Inhibition of Localized Corrosion of Hot Dip Galvanised Steel by 1 Phenylphosphonic Acid 2

This item is brought to you by Swansea University. Any person downloading material is agreeing to abide by the terms of the repository licence. Copies of full text items may be used or reproduced in any format or medium, without prior permission for personal research or study, educational or non-commercial purposes only. The copyright for any work remains with the original author unless otherwise specified. The full-text must not be sold in any format or medium without the formal permission of the copyright holder.


Introduction
With current legislative pressure to eliminate Cr(vi)-based inhibitors from protective coating formulations, there is a pressing need to identify effective, environmentallyfriendly alternatives.Chromate-free passivation treatments have been studied extensively [1][2][3][4] and the most widely used alternative to chromate is a phosphate passivating treatment.Such a system facilitates the precipitation of protective zinc compounds via the supply of soluble anions that react with the zinc ions generated in metal dissolution 5 .A phosphate passivation treatment generates a Zn 3 (PO 4 ) 2 barrier to inhibit the anodic process [4][5] , the formation of which can act to extend the pH stability of the surface layer down to values of around pH 6.However, as zinc phosphates are 1000 times more soluble than Cr (III), inferior corrosion inhibition performance is observed 7 .The corrosion of a zinc surface generates a wide range of pH zones, whether in bulk electrolyte conditions, atmospheric conditions or, most notably, under a water droplet 8,9 .Corrosion-driven cathodic coating delamination phenomena produces an alkaline underfilm environment where values of pH 10-11.5    have been recorded previously on a coated zinc surface 10 .In order to compete with an adsorbed chromium (III) oxide layer, an effective corrosion inhibitor system would have to be stable in a range of pH 4-14 7 .It is therefore of great importance to examine potential chromate replacement inhibitors in non-neutral conditions.
In a recent Scanning Kelvin probe study, we showed that in-coating additions of H 2 PP increased the initiation time for underfilm cathodic coating delamination by up to 20 h and reduced the delamination rate by up to 94% on hot dip galvanised steel (HDG) surfaces 11 .An 'etch' effect was shown to be the dominant inhibition mechanism, whereby a reaction occurs at the point of first contact of the wet coating with the substrate producing an interfacial metal phosphate salt layer.We hypothesise that a second mode of inhibition occurs whereby the leaching of dissociated PP 2-ions from the coating into the substrate/coating interface forms a ZnPP layer, blocking the anodic reaction.H 2 PP has been shown previously to strongly adsorb onto surface oxide films, disfavouring chloride 12 .The principal aim of this work is to determine the efficiency of H 2 PP as an inhibitor for HDG surfaces in bulk immersion conditions.
The localised corrosion activity occurring over a commercial grade HDG surface immersed in aqueous sodium chloride electrolyte has been mapped using in situ SVET, which has been widely used in previous corrosion studies [13][14][15][16][17][18][19][20][21][22] .The effect of concentration and pH on corrosion inhibition by H 2 PP and an inorganic inhibitor, namely sodium phosphate (Na 3 PO 4 ), are assessed.A pH of 11.5 has been chosen to mimic the underfilm conditions found in a corrosion-driven delamination cell.Neutral and acidic, pH 2, conditions have also been assessed.

Experimental
Materials: All HDG steel was supplied by Tata Steel UK and comprised of 0.7 mm gauge mild steel coated on both side with a 20 µm zinc layer cut into 50 mm x 50 mm square coupons.All chemicals were supplied by Sigma-Aldrich Chemical Co. and were of analytical grade purity.

Sample preparation:
Prior to each experiment, abrasive cleaning was carried out using an aqueous slurry of 5 µm polishing alumina followed by washing with aqueous surfactant and finally rinsing with distilled water and then ethanol.For SVET and open circuit potential (OCP) experiments, samples were then prepared for immersion by being completely covered using 90 µm thick extruded PTFE 5490 tape (3M Ltd.), leaving exposed only the 9 mm x 9 mm experiment area on one face.All experimental electrolyte was prepared using analytical grade reagents obtained from Aldrich Chemical Co. and distilled water.Solution pH was adjusted by the drop-wise addition of either HCl (aq) or NaOH (aq).
Methods: Measurements using the SVET were carried out using a probe comprising a Pt wire sealed in a glass sheath with a total tip diameter of ~ 225 µm and a 125 µm diameter Pt micro-disc electrode as the active region.A probe vibration frequency of 140 Hz and peak-to-peak vibration amplitude (A pp ), as measured stroboscopically in air, was 30 ± 5 µm.SVET instrument design, mode of operation and calibration can be found elsewhere 23 .Briefly, by Ohm's law, the peak-to-peak SVET voltage signal (V pp ) is related to the current flux density along the axis of probe vibration (j z ) by: where κ is solution conductivity such that a quantity G = κ/A pp may be defined as the

Corrosion inhibition of hot dip galvanised steel by phenyl phosphonic acid in neutral conditions
A baseline for the current density distributions, as a function of time, above the surface of unpolarised HDG samples freely corroding in uninhibited 5% wt/v NaCl 1(f) shows white rust corresponding to the anodic region observed on the current density maps.The principal anodic and cathodic processes that occur on a freely corroding Zn surface in solutions of neutral pH are: Respective current density maps for SVET experiments carried out in electrolyte with H 2 PP concentrations 1x10 -3 mol dm -3 , 1x10 -2 mol dm -3 , and 5x10 -2 mol dm -3 in neutral conditions are given in Figures. 2 and (Figure 1) (Figure 2) (Figure 3) Area-averaged, integrated SVET-derived anodic current density (J at ) data, plotted as a function of time for all H 2 PP concentrations tested, is given in Figure 4(a).The data was obtained by the numerical area integration of j z distributions to give an estimation of time-dependent total local corrosion currents.A progressive decrease in J at with increasing concentrations of H 2 PP can be observed.Using J at values, the total equivalent Zn loss is calculated using: .

. ( )
where F is Faradays constant.Zn loss (gm -2 ) measurements after 24 h immersion are presented in Figure 4(b) showing a progressive reduction of Zn dissolved from the sample with increasing H 2 PP concentration.
Inhibition efficiency (%) of H 2 PP can be calculated using the following equation: where 'mass loss c ' and 'mass loss i ' are the total mass of Zn lost after the 24 h immersion time in control conditions and inhibited conditions respectively.An efficiency of ca.96% is calculated for H 2 PP additions at a concentration of 5x10 -2 mol dm -3 .
(Figure 4) Figure 5 shows data plots for open circuit potential (OCP) measurements of bare HDG in 5% wt/v NaCl (aq) solution at neutral pH where H 2 PP was present in concentrations ranging from 1x10 -3 to 5x10 -2 mol dm -3 .The eventual value of the control result, represented by the dashed line, displays an E corr value near the equilibrium value of Zn.In the presence of H 2 PP, at the lowest concentrations of 1x10 -3 mol dm -3 , no significant effect on the final potential is observed.
(Figure 5) For H 2 PP concentrations of ≥ 1x10 produced by reaction (1) can migrate to the sites of local cathodic activity, combining with free PP 2− anions, to form an insoluble Zn(PP) film thus stifling ongoing interfacial electron transfer.Figure 6(a-c) gives a schematic representation of these processes.
(Figure 6) pH Dependence of a phenyl phosphonic acid inhibitor The inhibition efficiency of H 2 PP added to bulk electrolyte at a concentration of 5x10 - 2 mol dm -3 and adjusted to pH 2 and pH 11.5 is investigated.To obtain baseline information, the corrosion occurring on bare HDG in uninhibited conditions is first characterised.Results from experiments carried out at pH 2 are given in Figure 7.In contrast to the localised corrosion activity observed in neutral conditions, general corrosion activity can be observed.Due to a lack of OH -in the acidic electrolyte, no 'white rust' corrosion product is observed in the photographic image given in Figure 7(d) showing the sample surface after 24 h immersion.Instead, a tarnishing of the surface corresponding to the anodic regions visible in the current density surface maps is evident.In acidic solutions, the active form of corrosion occurring is controlled by the cathodic reaction.This is predominantly hydrogen evolution 9 , as per the following reaction: (Figure 7) Likewise, SVET surface plots for the inhibited case shown in Figure 8 indicate general corrosion by the highly interspersed anodic and cathodic regions.Anodic regions appear to be more intense than the uninhibited case.
(Figure 8) Summary plots of area-averaged, integrated SVET-derived J at vs. time profiles for samples immersed in electrolyte containing no inhibitor, 1x10 -2 moldm -3 H 2 PP and 5x10 -2 mol dm -3 H 2 PP at pH 2 and pH 11.5 are given in Figure 9(a) and (b) respectively.For experiments carried out in pH 2, at the highest H 2 PP concentration, J at is initially less than that of the uninhibited experiment but, after approximately 17 h, J at progressively increases to extremely high values.At the lower H 2 PP concentration very little corrosive activity was observed.Due to the generalised nature of the corrosion at this pH, the accuracy of these findings is in question and will be discussed in more detail later.
(Figure 9) In uninhibited conditions adjusted to pH 11.5, a single cathode and single anode situation prevails as with neutral conditions.However, as illustrated in Figure 10, after a 5 h period of a fixed local peak, anodic activity can be seen to spread in a wave-like manner across the substrate so the anodic and cathodic regions each occupy approximately half the experiment area.The appearance of the sample surface is depicted in the photographic image given in Figure 10(d) where the paler areas correspond with the anodic region shown in Figure 10(a-c).At pH 11.5, there is an increased thermodynamic likelihood that precipitation of an insoluble corrosion product will occur, such as zinc hydroxoxychlorides, resulting in a relocation of the anodic and cathodic sites 9 .
(Figure 10) Inhibited experiments carried out in pH 11.5 show that a strong anodic peak forms on the sample surface then passivates after ca. 6 h (Figure 11).Further anodic activity is shown to occur at ca. 22 h as shown in Figure 11(c).The photographic image given in Figure 11(d) reveals the substantial amount of corrosion product formed on the sample surface.The J at summary plot given in Figure 9(b) shows that, as the H 2 PP concentration is increased, the J at is progressively decreased.However, at the higher concentration of H 2 PP, it can be observed that intense localised anodic activity occurs in between periods of total inhibition.This is shown by the anodic spikes that initiate at immersion times of approximately 2 h and 22 h and remain for approximately 3 h.
We suggest that, in alkaline conditions, H 2 PP acts as an adsorption inhibitor whereby the PP The following equation can be used to calculate the [Zn 2+ ] threshold at which ZnPP will precipitate: where ksp is the solubility product calculated to be 1x10 -6 mol 2 dm -6 through a series of titrations.
Under aqueous conditions, the pH-dependent concentration of PP 2-([PP 2-]) is given by: where [PP] tot is the total concentration of the phenyl phosphonate species.From Equation (10) it can be shown that [PP 2-] = 6.8 x 10 -3 mol dm -3 [PP] tot at pH 7. By applying this value and the ksp value to Equation ( 9) a [Zn 2+ ] threshold can be calculated for the precipitation of solid Zn(PP) where a value of 1.46 x 10 -4 mol dm -3 is obtained.
According to the Pourbaix diagram for Zn 26 , when Zn is immersed in solution of low pH, an abundance of Zn 2+ ions would be present.By incorporating H 2 PP in the same conditions, the HPP -species would be prevalent, according to the pk a value given in reaction (7).As such, it is expected that PP 2-levels would be insufficient to form a solid film despite the level of Zn 2+ ions.This accounts for the drop in efficiency with increasingly acidic pH.At high pH levels, an increased supply of PP 2-would be expected in the electrolyte, according to Equation (8).However, the corrosion product formed at high pH, according to the Pourbaix diagram, is the highly soluble zincate anion (ZnO 2 2-).Therefore, even with sufficient PP 2-, without the presence of Zn 2+ ions it would not be possible for a solid Zn(PP) film to form.As such, a decreased inhibition efficiency of H 2 PP would be expected at increasingly alkaline pH, the results indicate this.
Corrosion inhibition of hot dip galvanised steel by sodium phosphate (10)     Having identified an optimum H 2 PP concentration value of 5x10 -2 mol dm -3 for the effective inhibition of a HDG surface, a comparative investigation assessed the efficiency of a sodium phosphate (Na 3 PO 4 ) inhibitor at this concentration.The photographic image given in Figure 12(b)(ii), taken at 24 h immersion time, shows that only partial inhibition is achieved in neutral conditions.A protective layer is visible over the whole surface but localised corrosive metal attack can be observed in a small, isolated area.This is confirmed by the anodic region shown in the corresponding corner of the SVET-derived surface map of Figure 12(b)(i).Figure 13[a-c] summarises J at vs. time data and it can be observed that values measured at pH 7 (Figure 13(b)) in the presence of Na 3 PO 4 are much lower than those measured in uninhibited conditions throughout the 24 h period.
As with H 2 PP, incomplete inhibition was observed for experiments carried out in alkaline conditions.Figure 12(c)(ii) shows the hydrolysed corrosion product formed on the sample surface after 24 h when immersed in solution at pH 11.5.As with experiments carried out at pH 7, all visible corrosion product is constrained to an isolated region and the remainder of the surface is corrosion-free.However, the SVET surface map shows several local anodes in a region covering half the experiment area.Extremely low initial J at values were recorded in the presence of Na 3 PO 4 (as shown in Figure 13(c)(ii)), this progressively increased to values matching those measured in uninhibited conditions at ca. 15 h and continued to increase to values of ca.3.4 Am -2 over the 24 h period.This was in contrast to the uninhibited experiment where J at values were consistently in the region 1.25 Am -2 for the duration of the 24 h experiment.Unlike the same experiment carried out with H 2 PP, in the presence of Na 3 PO 4 no re-passivation of the surface is observed after the occurrence of anodic activity where J at values return to approximately 0 Am -2 .
No sign of inhibition is observed where experiments were carried out at pH 2 as the J at values, given in 13(a), are substantially higher than that of the uninhibited case throughout the 24 h time period.
(Figure 12) (Figure 13) SVET-derived mass-loss data (Figure 14(a)) calculated using Equation ( 4) indicates that, in all cases, Zn mass loss is at a minimum when experiments were carried out in neutral conditions.In all pH conditions, additions of H 2 PP appear to reduce Zn loss when compared with the uninhibited experiment.H 2 PP additions were least effective in acidic conditions.
(Figure 14) The results suggest that, at pH 7, Na threshold for the precipitation of solid Zn 3 (PO 4 ) 2 is calculated.As described above, phosphorus oxyacids undergo a series of stepwise deprotonations and a progressively higher pK a is exhibited with each step.The first, second and third pK a values are 2.0, 6.9 and 12.3 respectively for Na 3 PO 4 25 .The following equation can be used to calculate the [Zn 2+ ] threshold at which Zn 3 (PO 4 ) 2 will precipitate: where ksp is 9x10 -33 mol 2 dm -6 27 .Under aqueous conditions, the pH-dependent concentration of PO 4 3-([PO 4 3-]) is given by: where [PO 4 ] tot is the total concentration of the phosphate species, which is 5x10 -2 mol dm -3 in this case.Equation ( 11) may be used to show that [PO 4 3-] = 8.8 x 10 -8 .By applying this value and the ksp value to Equation ( 10) a [Zn 2+ ] threshold can be calculated for the precipitation of solid Zn 3 (PO 4 ) 2 where a value of 5.38 x 10 -5 mol dm -3 is obtained.This is lower than the predicted [Zn 2+ ] threshold value for the precipitation of ZnPP, which is 1.46 x 10 -4 mol dm -3 ; this result suggests that Na 3 PO 4 should be the more effective inhibitor.However, it is proposed that the incomplete inhibition observed in this case, where availability of PO 4 3− is not limited, results from precipitation of Zn 3 (PO 4 ) 2 in solution above the sites of anodic Zn dissolution and not directly on the corroding surface.This is represented schematically in Figure 15.
Zn 2+ , therefore, does not migrate to sites of local cathodic activity, as is the case with ZnPP (illustrated in Figure 6(b)), to instigate the deposition of blocking films of Zn 3 (PO 4 ) 2 directly on these regions.This was observed previously by Williams et al where the effect of a phosphate inhibitor on a corroding magnesium surface was assessed 22 .

Gravimetric mass-loss experiments
Results for gravimetric mass-loss experiments are given in Figure 14(b) generally show good correlation with most SVET-derived data.One exception is the uninhibited experiments carried out at pH 2 where a significantly higher value of Zn loss was recorded when compared to the SVET-derived data given in Figure 14(a).In this instance, it appears that the anode-cathode separation distance is less than the 100 µm SVET probe height such that the current flux lines do not cross the plane of scan and are not detected by the SVET 28 .If Zn-loss data in uninhibited conditions at pH 2 has been underestimated by the SVET-derived data, it may be more realistic to suggest that Na 3 PO 4 is simply ineffective at low pH rather than an accelerant of corrosion, as the results from the previous section suggest.Although SVET-derived J at values obtained in uninhibited pH 2 conditions are not truly representative, the observation that a change from localised to general corrosion occurs is still an important result.
Gravimetric mass loss experiments carried out with 5x10 -2 mol dm -3 H 2 PP additions at pH 11.5 also showed significantly higher values (Figure 14(b)) when compared with the SVET-derived data (Figure 14(a)) obtained after 24 h immersion.The J at vs. time profile given in Figure 9(b)iii indicates that the surface re-passivates for 17 h following intense and highly localised anodic activity, as shown in the current density surface maps given in Figure 11(a-c).Samples in the gravimetric mass loss experiment must be immersed in the corrosive environment for one week in order to make discernible weight-change measurements.As such, it is likely that, over this longer time period, the protective barrier is broken down to the point that no further passivation occurs leading to further corrosion activity.The photographic image of the sample surface (Figure 11(d)) after the SVET experiment shows hydrolysed corrosion product covering areas of the sample surface that do not correspond with any anodic features in the current density maps given in Figure 11(a-c).We suggest that either general corrosion is occurring in these areas and anodic activity has gone undetected by the SVET, or corrosion product has been produced from one small and highly intense anodic region and subsequently moved through the electrolyte.In the case of the latter, there would be no reason to suggest that the SVET-derived data is not a true representative of the corrosion activity occurring on the sample surface over 24 h.
Inhibition efficiency (%) values, calculated using Equation 5 are listed in

Acknowledgements
The authors recognise the financial support of TATA Steel UK and the United Kingdom Engineering and Physical Sciences Research Council (EPSRC), Welsh Government and Innovate UK for the SPECIFIC Innovation and Knowledge Centre (grant numbers EP/I019278/1, EP/K000292/1, EP/L010372/1).

List of figures
SVET calibration factor.The SVET calibration was checked galvanostatically in NaCl (aq) electrolyte of different concentrations using a specially devised twocompartment cell.Each compartment contained a 1 cm 2 Pt electrode and the two compartments were linked by a vertically orientated, electrolyte-filled glass tube of length 70 mm and of internal diameter 5 mm.During calibration, the SVET probe was inserted a distance of ca. 5 mm downward into the tube lumen.At this position, the current flux density was constant across the tube diameter and equal to the cell current divided by the internal area of cross section (minus the cross-sectional area of the SVET probe

(
aq) at pH 7 was established using an in situ SVET.Representative current density maps obtained at times between 1 h and 24 h, after initial sample immersion, are shown in Figure.1.It can be observed that corrosion activity occurs with the development of a single anode and a single cathode.For the duration of the 24 h experiment, the large cathode occupies roughly one third of the experiment area and the anode remains localised throughout.The photographic image presented in Figure.

3 .
Although anodic activity is significantly reduced in comparison with the control experiment, for the lower H 2 PP concentrations of 1x10 -3 mol dm -3 , 1x10 -2 mol dm -3 , inhibition is incomplete.This is illustrated in both cases by the presence of a large local anode.However, the anode size and intensity appears to reduce with increasing H 2 PP concentration.Photographic images, given in part Figure2(d) and Figure3(a)ii, show that the Zn surface has been stripped away in a small region corresponding to the local anodic activity on the current density maps.In both cases a protective film can be observed on the non-corroded area of the sample.At a concentration of 5x10 -2 mol dm -3 H 2 PP, complete inhibition was achieved where no corrosion activity can be observed on the current density maps (Figure.3(b)i) and no surface tarnishing is visible in the corresponding photograph (Figure 3(b)ii).

Figure. 1 .
Figure. 1. SVET-derived current density surface maps of unpolarised HDG obtained following immersion in aerated 5% (w/v) NaCl (aq) at pH 7 at time a) 1h b) 2 h, and c) 6 h where d) shows a photographic image of the sample after 24h immersion.

Figure. 2 .
Figure. 2. SVET-derived current density surface maps of unpolarised HDG obtained following immersion in aerated 5% (w/v) NaCl (aq) at pH 7 containing 1x10 -3 mol dm -3 H 2 PP at times a) 30 min b) 6 h c) 24 h where d) is a photographic image of the sample after 24 h of immersion.

Figure. 3
Figure.3.i) SVET-derived current density surface maps of unpolarised HDG obtained following immersion in aerated 5% (w/v) NaCl (aq) at pH 7containing a) 1x10 -2 mol dm -3 and b) 5x10 -2 mol dm -3 H 2 PP at time 20 h where ii) is a photographic image of the sample after 24 h of immersion.

Figure. 5 .
Figure.5.Plot of E corr with respect to time for HDG immersed in aerated 5% wt/v NaCl (aq) electrolyte at pH 7 containing (i) 5x10 -2 mol dm -3 (ii) 1x10 -3 mol dm -3 (iii) 1x10 -2 mol dm -3 H 2 PP.The dashed line represents the eventual value obtained in the absence of inhibitor.Figure.6.Schematic representation of the mechanism of zinc corrosion inhibition by aqueous phenyl phosphonate ions at neutral pH, showing (a) localisation of the early stages of corrosion, (b) phosphonate speciation in the vicinity of the local cathode and (c) the deposition of an insoluble film.

Figure. 7 .
Figure. 7. SVET-derived current density surface map of unpolarised HDG obtained following immersion in aerated 5% (w/v) NaCl (aq) adjusted to pH 2 at time a) 1h b) 14 h, and c) 24 h where d) shows a photographic image of the sample after 24h immersion.

Figure. 9 .
Figure. 9. Area-averaged, integrated SVET-derived anodic current density vs. time profiles obtained for HDG immersed in aerated 5% (w/v) NaCl (aq) with i) no H 2 PP additions and H 2 PP additions of ii) 10 -2 mol dm -3 and iii) 5x10 -2 mol dm-3  where the bulk pH of the experimental electrolyte has been altered to (a) pH 2 and (b) pH 11.5.Figure.10.SVET derived current density surface map of unpolarised HDG obtained following immersion in aerated 5% (w/v) NaCl (aq) at pH 11.5 at time a) 1 h b) 6 h, c) 14 h where d) shows a photographic image of the sample after 24 h immersion.

Figure. 11 .
Figure.11.SVET derived current density surface map of unpolarised HDG obtained following immersion in aerated 5% (w/v) NaCl (aq) at pH 11.5 containing 5x10 -2 mol dm -3 H 2 PP at (a) 2 h, (b) 3.5 h, c) (e) 22 h where (d) shows a photographic image of the sample after 24h immersion.Figure.12.i) SVET-derived current density surface maps of unpolarised HDG obtained following immersion in aerated 5% (w/v) NaCl (aq) containing 5x10 -2 mol dm -3 Na 3 PO 4 after 24 h and ii) a photographic image of the sample after 24 h immersion where the bulk electrolyte was adjusted to a) pH 2, b) pH 7 and c) pH 11.5.

Figure. 15 .
Figure. 15.Schematic representation of a locally corroding HDG surface in the presence of phosphate ions.
).Furthermore, the current flux was aligned vertically i.e. parallel with 2-anion adsorbs onto the HDG surface and reinforces the Zn(OH) 2 layer.This layer disfavours Cl -adsorption preventing attack.The periods of intense, highly localised, anodic activity indicate that this layer is only partially protective.
Table1for both SVET-derived and gravimetric mass-loss values measured in p H 2, pH 7 and pH 11.5.The SVET-derived results show an efficiency of 96.3% for experiments carried out in neutral conditions and, as would be expected, the general trend is for efficiency to be reduced where pH is above and below neutral.Table.1.Values of inhibition efficiency when bare HDG samples are added to a 5% (w/v) NaCl (aq) electrolyte adjusted to pH 2, pH 7 and pH 11.5 containing additions of 5x10 -2 mol dm -3 H 2 PP or 5x10 -2 mol dm -3 Na 3 PO 4 .PP efficiency is shown to be higher than that for sodium phosphate (Na 3 PO 4 )where an SVET-derived efficiency of 75% was calculated for additions at the same concentration.The substantially lower inhibition efficiency of Na 3 PO 4 suggests that precipitation of Zn 3 (PO 4 ) 2 does not occur directly on the corroding surface but rests in solution in the regions above.(d)H 2 PP is shown to be less effective as a corrosion inhibitor in non-neutral conditions.At low pH, an abundance of Zn 2+ ions are present but the concentration of PP 2-is insufficient to form a solid film.At high pH a soluble zincate anion (ZnO 2 2-) tends to prevail.Despite an abundance of available PP 2-anions, Zn 2+ levels are insufficient to combine with PP 2-anions to establish a homogenous protective layer.