Composition,

Testing,
Dip Galvanizing
Flux

and Control

of Hot

by Dr. Thomas H. Cook

HC 52 Box 120-B, Hot Springs,

South Dakota

57747;

t is well known that pretreatment

is key to the
success of virtually every metal finishing operation. In hot dip galvanizing the fluxing operation
is a particularly critical step that directly affects the
quality and uniformity of the adhesion of the zinc to
the steel as well as the amount of zinc used. This
article summarizes a wide range of demonstrated
best practices and innovations
in managing
flux
chemistry and the fluxing process.
EFFECTS

OF VARYING

THE GUADBAFLUX

FORMULATION

An effective flux solution must be periodically tested and adjusted to attain required galvanizing
results. Among the most important
measures of
results is the percent gross zinc usage (%GZU). It is
calculated:
%GZU = (zinc used/steel galvanized)

x 100%

For steel averaging 3/16 in. thick to l/4 in. thick

(about 5 mm to 6 mm) a %GZU of about 5% can be
attained using an effective flux solution. For steel
which is on average thicker, an even lower %GZU
can be attained because of less surface area. For
%GZU no credits for ash or dross are allowed.
EFFECTS

OF VARYING

GUADBAFLUX

FORMULATION

Over the last 30 years the most effective flux solution for dry kettle galvanizing has been found to be
quadraflux with the composition shown in Table I.
For a 12.5 degree Baume flux solution 2 lb/US gal
(240 g/L) of quadraflux is required. Since quadraflux
is not normally available in the marketplace,
zinc
chloride, or double salt, or triple salt can be mixed

22

E-mail, ,[email protected]

with ammonium chloride. It is very important that

for each pound of zinc chloride there are 1.60 lb of
ammonium chloride. Because zinc chloride picks up
moisture, any partially used bags or barrels must be
properly resealed.
A Baume lower than about 11 causes black spots
(bare spots) on the steel. A Baume lower than about
12 causes the flux to release poorly from the zinc
surface resulting in excess zinc clean-up.A Baume
higher than about 14 gives excessive zinc spatter,
excessive ash, more smoke, and thicker zinc coatings
on the steel. A Baume of 21 is twice the concentration of a 12.5 degree Baume flux and gives terrible
zinc spatter at the kettle because there is a double
layer of flux on the steel. The outer layer is thin and
dry, whereas the inner thick layer is wet and mushy
This wet, mushy condition requires very slow entry
of the steel into the molten zinc. This produces a
very thick zinc coatings on the lower portion of the
steel products.
AMMONIUM

CHLORIDE

NUMBER

The Ammonium
Chloride
Number
(ACN) was
coined by the author in 1980. It is a mathematical
relationship
that equals the ammonium
chloride
concentration
divided by the concentration
of all
other components in the flux solution. In this case
we calculate it:
ACN = %Ammonium

Chloride/%Zinc

Chloride

It should be noted that because ACN is a ratio having ammonium chloride in the numerator and zinc
chloride in the denominator, units cancel out making
ACN unitless, independent of concentration, and the

Metal

Finishing

same number regardless of units. Thus it has the

same value whether one is working in English or
metric measures. The names, formulae, ACNs, and
common uses for galvanizing
fluxes are shown in
Table II. Fluxes with ACNs lower than about 1 do not
dry well and collect moisture from the air. These wet
fluxes cause moisture explosions as the steel enters
the molten zinc. It is known that the lower the ACN,
the more thermally stable the flux. A flux used in a
high-temperature
preheater or drying oven therefore
must have a low ACN. Fluxes with ACNs higher than
about 2 are thermally unstable and give loud, chemical explosions when immersed into the molten zinc. In
either case of an ACN lower than 1 or higher than 2,
the required slow entry into the molten zinc, causes
lower production and thicker zinc coatings on the
lower portions of the steel product.
The ACN of a flux solution usually does not remain
constant. It is common practice to reflux galvanized
rejects (with bare spots) rather than strip the zinc
(with acid). During the refluxing of galvanized rejects
some of the zinc dissolves in the flux solution forming
zinc chloride. This refluxing process lowers the ACN,
raises the pH, and raises the Baume. In some plants
zinc bars, dry ash, or even dross are added to the flux
solution to neutralize excess carried-over acid from
pickling. Also racking fixtures may be galvanized from
a previous cycle and enter the flux tank. Spin baskets
for centrifuge
work may be cleaned (partially
stripped of zinc) in the flux solution. All of these latter
procedures also increase zinc chloride that lowers
ACN. In moderation, all of these practices are fine as
long as ACN and pH are readjusted as needed and the
Baume does not get too high. For some galvanizers
this system works so well that they have not added
commercial flux (containing zinc chloride) for several
years. They just add ammonium chloride as needed
and strip rejects, racks, or baskets in order to get the
required zinc chloride.

24

Thus a galvanizer initially using triple salt with

an ACN of 1.18 (Table I) who introduces zinc into his
flux will, in time, have a flux solution with a lower
and lower ACN. In 1975, a galvanizer using triple
salt had his flux tested and it was found to contain
0.55 lb/gal ammonium
chloride and 0.72 lb/gal of
zinc chloride. Thus the ACN was 0.76 (0.55/0.72).
After adding 1,500 pounds of ammonium chloride to
this 5,000-gal flux tank (0.30 lb/gal) an ACN of 1.18
was re-established
[(0.55+0.30)/0.721. After adding
the ammonium
chloride, the rejects (steel going
through the zinc kettle twice) dropped from 3.1% to
0.2%. Even more importantly,
the average zinc
thickness on the steel dropped from 5.7 mils (145
microns) to 2.4 mils (61 microns). Zinc icicles that
were like pencils hanging from the bottom end of the
product became small, thin, and flat like silver gum
wrappers. Zinc clean-up labor was reduced by half
The zinc coating was also remarkably brighter.
In 1977 in Pakistan (where zinc chloride is expensive) ammonium chloride alone was used for a flux
solution. As a result of stripping zinc from rejected
parts, enough zinc chloride accumulated in the flux
to form a flux solution having a Baume of about 5
and an ACN of 5. The g-meter steel electricity towers took 4 minutes to become submerged into the
zinc due to intense chemical decomposition
of the
thermally unstable flux on the product. Fortuitously,
adding 1 ton of zinc chloride gave a Baume of 12 and
an ACN of about 1.6. With this improved flux solution, the electricity towers could easily be lowered
into the molten zinc in 30 seconds. Production
increased, zinc usage was lowered by 20%, and the
product quality was greatly improved. Most North
American galvanizers achieving a %GZU of around
5% use an ACN between 1.4 and 1.6. Some automatic pipe galvanizers
use an ACN of about 1.9
because it causes much thinner zinc coatings, especially on the inside of the pipes.

Metal

Finishing

Titanium racksfor anodizingand

electropolishingareavailablefrom a lot
of places,for usein yourjob shop,captive
operationor the newline youreplanning.
But while manyracksarefabricated,only a
few arereally designed.Factis, a poorly
designedrack costsyou more in lost
productionandrejectsthanits worth.
At Servi-Sure,our engineersknow anodizing
becausethe designandfabricationof titanium
racks...standard
andcustom...areour
only business.
Thatswhy, whenquality matters,onename
keepsrising to the top...Servi-Sure.
()ur catalogdetailsa full line of standard
productsaswell asour customengineering
oervices.PleasecontactServi-Surefor your
I opy today.
Email: [email protected]
Circle

090 on reader

information

card

or go to www.thru.to/webconnect

REDUCING

DROSS

BY

CONTBOLLING

DISSOLVED

IRON

Dissolved iron (ferrous; Fe+2) iron is carried from the

pickling solution through the rinse into the flux solution. Iron (Fe+2) in the flux does not cause dross in a
dry kettle. In sharp contrast, iron (Fe+2) in the flux
solution does cause dross in a wet kettle. The essential first step to removing the unwanted
dissolved
iron (Fe+2) is to determine accurately its concentration. The most accurate and reliable test is a volumetric titration using a standard potassium dichromate solution. After accurately testing the pH of the
flux solution, it is adjusted up to pH 5 using 26 degree
Baume ammonium hydroxide solution. Next the calculated volume of ammonium hydroxide is added [to
supply hydroxide
ions (OH-l) for the insoluble
Fe(OH), that will be formed] and then the calculated
and diluted volume of 35% hydrogen peroxide is slowly added. Mixing is done by air bubbling. The flux
solution is kept hot during additions. It normally
takes about 2 days for the insoluble iron III hydroxide to properly settle and a suitable flocculent can be
added to speed up the settling process. The orange
sludge formed is normally put through a filter press
treated with diatomaceous earth to reduce the volume of sludge by about ten-fold. Additional peroxide
is required if the flux contains suspended oil, which
will be degraded to free carbon.
Air bubbling to oxidize soluble ferrous iron (Fe+2)
to insoluble ferric iron (Fe+3) is not possible at the
normal operating
pH of the flux. Such bubbling
wastes heat energy and stirs up unwanted
sludge
that gets on to the steel product. There is no reason
to test for total iron (Fe+2 and Fe+3) or insoluble
ferric (Fe+3) because the ferric will settle to the bottom of the flux tank as insoluble Fe(OH),. Only iron
(Fe+2) is of interest and it is the iron that is reacted
with ammonium
hydroxide and hydrogen peroxide
during purification
[to form insoluble Fe(OH),l. Inplant testing methods for iron (Fe+2> in the flux are
required so that a person will know when all the
soluble iron has been removed.
REDUCING

BLACK

SPOTS

BY CONTROLLING

SULFATES

Sulfate ion (SOam2)is carried into the flux solution in

the case of sulfuric acid pickling. In general, galvanizing a sulfate concentration
over about 1% can
cause black spots. In galvanizing
where steel
immersion times in the molten zinc are short, such
as sheet, wire, and pipes, a much lower sulfate concentration
- 0.25% or even less - causes black
spots. A volumetric titration using barium chloride
gives the best sulfate test result. Because barium
chloride is extremely
toxic, the crystal dihydrate
should be used for purification;
and proper protec26

tive clothing and air filter masks are required. As

with iron removal, airlines are used to dissolve the
barium chloride and for reaction of barium ion with
sulfate ion. Anhydrous
barium chloride (without
water) must be avoided because the dust can be
lethal if inhaled. If excess barium chloride is used,
then the flux solution will be extremely toxic. An inplant test for sulfate is required to know when the
sulfate has been removed and so that excess barium
chloride can be avoided in the flux solution.
CONTROLLING

pH

FOR

OPTIMIZED

FLUXING

The correct pH for fluxing steel prior to hot dip galvanizing is 4.2. My experience is that the traditional literature citing a pH of 5 is wrong. At a pH of 4.2,
the flux provides a final cleaning of the pickled steel,
giving a clean and bright surface ready to fuse
chemically with the molten zinc. Levels of pH below
about 3.5 attack the steel excessively causing soluble iron (Fe+2) to build internally
in the flux solution. It is true that the lower the pH of a flux, the
stronger it is. Values of pH higher than about 4.7 are
ineffective in cleaning the steel surface. Recalling
that the pH scale is logarithmic,
a pH of 4 has ten
times higher concentration
of hydrogen ions than a
pH of 5.
In my experience, pH meters do not work in flux
solution, tending
to give readings
substantially
lower than reality. Some, but not all, brands of pH
papers can be useful. Certain organic dye indicators
that have a distinct color change near pH 4.2 work
extremely well in maintaining
proper flux pH. In
testing and controlling pH, I prefer to use the following procedure: To a specific volume of settled or
filtered flux solution is added a suitable organic pH
indicating
dye. Using a plastic syringe filled with
either 1:lO hydrochloric
acid or 1:lO ammonium
hydroxide (as titrant), the flux solution is titrated to
a pH of 4.2. The calculated volume of either concentrated hydrochloric acid or concentrated ammonium
hydroxide is then added to the process flux tank. In
some plants, pH adjustment
of the flux solution is
required daily.
It is important
to keep in mind that the goal of
testing via a titration
is to determine
how much
hydrochloric
acid or ammonium
hydroxide to add
to maintain a pH of 4.2. Knowing the pH - but not
knowing what to add or how much to add - is not
useful. I am convinced that a pH meter should
never be used. I also contend that zinc slabs, zinc
ash, or dross added to the flux solution causes the
pH to slowly go too high and should also never be
used.
A fabricator and galvanizer having a flux solution
Metal

Finishing

with 1.7% sulfate and 0.1% iron (Fe+2) was having

many black spots on the product. It was this galvanizers busiest season and the 3 to 4 days for purification were just not available. Making matters even
worse, neither flocculating
agents nor a filter press
were available to speed up purification.
Because the
dissolved iron was only O.l%, it was suspected that
the pH was too high. Usually the percent iron is
from one-third to one-half the percent sulfate. A pH
titration was completed and it was determined that
25 gallons of concentrated
hydrochloric
acid were
needed to lower the pH to 4.2 for the 5,000-gal flux
tank. After the hydrochloric
acid was added all the
black spots immediately
went away. Normal flux
purification
was done a few months later.
CONTROLLING

FLUX

TEMPERATURE

The correct temperature

of the quadraflux
solution
with the composition
shown in Table I is 160F
(71C). This flux solution does not function properly
at temperatures
below 150F. Below 150F the flux
solution is not aggressive enough to clean the steel
and an insufficient
amount of flux is deposited on
the steel surface. Temperatures
above 170F deposit
excess flux on the steel surface causing a double
layer of flux (outer dry layer and inner wet, mushy
layer) giving excessive zinc spatter, thick zinc coatings, and excessive dry ash. Also, higher temperatures cost more to maintain due to greater heat losses. Typically, a temperature
of 180F requires twice
the energy as a temperature
of 160F.
In addition to maintaining
160F in the flux solution, the steel must be left in the flux solution at
least 3 to 5 minutes. This time allows final cleaning
and enables the steel to fully attain the flux temperature
of 160F. Thus, when the steel is withdrawn from the flux solution the flux dries from the
inside out because of the hot steel.
During the energy crisis 25 years ago, a galvanizing workshop participant
asked, Why do we have so
many rejects on Monday mornings?
The answer
was simple. Heat to the flux tank was turned off on
Friday nights and not turned back on until Monday
mornings.
The cold (140F) flux solution did not
function properly. Thus, the galvanizers savings of
$50 on energy was actually spoiling $30,000 of galvanized product.
All flux solutions
should not be at 160F. For
example, an automatic pipe galvanizer doing X-in.
diameter
pipes must use about
122F (50C)
because each bundle contains a few hundred pipes
and there would be corrosion touch lines within the
bundle on the outside of the pipes. Thus, the first
half of the bundle would galvanize fine, but all the
July/August

2003

second half would be rejects. To compensate

for
lower temperature,
a higher Baume of about 20 to
25 is normally used. Conversely, for spin galvanizing, a higher temperature
and lower Baume allows
the product to enter the kettle with less zinc spatter.
If the steel product is packed tightly together during
the drying of the flux, the hotter the steel, the faster
it corrodes.
CONTROLLING

WETnNG

FOR

OPllMUM

ZINC

ADHESION

About 50 surfactants
(wetting
agents) have been
tested for effectiveness in flux solutions. There are
three main reasons or objectives for adding a surfactant to a flux solution. Most importantly,
a surfactant needs to help make the liquid flux drain
away from touching or closely spaced products such
as welds and threads. In basket work this can
reduce by three-quarters
the volume of liquid flux
that will be carried into the molten zinc, greatly
reducing zinc spatter and dry ash. To achieve this
objective, the surfactant
needs to be able to reduce
surface tension from 80 dynes/cm2 to 30 dynes/cm2
at a concentration
of 0.03% or less.
Increased corrosion protection is a second reason
to use surfactants. Again, the target concentration
of
the wetting agent is 0.03% or less.
The third reason to add a wetting agent is to produce a one-inch layer of bubbles (foam) on the flux
solution surface in order to half heating costs. This
foam is established and maintained
with a small air
line near the tank heating coil. The surfactant concentration needs to be about 0.06% to generate this
foaming effect.
Many surfactant producers claim to have products
that address all three of these objectives, but in my
experience few actually deliver the promised performance. In evaluating the effectiveness of wetting
agents, it is easy to measure surface tension with a
simple drop size test. A rounded glass eyedropper
tip will produce about 20 drops/ml
with distilled
water. With an effective surfactant at 0.03% or less
this same tip will produce about 40 drops/ml.
OPTIMIZING

FLUX

COMPOSITION

Reducing
Rejects by Eliminating
NaCl
In the mid to late 1970s it was discovered that some
commercial
hot dip galvanizing
fluxes contained
anywhere from 1% to 20% of the nonvolatile
compound sodium chloride (NaCl dry ordinary
table
salt). A galvanizer using a flux with 4% sodium chloride was having 4% rejects. When he changed to a
flux free of sodium chloride his rejects went away. A
galvanizer had a flux with 14% sodium chloride and
27

had 14% rejects, but when he switched to sodium

chloride free flux his rejects went away A pipe galvanizer had a flux with 20% sodium chloride and
had 20% rejects. He even had a special crane system
to return rejects to the kettle. With sodium chloride
free flux the rejects disappeared
and the special
crane system was taken down.
A U.S. galvanizer had the caustic rinse and acid
rinse interconnected
for water treatment
reasons.
After about 8 months, the sodium ion from the caustic soda in the caustic tank built up so much in the
acid rinse and also in the flux solution that this galvanizer got thicker and thicker zinc coatings on the
steel. These increasingly thicker zinc coatings were
a result of longer and longer cook-off times in the
kettle
due to the nonvolatile
sodium chloride.
Another galvanizer
has only one rinse tank used
both to rinse after caustic as well as after acid. In a
few months this galvanizer had permanently
poisoned his flux solution with sodium chloride. His
only choice was to change out the flux solution
because there is no purification
method to remove
sodium chloride.

Controlling

Smoke...at

a Price

Both sodium chloride and potassium

chloride do
have a potential beneficial use in flux solutions namely to reduce smoke. In ancient time, during the
building of the great pyramids, sodium chloride was
added to oil lamps to reduce smoke inside the pyramids. In Germany, smoke reduction in the galvanizing industry
has been studied scientifically
and
NaCl or KC1 is sometimes deliberately recommended. It is effective. However, for most galvanizers it
carries the unfortunate
economic price of higher
zinc usage. In any case, the NaCl and KC1 concentrations
are lower than in the cited cases. This
smoke reduction will be addressed later in connection with another technology.
Some flux producers offer products with a few percent KC1 for use as a top flux on a wet kettle. I am
unaware of any scientific or practical plant data to
show any advantage. Considering the very similar
chemical and physical properties of NaCl and KCl,
the use of KC1 in a flux solution is probably not
desirable.

Other

Flux

Components

Mostly

Unwanted

Other potential
components
in a flux solution
include barium chloride for sulfate control and zinc
oxide, zinc powder or zinc hydroxide for pH control.
These control additives cannot possibly exactly
match every galvanizers needs and need to be used
very rarely and with an awareness of their side
28

effects. Fluxes containing barium chloride are made

suddenly highly toxic, whereas without barium chloride, they are merely hazardous.
Excess barium
chloride in a flux solution can quickly kill many
workers. If a new flux solution were made up with
sulfate control additives, the excess barium chloride would be there on day one. All too frequently,
pH control flux gives a flux with excessively high
pH. Based on my experience, I cannot recommend
these controller additives.
In addition to these deliberate additives of questionable value, contaminants as well as calcium carbonate
and magnesium carbonate may be unavoidably present due to using hard water. All are unwanted.
DRAMATIC
EXAMPLE

REAL-WORLD

SAVINGS

- A DETAILED

In 1982 a South African galvanizer started using

quadraflux
and had excellent results. This company
was galvanizing 70,000,OOO pounds of steel annually. Their old flux solution was: 32 deg. Baume; ACN
0.62; iron (Fe+2) 0.8%; pH 5.5; temperature
70F
(unheated); and no surfactant. To make quadraflux,
about 3/4 of their old flux solution was pumped out
to a storage tank, several tons of ammonium
chloride were added to the remaining
flux solution in
their flux tank and dissolved with an air line. About
45 gallons of concentrated
hydrochloric
acid were
added to lower the pH to 4.2. A few gallons of effective surfactant were dissolved in hot water and then
added to the flux tank.
Stainless
steel pipe
(although titanium is the material of choice for use
with flux solutions) was screwed together to make a
steam coil to heat the flux tank to 160F. Finally,
water was added to bring the flux solution up to the
normal working depth. In short the final solution
matched that of Table I

Quantifying

the Financial

Benefits

For this galvanizer

the change to quadraflux
had
several major benefits. A dramatic
improvement
in quality
allowed
a 20% price increase while
simultaneously
decreasing
labor costs by 20%.
Much of the labor savings came from eliminating
the need for 40 laborers
assigned to clean up
work. Another
major savings came from a dramatic drop in the reject rate from 4.4% to 0.3%.
All of these benefits from changing to quadraflux
are quantitatively
presented
in Table III. The
total economic
benefit
for this operation
was
$8,210,000
annually
based on a 250-day work
year and an improvement
of $32,840 per work
day. All of these benefits occurred instantly
upon
conversion to quadraflux.
Metal

Finishing

If your pretreatment line is

more than two years old,

theres a nood chance

your wa&er may be
costing you money.
The new ProMaxB Quick-Connect
Nozzle System from
Spraying Systems Co. provides faster maintenance,
less
downtime
and ease of operation. Which provides you
with a better bottom line.

Standard features of the ProMax

Quid+Connect Nozzle System indude:
l

Easy, no-tools installation

Fits 1: l- l/4, 1- 1J2 and 2 pipe sizes
Uses standard, high-quality
ProMax spray tips
Quick, l/4 twist installation
and removal of spray tips
Color-coded
spray tips for easy identification

Want more information

on the money-saving
spray
technology
of ProMax Quick-Connect
Nozzle Systems or
other ways Spraying Systems Co. can help you cut costs?
Contact a sales engineer today at l-800-95-SPRAY
or
learn more at www.sprav.com.

VAUPM Variable Spray

ProMax Quick&

Nozzles

WindJet Air Control Produck

Spraying Systems Co.@

P.O. Box 7900
Wheaton, Illinois 60189-7900 USA
Phone l-800-95-SPRAY
Fax I-888-95-SPRAY
Outside the U.S.. Phone l(630) 665-5000
Fax l(630) 260-0842
V&if uur Web Site: hffp:Ywww.sprffycom
l

rd or go to www.thru.tolwebconnect

Producing
a Better Product
In Table III the original average galvanizing price
was $0.20llb and the cost of galvanizing was considered to be $O.lO/lb. The 20% higher price gave a new
average price of $0.24/lb that was set and sustained
by true higher quality. Esthetically, the product was
brighter with near electroplate
quality. Even more
importantly,
the zinc coatings were much thinner
and more conforming
to the contours of the steel.
Thus mating surfaces fit and bolt holes accepted
bolts without reaming. In addition, most zinc coatings were under 6 mils (150 microns) so that the
zinc coating did not chip off during rough handling
and shipping. This galvanizer
successfully - and
profitably
- created a differentiated
product in a
traditionally
nondifferentiated
industry.
The 20% higher production
(first row, Table III)
was a surprise to everybody including the author.
With quadraflux,
the product entered the molten
zinc with almost no zinc spatter. Thus, the workers
no longer had to hide behind building supports. The
time the product entered the molten zinc and the
time the product was in the kettle were both cut in
half. Ultimately
the middle and lower resistance
electric heating elements for the zinc kettle furnace
had to be replaced by higher kilowatt units in order
to keep up with the increased production.
Increasing
Production
and Reducing
Zinc
Consumption
In the second row of Table III, the $O.l4/lb of total
profit reflects the 20% increased price and the
14,000,OOO lb reflects the 20% increased production,
both due to quadraflux.
It is interesting
that this
plant, as most galvanizing plants, regarded the kettle as the most critical step limiting
production.
However, it became increasingly
obvious that the
actual limitation
on increased production
throughput was not the kettle, but the old flux solution that
was the real bottleneck.
In the third row of Table III, the $0.85 was the
high price of zinc in 1982. The 0.0269 is the old
%GZU of 8.19% (using the old flux) minus the new
%GZU of 5.50% (using quadraflux)
all divided by
100 to get a fractional number. The reduction in zinc

30

consumption was the result of thinner zinc coatings

on the steel, less dry ash (a drop from four to only
one barrel per day), and fewer rejects that needed to
be stripped and redone.
Slashing Reject Costs
In the forth row of Table III, the $0.30/lb is the total
galvanizing
cost to process rejects, calculated
at
$O.lO/lb the first time through
the kettle and
$0.20/lb the second time. Considering the zinc stripping cost and time, the tripling of galvanizing cost is
reasonable. The 0.041 is the old reject rate (4.4%
using old flux) minus the new reject rate (0.3%
using quadraflux)
all divided by 100 to get a fractional number. It was amazing how many of the old
reject types were eliminated
using quadraflux.
Welding slag rejects were virtually
all eliminated.
Somehow quadraflux
got under the welding slag
and fully galvanized the product. Black spots virtually disappeared.
Basically the only remaining
rejects (0.3%) were
due to paint identification
markings on the steel.
This galvanizer tested various paint pigment marking pens and found a Japanese aluminum paint pigment pen that was removed both by caustic and by
acid due to the amphoteric
nature of aluminum.
This galvanizer
purchased
a case of the pens
imprinted
with the galvanizers advertisement
on
them
and gave them away to their customers to
mark their steel.
The fifth row of Table III calculates the yearly
wage (with 25% allowance for fringe benefits) of
forty zinc clean-up laborers who were laid-off within
a few days after changing
to quadraflux.
With
quadraflux
the hanging zinc icicles that originally
looked like the point of a pencil became small, very
thin, flat foil, like a silver gum wrapper that practically blew away with the wind. Zinc clean-up
changed from intense, high-speed side grinding to
very light, to no hand filing.
Obviously the South African galvanizer was very
pleased with the $8,210,000 annual benefits. When
he ran out of surfactant, he had it air shipped from
the U.S. to South Africa at a cost of $6,000 per barrel - $1,000 for the surfactant
and $5,000 for air

Metal

Finishing

shipping. Changing
to quadraflux
cost only a few
thousand dollars and the investment has been paid
off many, many times over the last 20 years.
TESTING

FLUX

CHEMISTRY

A galvanizer should now be able to initially make up

a quadraflux
solution as shown in Table I. As the
flux is used, the composition
of this solution will
change. The Baume, ACN, iron (Fe+2), sulfate (SO,-2)
(sulfuric acid pickle), pH, and surfactants
will all
change. Accurate
testing
to control the flux is
required. A galvanizer needs fast, accurate, and low
cost test results.
Methodology
The author wanted to evaluate the performance,
cost, speed, and accuracy of a wide range of laboratories performing
this critical analysis. To create a
standardized
test, the author made up 8 liters of a
standard flux. This standard solution was made up
by dissolving
10 mL of concentrated
(37%)
hydrochloric acid, 21.05 grams of anhydrous ferrous
sulfate, 80 grams of ammonium
chloride, 80 grams
of zinc chloride all dissolved in each liter of solution.
The resulting density is 13.4 degrees Baume (1.102

July/August

2003

g/mL). The solution has an ammonium chloride concentration of 7.26%, a zinc chloride concentration
of
7.26%, an ACN of 1.00, an iron (Fe+2) of 0.70%, and
a sulfate (S0,-2) of 1.21% as shown on the first row
of Table IV marked STD.
Lab Results
The second, third, and fourth rows of Table IV are
test results for Bob, Tom (the author),
and
Erik (see acknowledgements.).
For the second,
third, and fourth rows, the zinc chloride concentration was determined
mathematically
by accurately knowing the density, the iron (Fe+2), and the
sulfate (SObm2). This methodology
works very well
as shown in Table IV because the worst zinc chloride concentration
of 7.35% is only 1.2% higher
than the true value of 7.26%. For rows two, three,
and four the, ACN is directly measured
by two
methods within a single test procedure. The first
way was based on heat of solution knowing that
ammonium
chloride gets cold when dissolved in
water and zinc chloride gets hot when dissolved in
water. The second way was by density knowing
that ammonium
chloride forms low density solutions and zinc chloride forms high-density
solu-

31

Just circle the corresponding number - from the categories listed below - in the Information Service Section of our Reader
Service Card. Information will be sent directly to you from the manufacturers and suppliers of products and services in these
categories who advertise in the pages of Metal Finishing. The categories listed include:
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530

Adhesion Testers
Analytical Instruments
Anodes
Anodizing
Barrel Plating Equipment
Blackening &Antiquing
Blasting
Buffing/Brushing/Polishing
Centrifugal Dryers
Chemical Feeders
Cleaners, Aqueous

Cleaners,Solvent
Cleaning & Pretreatment Equipment
Conversion Coating Processes
Conveyors & Hoists
Corrosion Testing
Electroless Plating
Electrolytic Recovery Equipment
Evaporators & Stills
Filtration Equipment & Media
Heating & Cooling Equipment
Ion Exchange
Masking
Mass Finishing
Membrane Separation Processes
Mist Suppressants
Ovens
Paint Overspray Filters
Painting Equipment & Systems
Paints & Coatings

531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559

pH/ORP Meters & Controls

Phosphate Coating Processes
Pickling 81Bright Dipping
Pipes, Fittings & Valves
Plating Equipment & Systems
Plating Processes, Copper & Alloys
Plating Processes, Nickel & Chromium
Plating Processes, Precious Metals
Plating Processes, Tin & Alloys
Plating Processes, Zinc & Alloys
Powder Coating
Power Supplies
Pumps
Racks & Baskets
Refining Systems
Rinse Control Systems
Sludge Handling
Spraybooths
Spray Guns
Strippers, Metal
Strippers, Paint
Tanks & Tank Liners
Test Cells & Panels
Thickness Testers
Training Courses
Vacuum Coating
Ventilation & Air Pollution Control
Waste Disposal & Recycling Services
Water Pollution Control

1,1:, &oduct and techno

@anfonnect to websitesby
11still have the capabil
quiry System(RIS).

usly...from ONE place.

32

Metal

Finishing

:j

That your scheduleand budget are tight.

l That you need if right the first time.
l We understandthat expunding, modifying, consolidating or renovating your
mefol finishing processingcapabilifies is more than a drag and drop
proposition.
l We understandhow these changesaffect your:
- flectrical
- Processand MechanicalSystems
- Structural
- Industriul Wastewater TreafmentSystems
- Environmental Permiffing.
l

Burns& McDonnellprovides full support services,from consulfing, futility layout

planning, estimating and environmental permiffing to engineereddesign and
constructionof facilities, to froining, operating and monaging those facilities.

Circle

002 on reader

information

card

or go to www.thru.to/webconnect

tions. The ACN is found directly on the tables by

comparison of other standard solutions.
These ACN tests developed by the author have
been very successful for about 20 years. For rows
two, three, and four the ammonium chloride concentration is calculated by knowing the zinc chloride
concentration
and the ACN. The worst result of
6.91% for ammonium
chloride is 4.8% lower than
the true value of 7.26%. In rows two, three, and four
the iron (Fe+2) and sulfate (SO,-2) were determined
by modified dropper bottle test kits developed by the
author 30 years ago. The modifications
are putting
the titrants into l-cc plastic syringes and using a
specific volume of flux such that 1 cc of titrant
equals 1% of impurity As shown in Table IV (rows
two, three, and four) the worst result is 0.72% for
iron (Fe+2), which is 2.9% higher than the true value
of 0.70%. The worst value for sulfate, 1.19% is only
0.017% lower than the true value of 1.21%.
In Table IV, rows five through seventeen give test
results from thirteen labs, some of which have tested galvanizing fluxes for many years. Prices charged
for testing
this standard
flux solution
ranged
between $0 and $100. Row eighteen gives the number of results from a possible of thirteen that are
within 5% of the true value.
As Table IV shows only labs F and G got within
5% of the true value of ACN (which was 1.00).
Unfortunately
for lab G the very high zinc chloride
result of 15.45% makes the calculations blow up prescribing that 830 pounds of ammonium chloride/l,000
gal of flux should be added to raise the ACN from 1.00
to 1.60. This would actually produce an ACN of 2.24,
which is not at all desirable. This ammonium chloride
addition is more than twice the real value of 400 lb of
ammonium chloride/l,000
gal of flux.
Lab F got a usable value for ACN, zinc chloride,
and ammonium chloride to properly adjust the ACN
to a target value of 1.60. The results from lab F for
iron of 0.57% (true value 0.70%) and for sulfate of
1.36% (true value 1.21%) does not justify a high level
of confidence in their work. The values reported by lab
F are 18.6% lower and 12.4% higher for iron and sulfate respectively. This standard flux solution was the
first galvanizing flux sample that lab F ever tested.
A different standard flux solution has been sent to lab
F to see if they can get additional good results for
ACN, zinc chloride, and ammonium chloride. A second
sample (also with an ACN of 1.00) was sent to lab H
and they returned a result of 1.20, which is worse than
their original 1.12 ACN for the standard flux sample
dealt with in this article.
As Table IV shows only labs B and H got iron
(Fe+2) correct within the 5% error margin. In the case
34

of low results for iron, insufficient volumes of ammonium hydroxide and hydrogen peroxide would be used
during purification
and all the iron would not be
removed. In the case of reporting high results for iron
the result would be that too much ammonium hydroxide and hydrogen peroxide would be added. This
would have resulted in too high pH and an excess
hydrogen peroxide which would bubble for days, not
allowing insoluble iron III hydroxide to settle.
From Table IV only labs B, G, K, and L got
within 5% of the true value for sulfate. In the case of
low results being reported for sulfate, an insufficient
amount
of barium
chloride would be added for
purification.
This would result in leaving unwanted
sulfate in the flux solution. In the case of reporting
too high results for sulfate, an excess of barium chloride would be added giving a very highly toxic flux
solution.
As Table IV clearly shows, no single lab is capable
of usable results for ACN, ammonium
chloride, zinc
chloride, iron (Fe+2), and sulfate. In addition, if the
pH of the flux is high, and there is an air bubble in
the sample bottle, the iron (Fe+2) will oxidize during
transit to the lab. If cold shipping conditions exist,
ferrous sulfate can crystallize from solution when
the sulfate is about 2% or higher.
Testing Summary
Turn-around
times for labs A through IVY were 2
to 6 weeks, which is far too long to keep up with
needed flux adjustments.
None of the labs indicated
having any problem in their testing. In view of the
results from labs A through M, it is the authors
opinion that it is pointless for a galvanizer to send a
flux solution sample to any of these labs.
KEEPING

THE

FLUX

SOLUTION

OPTlNllZED

Getting accurate analysis of the flux is only a first

step. The real challenge is to use the analysis to calculate the correct amounts of chemicals to add to the
flux solution. Returning to the South African example, a few months after the successful conversion to
quadraflux, zinc consumption (%GZU) began to rise.
The reason was easily found in the daily log book
and it was a straightforward
task to write a computer program, which freed individuals from having
to make detailed calculations manually, The %GZU
went back down and stayed down. Currently, this is
being converted by the author to a more accurate
fluxcalc EXCEL computer program.
GElTlNG

STAKTED

This article is not a universal blue print to make up

a universal quadraflux
formula to improve galvaMetal

Finishing

nizing results. However, it does hope to have summarized a vast amount of real-world experience and
identified those variables that have a direct effect on
performance,
cost, and quality results. Controlling
flux composition,
temperature,
and use requires
knowledge,
understanding,
and experience.
The
impact on the bottom line, however, justifies all the
time and all the effort.
It goes without saying that the starting materials,
such as ammonium
chloride, zinc chloride, and commercial flux, must be of adequate purity. Beyond
that starting point, one has to ask where can these
materials be tested? Will the test results be correct?
What are my surfactant
requirements
and what
surfactant
should be used? How will pH be adjusted? What adjustment
may be necessary each day?
How will ACN testing be done? How much ammonium chloride should be added and how often? When
and how should the flux be purified? Will there be
more or less smoke from quadraflux?
There are a lot of questions
that need to be
answered
in the context of ones own operation.
There is no substitute
for hands-on testing under
your actual working conditions. For small scale testing of quadraflux,
I recommend making up 4 gallons

(15 liters) in a &gal ceramic or porcelain coated

steel container (a canning pot). This allows you to
make side-by-side
comparisons
with the old flux
solution. Remember that 160F (71C) is required
for quadraflux
and the steel must remain in the
quadraflux
3 to 5 minutes. Compare zinc spatter as
the fluxed steel enters the molten zinc, compare
cook off times (when bubbling stops), compare dry
ash formed, and evaluate galvanizing
quality. Use
shorter and shorter kettle times until quadraflux
gives too thin zinc coatings or the old flux produces
black spots. Take the time to get some experience
with a flux that can generate more profit. If you are
like the South African galvanizer, a delay in making
this decision costs $32,840 a day.
ACKNOWLEDGEMENTS

The author thanks Bob Olsen of T and L Irrigation

in Hastings,
Nebraska,
and Erik Jacobs of Ace
Galvanizing in Seattle, Washington, for testing the
standard flux solution in Table IV. The author also
thanks long-time Metal Finishing
editor, Michael
Murphy, for encouragement
in writing galvanizing
articles for this magazine.

DOES YOUR WATER TREATMENT SYSTEM ALWAYS KEEP YOU IN COMPLIANCE?

TIRED OF WALKING THAT FINE LINE OF BEING IN OR OUT OF COMPLIANCE?

ELIMINATE

THESE PROBLEMS

AND WORRIES

IN 128 SECONDS!

In just 128 seconds Octolig@will

+ Makesurethatyoumeetthe limits.
+ Operateconsistentlyandefficiently.
+ Enableyouto increaseyournetprofits.
6 Eliminateenvironmental
complianceissues.

Lcfie%?
4 fEEr+

So just how important are these 128 seconds?

METRE-GENERAL,

INC.

3771 MonarchStreet, UnitA-P.O. Box 1149

Frederick,CO 80530
Fax: (303)833-1463
Phone: (303)833-1432
www.octolig.com and then give us a call.
Visit us at our web site Circle
July/August

2003

065 on reader

information

card

or go to www.thru.to/webconnect
35