Pre-treatment of Steel

TECHNICAL FIELD OF THE INVENTION

The present invention relates to the pre-treatment of steel articles, and zinc and zinc alloy surfaces such as galvanised steel, prior to application of one or more paint layers.

BACKGROUND ART

Galvanised steels with organic coatings, such as paint, are of major commercial importance in a number of applications. A steel substrate 1 provides the strength required of the product, and is generally coated with metallic layers 2a and 2b of zinc or zinc alloy on either side. The metallic coating is then covered by pre-treatment layers 3a and 3b, an organic primer layer 4a and 4b containing corrosion inhibiting pigments, and a topcoat 5a and 5b which can be either a second organic paint layer or a laminate film. This is shown schematically in fig 1 .

Organic paints do not adhere to a metallic substrate sufficiently well to meet the demands of the various applications, which require a product that can be formed into a variety of shapes, with no loss of adhesion to the coatings or reduction in the corrosion resistance. Therefore, before the organic coatings are applied, the metal substrate is generally treated with adhesion promoters to improve the bonding of subsequent coatings, and the corrosion resistance of the finished product. The pre-treatment systems presently in use are effective at providing adhesion and corrosion resistance, and typically involve the use of phosphates, chromates, or mixed metal

oxides to bond to the metal substrate. These coatings are then usually exposed to an acidic chromate solution, which forms a second pre-treatment layer.

This system is able to produce a technically satisfactory product, but has the serious disadvantage that some of the chemical agents involved are the subject of significant environmental concern. This applies in particular to the Chromium (VI) compounds. There is a definite prospect that such compounds will be the subject of environmental control legislation.

The present invention is the result of the inventors' efforts to find an alternative pre-treatment process which does not rely on the above environmentally sensitive compounds. The phosphonate group of chemicals have been identified as having potential for use as pre-treatments in this context. This group has the advantage that the molecules can be designed so that they have the ability to form chemical bonds with both the metal 2, and with the organic coating 4. This is shown schematically in fig. 2, where a single layer 3 has been shown for simplicity. The actual coating may be several molecular layers in thickness.

Phosphonates have been used previously within the water treatment industry, such as in the oil and gas fields, where a large amount of brine is produced and hence control of scale and corrosion is required. Research on phosphonates and their complexes in aqueous media using electrochemical and gravimetric methods has been carried out to show that these inhibitors slowed down both the cathodic and anodic reactions on low carbon steel.

A number of patents have been published concerning the use of phosphonates on metallic substrates, including aluminium, steel and galvanised steel. These disclosures discuss adhesion and corrosion resistance. In addition, there are several papers describing the use of

aminophosphonic acids such as nitrilotris(methylene)triphosphonic acid to prevent the conversion of aluminium oxide surfaces to aluminium hydroxide, without reducing the mechanical linkage to the oxide by organic materials such as adhesives and paints.

One example is US-A-4,777,091 which proposes the treatment of steel or galvanised steel with aminophosphonic acid compounds prior to coating with adhesive compositions.

Another example is US-A-4,308,079 which discloses a phosphonate- based pre-treatment for aluminium surfaces. It states that the phosphonate acts as a hydration inhibitor for the aluminium oxide layer, ie preventing conversion to aluminium hydroxide, and therefore allows a good mechanical key to form between the irregular anodised oxide surface and the paint layer.

In the WO 93/20258, a method of treating a nonferrous metal substrate such as aluminium with an activating agent such as HF followed by treating with an organophosphate or organophosphonate is suggested. It recommends that this treatment provides for improved adhesion and flexibility as well as resistance to humidity, salt spray corrosion and detergents of subsequently applied coatings.

When aluminium alloy was anodized in the presence of vinylphosphonic acid (VPA) for 30s or less, Nitowski et al (Proc. Adhesion Soc, 1 995 Feb p24-26) claimed that the oxide formed was covered with a layer of the reaction product of VPA and AI ₂ O ₃ . They postulated that the organic part of the VPA prevents the dissolution of the AI ₂ O ₃ and the layer formed is hydration resistant due to the formation of the hydrolytically stable A1 -OOP bonds. The vinyl group is then available for chemical reaction with the coating. This chemical reaction is not possible with a non-reactive group such as phenylphosphonic acid, which was not as successful.

SUMMARY OF INVENTION

The present invention primarily relates to the complex formable in solution by immersion of a metallic species in a solution of the compound R- PO(OH) ₂ where R includes an organic group, and also to the use of that complex as a coating agent. The metal can be in the native form or as a salt.

The present invention further relates to a solution of the above- defined complex, preferably in the substantial absence of solid metal. This solution can be used as a pre-treatment bath for steel items without causing an initial weight loss that the present invention shows to be associated with phosphonic acid treatments.

The present invention further relates to a method of coating a steel article, comprising the steps of; (i) providing a solution of the compound R-PO(OH) ₂ where R includes an organic group ; (ii) contacting the solution with a metal species or a salt thereof other than the steel article thereby to form a solution of a metal-phosphonate complex; (iii) optionally, removing the metal, if any remains;

(iv) optionally, providing a metallic coating on the steel article;

(v) applying the thus formed complex solution to the surface of the steel article; and (vi) coating the steel article with an organic coating.

A suitable metallic coating at step (iv) is a zinc coating, for example as provided by a galvanising process.

The present invention also proposes the use of the compound R- PO(OH) ₂ where R includes an organic group but not an amino group, or a

metallic complex thereof, as a surface treatment for a steel article prior to application of a paint or other organic layer thereto.

The present invention envisages the production of an article comprising a steel substrate on which is formed, successively, an optional metallic layer such as a zinc or zinc alloy galvanising layer, an intermediate layer, and one or more paint layers, wherein the intermediate layer is formed by immersion of the steel substrate with optional zinc layer in a solution of the compound R-PO(OH) ₂ , or a metallic complex thereof, where R includes an organic group to which the paint layer(s) can bind.

Preferably, the bonding between the R group and the paint layer is by way of a chemical bond.

Finally, the present invention relates to a method of finishing a steel item comprising the steps of;

(i) optionally, metallic coating the steel item;

(ii) applying in aqueous form the compound R-PO(OH) ₂ , optionally in combination with a source of metal cations; where R includes an organic group to which a paint layer can bind; and (iii) applying at least one paint layer.

The metal cations, when provided, can be present through the formation of a complex with the compound.

Preferably, the compound is present in a solution. This solution can be applied to the said steel item by at least substantial immersion, by spraying, or by roller coating, or any combination thereof.

A suitable metallic coating can be provided by a galvanising operation.

The process can be accelerated by applying a suitable current density or voltage to the article.

Preferred forms of the invention according to all the above aspects are ones in which the R group includes at least one epoxy group, or at least one hydroxy group, or at least one unsaturated hydrocarbon group, or at least one amide group, or at least one carboxylic group, or a combination thereof. Such groups bind well with most paints. Suitable unsaturated hydrocarbon groups include alkenes, vinyl groups, or acrylate groups. Other preferred forms are those in which the R group includes an aliphatic chain with which the paint layer can form an interlocking network. In this case, the R group becomes strongly associated with the paint layer but not directly chemically bonded. Such chains should be at least C ₄ long preferably larger than C ₈ and more preferably larger than C, ₂ . Such chains may be substituted, for example with epoxy, vinyl, hydroxy or other groups, or may contain one or more unsaturated regions. Another preferred from for the R group is one in which amino groups are not present.

Further such preferred forms are those in which the solution of the said compound or complex includes an accelerator compound. Preferably, the latter is present in a catalytic amount. Suitable such compounds include peroxides, nitrates, nitrites, chlorates and organic nitro-compounds.

The present invention also relates to an apparatus suitable for treating steel articles by use of the above complex or solution.

BRIEF DESCRIPTION OF DRAWINGS

The present invention will now be described in more detail by way of example, with reference to the accompanying figures, in which;

Figure 1 is a schematic view showing the structure of a typical

organic coated steel product;

Figure 2 is a schematic illustration of the phosphonic acid molecule to which the present invention relates;

Figures 3 to 23 are graphs and SEM micrographs showing the result of investigations into suitable phosphonic acid-based compounds.

DETAILED DESCRIPTION OF EMBODIMENTS

The present invention results from work carried out to investigate the effect of phosphonic acids on a zinc substrate, and to establish their potential as pre-treatments for organic coated steels. The first set of phosphonic acids conform to two groups of chemicals, which are similar, but with one carbon difference in the backbone of the molecule. The first pair are 2-carboxyethylphosphonic acid (2-CEPA) and phosphonoacetic acid (PAA), which are capable of forming bonds with zinc at both ends of the molecule, either by the carboxylic acid group, or by the phosphonic acid group. The second pair are 1 ,3-propylbisphosphonic acid ( 1 ,3-PBPA) , and 1 ,2-ethylbisphosphonic acid ( 1 ,2-EBPA), which are also capable of bonding at either end of the molecule to zinc, by the phosphonic acid groups. The structures are:

HOOC-CH ₂ -CH ₂ -PO(OH) ₂ (HO) ₂ OP-CH ₂ -CH ₂ -CH ₂ -PO(OH) ₂ 2-CEPA 1 ,3-PBPA

HOOC-CH ₂ -PO(OH) ₂ (HO) ₂ OP-CH ₂ -CH ₂ -PO(OH) ₂ PAA 1 ,2-EBPA

The second set of phosphonic acids studied were vinylphosphonic acid (VPA) , nitrilotris(methylene)triphosphonic acid (NMTP) and phenylphosphonic acid (PPA) . The structures of these are:

H ₂ C = CH-PO(OH) ₂ N(CH ₂ PO(OH) ₂ ) ₃

VPA NMTP

C ₆ H ₅ -PO(OH) ₂ PPA

EXPERIMENTAL DETAILS

The zinc foil used was purchased from Goodfellows; thickness, 0.5mm, purity, 99.95 + %, with a typical analysis of (ppm) : Ca 1 , Cd 20, Cu 1 5, Fe1 0, In 1 0, Mg < 1 , Ni 1 , Pb 1 00, Si 2, Sn 8.

2-Carboxyethylphosphonic acid (2-CEPA), phosphonoacetic acid (PAA), vinylphosphonic acid (VPA), nitrilotris (methylene) triphosphonic acid (NMTP) and phenylphosphonic acid (PPA) , were purchased from Aldrich. 1 ,2-ethylbisphosphonic acid (2-EBPA) and 1 ,3-propylbisphosphonic acid 1 ,3- PBPA were synthesised by the well known Arbusov reaction (see Kosalapoff, J. Am. Chem.Soc. 1 944,66, 1 51 1 ) of triethylphosphite with the appropriate dibromoalkane, followed by hydrolysis of the phosphonate using the method of GB-A-1497992, which involves reaction with formic acid, and distillation of ethyl formate.

Anal. Calcd for (C ₂ H ₈ O ₆ P ₂ ) : C. 1 2.6% : H.4.2% found: C.1 2.5%: H.4.5% Η NMR (D ₂ O) : + 4.8 (s,OH) : + 1 .8(d.CH ₂ ) ^{3 l} P{ ¹ H}NMR(D ₂ 0) : + 28.6 Anal. Calcd for (C ₃ H ₁₀ O ₆ P ₂ ) : C. 1 7.6%:H.5.0% found: C.1 7.7%:H.5.4% ' H NMR (D ₂ O) : + 4.8 (s,OH): + 1 .8(m.CH ₂ ) ^ P HjNMRfCDC I _; ,) : + 30.0

A series of experiments were carried out in which pieces of zinc foil were immersed in aqueous solutions of phosphonic acids under various conditions, and then rinsed in distilled water, before drying at room temperature to constant weight. For the weight loss/gain studies, unless otherwise stated, the phosphonic acid solutions were prepared at a concentration of 0.5%w/w using distilled water, and kept in a water bath at 60°C, and the pieces of zinc foil had a total surface area of 1 2cm ² . For the

results in fig 3 and fig 4, only one zinc piece was used for each experiment, which was returned to the solution after each weighing. For the remainder of the work, however, a fresh solution and a fresh piece of zinc foil was used for each result.

The solutions were analysed for zinc content using a Perkin Elmer Plasma 400 Inductively Coupled Plasma Spectrometer. The zinc surfaces were analysed using a Jeol 35C SEM with an Oxford Instruments Link ISIS EDX system and Autobeam attached. The samples were analysed at 20kV, x 550 magnification, and the results quoted are an average of five analyses of different areas of the sample. ^] H NMR were recorded on a Bruker WM- 360 operating at 360 MHz, and ³¹ P{ ¹ H} NMR were recorded on a Jeol FX- 90Q operating at 36 MHz. All NMR spectroscopy data is quoted in ppm. ³¹ P { ¹ H} NMR spectra were referenced to 85% H ₃ PO ₄ (Oppm), and 'H NMR spectra were referenced to TMS (Oppm) . An insert containing triphenylphosphone in deuterated acetone, which gave an observed peak at - 2.81 ppm was used as an external reference to study the position of the phosphonic acid peaks, and to enable semi-quantitive analysis of the solutions. For the semi-quantitive analysis, a pulse delay of 5 seconds was used for 325 scans. This pulse delay was determined to be adequate for comparison of the levels of phosphorous present in the solutions of these compounds, as further increase in pulse delay did not significantly influence the results.

EXPERIMENTAL RESULTS

The change in the weight of the pieces of zinc foil after immersion in aqueous solutions of 2-CEPA at temperatures from 20 °C to 60°C are shown in fig. 3. All of the pieces of zinc foil lost weight, until a certain level of zinc in solution was reached. After this weight loss had occurred, the pieces of zinc foil then gained weight. Both the rate of weight loss and the rate of weight gain increase as the temperature of the solution is raised.

Figure 4 shows the changes in weight that took place when pieces of zinc foil were added to three solutions of 2-CEPA in a water bath at 50°C. The control shows the previously described weight changes. After 7 hours, when the piece of zinc foil in a second solution had started to gain weight, a further portion of 2-CEPA was added. The zinc foil lost weight rapidly, before once again starting to gain weight, at a slightly faster rate than the control. The pieces of zinc foil in the third solution were replaced with fresh samples after 7 and 1 6 hours. Both of the fresh pieces of zinc foil increased in weight after they had been in solution, although the rate of weight gain was slightly slower than that of the control sample.

Other experiments included changing the solutions every hour, but using the same piece of zinc foil, which resulted in continued loss in weight of the sample, giving a straight line graph. The concentration of the solutions was also varied between 0.25 and 0.75%w/w. This concentration range did not influence the time taken to reach the maximum zinc loss of the samples, although it did affect the amount of weight loss, which increased as the concentration of phosphonic acid in solution was increased. Increasing the size of the zinc pieces however, did result in a reduction in the time taken for the maximum loss in weight to occur.

In Figure 5, the changes to the weight of the pieces of zinc foil in solutions of 2-CEPA are shown alongside the zinc content in the corresponding solution. The initial loss in weight of the pieces of zinc foil is mirrored by the increasing zinc content in solution. After two hours there was a reduction in the amount of zinc present in solution, and, the pieces of zinc foil start to gain weight.

The phosphorus content in the solutions of 2-CEPA, measured by NMR, (fig. 6) decreases after 3 hours, but of more interest is the change in the position of the phosphorus peak with time, when compared to the position of the peak due to reference triphenylphosphine (external) . The

peak moves upfield for the first hour, but there is less movement in the second hour, and after four hours there is a slight shift back downfieid.

Analysis of the surface of the pieces of zinc foil by SEM, figs 7 and 8, show that over the first hour there was a reduction in the level of carbon present on the surface, with little change over the following hour, and an increase in the level detected after two hours. The amount of zinc detected increased over the first half and hour, and then stayed at a similar level until after the second hour, when there was a rapid decrease in the amount detected. The levels of phosphorus and oxygen both increased after two hours.

Figure 8 shows SEM micrographs of the zinc surface after varying time in solution. The surfaces showed increased pitting with time spent in solution for the first hour, and flat rectangular platelet type crystals, growing away from the surface, were observed after one hour. The surface coverage by these crystals increased with time, but even after 8 hours, there appeared to be some small areas of the zinc surface visible.

PAA

The weight loss of the pieces of zinc foil in solutions of PAA, fig. 9 is accounted for by a corresponding increase in zinc content of the PAA solutions. There was no obvious increase in the weight of the samples, as there was for those in solutions of 2-CEPA, however, there was a slight increase in weight after 6 hours, although this was of less significance. There was also no detectable reduction in the zinc content of the solutions. The amount of phosphorus in solution measured by ³¹ P NMR, remains relatively constant, although the position of the peak does change, fig. 10. The phosphorus peak moved upfield for the first hour, then it remains in the same position for the next hour, before moving back downfieid, to a position further downfieid than the initial position of the peak.

Analysis of the samples by SEM, fig 1 1 , showed similar results to those for 2-CEPA, with increased pitting of the surface seen as the time in solution was increased. No coating was observed on the surface until the pieces of zinc had been in solution for 6 hours, after which time there was a large increase in the amount of phosphorus detected. The amount of zinc detected after the coating had formed was higher than the amount detected on the samples that had been in solutions of 2-CEPA for the same length of time. The coating completely covered the surface in a flat layer, with small spherical crystal growths randomly dotted on top of the coating. It is possible that the 'crazy paving' appearance of the surface is caused by the electron beam.

1 .3-PBPA

The weight changes of the pieces of zinc foil in solutions of 1 ,3-PBPA, fig. 1 2, are similar to those for PPA, with weight loss occurring over several hours, but then little or no weight gain. Phosphorus was detected on the surface after 2 hours, however, a visible surface coating fig. 1 3 was only seen after 4 hours. The crystal growth was the same as that for PAA, a flat coating with a 'crazy paving' appearance possibly due to damage caused by the electron beam, with spherical growths randomly positioned.

1 .2-EBPA

The pieces of zinc foil in solutions 1 ,2-EBPA, fig. 14 lost weight over the first hour, and then showed little change in weight for the next two hours. After three hours, they started to gain weight, with a corresponding decrease observed in the zinc content of the solutions. Surface analysis, fig. 1 5 showed an increase in the amount of phosphorus detected on the zinc foil after one hour, and small round crystal growths were also observed after this time. The number and size of these growths increased as the piece of zinc foil was in the solution for a longer time. After 8 hours they had merged to form an almost continuous coating, although there were still some

small areas of the zinc substrate visible.

The weight changes with time for the pieces of zinc foil in solutions of the above four phosphonic acids are shown for comparison in figs. 1 6 and 1 7. All show an initially similar rate of weight loss which reaches a maximum and then levels off. The pieces of zinc foil in solutions of 2-CEPA and 1 ,2-EBPA then start to increase in weight after two hours. This weight gain also reaches a maximum and then levels off, with 1 ,2-EBPA showing the biggest increase in weight. For the zinc pieces in solutions of 1 ,3-PBPA and PAA it is not so obvious if any weight increase has taken place. There is possibly a slight weight increase for the zinc pieces in solutions of 1 ,3- PBPA occurs after 3 hours, and for PAA after 7 hours. The pH changes for the solutions of 2-CEPA and PAA (fig. 1 8) were very similar, as they both increased to around pH 4 after three hours.

VPA

Figure 1 6 includes details of the weight changes observed for zinc foil in a VPA solution. Weight loss was observed for the first 5 hours, followed by a period of weight gain.

NTMP

Figure 1 7 includes details of the weight changes observed for zinc foil in NTMP solution. An initial weight loss was observed over the first hour, which then ceased. No further increase or decrease in weight was then detected.

PPA

Figure 1 6 includes details of the weight changes observed for zinc foil in PPA solution. A slight initial weight loss was detected over the first 2

hours but was reversed over the subsequent 3 hours.

Over the concentration range used in these experiments, the first process that occurs with all of the phosphonic acids involves a reaction with the zinc to form a soluble complex. The process continues until a certain level of this complex in solution is achieved, as shown in figs. 7, 9, 1 2 and 14. The formation of a coating on the surface of the zinc substrate can occur concurrently with the removal of zinc from the substrate as was seen for the zinc pieces in solutions of 2-CEPA. A few crystals were observed on the surface after one hour, fig. 8, whilst weight loss continues for another hour. The bulk of the coating formation, however, only occurs after a certain level of zinc in solution has been reached. This was shown by replacing the phosphonic solution every hour, which produced a continual weight loss in the zinc piece, by adding a further portion of phosphonic acid after weight gain had started, which caused a further loss in weight in the sample, and by changing the zinc piece after weight gain had started, which showed an increase in weight with no previous weight loss, fig 4. Therefore, a solution can be produced which starts for form a coating as soon as the substrate is immersed.

When the weight changes that occur to the zinc foil substrates in solutions of the series of phosphonic acids are compared , Figs 1 6 and 1 7 it can be seen that after the initial weight loss, which the samples in all of the acids experience, the pieces of zinc foil in three of the phosphonic acids, 2- CEPA, 1 ,2-EBPA, and VPA then undergo a marked increase in the weight over the time period investigated. For the zinc foil pieces in solution of the other four phosphonic acids, PAA, NTMP, 1 ,3-PBPA, and PPA, no similar large increase in the weight of the samples was observed. When weight loss was experienced by the zinc substrates, there was a corresponding increase in the zinc content in the relevant solution. As some of the substrates gained weight, there was a corresponding reduction in the zinc content of the solution, as shown when the solution used contained 2-CEPA

in Fig 5. The analysis of the zinc surfaces which had been immersed in the solutions of 2-CEPA show the changes in surface composition with time, Fig 7.

Changing the temperature of the solution, fig. 3, results in a change to both the rate of reaction of the phosphonic acid with the zinc, and the rate of coating formation. The time taken to reach maximum weight loss of the zinc piece was not affected by changing the concentration of the phosphonic acid between 0.25 and 0.75%w/w, however, increasing the surface area of the zinc piece did reduce the time taken for this to occur.

There are considerable differences in the time taken for these events to occur, and the type of surface coatings formed, depending on the phosphonic acid used. The phosphonic acids 2-CEPA and PAA, both have a phosphonic acid group at one end of the molecule, and a carboxylic acid group at the other, but 2-CEPA has three carbons in the backbone of the molecule, and PAA has only two. The phosphonic acids 1 ,2-EBPA and 1 ,3- PBPA, have two phosphonic acid groups at either end of the moiecule, and there is also a difference of one carbon in the backbone. All four phosphonic acids therefore have groups capable of forming chemical bonds to zinc at either end of the molecule. ^' If the same structure of salt was formed for both pairs of phosphonic acids, and if just the solubility of the salt determined the coating weight, it would be expected that 2-CEPA and 1 , 3-PBPA would produce the highest coating weights, as they are the largest molecules, but it is 1 , 2-EBPA which shows the biggest increase in coating weight. Another possibility is that the types of coatings formed by 2-CEPA and 1 ,2-EBPA encourage film growth by having readily accessible functions for attachment to either molecules in solution, whereas those formed by PAA and 1 ,3-PBPA do not. The zinc salt of 2-CEPA,

Zn[Zn(0 ₃ PCH ₂ CH ₂ CO ₂ ) ₂ .3H ₂ O] was prepared from zinc chloride, and was structurally characterised. This study found that the structure contains one set of zinc atoms four co-ordinated by oxygen atoms of the phosphonate

groups and another set five co-ordinated by oxygen atoms of the carboxyl groups and lattice water molecules. If the movement of the phosphorus peaks in solutions of PAA was due to chelate formation then this would be likely to produce a different type of coating, and may account for the differences.

Digital images showing the effect of some of the different phosphonic acids on the zinc surfaces are shown in Figs 8, 1 1 , 1 3 and 1 5. The structure of the coating deposited can be split up into two types. The first type is composed of flat rectangular plate-like crystals, which seem to grow away from the surface of the zinc substrate. The second type of coating formed consists of a uniform continuous coverage of the substrate. The phosphonic acids can be broadly grouped into these two types as follows:

Continuous coverage - NTMP, PAA, 1 ,2-PBPA

Flat regular shaped crystals - 2-CEPA, 1 ,2-EBPA, PPA, VPA

It is then a simple matter to adopt a chemical structure based on one or more of the above but which includes a group to which an organic layer such as paint can adhere. Such groups include epoxy groups and/or vinyl groups with which paints can react, or short to medium chain compounds with which paint layers can form an interconnecting network. These latter compounds can be substituted, for example with one or more of the former, and/or contain unsaturated areas.

The present invention also envisages the inclusion of accelerator or catalyst compounds. Such compounds may be able to ensure completion of the reaction quickly enough to allow the invention to be used as part of a continuous process. A suitable compound is a peroxide, which need only be present in catalytic amounts,

An important development allowed by the invention is to form a

solution of the relevant phosphonic acid compound and a metallic species, and to use this complex solution for the pre-treatment step. This would mean that the weight loss experienced by the experimental samples above would be eliminated and the reaction would proceed in place in a correspondingly shorter time.

Using an applied current density as an accelerator in solutions containing a phosphonic acid that produces a flat coating, and sufficient zinc salts to produce a coating, at an appropriate temperature, by varying the pH of the solution, Fig 1 9, it was seen that the most useful pH range was pH 3.5 to 7. Below pH 3.5, dissolution of the substrate occurred, and above pH 7, although the samples achieved a higher coating weight, the coating formed was only loosely adhered to the surface.

To show the effect of applied current as an accelerator, chronopotentiometry was carried out using a solution containing a phosphonic acid that produced a flat coating, and sufficient zinc salts to produce coating formation, within the coating pH range, and at a sufficiently high temperature, the applied current was varied. Within the range used, as the applied current was increased, the coating weight also increased, Figs 20 and 21 .

Using the same method, peroxides and other accelerators can be shown to increase coating weight. The use of a few drops of hydrogen peroxide in 500 mis of a suitable solution, with no other acceleration process, was sufficient to produce a coating which was detectable using a scanning electron microscope after just a 1 0 second immersion of the zinc substrate, Fig 22.

The use of a solution containing a phophonic acid that produced a flat coating, and a sufficient amount of zinc salts, at an appropriate pH and temperature can be used to produce coatings on a range of metallic

substrates. Examples are shown in Fig 23. The substrates used were:

HDG, a steel substrate coated with a layer of zinc by immersion of the steel in molten zinc, giving a surface composed almost entirely of zinc,

EZ, a steel substrate coated with zinc during an electrochemical process, once again giving a surface composed almost entirely of zinc,

IZ, produced by the same process as HDG, however, after coating with zinc, the material is passed through a furnace. The result is that iron diffuses through the entire coating so that the surface is composed of about 90% zinc, 10% iron,

GA, also produced by a hot-dip coating process during which the steel substrate is coated with an alloy containing 95% zinc and 5% aluminium.

By preparing the solution so that it contains the required amount of phosphonic acid and zinc salt, and is at a sufficiently high temperature, and at the correct pH, it is possible to produce a coating that can be analysed using a SEM within 1000 seconds. If accelerators are used, this time can be reduced to less than 10 seconds. It may be possible to reduce time further if the pre-treatments are only required as monolayers, or several layers in thickness.

It is envisaged that the metal-phosphonate complex made possible by the above would be applied as part of a continuous process. Such a process could be carried out on apparatus comprising, in order, a galvanising station, a dip or spray station to apply the complex or solution thereof, and a paint application station.