FIELD OF THE INVENTION This invention relates to a galvanizing alloy and process and, more particularly, relates to a galvanizing alloy and an immersion galvanization process adapted to control the undesirable effects associated with galvanizing reactive steels.

BACKGROUND OF THE INVENTION

The conventional process for hot dip galvanizing of low carbon steels comprises pretreatment of said steels in a 25 to 35% by weight zinc-ammonium-chloride (ZnNH ₄ Cl) pre-flux, followed by immersion in molten zinc or zinc alloy baths. The 'normal' or "N' coating structure produced on low reactivity steel by conventional hot dip galvanizing processes has well defined, compact alloy (intermetallic) layers. The predominant growth mode in this type of coating is by solid-state diffusion of iron and zinc, and thus well established intermetallic (delta and zeta) layers control the rate of the galvanizing reaction. The diffusion reaction rate decreases as the coating thickness increases, thus permitting predictable, consistent coverage. The normal coating has a bright metallic lustre.

Recent developments in the manufacture of low alloy high strength steels include continuous casting. In the continuous casting process, it is necessary to add elements that 'kill' or deoxidize the steel, ie., prevent gaseous products which produce porosity. Silicon is commonly employed for this purpose. These steels as a result generally contain between 0.01 to 0.3% by weight silicon but may include up to or more than about 0.5% silicon and are known as 'reactive steels' or silicon steels.

Steels containing phosphorus are also reactive steels having an accepted measure of reactivity relative to the silicon content of Si + 2.5 P (the silicon content plus 2.5 times the phosphorus content). This is called the effective silicon content.

Silicon steels that have high reactivity pose problems to the galvanizing process, producing thick, brittle and uneven coatings, poor adherence and/or a dull or marbled appearance. These coatings are known as 'reactive' or 'R' coatings. The high reactivity of the silicon steels also causes excessive zinc consumption and excessive dross formation.

Silicon released from the steel during galvanizing is insoluble in the zeta layer. This creates an instability in the zeta layer and produces thick, porous intermetallic layers. The microstructure is characterized by a very thin and uneven delta layer overlaid

by a very thick and porous zeta layer. The porous intermetallic layer allows liquid bath metal to react near the steel interface during the entire immersion period. The result is a linear growth mode with immersion time that allows the formation of excessively thick coatings. These coatings are generally very rough, undesirably thick, brittle and dull in appearance.

Steels with silicon levels between 0.05 to 0.15 (i.e. around the "Sandelin peak" area), may also develop a 'mixed' reactivity or 'M' coating. This coating is characterized by a combination of reactive and non-reactive areas on the same steel which is believed to be due differences in localized silicon levels on the surface of the steel.

It is known in the prior art to control reactivity by reducing bath temperature and immersion time at a rate inversely proportional to the silicon content of the steel. Lower bath temperatures, in the order of 430 °C, and reduced immersion times, tend to control reactivity on high silicon steels. However using low bath temperatures and times on low silicon steels produces unacceplably thin coating thicknesses. Thus the galvanizer must know the silicon content of the steel beforehand and adjust the hot dip parameters accordingly. This approach cannot be implemented if steel reactivity is not known or if components to be galvanized comprise parts of different reactivities welded together. With low-temperature galvanizing, productivity can be poor because of the need to increase immersion times.

It is also known to control steel reactivity by adding alloy elements to the zinc galvanizing bath. One such addition is nickel in a process known as the Technigalva™ (or Nickel-Zinc) process. A nickel content of 0.05 to 0.10% by weight in the zinc bath effectively controls reactive steels having up to about 0.2% by weight silicon content. For steels having silicon levels above approximately 0.2%, this nickel-zinc process is not effective and thus it is only a partial solution to the reactive steel galvanizing problem. Low reactivity (normal) steels, when galvanized by the nickel-zinc process, pose the same difficulty as seen in low temperature galvanizing in that coating thickness may be unacceptably thin. With this process, it is thus preferred that the galvanizer know the reactivity of the steel beforehand and adjust galvanizing conditions accordingly, both of which are difficult to accomplish in practice. Under some conditions, this process also

15696

produces dross that tends to float in the bath and be drawn out on the workpiece, producing unacceptable coatings .

Another alloy used to control reactivity is that disclosed in French Patent No. 2,366,376 granted October 27, 1980 for galvanizing reactive steels, known as the Polygalva™ process. The alloy comprises zinc of commercial purity containing by weight 0.1 to 1.5% lead, 0.01 to 0.5% aluminum, 0.03 to 2.0% tin and 0.001 to 2.0% magnesium.

U.S. Pat. No. 4,439,397 granted March 27, 1984 discusses the accelerated rate at which the magnesium and aluminum are consumed or lost in this Polygalva™ process for galvanizing steel. Procedures are presented to overcome the inherent difficulty in replenishing deficient aluminum or magnesium in the zinc alloy galvanizing bath. The process has serious limitations in that the steel has to be meticulously degreased, pickled, pre-fluxed and oven-dried to obtain good quality product free of bare spots. Thus, in most cases, new high-quality installations are usually required.

U.S. Pat. No. 4,168,972 issued September 25, 1979 and U.S. Pat. No. 4,238,532 issued December 9, 1980 also disclose alloys for galvanizing reactive steels. The alloys presented include variations of the Polygalva™ alloy components of lead, aluminum, magnesium and tin in zinc.

It is known in the prior art that aluminum included in the galvanizing bath reduces the reactivity of the high silicon steels. A process known as the Supergalva™ process includes an alloy of zinc containing 5% aluminum. The process requires a special flux and double dipping not generally accepted by commercial galvanizers.

It is a principal object of the present invention to provide a process and alloy to effectively control reactivity on a full range of steels including low and high silicon steels. The process should also produce coatings of acceptable and uniform thickness over the full range of steels.

Another object of the invention is to provide an alloy and process which uses standard galvanizing equipment operated under normal conditions for galvanizing steels of mixed reactivity without the need to adjust for variations in steel chemistry.

SUMMARY OF THE INVENTION

The disadvantages of the prior art thus may be substantially overcome by providing a new galvanizing alloy and process which can be readily adapted to standard hot-dip galvanizing equipment.

In its broad aspect, the process of the invention for galvanizing steel, including reactive steels, by immersion comprises immersing said steel in a molten bath of a zinc alloy comprising, by weight, aluminum in the amount of at least 0.003%, and one of an element selected from the group consisting of vanadium in the amount of at least 0.06%, preferably at least 0.08%, titanium in the amount of at least 0.03%, preferably at least 0.5%, and both vanadium and titanium together in the amount of at least 0.02% vanadium and at least 0.02% titanium with a total of at least 0.04%, preferably at least 0.06% vanadium and titanium, the balance zinc. The alloy of the invention for galvanizing steel comprises, by weight, aluminum in the amount of at least 0.003%, and one of an element selected from the group consisting of vanadium in the amount of at least 0.06%, preferably at least 0.08%, titanium in the amount of at least 0.03%, preferably at least 0.5%, and both vanadium and titanium together in the amount of at least 0.02% vanadium and at least 0.02% titanium with a total of at least 0.04%, preferably at least 0.06% vanadium and titanium, the balance zinc.

In another aspect, the invention relates to the pre-treatment of the steel to be galvanized by applying a flux comprising about 15 - 20% ZnNH ₄ Cl in an aqueous solution.

BRIEF DESCRIPTION OF THE DRAWINGS

The process of the invention and the alloy produced thereby will now be described with reference to the following drawings, in which:

Figure 1 to 3 are graphs illustrating galvanized coating thickness of a variety of galvanizing coatings on steel surfaces having a silicon content ranging from 0 to 1.0 wt% under conditions of eight-minute immersion at 450 °C, Figure 1 being a graph showing average coating thickness versus silicon content in a galvanizing bath of titanium with Prime Western (PW) zinc, Figure 2 being a graph showing average coating thickness

versus silicon content in a galvanizing bath of vanadiam with PW zinc, and Figure 3 being a graph showing average coating thickness versus silicon content in a galvanizing bath of vanadium and titanium together with PW zinc;

Figure 4 is a graph showing average coating weight versus silicon content in steel coupons produced by a four-minute immersion in a galvanizing bath of vanadium with PW zinc;

Figure 5 is a graph showing average coating weight versus silicon content in steel coupons produced by an eight-minute immersion in a galvanizing bath of vanadium with PW zinc;

Figure 6 is a graph showing average coating weight versus silicon content in steel coupons produced by a four-minute immersion in a galvanizing bath of titanium with PW zinc;

Figure 7 is a graph showing average coating weight versus silicon content in steel coupons produced by an eight-minute immersion in a galvanizing bath of titanium with PW zinc;

Figure 8 is a graph showing average coating weight versus silicon content in steel coupons produced by a four-minute immersion in a galvanizing bath of vanadium plus titanium with PW zinc; and

Figure 9 is a graph showing average coating weight versus silicon content in steel coupons produced by an eight-minute immersion in a galvanizing bath of vanadium plus titanium with PW zinc.

DESCRIPTION OF THE PREFERRED EMBODIMENTS With reference first to Figures 1 , 2 and 3 of the drawings, curve 10 typifies the variation of thickness in microns of a coating of zinc of commercial purity, such as conventional Prime Western (PW), on a steel surface as a function of the silicon content of the steel. The term "commercial purity" used herein will be understood to include Prime Western, High Grade and Special High Grade zinc. Under these conditions of bath temperature (450°C) and immersion time (8 minutes), the thickness of zinc coating peaks at a thickness of about 260 microns at a silicon content of about 0.15 wt%, decreases to a thickness of about 175 microns at a silicon content of about 0.2 wt%, and then increases to a maximum thickness of about 375 microns at a silicon content of about 0.5 wt%,

decreasing i„ thickness slightly to a silicon content of 1.0 wt%. This curve 10 will be recognized as being very similar to the well-known Sandeli ⁱ n curve. The composition of the steels used is listed in Table I following.

Table I

/15696

In accordance with ASTM standards, e.g. the ASTM A- 123 Standard (610 g/nr or 86 microns for 3.2 to 6.4 mm thick steel plate), a uniform coating thickness of about 100 microns is desired in order to meet minimum thickness requirements while avoiding the expense and waste of thick coatings. Also, excessive thickness of zinc coatings on reactive steels and steels of mixed reactivity due to high or variable silicon contents usually produce rough, porous, brittle and generally unsightly coatings which can have poor adherence to the underlying steel surface.

It is generally accepted that the addition to the galvanizing bath of strong suicide formers may neutralize the influence of silicon in reactive steels. It has been found that vanadium is an effective alloying element for reducing the reactivity of high silicon steels. Generally, galvanizing baths containing 0.08 to 0.12% vanadium effectively control excessive reactivity. However, galvanized steel from a galvanizing bath containing an alloy of zinc with vanadium rapidly forms a coloured vanadium oxide layer on the surface and has excessive coating roughness and occasionally bare spots. The effect of aluminum in a zinc/vanadium bath for avoiding undesirable coloration has been found to be first evident at 0.003% and, when the aluminum level is increased to 0.004%, vanadium oxide formation is essentially reduced and the bath surface retains a grey, metallic surface for an indefinite time.

Tests have shown that in a galvanizing bath containing 0.08 wt% vanadium, 0.005 wt% aluminum and the remainder zinc of commercial purity, reactivity is controlled in steels having a silicon content up to about 0.25 wt% by the presence of at least 0.06 wt% vanadium, as shown in Figure 1. Vanadium in the bath is believed to combine with the silicon to form vanadium suicides as inert particles that become dispersed in the zeta layer. The silicon-free iron can then react with zinc to form a very compact and smooth layer that prevents liquid bath metal from reaching the delta layer. In essence, the vanadium effectively suppresses reactivity by stabilizing the growth of the zeta layer in the coating, which controls the growth rate by a diffusion process.

The coating thickness for high silicon steels matches those of non-reactive, low silicon steel subjected to conventional galvanizing procedures. However, coatings on the galvanized steels may have bare spots. These bare spots are attributed to a reaction of the aluminum and vanadium with normal commercial ZACLON™ pre-flux comprising 25

„,, _. _ _:.. - _: 97/15696

to 30% by wt zinc-ammonium-chloride (ZnNH^Cl). The reaction oxides subsequently deposit on the surface, preventing wetting of said surface. We have found that a pre-flux solution reduced to 15 to 20% by weight ZnNII^Cl completely eliminates bare spots.

In an alternative embodiment of the process of the present invention, titanium is used in place of vanadium. Tests have shown that a galvanizing bath containing 0.05 wt% titanium, 0.005 wt% aluminum and the balance zinc of commercial purity, the presence of at least about 0.05 wt% titanium effectively controls reactivity in varying degrees in steels having silicon contents up to about 0.5%, as shown by Zn-Ti curve 1 in Figure 2. However, galvanized coatings on steels containing 0.3% to 0.5% silicon had a "mixed reactivity" growth, producing rough coatings that would likely be commercially unacceptable.

The titanium addition to the bath modifies the Fe-Zn intermetallic layers on the reactive steels to produce the more compact and even delta and zeta layers found on the non-reactive steels. This suggests that much like the vanadium, titanium is a strong silicide former that ties up the silicon released from the steel during galvanizing, allowing the zeta layer to stabilize. However, unlike vanadium, titanium forms a ternary Zn-Fe-Ti intermetallic layer at or near the steel surface where there is iron-enrichment. The intermetallic particles are trapped in the eta layer (outer layer) and hinder zinc drainage, thus producing a thicker coating.

The coating microstructures produced by Zn-Ti coating alloys show clearly that the thicker coatings obtained with the titanium alloys are due to Zn-6%Fe-3%Ti intermetallic particles that are present in the eta layer. A beneficial side effect of the thicker eta layer is that low-silicon steels have coatings that meet ASTM thickness standards such as the ASTM A- 123 Standard.

These intermetallic particles however cause higher levels of dross production and increased zinc consumption.

It has been found that optimum results are obtained when vanadium and titanium are combined in the galvanizing bath. In accordance with a preferred embodiment of the present invention, the process for hot dip galvanizing includes a galvanizing bath comprising aluminum and both vanadium and titanium in an amount of at least 0.02 wt% of one of vanadium or titanium and sufficient of the other for a total of at least 0.06 \vt%,

the balance zinc of commercial purity. More specifically, tests show very good results with an alloy comprising 0.04 wt% vanadium, 0.05 wt% titanium, 0.005 wt% aluminum and the remainder zinc of commercial purity, as illustrated in Figure 3, Zn-V-Ti curve 16. In this case, the good coating thickness control was retained with up to almost 1.0% silicon in the steel. The results indicate that the combination of vanadium and titanium together outperforms the single element additions of concentration levels higher than the sum of the two elements.

The process of the invention preferably includes pre-treatment of the steel surface in a reduced-strength pre-flux aqueous composition of zinc-ammonium-chloride (ZnNH ₄ Cl), specifically 15 to 20% by weight Zι_NH ₄ Cl. The delta and zeta layers of this process embodiment of the invention for a zinc bath containing vanadium and titanium are compact, even and very thin. The eta layer usually has a fine dispersion of intermetallic particles and the thickness is about 10% thicker than obtained from the vanadium alloy. Coating thickness meets ASTM requirements for all steels including low silicon steels. This combination of vanadium and titanium controls reactivity on steels with up to at least 0.5% silicon content and provides bright metallic coatings void of bare spots.

The alloy composition and process of the invention will now be described with reference to the following non-limitative examples.

EXAMPLE 1 Immersion galvanizing of reactive steels in a zinc alloy containing vanadium, showing effects of aluminum on colour suppression.

Preliminary trials were conducted to establish a standard control based on the zinc-vanadium alloy process. The experimental melts weighed 25 kg and were prepared in a ceramic crucible that was electrically heated with external radiant tubes. The crucible provided a 150 mm diameter surface for galvanizing.

A PW bath, saturated with iron, was prepared and an addition was made of 0.002% aluminum brightener. Small steel coupons (77 mm x 39 mm x 3 mm) were dipped for 4 minutes at 450°C, to develop a dipping procedure and to produce control

/15696

10 samples. The steels used were selected from the group listed in Table II below. Galvanized coatings were of good quality, and were normal for the particular silicon- level in the steel.

Table II

STEEL COMPOSITIONS

N.A. = Not Available

/15696

11

An addition of 0.04% vanadium was made to the bath when baseline galvanizing conditions were established. The vanadium was added as a master alloy containing 2.3% vanadium. Immediately after the vanadium alloy was introduced into the bath, the surface became covered with a yellow oxide layer (as opposed to the matte-grey that formed before the addition of vanadium). The surface was skimmed, but the yellow oxide appeared again within seconds. The oxide layer became thicker with time and changed in colour from yellow to purple, to dark blue, within a period of a couple of minutes. A few coupons were dipped to assess galvanizability under these bath conditions. The galvanized coupons had a yellow appearance, presumably because of the very thick and rough vanadium-oxide layer. The trial continued by making additions of aluminum at 0.0005% increments. As the aluminum level in the bath increased from 0.002 to 0.003%, the time for the surface to form the yellow oxide also increased. At 0.004% aluminum, the bath surface retained a metallic grey sheen for about 5 minutes after which it would gradually change to a light yellow colour. When the aluminum level was increased to 0.005%, the bath surface retained the grey/metallic surface for an indefinite period of time.

The vanadium in the bath was increased to 0.12% to ensure that 0.005% aluminum was sufficient to control surface oxidation even at the higher vanadium bath level. The grey surface was maintained and oxide formation was controlled. A few coupons galvanized in this bath had a grey metallic sheen.

EXAMPLE 2 Pre-flux treatment of reactive steels to be galvanized with a zinc-vanadium alloy.

Although galvanized coatings from vanadium-alloy baths with 0.005% aluminum had a normal (metallic-grey) appearance, they also contained many small bare spots. Close examination of these defect areas showed that the steel surface had not been wetted, and the surface was covered with black residues. Scanning Electron Microscope and Energy Dispersive Spectrometer examination of these residues detected high levels of aluminum, vanadium and chlorine. The flux had reacted with the aluminum and vanadium in the bath, and the reaction products deposited on the surface were preventing wetting of the steel.

15696

12

In the ensuing dipping trials, a number of processing parameters were systematically varied to determine their effects on surface wetting and to establish a galvanizing procedure that would produce defect-free test coupons. The list of parameters examined includes steel type and size, bath temperature, immersion and withdrawal rates, pickling time, pre-flux strength and type, and pre-flux drying time and temperature.

Varying the pickling times and immersion and withdrawal rates did not improve wetting. The occurrence of bare spots was lower on the high reactivity steels. Increasing the bath temperature eliminated the problem, since the thicker steel would heat-up faster and release the pre-flux at, or near, the surface. The galvanizing temperature was raised to 480°C and the thicker (3 mm) coupons were galvanized. The coatings produced were totally free of bare spots.

Trials were conducted with the following preflux composition: 30% by weight Zaclon™ F (normal commercial composition), 10% by weight Zaclon™ F, and 25% of a low-fuming flux. Pre-flux drying temperatures and times were also varied from 90° to 110°C and 5 to 10 minutes, respectively.

To evaluate the effect of the different pre-flux conditions, steel coupons were galvanized at a bath temperature of 455 °C and an immersion time of 4 minutes. The results showed that the most influential factor in preventing bare spot formation was the pre-flux concentration. The normal galvanizing flux (30% Zaclon) produced the most bare spots. The low-fuming flux produced fewer bare spots. The 10% Zaclon™ flux produced a bright shiny coating without any bare spots. Drying temperatures and times were found to have a lesser effect and were set at 100 °C and 5 minutes, respectively.

In later trials, the Zaclon flux concentration was increased to as much as 20% before bare spots would start to occur. A higher flux concentration (greater than 10%) may be required in a galvanizing plant to ensure that poor wetting due to inadequacies of the up-stream prc-cleaning operations (such as pickling and rinsing) can be overcome by the pre-flux.

EXAMPLE 3 Galvanizing Trials

Eight bath alloys were prepared for the galvanizing trials. The alloying additions were made to PW grade zinc.

Bath samples were taken before and afier the day's galvanizing trial and analyzed by atomic absorption analysis to ensure that the aluminum and vanadium levels were maintained close to the nominal composition and to determine the losses from galvanizing. The analysis results for the various experimental baths are listed in Table III.

Table UI BATH ALLOY COMPOSITIONS

N.Λ. - Noc Λviilible

96

14

The typical composition of PW is shown in Table IV.

Table IV

COMPOSITIONS OF PW ZINC

15

The vanadium additions were made with a 2.3% master alloy. The appropriate amount was stirred into the bath and dissolved readily at the galvanizing temperature of 455 °C. The titanium was added directly to the bath at a temperature of 550 ^C C. The bath was maintained at that temperature for about 3 hours until the titanium was dissolved. The temperature was then reduced to 455 °C before galvanizing began. All experimental baths were saturated with iron and a 5% aluminum master alloy was added to maintain a 0.005% aluminum level in the bath.

A bench scale line was set-up to process the test samples consistently. The following steps were used:

- Degreasing: 0.25 g/cc NaOH solution at 70 °C with agitation for 10 minutes.

- Rinse: Tepid flowing water.

- Pickling: 15% HCI at room temperature, inhibited with Rodine™ 85 (' 4000), for 20 minutes.

- Pre-flux: 10% Zaclon™ F (ZnNILCl) at 60 °C, for 2- minute immersion.

-Drying: Oven dried for 5 min. at 1 10°C.

Twenty-five kg melts were prepared in a SiC crucible that provided a galvanizing surface of 150 mm diameter. The crucible was heated in a radiant tube furnace.

The galvanizing temperature was 455 ± 2 ^C C. The melt surface was skimmed prior to immersion and just before the test coupons were withdrawn. The test coupons were dipped for 4-minute and 8-minute immersions. The immersion rate was 40 mm/sec, while the withdrawal rate was 160 mm/sec. The samples were air cooled at room temperature (no quenching).

Hot rolled, low-carbon silicon-killed steel coupons, measuring 77 mm x 39 mm x 3 mm were used. The six steel compositions, with silicon levels ranging from 0.01% to 0.29%, are listed in

Table II. This table includes the respective Si + 2.5P level for the steels, which takes into account the weighted effect of phosphorus as it relates to the reactivity behaviour of the steel.

696

16

The galvanized coatings produced in the experiments were evaluated by the following methods.

Costing Appearance: The test coupons were photographed and classified under one of the three following categories: Normal (N), Reactive (R) and Mixed (M). A description for each type of coating appearance is as follows: Normal - The typical coating of a low-reactivity steel, usually bright and relatively smooth with visible spangle. Reactive- The typical coating of a reactive steel, usually matte-grey with no visible spangle. Mixed - The typical coating of a steel that has both reactive and non- reactive areas. The coating is usually very rough and varies from thin in low-reactivity areas to thick in the reactive areas. Coating Weight and Thickness: Coating weights were determined by the chemical weigh-strip-weigh method according to ASTM A90 Standard. Only a portion of the test coupons, measuring 25 mm x 25 mm, was used for this test. Results are reported in gm/m ² . Corresponding thicknesses (in microns) were also calculated from the coating weights.

Optical thickness measurements were taken from the metallographic sections. Average, maximum and minimum thicknesses from each section examined were recorded.

Metallography: 25-mm long pieces were cut from representative areas of the test coupons and prepared by conventional metallographic techniques for microscopic examination. All test samples were examined by optical microscopy. Selected samples were examined with a scanning electron microscope (SEM) and energy dispersive x-ray micro-analysis (EDS) was performed on selected samples as required.

5696

17

The appearance of the test coupons is summarized in Table V. The three general

categories used to describe the samples are Normal, Reactive, and Mixed.

Table V

COATING APPEARANCE OF TEST COUPONS

Appearance Categories: N = Normal PW - Cominco Prime Western Zinc R = Reactive M = Mixed Reactivity

„,„_. 15696

18

The coating weight thickness results are presented in graph form in Figures 4 to

9. The graphs show the average coating weights developed on the steel coupons in the

various galvanizing trials. Average coating thicknesses, calculated from the weights, are

also shown on the graphs. Because the phosphorus levels of the steels used in these trials

were very low, the Si + 2.5 P values did not significantly vary from the percentage of

silicon values in the steels. Thus, to simplify matters, the reactivity curves in Figs. 4 to

9 were plotted using the silicon content of the various steels.

Figures 4 and 5 show the results of the vanadium trials with PW zinc for the 4-

and 8- minute immersions of test coupons at 455 °C. Figures 6 and 7 show the titanium

trials with PW zinc for the two immersion times of test coupons at 455 °C; the 4-minute immersion of Figure 6 was done with the 0.05% titanium only. Figures 8 and 9 show the

results of the vanadium plus titanium trials with PW zinc for 4- and 8- minute immersion

times of test coupons at 455 °C.

Galvanizing with PW zinc produced normal or non-reactive type coatings on Steels No. 1 and 2 (silicon levels of 0.01 and 0.05). Steels No. 3, 4, 5 and 6 (silicon

levels of 0.12 to 0.29) developed reactive-type coatings.

Galvanizing in baths containing 0.04% vanadium produced normal coatings on

Steels No. 1 and 2, mixed reactivity coatings on Steels 3 and 6, and reactive coatings on

Steels 4 and 5. Galvanizing in baths containing 0.12% vanadium produced normal (non-

reactive) coatings on all six steels. Inconsistent results were obtained with Steel No. 5

(0.22% silicon), where in some instances, mixed reactivity structures were obtained. This

was later determined to be due to abnormal roughness of the steel surface.

_.,„, /15696

19

A noticeable difference between the non-reactive coatings produced in baths

containing vanadium is in the delta-to-zeta ratio. The PW coatings had a 1 :3 delta-to-zeta

ratio while the vanadium coatings had a 1.1 ratio.

The 0.05% titanium addition to the PW bath produced non-reactive

microstmctures on all but the No. 5 Steel. The 0.1% titanium produced non-reactive

microslructures on all six steels. Although the intermetallic layers of the 0.1% titanium

coatings were significantly thinner than those of the 0.05% titanium coatings, overall the

0.1 % titanium coatings appeared to be thicker than the 0.05% titanium coatings. This

was because the eta layer of the 0.1% titanium coatings was thicker. The reason for the

thicker eta layer is the relatively large intermetallic particles contained in the layer that

affected drainage.

Intermetallic particles were also present in the eta layer of the 0.05% titanium

coatings. However, their size was much smaller than those in the 0.1% titanium coatings

and did not interfere as much with the drainage of the eta layer. EDX analysis of these

intermetallic particles determined their composition to be Zn-6%Fe-3%Ti.

Coating microstmctures on the 0.08% vanadium plus 0.05% titanium and on the

0.04% vanadium plus 0.05% titanium samples were very similar to those produced with

the 0.05% titanium (except Steel No. 22) and 0.1% titanium baths. The reactive-steel

coatings were modified by the alloy additions to produce the compact alloy layer

structure of the non-reactive steels. As was seen in the vanadium coatings, the zeta layer

was thinner, giving a delta-to-zeta ratio of about 1 :1. The small Zn-Fe-Ti intermetallic

particles that were observed in the 0.05% titanium coatings were also present in these

5696

20 coatings. EDS analysis of the particles determined that their composition was similar to

that of the titanium coatings.

Referring now to the PW zinc reactivity curves in Figures 4 and 5, it can be seen

that reactivity was low with the 0.01% and 0.05% silicon steels. Reactivity increased to

a maximum with the 0.12% silicon steel, followed by a drop with the 0.16% silicon steel.

Reactivity increased again with the 0.22% silicon steel and decreased with the 0.29%

silicon steel. This decrease was unusual for a steel of this silicon level, 0.3% silicon

steels normally performing as shown in Figure 1.

Additions of 0.04% vanadium to PW zinc decreased coating weight by an average

of 10-30% on all experimental steels, except for Steel No. 3 (0.12% silicon), which

showed a substantial 35% and 48% coating thickness increase for the 4-minute and 8-

minute immersions, respectively. The 0.04% vanadium did not significantly modify the

coating microstructures of the reactive steels (Nos. 3, 4, 5 and 6).

Additions of 0.08% vanadium and 0.12% vanadium to PW zinc reduced the

coating thickness of all the steels tested. The most dramatic reduction was for the

reactive steels (Nos. 3, 4, 5 and 6), whose coating thickness now matched those of the

non-reactive, low-silicon steels. Coatings for the low-silicon steels (Nos. 1 and 2), at 4-

minute immersion, were marginally thinner than required by ASTM A- 123 Standard of

610 g/m ² or 86 microns.

Some inconsistency in controlling the reactivity of Steel No. 5(0.22% silicon) was

experienced at both the 0.08% vanadium and 0.12% vanadium levels due to the

roughness of the steel and, therefore, the observations for Steel No. 5 were disregarded.

It is believed that excessive roughness of the steel surface, the presence of microcracks,

15696

21 or the stress effects of previous cold working can result in inconsistency in the control

of reactivity.

Examination of the microstructures obtained from the 0.08% and 0.12%

vanadium trials showed that the thick, porous intermetallic layers normally produced on

the reactive steels (Nos. 3, 4, 5 and 6) were modified by the vanadium addition to

produce the compact and even alloy layers that were produced on the non-reactive steels

(Nos. 1 and 2). One noticeable difference between them was that the delta-to-zeta alloy

layer ratio of the vanadium-modified coalings was higher than that of the normal non-

reactive coating structure. EDS analysis showed that the Fe-Zn intermetallic layers

produced in the vanadium coatings had the same composition as the conventional delta

and zeta layers of the non-reactive coating. Vanadium was detected in discrete pockets

at the zeta/eta interface and on the outer surface of the coating or eta layer. Since the

solubility of vanadium in the zeta phase is very low, it is rejected at the zeta and delta

interfaces. No vanadium was detected in the alloy layers.

The results of investigations into the causes of silicon-induced reactivity in steels

can provide a logical explanation of how the vanadium affects the microstructure and

produces a non-reactive coating. Silicon released from the steel during galvanizing is

insoluble in the zeta layer. This creates an instability in the zeta and prevents the

formation of a compact, pore-free layer. Vanadium in the bath combines with the silicon

to form vanadium silicides, inert particles that become dispersed in the zeta layer. The

silicon-free iron can then react with zinc to form a very compact and smooth layer that

prevents liquid bath metal from reaching the delta layer. The growth of the zeta and delta

5696

22 layers are then diffusion controlled. Since the availability of iron is higher at the

delta/zeta interface, the delta layer grows thicker than in normal coatings.

With reference to Figures 6 and 7, the addition of 0.05% titanium to the PW baths

reduced coating thicknesses on all but the No. 5 (0.22% silicon) Steel. Additional

coupons of the silicon steel were dipped to ensure that the results were consistent, since

the performance of this steel was found to be questionable in the vanadium trials. The

results showed that 0.05% titanium level could not consistently control the reactivity of

this steel. Observations for Steel No. 5 accordingly were disregarded. Later trials,

reported above with reference to Figure 2, confirmed that steels with the same silicon

level (0.22% silicon) were controllable with 0.05% titanium. Although 0.1% titanium

reduced the thicknesses of the coatings on all six steels, coating thicknesses were 7-32%

greater with 0.1% titanium than with 0.05% titanium (excluding No. 5 Steel). Compared

to the coating thicknesses obtained with the 0.08% vanadium alloy, the 0.05% titanium

and 0.1% titanium coatings were on average about 20% and 45% thicker, respectively.

The coating microstructures show clearly that the thicker coatings obtained with the

titanium alloys were due to the Zn-6%Fe-3%Ti intermetallic particles that were present

in the eta layer.

The titanium addition to the bath modified the Fe-Zn intermetallic layers on the

reactive steels to produce the more compact and even delta and zeta layers on the non-

reactive steels. This suggests that, much like the vanadium, titanium is a strong silicide

former that ties up the silicon released from the steel during galvanizing, allowing the

zeta layer to stabilize. However, unlike vanadium, titanium forms a ternary Zn-Fe-Ti

intermetallic in the bath at the coating/melt interface where there is iron-enricliment. The

15696

23 intermetallic particles are trapped in the eta layer and hinder zinc drainage. A beneficial

side effect of the thicker eta layers in the titanium samples was that they ensured that the

low-silicon steels did not have coatings that were thinner than specified by ASTM

standards.

The larger intermetallic particles in the 0.1% titanium coatings were very similar

to those found in coatings galvanized in the nickel-zinc alloy baths containing about 0.1 %

nickel (Figure 2). These large intermetallics produce higher-than-normal amounts of

dross and, hence, increase zinc consumption and operating costs.

The effectiveness of the 0.08% vanadium plus 0.05% titanium alloy in controlling

steel reactivity as shown in Figures 8 and 9 was not totally unexpected, since the 0.08%

vanadium alloy, on its own (Figures 4 and 5), was effective. However, the 0.04%

vanadium (Figures 4 and 5) or 0.05% titanium (Figure 2) alloys alone could not fully

control reactivity over the full range (0-l%Si) of steels tested. But together, they

successfully controlled reactivity of all steels tested, indicating a synergism in the

performance of the two elements together.

The microstructures of both the 0.08% vanadium plus 0.05% titanium and 0.04%

vanadium plus 0.05% titanium appeared very similar to those of the 0.05% titanium

alloy. The delta and zeta layers were compact, even and very thin. The eta layer had the

fine intermetallic compound seen in the 0.05% titanium coatings. Coating thicknesses

from both alloys were on average about 10% thicker than those obtained from the 0.08%

vanadium alloy.

The 0.04% vanadium plus 0.05% titanium alloy (Figure 9) has an advantage over

an 0.08% vanadium or 0.12% vanadium alloys (Figure 5) since it does not produce

696

24 coatings that are below specification with the low-silicon steels. It also has an economic

advantage since substitution of some of the expensive vanadium with the less expensive

titanium reduces the cost of the alloy. Referring now to Figures 1-3 representing later

trials, it was found that as little as 0.02% vanadium and 0.05% titanium together were

sufficient to effectively control reactivity in up to the 0.05% silicon levels in steel. Alloy containing 0.04% vanadium and 0.02% titanium was also found to perform similarly to the 0.02% vanadium, 0.05% titanium alloy. It was determined that vanadium with

titanium in the amount of at least 0.02% of each vanadium and titanium and at least

0.04%, preferably at least about 0.06%, of a total of vanadium and titanium, will control

ractivity to at least 0.5% silicon in steel.

The invention provides a number of important advantages. Galvanized coatings produced in accordance with the invention are complete and uniform and of desired thickness on low and high silicon steels including steel having silicon content from 0.01

to at least 0.5%. The coalings produced also have a bright metallic lustre. The process can

be easily adapted to conventional galvanizing production equipment using normal galvanizing temperatures and immersion times.

It will be understood, of course, that modifications can be made in the

embodiment of the invention illustrated and described herein without departing from the

scope and purview of the invention as defined by the appended claims.