In
each step, the mill-finished, unprotected metal is handled with
hands (clean or greasy, with or without gloves), stacked in contact
with wood or cardboard (wet or dry), stored outside exposed to
the elements or inside close to the anodizing line with its inherent
fumes, and eventually racked (again with gloved or bare hands).
In one case, an individual had been working on a greasy crane
motor and came over to "help" the rackers for a few
minutes. Of course, every piece that he handled ended up after
anodizing with his gloved finger and/or palm prints. The grease,
which was not only heavy but impregnated with fine metallic particles,
left a residue and surface blemishes that the cleaning solution
could not remove, so selective rather than uniform etching took
place. Generally, any type of hand prints will show up as etch
patterns (distinct or unrecognizable) after anodizing. Again,
the anode coating cannot hide such blemishes.
Another problem is in the storage of mill-finished material. Over time, a normal, uniform oxide film will build up, and this usually does not present a problem. However, in contact with moisture, a humid atmosphere or chemical fumes, the surface may become more oxidized in certain spots. Cleaning/degreasing alone cannot deal with this type of oxidation, and the result after etching is pitting or a galvanized appearance. Once this occurs, there is little that can be done to salvage the metal. Passing the load to the desmutting solution (using it as a deoxidizer) prior to degreasing sometimes can reduce the incidence of the problem; however, it is not a cure-all for poorly handled or stored material.
Racking Procedures
It is important to remember that racking is used to make electrical
contact, not just to suspend parts for passage through various
solutions. Good racking practices cannot be overemphasized. The
contact, once made, has to endure severe attempts to break it
loose, not the least of which occurs during the anodizing step
with the tremendous force exerted by the growing aluminum oxide
film, which can then act as an electrical insulator.
The results of poor contact are readily measurable using coating
thickness testers or by observing the contact point on the end
of the part. Burned or gray-looking areas may indicate overheating
(Fig. 3). Thickness non-uniformity on a load of parts is an indicator
of random rack contact. (In electrolytic coloring, this is readily
visible as color variation, or "light ends," on workpieces.)
Another source of problems is parts spacing. Anodizing followed
by coloring must conform to certain spacing rules if uniform coating
thickness and color are to be achieved. Either single- or double-spine
designs may be employed, optionally with a center cathode in the
anodizing and electro-coloring tanks. For clear anodizing, racking
is relatively dense with multiple parts across. But this can lead
to non-uniform coating thickness because the inside parts receive
less energy for film growth with distance from the cathodes and
because they are shielded by outside pieces. Thus, parts in the
center will have a thinner coating than those at the outside.
For most applications of commercial and residential clear-anodized
products, the thickness may be adequate to protect the aluminum,
though it may not actually meet the required specification.
Returning to electrical contact, a significant problem that can
occur in a very random fashion, depending on the operators attention
to detail, concerns inadequate rack stripping between loads, especially
if Class 1 and Class 2 architectural coatings are being produced.
If any film is left on the spline, it will act as an insulator
for the next load of parts, preventing contact altogether or causing
a high resistance contact point that can result in the same burning
or discoloration problems discussed above.
Pretreatment Line
Deoxidizing, degreasing, etching, and desmutting are among the
chief pretreatment steps that have an impact on finish quality
and appearance (Fig. 4). We will consider each one in this section.
Deoxidizing: As pointed out previously, if handling or
storage of the aluminum workpiece results in hand prints or other
problematic conditions (e.g., corrosion), it is a good idea to
add a deoxidizer step at the beginning of the pretreatment sequence.
Cleaning/Degreasing: The importance of this step in the
chain of events is often overlooked. If the temperature is too
cold, the concentration of the soap additives too weak, or the
formulation not correct for the needs of the plant, cleaning will
be inadequate. The result is selective etching, leaving mottled
patterns on the metal surface.
If allowed to become too old and contaminated, the soap solution
can cause a pitting action if the load is left in for more than
5 min. When working properly, it should be possible to leave loads
in the solution for up to 30 min without experiencing problems.
The degreasing agents in cleaners are formulated to be non-etching,
but are either slightly caustic (common for architectural anodizing)
or slightly acidic (usually for special applications such as automotive
anodizing where buffing compounds are difficult to remove with
a single cleaner).
Etching: The basic purpose of etching is to dissolve the
surface of the metal to: remove ground-in impurities, diminish
or eliminate extrusion die lines and mild scratches, impart a
smooth, uniform appearance, and change the natural brightness
to a mat condition. The important control criteria are: solution
temperature; concentration of caustic, chelator, and grain refiner;
degree of air agitation; and rate of load transfer.
Variations in chemical concentrations and temperature can result
in insufficient etching (apparent by brightness and appearance
of die lines) to over-etching, which results in a very grainy
texture and too dull a finish. Air agitation and transfer time
can cause streaking from the abrasive action of air bubbles traveling
upward or from rundown and continued etching during transfer.
It is important to use well-distributed, low-pressure air, to
keep the temperature moderate to avoid dry-on during transfer,
and to transfer quickly.
If dry-on occurs, salt deposits resultóspecifically, aluminum
hydroxide, which is not removed in the rinse tank or the desmutting
solution. These deposits not only look bad on the finished product
but represent sites where anodizing will not occur (similar to
the effect of metallic inclusions) and can become corrosion zones
on exterior exposure. Many etch formulations, referred to as "never
dump" solutions, are designed to function with a high rate
of drag-out to control the buildup of aluminum in the solution
and allow for rapid transfer.
Desmutting: Most of the principal alloying constituents (silicon, iron, copper, and manganese) are not soluble in or reactive with the etch solution. As a result, particles of these elements are left on the surface in a form called smut. It is necessary to remove these particles because they will contaminate the anodizing tank.
The solution formulated to accomplish this is based on mineral
acid, which readily dissolves all but the silicon.
This would seem to be straight forward and, in fact, is probably
the least problematic step in the process. However, if thorough
rinsing is not carried out before the workpiece enters the desmutting
solution, the reaction between the acid and the caustic can result
in the formation of aluminum hydroxide scum, which entraps smut
particles. At the least, this increases the time required to dissolve
the smut in the desmutting solution. At worst, it leaves a film
that may selectively dissolve in the anodizing solution, leaving
insoluble particles such as silicon on the surface. These particles
will not prevent anodizing as theyre not in the metal. However,
they will neither dissolve in the anodizing solution nor be incorporated
in the cellular structure of the anode coating, so the particles
end up as a film or residue on the finish.
With the standard anodizing-quality alloys, the silicon concentration
is controlled or coupled with magnesium such that the silicon
is removed with the other elements in the desmutting solution.
However, primary alloys like 6061 or secondary 6063 alloys, which
have a high silicon concentration and which have been heavily
etched, can present problems. Clear finishes may not show this
residue or may exhibit a slightly gray appearance, but this can
be a serious problem for colored finishes, and removal is difficult.
Anodizing Concerns
In this section, we will look at anodizing related concerns involving
contact, coating thickness, and current distribution.
Contact: There are three important contact systems, the
first of which was discussed briefly under "racking."
There must be good contact between the parts and the splines,
between the splines and the flight beam, and between the flight
beam and the power saddles located on the tank. Well refer to
this as the workload. The workload has to receive power from the
rectifier via bussing, which may have several connections between
it and the power saddle. This constitutes the supply. The workload
then has to pass the power through the electrolyte to the cathodes,
which are connected to a system of bussing that (also having several
connections) returns the power back to the rectifier, thus completing
the circuit. This well call the return.
Breakdowns at any of these connection points will cut off the
circuit and no power will flow. The obvious result is no anodizing,
resulting in the need for immediate action to get the power flowing
again so that production may resume. (There is, however, a far
more subtle problem, the insulation effect, which can lead to
serious and varied problems in coating uniformity. Well talk
about that effect in our discussion of current distribution.)
Film Thickness: One of the most simple and useful tools
in anodizing is the non-destructive thickness tester that can
be used right on the production line. With it, the operator can
determine how thick the coating is and how well it is distributed
over the workload. The most common problems in this category involve
thickness variations from one workload to another, from top to
bottom, from end to end, and from inside to outside. Some of these,
as suggested above, are the result of racking techniques, but
they can also occu r d ue to contact problems.
Current Distribution: Coating variations within a workload
are generally a function of how the current is distributed once
the parts enter the tank. The current flows off the workload through
the electrolyte to the cathodes, which have to receive it in the
proper manner. It must be remembered that the current doesnt
care where it goes; it simply follows the path of least resistance.
An example of this is when thicker coatings form on parts at the
bottom than on parts at the top of a workload. The problem source
may be a lead-lined anodizing tank or a large cathode surface
area at or near the tank bottom. In these cases, more current
is drawn through parts near the bottom of the load, generating
a thicker anodic film. It is advisable to mask off or cover the
bottom, ends, and portions of the side walls of a metal-lined
tank to obtain better current distribution.
In designing new systems, the tank lining is usually made of a
special inert rubber, plastic or fiberglass material with cathodes
suspended along the sides. Experience has shown that the cathode
area, properly distributed along the side walls, can be one-third
to half the surface area of the largest routine workload and still
be sufficient for most anodizing needs, especially if made of
aluminum. In cases where film uniformity is critical for inside
surfaces, a center cathode can be added.
Another problem, not as common but far more serious when it does occur, is end-to-end or totally random variation in film thickness. The cause can be the contact between the cathodes and the bussing along the top of the tank. Acid solution aspirated by air agitation from below Of dripping off loads passing overhead can penetrate between buss surfaces and form oxides that ultimately insulate the connection. Current flow is then reduced or cut off entirely to that cathode
section. The current, seeking the path of least resistance, will
flow from workload zones closest to the cathode areas that are
functioning better. Quite often, the insulation effect develops
at the end of the tank where most of the drag-out drainage occurs
and on the side over which the workload passes for rinsing.
This problem was encountered in a plant producing Class 1 films,
which are supposed to be 0.7 mil thick. Workloads exhibited variations
of 0.3 mil at the drainage end to 1.2 mil at the opposite end.
A check of the cathode plates demonstrated that less than half
were functioning properly. The problem arose because the cathode/buss
connections were unprotected from acid drainage of workloads.
For positive identification of the insulation effect in any shop,
the operator places a workload in the anodizing tank and allows
it to stabilize for about 5 min. At that point, the air agitation
is shut off and the evolution of gases from the cathodes observed.
Gassing should be uniform from all surfaces opposite the workload
and should exhibit equal flowing action from both sides to the
center of the tank. If this is not the case, the cathodes may
be partially or totally insulated.
In any event, steps should be taken to protect and regularly maintain
the cathode/buss connections and all others in the anodizing circuit
to prevent the insulation effect.
Constant Current
Many anodizers operate with constant voltage control, which is
used exactly the same way on every workload regardless of size
or shape. The assumption is that the result will be the same each
time.
For some applications, this may be adequate, but it certainly
will not give identical results. The problem is that the total
system (supply, workload, and return) is erroneously thought of
as an integral unit that has one large voltage drop. Actually,
the system operates as a series of small voltage drops that begin
at the rectifier and accumulate as the current passes through
every connection point. And, as we have discussed, the connections
that are in a fixed position can be attacked by solutions splashed
during transfer or from fumes and can suffer from the insulation
effect. Over time, resistance may build and the current available
for anodizing can be slowly reduced. So if constant voltage is
used, the anodizing time will likely have to be increased or the
coating thickness will gradually diminish as current flow is reduced.
In the same system, there are many connections (as in racking)
that change with every workload. If those connections do not have
the same voltage characteristics each time, there will be current
variation from workload to workload and consequent differences
in film thickness.
Because electrochemical oxidation in anodizing occurs as a function
of current and not voltage, maintaining the same current density
to each workload and freeing the voltage to respond to the resistance
variations results in a more uniform film from one load of parts
to the next. Control by constant current density is based on a
predetermined current per unit of surface area (e.g., ft2) and
requires a calculation of total workload area, including racks.
This is a somewhat tedious procedure; however, if there is a problem
of load-to-load non-uniformity, it may be advisable to consider
changing to constant current control, even if making workload
area calculations may not be the easiest way to go. Though tedious
at the outset, once put into practice as a routine part of the
job, it can become a very effective tool.
Sealing Process
The sealing process follows anodizing and requires contact with
hot water, hydrolyzing the aluminum oxide to form a hydrate that
increases the volume of the cell structure. It effectively closes
pores and prevents film penetration by staining or corrosive materials.
A destructive test for ascertaining seal quality (and coating
weight if necessary) utilizes a solution of chromic and phosphoric
acids at selected temperatures. Because it dissolves the film
but not the aluminum metal, the test can reveal if a problem originates
in the anodic coating or in the basis metal. The test can be set
up easily in the lab and used as an important measure of quality
control.
Factors that affect sealing include water quality, pH, temperature,
and time. Negligence in their control will impair seal quality
and generate some appearance problems resulting from smut, residue,
and powder.
Smut: If the temperature and/or pH of the seal are too
low (less than 200 F and 5.5, respectively), the hydrate that
forms is not the desired one. Instead of a hard, smooth, clear
finish, the result is a soft, rough, chalky or hazy film. If the
surface is then wiped, this causes the film to transform to a
white powder, which, if removed, considerably reduces the thickness
of the sealed coating.
This is true seal smut and is a direct result of sealing problems.
In recent years, with an understanding of this problem, conditions
have been controlled or low-temperature nickel based seals have
been substituted, almost eliminating it. As a result, "smut"
has come to be used to describe two other surface problems that
do not necessarily relate to the quality of the seal.
Residue: Deposits of alloying constituents such as silicon
can end up as a residue on the film after sealing. This is not
smut and is not generated by poor seal conditions, but may be
confused as such. The difference is that, although uniformly deposited
on all part surfaces, it can be removed by wiping. The surface
underneath, if sealed properly, will be hard and smooth .
Another source of residue comes from dissolved solids, salts or
minerals that remain on the surface after drying. These can accumulate
in the seal tank through drag-in contamination or prolonged use
of the seal water. This residue is also removable by wiping. The
surface underneath may be properly sealed or, because of the mineral
content, only partially sealed, but will still be hard, clean,
and smooth after wiping. This condition indicates the need for
a change of the seal bath.
Powders: The other type of deposit that can arise is loose
powder, which wipes off with very little effort. It results from
undissolved solids that come from the neutralization of soluble
metal salts to form insoluble metal hydroxides in the rinse tanks.
Powders can also be a problem with the precipitation of nickel
hydroxide in low temperature seal baths if conditions are not
properly controlled. Filtration of the seal bath can help control
this problem .
Rinsing Stages
It is unfortunate that rinses are often the least considered operation on the anodizing line, with the possible exception of their water quality as they leave the plant. Too little water flow leads to many residue and powder
problems as well as contamination of the various working solutions.
Excessive water flow is a waste and may result in unnecessarily
high water and sewer costs.
Most of these situations can be mitigated or ameliorated by using
counterflow rinses and conductivity probes in the rinse tanks
to control the chemical concentrations as a function of pH. For
example, many of the metal hydroxides remain water soluble at
a pH of 2.5 to 3.0, which is "clean" enough for a first
rinse after most operations. This can be followed by a second,
cleaner rinse maintained at a pH of 5.5 to 6.5 and which counterflows
to the first rinse. The water flow through both tanks is then
controlled by placing a conductivity probe in the first rinse
to actuate a valve that adds water to the second.
Besides the residue-type problems, over-contaminated rinse tanks
can lead to surface imperfections on the basis metal (Fig. 5).
For instance, pitting or galvanic corrosion may arise because
of the chemical interaction between the contaminating salts and
the bare, unprotected aluminum. Also, because rinse tanks are
considered inert, workloads are sometimes left hanging on the
crane without having been electrically isolated or left touching
metallic parts of the tank. This may result in a galvanic action
that can destroy the surface condition of the basis metal in a
very few minutes.
Troubleshooting
Three of the major steps in the troubleshooting procedure are
objective identification, source determination, and patient investigation.
Lets discuss each one briefly.
1. Objective Identification: How should a problem be defined,
isolated, and eliminated? First and most important, the anodized
(and colored) workload must be examined and tested carefully and
objectively so that the problem is accurately defined. This has
to be done without looking to place blame on any particular individual
or department. There have been many situations in which problems
were made more difficult to resolve because of someones unwillingness
to accept the responsibility for correcting the problem. This
can seriously interfere with making a good assessment. So be objective,
open minded, and not defensive!
2. Source Determination: Once determined, the location or source
must be pinpointed. This can be achieved by taking note of the
steps that can create the problem. For instance, pitting of the
aluminum under the final finish can occur before or during anodizing.
The type of pitting (e.g., star shapes, sharp pits or galvanic
frost) can give a clue, but not the answer.
The first step in solving the problem is to run another workload.
It may seem expensive to sacrifice a load of metal in this manner,
but if anything is changed at this point, it will be difficult
to determine if the condition was just a one-time occurrence or
if it will affect subsequent work. If the problem does not repeat,
it should be noted if it recurs in a random fashion. If it does
repeat, then it is necessary to start changing variables and running
more loads.
Do not change more than one variable at a time! This will only
lead to confusion about which variable caused or corrected the
problem.
With respect to a pitting problem, good, fresh metal should be
processed. If pitting is still evident, then an examination of
pitting sources on the anodizing line should be investigated.
A check of rinses and process solutions, as well as electrical
contacts between the workload and ground, should be checked. If
the workload emerges without pits, run some more loads. If pits
again appear on the workpieces, examine the storage practices
and age of the aluminum and consider using the deoxidizing step
before further processing.
3. Patient Investigation: An important factor in resolving anodizing
problems is patience. When under pressure to maintain production,
a problem is often dealt with hastily and is rarely properly resolved.
Troubleshooting requires time and a stepwise investigation. Once
the source and location of the problem have been determined, steps
should be taken to eliminate it. Unless completely corrected,
it cannot be expected that the anodizing line will run smoothly
and consistently.