The effects of lead on the solidification and preferred ...

Lehigh UniversityLehigh Preserve

Theses and Dissertations

1994

The effects of lead on the solidification andpreferred orientation of the zinc coating oncontinuously hot dipped galvanized sheet steelRichard E. FraleyLehigh University

Follow this and additional works at: http://preserve.lehigh.edu/etd

This Thesis is brought to you for free and open access by Lehigh Preserve. It has been accepted for inclusion in Theses and Dissertations by anauthorized administrator of Lehigh Preserve. For more information, please contact [email protected].

Recommended CitationFraley, Richard E., "The effects of lead on the solidification and preferred orientation of the zinc coating on continuously hot dippedgalvanized sheet steel" (1994). Theses and Dissertations. Paper 257.

http://preserve.lehigh.edu?utm_source=preserve.lehigh.edu%2Fetd%2F257&utm_medium=PDF&utm_campaign=PDFCoverPages

http://preserve.lehigh.edu/etd?utm_source=preserve.lehigh.edu%2Fetd%2F257&utm_medium=PDF&utm_campaign=PDFCoverPages

http://preserve.lehigh.edu/etd?utm_source=preserve.lehigh.edu%2Fetd%2F257&utm_medium=PDF&utm_campaign=PDFCoverPages

http://preserve.lehigh.edu/etd/257?utm_source=preserve.lehigh.edu%2Fetd%2F257&utm_medium=PDF&utm_campaign=PDFCoverPages

mailto:[email protected]

AUTHOR: 7

Fraley, Richard E., Jr. .

TITLE:" "- The Effects of Lead on the

/

"

Solldlficat-ion'-andPreferred

Orientation of the Zinc/

Coating on continuously

Hot dipped Galvanized

Sheet Steel

DATE: May 29,1994, ,

The Effects of Lead on the Solidification and Preferred Orientation of the Zinc Coating

on Continuously Hot Dipped Galvanized Sheet Steel

by

Richard E. Fraley, Jr.

A Thesis

Presented to the Graduate and Research Committee

of Lehigh University .

in Candidacy for the Degree of

Master of Science

ih

Materials Science and Engineering

Lehigh University

April 6, 1994

Acknowledgements

I would like express my gratitude and appreciation to my research advisor, Dr. A. R..Marder, for his support and encouragement throughout this investigation. I also wish to

thank Ms. Cathy Jordan for the extra effort she put into the Electron Microprobe

Analysis. Mr. George Rommal and Mr. Tom Suchy produced the high quality samples

..which made this study possible. The financial support for this program was provided by

the Bethlehem Steel Corporation through its Tuition Assistance Program.

iii

Table of Contents

Abstract

Background

The zinc-lead systemSolidification of thin zinc coatingsCrystallography of zincPreferred orientation of bulk zinc castingsPreferred orientation of thin zinc coatings

Experimental Procedure

Sample preparationChemical analysis and coating weightsElectron Probe MicroanalysisSpangle size measurementsZinc inverse pole figure analysis

Results and Discussion

Chemical analysis and coating weightsElectron Probe MicroanalysisSpangle size measurementsZinc inverse pole figure analysis

Conclusions

References

Tables

Figures

Biography

iv

Page 1

Page 2

Page 19

Page 25

Page 34

Page 36

Page 38

Page 48

Page 107

List of Tables

Table 1. Processing Parameters for the Eleven Samples.

Table 2. X-ray Diffraction Instrument Parameters.

Table 3. Calculated Intensities for Zinc with Chromium Radiation

Table 4. Sample Calculation of Normalized Intensity Ratio for a Random Zinc PowderSample.

Table 5. Results of ZIPF Verification with Random Zinc Powder.

Table 6. Chemical Analysis of the Primary Zinc.

Table 7. Chemical Analysis of the Primary Lead.

Table 8. Chemical Analysis of the Steel Substrate Coils.

Table 9. Zinc Coating Weights.

Table 10. Chemical Analysis of the Zinc Pot and Coating.

Table 11. Results of Spangle Size Measurements.

Table 12. Results of XRD ZIPF Analysis for the Eleven Coils.

Table 13. Average Normalized Intensity Ratios for the Eleven Coils.

v

'-

Figure 1.

Figure 2.

Figure 3.

Figure 4.

Figure 5.

Figure 6.

Figure 7.

Figure 8.

Figure 9.

Figure 10.

Figure 11.

Figure 12.

Figure 13.

Figure 14.

Figure 15.

Figure 16.

Figure 17.

Figure 18.

Figure 19.

Figure 20.

Figure 21.

List of Figures

The effect of lead on the surface tension of liquid zinc [8].

·The effect of lead on the viscosity of liquid zinc [8].

Plot of crystal nucleation totp.1 free energy [9].

Crystal cap and wetting angle, 8, for heterogeneous nucleation [9].

The effect of aluminum on the surface tension of liquid zinc [8].

The effect of aluminum on the viscosity of liquid zinc [8].

The zinc-lead phase diagram [12].

Schematic of constitutional supercooling [9].

The zinc-bismuth phase diagram [12].

The effect of aluminum and lead on spangle size [3].

Correction line estimation of soluable aluminum [19]."

Effect of soluable bath aluminum on coating aluminum [20].

Plot of grain size versus alloy concentration for zinc-lead and zinc-binaryalloys [4].

The first six zinc planes and the angle that the plane makes with the basalplane.

-Mirror-like, dimpled, feathery, and ridged spangle morphologies [28].

Plot of area percentage basal texture versus grain size [5].

Schematic of 6" Continuous Hot Dip Galvanizing Pilot Line at BethlehemSteel's Homer Research Laboratories.

Formation of continuous and discontn,uous cones of diffracted intensity [25]and the,effects of sample spinning. . .

Schematic of sample spinning stage.

XRD scan of a random zinc powder sample.

XRD scan of a sample with approximately four times random basal planes.

vi

FigtJre 22. Plot of bath lead versus coating iron.

Figures 23-33. LOM photomicrographs of the Fe-Zn alloy layers of the eleven. samples.

Figure 34. EPMA line scans of the coating of Sample 15.

Figure 35. EPMA electron image and lead map of a lead particle in the Sample 15coating.

Figures 36-37. EPMA element maps and electron image of Sample 15 coating.

Figures 38-39. LOM photomicrographs of representative spangle morphologies.

Figure 40. Plot of spangle size versus coating lead content.

Figure 41. Plot of spangle size versus lead content according to Kim et. al. [3] andFasoyinu et. al. [4] combined with the data from this investigation.

Figures .42-44. Planar EPMA photomicrographs and element maps of Samples 1, 9, and15.

Figure 45. Plot of coating iron versus spangle size.

Figures 46-51. Results of XRD ZIPF analysis versus lead level by zinc plane.

Figures 52-57. Averaged XRD ZIPF analysis results for ~~ch lead level.

Figure 58. Plot of basal planes versus spangle size.

Figure 59. Plot of basal planes versus coating iron content.

vii

---_.- - ~ _._------~-----'-- ~-

The Effects of Le(id on the Solidification and Preferred Orientation of the Zinc Coating

on Continuously Hot Dipped Galvanized Sheet Steel

Abstract

A series of eleven steel coils were continuously hot dip galvanized on Bethlehem Steel's

6 inch pilot line. The lead level of the zinc baths was varied from 0.002 to 0.1 O%l:>b so

that the effects of lead additions on the solidification and preferred orientation of the

galvanized coatings could be investigated. The samples were investigated by Chemical

Analysis, Electron Probe Microanalysis, Light Optical Microscopy, and X-ray Diffraction

Inverse Pole Figures. Lead levels greater than 0.04%Pb resulted in larger spangle sizes

because lead additions to liquid zinc significantly reduce the surface energy of liquid

zinc, thus inhibiting the heterogeneous nucleation of zinc crystals. Larger spangle siz,es

were also associated with heavier Fe-Zn alloy layers (coating iron contents greater than

1% Fe) which indicates that the Fe-Zn alloy layer also inhibits zinc crystallization. It is

proposed that. the heat of formation of the Fe-Zn phase remelts some zinc nuclei. The

study showed that lead levels in excess of 0.04% Pb in the galvanizing bath promoted

the formation of the Fe-Zn alloy layer by interfering with the formation of the Fe-AI

inhibition layer. Therefore, higher aluminum levels may be necessary to inhibit the

formation of the Fe-Zn alloy layer in leaded galvanizing baths. The formation of the Fe-

Zn alloy layer tended to randomize the preferred orientation of the zinc coating and,

lastly, the preferred orientation was not dependent on the spangle size or lead content

of the zinc coating.

Background

The purpose of this study was to investigate the effects of lead on the solidification and

preferred orientation of the zinc coating on continuously hot dipped galvanized sheet(

steel. Sheet steel has been coated with zinc since the mid-1 800's to protect it from

environmental attack. Zinc coatings protect steel two ways:

1) Barrier protection - the zinc coating separates the steel sheet from the

corrosive environment.

2) Galvanic protection - Iron is cathodic to zinc, therefore, the zinc coating

corrodes sacrificially to the steel sheet.

Until the mid-1970's, most primary zinc was refined in horiz~tal retort smelters. Zinc

produced in horizontal retorts was called Prime Western Zinc and contained a significant

level of lead as an impurity. The smelters were environmentally destructive and have

since been legislated out of existence. The current supply of primary zinc is

electrolytically'refined and contains very little lead.

The most prominent effect of lead in the zinc bath are the large zinc grains (spangles)

which develop when lead is present in liquid zinc. Traditionally, coaters have promoted

the bright well defined spangled appearance of their products as a mark of high quality

and a spangled appearance readily distinguishes hot dip galvanized products from other

coated and uncoated products. However, the spangled appearance is undesirable in

"painted applications because the spangle texture and pattern tnay show through the

paint. Since the elimination of Prime Western Zinc as source of primary zinc in the

1970's, hot-dippers have added lead to the zinc bath when a spangled finish is desired.

2

Further. environmental pressure against lead is forcing hot-dippers to remove lead from

zinc coatings altogether and some coaters have substituted antimony or bismuth for

lead ·in coating applications where a spangled appearance is desired. Antimony and

bismuth both promote the formation of spangles. It has become necessary to

-investigate the effects of lead on the preferred orientation and solidification of

'~ .

continuously galvanized zinc coatings in order to establish a baseline for future

comparisons of the effects of antimony, bismuth, and lead on the properties of zinc

coated steel sheet. Previous studies leg. 1-7] have utilized laboratory dipped and/or

commercially produced material for samples. The studies using laboratory dipped

samples could not replicate the commercial coating conditions of modern_hot dip coating

lines which continuously coat sheet steel at speeds up to several hundred feet per

minute. The studies that utilized commercially produced material have poorly

.characterized the sample material· with regard to production conditions and coating

I

characteristics. The samples used in this study were produced on the 6" pilot line at,

Bethlehem Steel's Homer Research Laboratorie$. The pilot line replicates commercial

coating conditions in a controled laboratory enviroment.

The Zinc-Lead System

Thresh [8] determined that lead significantly decreases the surface tension of liquid zinc

(Figure 1) and has little effect on the viscosity of liquid zinc (Figure 2). The surface

tension-of a Zn-0.9% Pb alloy is 30% less than pure liquid zinc, 530 dynes/em vs 770

dynes/em respectively. The effect of lead on the surface tension of liquid zinc is very

important because surface tension, which is proportional to surface energy, influences

crystal nucleation [9,10]. Homogeneous nucleation is controlled by the sum of the

positive free energy resulting from the creation of a surface and the negative free energy

3

resulting from the creation of a volume. The total free energy plot for nucleation is

shown in Figure 3. Two important features of the total free energy plot are the free

tnergy barrier, ~Gi * and critical radius, r*. Atom clusters smaller than r* are unstable

and redissolve while larger clusters continue growing to form crystals.

Equations 1, 2, and 3 are the mathematical expressions for ~Gi *, r*, and the rate of

homogeneous nucleation, I :

~Gi* = (16" cr Tm v;-1/ (3 ~H2 ~T2)

r* = (-2 aTm V.) / ( ~H ~T )

where: a = surface energyTm = equilibrium melting pointV. = molar volume~H = molar change in enthalpy~T = undercoolingDL = liquid diffusion coefficientDLm = liquid diffusion coefficient at Tm

B1 = constant

(1 )

(2)

(3)

Because the surface energy is· cubed in the expressions for ~Gi * and I, decreasing the

surface tension of liquid zinc greatly decreases ~Gi * and increases I. Thus,·

homogeneous nucleation of zinc crystals is promoted by the presence of lead which

ch!creases the surface tension of liquid zinc.

Heterogeneous nucleation is similar to homogeneous nucleation except that the

formation crystal nuclei is enhanced by the presence of a catalytic surface. The extent

to which nucleation is enhanced is controlled by the contact angle between the crystal

and the substrate, 8, which is a measure of how well the crystal wets the substrate,

as shown in Figure 4. Improvements in the crystal wetting the substrate decrease the

wetting angle and enhance heterogeneous nucleation. The enhancement is a function

4

fL·

'of 8 given by equation 4:I

f( 8 ') = 1/4( 2 +cos8 )( 1-cos8 )2 (4)

Equations 5 and 6'are the free energy barrier, bGci * and nucleation rate, IcJor

heterogeneous nucleation:

(5)

At equilibrium, the sum of the surface energies in the plane of the substrate is equal to

zero:

aCL + asc + ( cos8 )aSL = 0 (7)

Assuming that aCL and aSC remain constant, decreas,ing aSL will increase the contact

angle, 8, between the crystal and the substrate. The addition of lead to liquid zinc

decreases the surface ten'sion of liquid zinc, increases the contact angle between .

crystalline zinc and substrate, and inhibits the heterogeneous nucleation of zinc

crystals.

From a practical standPoi~ the addition of lead to liquid zinc should enhance the

wetting of the steel by the galvanizing bath. Sebisty et al. [1] conducted an

investigation of the effects of lead and aluminum additions to the galvani~ing bath. One

of their findings was that prominent bare spots developed in coatings dipped in 0.20%

AI baths and that the bare spots were alleviated when 0.5% or 1.00;.0 Pb was added to

the bath. Renaux et at. [11] corrobora!eg Sebisty et al.' s finding that 0.20% AI-Zn

baths do not wet steel well with their findings that the wetting force between an AI-Zn

bath and the steel substrate is a minimum at 0.2% AI. Renaux et al. partially attributed

this finding to the effect of aluminum on the surface tension of liquid zinc, however,

Thresh [8] showed that aluminum has little effect on the surface tension of liquid zinc

5

(

(Figure 5) but does effect the viscosity of liquid zinc (Figure 6).

The zinc-lead equi~ibrium phase diagram is shown in Figure 7 [12]. The melting point of

pure zinc is 419.8C. A prominent feature of the phase diagram is the miscibility gap

above 41 7. 8C from 0.9 to 98 % Pb. Liquid zinc does not dissolve liquid lead in this

region. The zinc-lead system has a monotectic point at 417 .8C and 0.9%Pb. At the

monotectic, liquid zinc solution transforms to pure solid zinc and a liquid 98%Pb lead

solution. In the range 0-0.9% Pb, lead is soluble in liquid zinc but. not in solid zinc.

Constitutional supercooling [13] is possible in systems such as zinc-lead where the

components are soluble in the liquid phase and have limited solubility in the solid phase.

Constitutional supercooling is an important mode of dendritic-solidification in the linc-

lead system. A diagram depicting constitutional supercooling is shoyvn in Figure 8. As

a zinc-lead liquid solution in the range 0-0.9%Pb solidifies, the first phase to· crystalize

will be pure zinc. Because lead is insoluble in solid zinc, it is rejected into the liquid

alloy, causing the enriched liquid to have a lower melting temperature then the original

alloy. Equation 8 is the mathematical expression for solute redistribution:

9L = Co (1 + (( '1-k) I k) exp{(-Rx' / DL)])

where: CL = composition of liquid as f( x' )Co = composition of original liquid alloyk = C·S/C·L

R = rate of movement of interfacex' = distance from in~erface

(8)

Depending upon the prevailing temperature gradient, the melting point of the enriched

liquid may be higher than the actual temperature of the bath allowing for a supercooled

liquid. This condition destablizes the solid-liquid planar interface resulting in the growth

of dendrites. The mathematical expression for interface stability is given by equation 9:\

6

(9)

where: GL = actual temperature gradient at the interface- mL = slope of the liquidus line

Therefore, decreasing k or increasing Co promotes dendritic growth by destablizing the

solid-liquid interface. Because lead is insoluble in solid zinc, k is very small and dendritic

solidification caused by constitutional supercooling can be expect~d in zinc lead alloys.

Solidification of Thin Zinc Coatings

Cameron et al. [2] conducted an extensive study of the solidification and formation of

spangles in' thin zinc coatings on steel. The study concentrated On baths containing

0.5% and 1.3% lead with tin rangin9 from 0 to 0.5%. Commercial cold rolled'and

annealed capped steel panels (4 in. X4 in., 0.020 and D.024 in. thick) were dipped into

the 440C zinc baths for 15 seconds and the results showed that lower cooling rates

produce larger spangles and that lead additions result in large spangles. ,Cameron et al.

also stated that bismuth additions result in large spangles although they did not actually

experiment with bismuth. The bismuth-zinc phase diagram [12], Figure 9, is similar to

the lead-zinc phase diagram. Both contain monotectics and miscibility gaps and both,-lead and bismuth dissolve in liquid zinc but not in solid zinc. Cameron et al. stated that

these elements coarsened the coating structure by inhibiting nucleation (however, they

did not elaborate on a mechanism). They also indicated that an auxiliary alloy element

was required, such as tin or aluminum, that had some solubility in both liquid and solid.

zinc. Cameron et al. po~tulated that the auxiliary alloy causes a metastable retention of

lead in solid zinc and the large lead atoms distort the zinc crystal lattice resulting in the

formation of subgrains and the spangled appearance. Cameron et al. also determined

that spangles grow dendritically in the < 100> direction based on Laue x-ray diffraction

(XRD) and that the primary arms were parallel to the plain of the sheet.

7

Cameron et al. [14] asserted that spangles nucleate at the zinc/steel interface based on

an experiment in which steel discs were dipped into molten zinc and centrifuged before

the coating completely solidified. Few details of the experiment were provided other

than that the disc diameter was 2 in. and the bath contained 0.5% Pb add 0.2% AI.

As a result of centrifuging the partially solidified coating, the liquid zinc overlay was

spun away leaving behind only the solid nuclei which were attached to the steel. The

results showed that at least some nucleation occurs at the zinc/steel interface thus

indicating that the nucleation of zinc crystals is heterogeneous.

Kim et al. [3] characttrized the solidification of thin zinc coatings on steel by dipping

panels in molten zinc baths with various Pb and AI contents. The lead range studied

was 0 -1.0% Pb and the resulting spangle sizes ranged from 0.5 to 3 mm as shown in

Figure 10a. The panels were 0.90mm thick AISI 1008 plain carbon steel which were

held at 954C for seven minutes, cooled for 1 minute, and dipped for 5 seconds in the

zinc bath which was at 454-460C. The AI content of the bath was 0.2% and the

experiment was conducted in a 20%H2-80%N2 controlled atmosphere. The surface

temperatures of the coated panel were monitored by an infra-red temperature sensing

device and Kim et al. reported that a significant supercooling, 23C below the melting

point of zinc, o'ccurred during solidification. The actual cooling rate for the experiment

was not reported but an approximate rate, 25C/sec, may be inferred from the total

cooling time after hot dipping, 15-17 seconds. Increasing the cooling rate to 60C/sec

,did not significantly effect the spangle size of the 0.2% AI- 0.1 % Pb panels. Quenching

the samples in water did have a significant refining effect on the structure. Kim et al.

- explained the effect of lead on spangle size by stating that lead additions to a molten

zinc bath lower the surface tension of liquid zinc and inhibit nucleation.

8

Kim et al. [3] also investigated the effect of aluminum in the absence of lead on the

spangle size. They reported a slight reduction in spangle size over the range of 0-0.20%

AI (Figure 10b). This is a very minor effect compared to aluminum's influence on the

brightness of galvanized coatings and its ability to inhibit the growth of iron-zinc

intermetallics. Aluminum additions brighten zinc coatings through the formation of a

1O's em thick surface film of AI20 3 [15]. The AI20 3 film prevents oxygen from

contacting the zinc coating, therefore the coating does not oxidize and remains bright.

The brightening effect is accomplished with aluminum levels on the order of 0.02% AI.

Significantly higher aluminum levels, approximately 0.15 % AI, are required to inhibit the

formation of iron-zinc intermetallics. Ghuman and Goldstein [16] found that the iron

zinc reaction was inhibited by an Fe-AI-Zn ternary compound which transformed to a

structure isomorphous with Fe2AI6 as it became richer in aluminum. Urednicek et al.

[17] asserted that Fe2AIs and FeAI3 form on the steel surface and prevent the formation

of the iron-zinc intermetallics. The effect is temporary, however, because zinc diffuses

through the iron-aluminum causing FeZn7 to form. It is important to recognize that the

total aluminum of the bath is not available for control of the iron-zinc intermetallic

. reaction. Some of the total aluminum will comoine with iron in the zinc bath to form an

Fe-AI-Zn ternary compound so that only the excess aluminum will be effective for plloy

layer control. Tang et al. [18] calculated that when the total aluminum content of the

bath is less than 0.14%, Fe-Zn intermetallic compounds are stable and the total bath

aluminum is equal to effective aluminum. When the total bath aluminum is greater than

0.14%, the bath aluminum must be corrected for Fe-AI-Zn ternary compounds to obtain

the effective aluminum. In order to correct the values -of total bath aluminum for the

contribution of Fe-AI-Zn intermetallics, the level of effective aluminum. can be estimated

9

following the procedure described by Gagne et al. [19]. The procedure assumes that

the composition of the intermetallics is 0.45 AI - 0.35 Fe - 0.20 Zn and utilizes

calculated curves of the solubility of iron in zinc, as a function of aluminum, for various

bath temperatures. To estimate the level of soluable aluminum:

1) Plot the actual pot analysis on the iron versus aluminum solubility curve asshown in Figure 11.

2) Construct a line with a slope equal to 0.78 (% Fe = 0.78 %AI) through theactual analysis.

3) The intersection of the correction line with the iron solubility line appropriateto the bath temperature gives the level of soluable aluminum in the bath.

Belisle [20] showed that at least 0.14% effective aluminum in the bath was necessary

to inhibit Fe-Zn alloy layer formation by plotting coating aluminum versus effective

aluminum as shown in Figure 12. The rapid increase in coating aluminum at 0.14%

effective aluminum is indicative of inhibition of iron zinc intermetallic formation and is in

good agreement with Tang et al.'s calculation.

Fasoyinu et al. [4,21] conducted an extensive investigation on the solidification of zinc

on steel sheet. Pregalvanized 1008 steel panels (8 in. X 3in. X 0.20 in.) were dipped in

varying zinc baths at 475C in order 10 study the effects of bath composition, melt

undercooling, cooling rate, alloy surface tension, and segregation. The cooling rates

were on the order of 4C/sec. Fasoyinu et 81. determined that melt undercooling was

negligible, Le. less than 1C, from thermocouples that were attached to the panels as the

zinc overlay solidified. These results on melt undercooling contradict those of Kim et al.

[3] who found a 23C undercooling. Fasoyinu et al. suggest that the difference is the

result of the temperature measurment technique. Kim et al. used an infra-red pyrometer

and the emissivity fluctuate_d as the overlay solidified which caused significant errors in

10

their measured te~peratures. Fasoyinu et al. state that based on their results, i. e.

there is no bulk undercooling of the liquid zinc overlay during solidification, spang!e

growth is not driven by large melt undercoolings. This finding eliminates thermal

dendrites as 'a possible solidification mode for the zinc overlay, Fasoyinu et al. also

. showed that the thermal gradient from the steel to the zinc-air surface is flat. The heat

flow between the steel and the zinc is by metallic conduction, thus resistance to heat

flow is low, and the thermal gradient is minimal. With regard to spangle nucleation,

Fasoyinu et al. proposed that spangles nucleate heterogeneously within the melt and not

necessarily at the steel or air surface. However, what the heterogeneous particles might

be was not indicated. Their nucleation hypothesis was based on autoradiography

examinations of several large spangles which indicated no overall segregation of

radioactive thallium ( 204TI) to the steel or air interfaces. Furthermore, they discounted

the idea that lead and bismuth produce large spangles by inhibiting nucleation and

stated that it appeared unlikely that lead and bismuth chemically dissolve the nuclei in

the bath. Fasoyinu et al. overlooked the influence of lead on the surface tension of

liquid zinc and the heterogeneous nucleation of zinc crystals.

Fasoyinu et al. proposed that a spangle's primary dendrite arms grow briefly in random

directions in the liquid zinc overlay until they meet a steel/zinc or zinc/air surface and

that the presence of the surface favors growth, therefore·the dendrites grow along the

steel/zinc or zinc/air surfaces. Fasoyinu' et al. draw a parallel between zinc and ice

which are both hexagonal close packed crystal structures. Fasoyinu referenced

Lindemayer [22] who showed that the growth of ice was faster on a surface.

Lindemayer attributed the faster growth to the surface acting as a heat sink, but, by

Fasoyinu's own calculations, the steel is in thermal equilibrium with the molten zinc and

11

\.

is not ameat sink. Therefore, the growth of zinc dendrites along the steel/zinc or

zinc/air surfaces may not be enhanced as proposed by Fasoyinu et al.

Fasoyinu et al. observed that larger spangles were associated yvith alloy solute additions

which have limited solubility in solid zinc and relatively low surface tensions. Steel

panels were dipped in zinc baths containing 0.2% additions of lead, bismuth, antimony,"

magnesium, cadmium, and tin. The additions with the lower surface tensions, i. e. lead,

bismuth, and antimony (average elemental surface tension of 420 dynes/em) produced

large spangles while the baths co.ntaining magn~sium, cadmium, and tin (average

elemental surface tension of 580 dynes/em) did not. Fasoyinu et al.'s findings for lead

and antimony are reproduced in Figure 13. All six additions are soluble in liquid zinc ana

have limited or no solubility in solid zinc so they are conducive to constitutional

supercooling. The absence of large spangles when magnesium, cadmium, or tin are

added to liquid zinc caused Fasoyinu et al. to conclude that constitutional supercooling

by itself can not account for the production of large spangles. Fasoyinu et al. proposed

that the alloy solute addition rej~cted at the dendrite tip during solidification results in

lower surface tension, sharper tip radius, and faster dendrite velocities. Dendrite

\velocities were calculated by Fasoyinu et al. in accordance with a model developed by

Nash et al. [23]. The dendrite velocities for the additions which result in large

spangles; Zn-Pb, Zn-Bi, and Zn-Sb were three times faster than for Zn-Mg and Zn-Cp;

1OX1 0-2 vs 3X10-2 em/sec, respectively. The dendrite velociW.-9J Zn-Sn, an addition

which results in small spangles, was comparable to the large spangle group, 11 X1 0-2

em/sec. Fasoyinu et al. did not explain the spangle behavior of Zn-Sn.

Fasoyinu et al.'s results could be explained in terms of the basic theories of

12

constitutional supercooling and heterogeneous nucleation which have been previously

discussed. Magnesium and cadmium have some solublityin solid zinc where as lead,

bismuth, antimony, and tin are completely insoluablein solid zinc, therefore the values

of k (C s */C,*ffor magnesium and cadmium are greater than for lead, bismuth, antimony,

and tin. Because a stable interface is more likely as k increases from equation 9,

dendritic growth is less likely to occur in Zn-Mg and Zn-Cd coatings which would result

in smaller grains for these systems. Tin is completely insoluable in solid zinc so

-<den~ritic growth is expected. No data is available for the effect of tin on the surface

tension of liquid zinc, however, if tin does not lower the surface tension of liquid zinc

and inhibit heterogeneous nucleation as does lead, the result would be small spangles

like Fasoyinu et al. observed.

Zinc Crystallography

This section provides background information about the crystallography of zinc because

familiarity with the crystallography of zinc is necessary to understand its preferred

orientation. The crystal structure of zinc is HCP. The lattice parameters are a =2.665

and c =4.947 with a cia ratio of 1.856 [24]. Because the ideal cia ratio for an HCP

arrangement of spheres is.l .633 and there is no reason to believe that the atoms are

not in contact, the zinc atoms must have anis'otropic shapes [25]. This suggests that a

high degree of anistropy might be expected in the properties of zinc.

Figure 14 is a sketch of the zinc unit cell. Six principle atomic planes are shown. The

(001) plane is. the most densely packed and is called the basal plal')e. The (110) and

'-(100) are the prism planes and are perpendicular to the basal plane. The other planes,

(103), (102), and (101) are pyramid planes andare grouped according to the angle that

-.13

they make with the basal plane. The (103) and (102) are low angle pyramid planes and

are inclined 35.5 and 47 degrees to the basal plane, respectively. The (1 01) is a high

angle pyramid plane inclined 65 degrees to the basal plane. The <001 > direction

which is normal to the (D01) plane and the < 100> which is not normal to the (100)

plane are also shown in Figure 14. Directions in a hexagonal lattice are not necessarily

perpendicular to planes with the same indices, as is the case in a cubic lattice.

Preferred Orientation of Bulk Zinc Castings

Edmunds et al. [26,27] studied the preferred orientation of bulk zinc castings. Pure zinc

was cast into molds of air, Transite, iron, and molten lead and the zinc crystal

orientation was determined in the chill and columnar zones. It was found that the

(001) plane was parallel to the mold wall in the chill zone for all mold materials which

was attributed to the effects of surface tension. Due to the attractive forces between

the atoms, they tend to assume a surface made up of the most close-packed atomic

plane, (DOl). It appears that the tendency for (D01) parallel to the mold wall in the chill

zone is not very sensitive to heat transfer rates since it was observed with such a

variety of mold materials. It was also found that the (001) plane was perpendicular to

the mold wall in the columnar region. This is consistent with < 100> being the

preferred dendrite growth direction for HCP materials as reported by Chalmers[13].

Chalmers observed that the dendrite growth direction is the axis of a pyramid of the

closest p~cked planes that can form a pyramid. ~This condition excludes the (001)

planes in the hexagonal lattice.

Preferred Orientation of Thin Zinc Coatings

Jaffrey et all [28] grouped spangles into four morphological categories: mirror-like,

14

feathery, dimpled, and ridged. Examples of each type appear in Figure 15. It was found

that the "mirror-like", "feathery" i·and "dimpled" spangle morphologies had the (001)

plane aligned with the plane of the sheet 'nut the "ridged" spangle morphology was not.

The crystal orientation determinations were by electron diffraction, however, only 17

spangles were investigated and diffraction conditions are relaxed in electron diffraction>

, thin films, thus precision suffers. Wall et al. [5] found evidence corresponding to

dendritic growth in the <001> and < 110> zinc crystal directions and their crystal,,'

orientation results agreed with Jaffrey when they examined some spangles by a Laue

XRD technique. On the contrary, Cameron et al. [2] and Fasoyinu et al. [4] determined

that spangle sector (001) planes were inclined to the plain of the sheet by an average of

45 degrees. Fasoyinu et al. suggested that this was evidenc.e of dendritic growth in

directions other than < 100> outside the basal plane. The results of individual spangle

and spangle sector orientation analysis have been contradictory, possibly due to

deficiencies in the number of spangles sampled. The techniques used to determine the

orientations are tedious and time consuming, therefore, only a few spangles are

investigated which may not be representative of the bulk coating.

Wall et al. [5] also used a zinc inverse pole figure (ZIPF) technique to determine the

preferred orientation of the zinc layer on commercially produced material. The material

was only characterized with respect ot spangle size, no other information such as

coating thickness, coating chemistry, etc. was provided. They found that within the

spangle size range 1-1 Omm, there was a strong tendency towards a basal orientation.

Wall et al. also investigated minimized spangle material produced by the Heurty process.

In the Heurty process, fine zinc powder is projected onto the molten zinc overlay which

causes a very high rate of zinc crystal nucleation. The spangle ,size of the Heurty

15

material investigated by Wall et al. was 0.1 mm and the basal orientation of the zinc

crystals was very strong. The results of Wall et al.'s study are reproduced in Figure 16.

The results present only the information concerning the percentage of basally oriented

zinc crystals and no information was provided regarding zinc crystals with other

orientations.

Leidheiser and Kim [6,29) used a chemical etch that colors zinc crystals with basal

planes oriented within 15 degrees of the sheet plane brown to determine the preferred

orientation of commercially produced material in their exhaustive study of /

crystallographic factors affecting the adherence of paint to deformed galvanized steel., .

The findings showed that deformation paint adherence increases with the per centage of

basally oriented zinc grains. Leidheiser and Kim also showed that some basal

orientation was always observed in regular spangle material and that a much stronger

basal orientation was observed in minispangle material. These results are consistent

with Wall et al [5).

Klang et al. [7) investigated the preferred orientations of commercial material and three

sets of lab produced zinc coatings. The commercial samples included regular and

Heurtey minimized spangle materials while the three sets of lab produced samples were

pure zinc, zinc with aluminum ranging from 0.12 to 0.20%AI, and zinc with lead ranging

from 0.002 to 0.20% Pb. The bulk coating orientation of the samples was determined

qualitatively by partial (001) pole figure analysis. Half of the pure zinc samples had

strong basal preferred orientations while the other half did not. Because the technique

"considered only the (001) zinc planes, its unknown whether the samples without a basal

preferred orientation were randomly oriented or had some other orientation. Most of the

16

zinc-aluminum samples were not basally oriented, however a moderate basal preferred

J .

orientation was observed in the zinc-lead samples. The basal preferred orientation in the

zinc-lead samples was never as strong as that observed in the pure zinc samples when

the pure zinc samples were basally oriented.

Klang et al. explained their results by proposing a negative temperature gradient from

the steel/zinc to the zinc/air interface. The gradient would be affected by the formation

of the iron-zinc alloy layer, an exothermic reaction, and the preferred orientation would

depend on where the spangles nucleate. If nucleation occurred at the zinc/steel

interface, the dendritic growth would be "dl3.wn hill" in the < 100> crystal direction to

the zinc/air interface and the basal orientation would not develop. If nucleation occurred

on the zinc/air interface, growth would be along the surface, not "up hill", and a strong

basal texture would resul.t. Klang et al. suggested that nucleation occurred at the

zinc/air interface in some of the pure zinc samples and at the steel/zinc interface in

others, thus accounting for the all or none basal preferred orientation. In the case of the

zinc-aluminum samples, they note that aluminum inhibits iron-zinc alloy layer formation

ithus eliminating the iron-zinc reaction as a source of heat and lowering the gradient, so

nucleation can occur at the steel/zinc interface, thus basal textures are not observed.

They state that the effect of lead is to inhibit nucleation, therefore dendritic growth

occurs in a more supercooled enviroment which promotes a basal orientation and that'- .

lead might somehow promote nucleation at the zinc/air interface. Klang et al. might

have corroborated -their theory that the absence of a basal orientation is the result of

zinc crystal nucleation at the steel/zinc interface with additional preferred orientation

studies. The additional studies could prove that the basal plane is perpendicular to the

sheet in coatings without a basal texture.

17

..

Spittle et al. [30] criticized Klang et aI.'s theory by" arguing that grains nucleating at the.

zinc/steel interface could also grow laterally. They proposed an alternative theory which

is very similar to one proposed by Wall ~t al. [5]. Wall et al. noted that the most

commonly observed dendrite growth direction in zinc is < 100> and that the < 100 >

direction is in the (001) plane. They state that there is a natural tendency for zinc

coatings to assume a basal preferred orientation. Starting with randomly oriented nuclei,

those nuclei oriented with the < 100> parallel to the steel substrate will grow wider

. than nuclei with unfavorable orientations. In the case of the Heurtey minispangle

material which has a strong basal preferred orientation, Wall et al. state that "processes'--~l"

which minimize the spangle, by enhancing nucleation, increase the number of favorably

oriented nuclei." •Spittle et al. [25] proposed that the extent of the basal orientation is

directly related to the degree of supercooling prior to nucleation and growth of the

crystals.

With respect to lead, both Wall et al. and Spittle et al. state that lead inhibits zigc

crystal nucleation (without elaborating on the mechanism) which increases supercooling

prior to nucleation and growth. Additionally, increasing supercooling enhances dendritic

growth. Therefore, Wall et al. and Spittle et al. state that some degree of basal

orientation can always be expected when lead is present in the zinc bath.

18

Experimental Procedure

Sample Preparation

The samples used in this experiment were produced on Bethlehem Steel's Six Inch

Continuous Hot Dip Galvanizing Pilot Line at the Homer Research Laboratories (Figure

17). The line replicates a comJ;{lercial production line in a laboratory environment. The

line incorporates alkaline cleaning and brushing to prepare incoming strip, an induction

furnace to preheat the strip, a zinc pot for coating, wipers (nitrogen or air) for controlling

coating weight, and a set of blowers for enhanced strip cooling rates.

Coils of cold rolled low carbon &feel strip (0.020 in. thick) were galvanized in zinc baths

with eleven lead levels ranging from 0.002 to 0.1 O%Pb. The coils were stored in a

controlled atmosphere prior to coating in order to prevent oxidation. All process

variables were carefully controlled throughout the experiment so that they would remain

constant. The strip was preheated to 550C in the induction furnace and entered the pot. ,

at 465C. The line speed was 100ft/min and the pot temperature was 460C for all

samples. The coated strip was nitrogen wiped in order to maintain a 91.5 g/m2

minimum coating weight and blowers forced air against the strip to achieve enhanced, ~

cooling rates. The actual processing parameters for the eleven coils are tabulated in

Table 1. Previous experimentation under the same process conditions indicated that the

average cooling rate during solidification of the zinc overlay was on the order of

45C/sec.

,Chemical Analysis and Coating Weights

The steel coils used in this study were chemically analyzed by routine methods. The

19

zinc coatings were stripped with inhibited Hel and the solutions were analyzed by

Atomic Absorption Spectrophotometry for total lead, aluminum, and iron. The zinc

coating weights were determined by a weigh-strip-weigh technique. A sample of each

zinc bath was milled, dissolved in acid, and analyzed by Atomic Absorption

Spectrophotometry for total lead, aluminum, and iron.

Electron Probe Microanalysis (EPMA)

Electron microscopy was performed on two separate" JEOL 733 Superprobes located at

"Lehigh University and Bethiehem Steel's Homer Research Laboratories. Both

instruments were operated at 17Kv and -50 Na. Iron, zinc, aluminum, and lead EPMA

Iinescans and element dot maps were generated in the wavelength dispersive mode

using L1F, TAP, and PET crystals.

Spangle Size

Spangle sizes were determined by the Jeffries Planimetric method [31]. Planar samples

from the top and bottom of each coil were metallographically prepared to remove the

surface topography. The polished samples were etched in a 10% NaOH solution to

reveal the grains and the number of grains within a known area (5000 mm 2) were

counted.so the number of grains per unit area ( NA ) could be calculated. The number of

grains per unit area were converted to spange diameters ( d ) by assuming circular

grains:

d = ( 4 / NA ) 1/2

Zinc Preferred Orientation by X-Ray Diffraction

(10)

Zinc crystal orientations were determined by a Zinc Inverse Pole Figure technique (ZIPF)

20

('

[32]. The procedure is very similar to the In-Sheet-Plane Percentage Method described

by Shaffer et al. [33]. The crystal orientation is represented as the ratio of the actual

relative intensities of zinc planes parallel to the sheet surface versus what is expected

from a random distribution of zinc crystals. The method assumes that the orientation of

the grains is radially symmetric about the sheet normal axis (fibre texture). ZIPF crystal

orientation measurements are adequate for studying two dimensional properties such as

corrosion and paint adhesion but they lack the third dimension (rotation about the sheet

normal) that may be necessary for analyzing a three dimensional property such as

formability.

'--

A Diano XRD 8535 x-ray diffractometer equipped with a chromium source was used in

this experiment. The instrument settings are given in Table 2. Chromium (K-

alpha =2.291 A) has several advantages over the more common copper (K-

alpha = 1.54A) sources for the analysis of coated sheet steel. The most important is

that copper fluoresces iron and chromium does not, therefore chromium provides a

significant signal to noise advantage over copper. Secondly, because chromium has a

longer wavelength than copper, chromium penetrates coatings less than copper. This

means that the specimen transparency correction is less significant for chromium than

for copper. The depth of interaction between the x-ray beam and the specimen varies

with the Bragg angle, 8, and equation 11 gives the fraction, Gx,-of the XRD signal

arjsjngJLom_athin layer located aLa_depth x_below the surface of the coating as a

function of the Bragg angle [25].

Gx= ( 1 - exp[ -2J1x / sin8 ))

where: J1 = mass absorption coefficient8 = Bragg angle

(11 )

XRD infinite thickness is calculated by substituting 0.9999 into equation 11 for Gx and

21

as JJ decreases, x increases. Therefore, copper (JJzn = 8.3 ) penetrates zinc coatings

more than chromium (JJzn = 23.7). If the thickness of the coating is known and

substituted for x, then the fraction, Gx ' of the XRD signal arising from a thin layer

located at a depth x below the surface can be back calculated. Gx can then be used to

correct the measured integrated intensities for specimen transparency.

Lastly, because of its longer wavelength, the inherent angular resolution of a chromium

source is greater than for a copper source. The range of d-spaces used in this

experiment were spread across 80.4 degrees ( 28 ) by using a chromium source versus

40.7 degrees with a copper source. Therefore, the use of a chromium source resulted

in a twofold improvement in angular resolution.

X-ray techniques have two significant limitations with respect to analyzing coarse

grained samples such as the larger spangle samples in this study. Leidhieser recognized

that in the larger spangled'materials, only a few grains were sampled by the x-ray beam

[6,29]. The second limitation is that the cones of diffracted intensity are disc;ontinuous

in large grained material~ (see Figure 18) while the cones are continuous in fine grained

samples [25]. When a diffractometer scan slices through the cones of diffracted\'

intensity, the measurement may not be representative if the cones are discontinuous.

These two deficiencies were addressed with a new sample stage which was

manufactured at Homer Research Laboratories (Figure 19). The new stage spins

samples in the form of three inch discs at 60 rpm. The axis of rotation~s offset 25 mm

from the b~am so that a 20 cm 2 sample area is scanned. This is a ten fold improvement

over a stationary sample. Spinning the sample effectively imparts an arc motion to the

22

diffraction cones, thereby smoothing any discontinuous cones of diffracted intensity.

The relative diffraction peak intensities for a random distribution of zinc crystals can be

calculated from zinc crystal data as described by Cullity [25]. Sample calculations for

chromium K-alpha are shown in Table 3. The relative peak intensity is a function of the

wavelength of the radiation source and relative intensities from the JCPDS-ICDD cards

may not be applicable. Calculated intensities should always be used and they should

always be checked by scanning a powder sample on the particular diffractometer being

used in the experiment.

The results, of the ZIPF are usually presented in the form of a bar chart. The chart

shows the normalized ratio of the actual relative intensity to the theoretical relative

intensity of the zinc planes of interest. The ratios are normalized to the number of zinc

planes surveyed for the analysis. In this work, the first six zinc planes were surveyed.

They are Zn (002), Zn (100), Zn (101), Zn (102), Zn (103), and Zn (110). Planes

distributed randomly would have a normalized ratio equal to 1. If the sample was

entirely basally oriented, the normalized ratio of the Zn (002) Would be 6 and the other

zinc planes would be O. To convert from a normalized ratio to the percentage of grains

with that plane parallel to the surface, divide the plane's normalized ratio by the total

number of planes, 6. Sample calculations are given in Table 4 and Figures 20 and 21

compare the XRD patterns of a randomly oriented zinc powder sample to a sample with

a basal orientation approximately four times random.

The x-ray procedure and new sample stage were verified by mounting a random zinc

powder sample on a three inch disc and determining its ZIPF in triplicate. The ZIPF

23

verification is pres~·nte(TinTabTe·5~--Excell8nt-agre·emenf-was·-6tmnned-b-etwe-enl:he--

three runs and the normalized ratios ranged from 1.14 for Zn (002) to 0.88 for Zn

(101) and (100). The slight positive bias in the case of Zn (002) may be attributable to

ZnO contamination because the strongest line of ZnO, (101), overlaps with Zn (002).

The,ZIPF verification was conducted with instrument settings identical to those used in

the experiment.

A total of eighty eight ZIPFs were 'determined for this experiment. They included four

tops and four bottoms from each of the eleven coils. The results were av~raged for

each coil in order to better represent the overall preferred orientation of the zinc coating,

especially in the case of the larger spangled material. A total area of 160 cm2 was

analyzed for each coil which included approximately 3250 of the larger spangles versus

81,600 of the smaller spangles.

24 r

Results and Discussion

Chemical Analysis and Coating Weights

The chemical analysis of the primary zinc and lead added to the bath and the sheet steel

coils used in this investigation are given in Tables 6-8. Samples 1 and 4were produced

with Coil 1 and the balance of the samples were produced with Coil 2. Both coils were

low carbon cold rolled steel of similar composition. The zinc coating weights are given

in Table 9 and ranged from 105 (Sample 14) to 123 g/m2 (Sample 15) zinc per side.

The lead, iron, and aluminum contents of the zinc pot and the zinc coatings are given in

Table 10. The lead content of the pot was greater than the lead content of the coatings

in most cases. This is because lead is insoluble in solid iron and solid zinc and therefore

is not expected to be soluble in the Fe-Zn alloy layer which is included in the coating

weight and chemical analysis determinations. The inclusion of the Fe-Zn alloy layer in

the calculations dilutes the coating I,ead conter;lt: The iron content of the pot ranged_.. - ---- - --_. - -- .. _--\ -. - - _.. -

from 0.03 to 0.05 %Fe and the total aluminum 'content ranged from 0.11 to 0.15 %AI,

therefore the liquid zinc was saturated with iron in all cases and either Fe-AI-Zn or Fe-Zn

intermetallics will form. The aluminum content of the Fe-AI-Zn phase is 45%AI,

consequently, the formation of the Fe-AI-Zn phase consumes significant quantities of

bath aluminum that otherwise would be available to react with the steel sheet to form

an Fe-AI inhibition layer [19]. Therefore, when the Fe-AI-Zn phase is stable, the total

aluminum content of the bath must be corrected for the aluminum content of the Fe-AI-

Zn phase to obtain the soluble aluminum content of the bath. The solubility of

aluminum in the Fe-Zn phase is very low, less than 3%AI, so when the Fe-Zn phase is

stable, the soluble aluminum content of the pot is equivalent to the total aluminum

25

content of the pot [18]. According to Tang et al. [18 ], Fe-Zn intermetallics are stable

when the total pot aluminum content is less than 0.14%AI. Sample 14, with 0.15%

total aluminum, was the only sample with more than 0.14% total aluminum,.accordingly, the total pot aluminum for sample 14 was corrected for the contribution of

Fe-AI-Zn intermetallics to yield 0.14% soluble aluminum.

~The soluble aluminum contents of the pots varj~d from 0.11 to 0.14%AI and, according

to Belisle [20], should have a constant effect on the Fe-Zn alloy layer. Belisle plotted

coating aluminum versus soluble aluminum (Figure 12) to show the effect of soluble

aluminum on the formation of the Fe-AI inhibition layer. His data was for temperatures

and immersion times comparable to this study (465C bath, 6 second immersion) and he

showed that coating aluminum was constant over the range of soluble aluminum,

0.09% to 0.13%AI. Belisle's study found that the formation of the Fe-AI inhibition layer

was indicated by a rapid pickup in coating aluminum corresponding to 0.14% soluble

bath aluminum. The pickup in coating aluminum was not observed in this experiment,

however, the samples in Belisle's investigation were lead free (15 ppm Pb) while all the

samples in this investigation contained lead (O.002-0.10%Pb). This suggests that lead·

in the zinc bath prQmotes the formation of the Fe-Zn alloy layer and therefore increases

the level of soluble aluminum necessary for the formation of the Fe-AI inhibition layer.

The iron content of the coating is a good bulk indication of the extent of Fe-Zn alloying

and plotting bath lead content versus coating iron content (Figure 22) indicates that

coating iron content does increase with coating lead content. Generally, the coating

iron content appears to be constant up to a level of 0.04%Pb, beyond which the

coating iron content increases significantly with lead content. Cross-sectional Light

26

Optical Microscope (LaM) photomicrographs of the eleven coatings showing the Fe-Zn

alloy layers are pr'esented in Figures 23-33. Alloying was heaviest in samples 13 and 14

which contained 0.060 and 0.109 pot lead, respectively. The Fe-Zn alloy layer

development in these samples was similar to Fe-Zn alloy layer development observed by(

Belisle at 0.09%AI soluble bath aluminum, thus, LaM corroborates the chemical analysis

data which indicates that lead in the zinc bath promotes the formation of the Fe-Zn alloy

layer.

Electron Probe Microanalysis (EPMA)

In order to further investigate the effect of lead on Fe-Zn alloy layer growth, the coating

-.of Sample 15 was studied by Electron Probe Microanalysis (EPMA). Sample 15 was

selected because it was the highest in coating lead (O.088%Pb) and coating iron

(2.24%Fe). Two possible ways that lead could enhance the growth of the Fe-Zn alloy

layer are that lead may interfere with the formation of the Fe-AI inhibition layer or that

lead may affect the ternary diffusion of zinc through the Fe-AI inhibition layer. If lead

affects the ternary diffusion of z~nc through the Fe-AI inhibition layer it might be

homogeneously distributed in the Fe-Zn alloy layer. Iron, zinc, aluminum and lead EPMA

line scans were performed across the coating of Sample 15 and the normalized weight

percents of two typical scans are plotted in Figures 34A and 34B with the vertical axis

located at the steel/coating interface. 'The line scans showed that the steel was 99%

iron and that the zinc overlay was 98% zinc as expected. Aluminum is soluble in iron

zinc intermetallics and was detected in the Fe-Zn alloy layer (Figure 348).

Two indications for lead were found in the zinc overlay (Figure 34B). One lead

indication corresponded to discrete lead particles found at approximately 12 microns

27

./

from the interface and shown in the EPMA photomicrograph and lead dot map of Figure

35. There were also trace indications of lead at the steel/coating interface and in the

Fe-Zn alloy layer, however, the validity of these indications was questioned as the lead

levels indicated were very close to the minimum detection limit (M. D. L.) of the EPMA

line scan analysis.

To verify the line scan lead indications, area mC!ps of the steel coating interface and

alloy Iqyer were conducted. Two dimensional area maps would clearly provide more

information about the lead level and distribution than the one dimensional line scans.

The results of EPMA mapping for lead~aluminum, and iron are presented in Figures 36

and 37 which show discrete submicron lead particles at the steel/coating interface. The

EMPA map results confirm findings by van der Heiden (34). Van der Heiden taper

p'olished 0.1 %Pb zinc coatings at a 20 degree angle to enhance the steel/coating

interface and found discrete Fe-AI particles and individual lead particles at the

steel/coating interface. The absence of a homogeneous distribution of lead in the Fe-Zn

alloy layer (as compared to aluminum) confirms that lead is not soluble in zinc, iron, or

iron-zinc compounds and thus it is unlikely that lead enhances Fe-Zn alloy layer growth

by affecting the diffusion of zinc through the Fe-AI inhibition layer. Lead may affect the

growth of the Fe-Zn alloy layer by hinderin'g the formation of the Fe-AI inhibition layer.

Van der Heiden's work showed that the continuity of the Fe-AI inhibition layer was

disrupted by discrete lead particles and would confirm the observation that lead in the

galvanizing bath retards inhibition and therefore enhances the formation of the Fe-Zn

alloy layer in gal~a[1ized coatings. Figure 22 shows that for bath lead additions up to

0.04%Pb, the coating iron content is constant at about 0.75%Fe. Ho'wever, when the

bath lead exceeds 0.05%Pb the coati!\g iron content increases continuously, again~

28

confirming the negative effect of bath lead content on inhibition.

Spangle Size

All of the zinc coatings were dendritic and the four spangle morphologies described by

Jaffrey et al. [28], mirror, feather, dimpled, and ridged, were observed in each.. -'

'"Representative LGM photomicrographs of each type. of spangle m6rphology are shown

in Figures 38 and 39. All of the zinc coatings were of high quality and no bare spots or

pin holes were observed, thus indicating good wettability of the steel by the liquid zinc.

The spangle size of the zinc coatings ranged from 0.5 to 2.5mm. The results are listed

in Table 11 and plotted against coating lead content in Figure 40. This data shows a

transition from small spangle size to large spangle size that is complete at approximately

0.04% Pb. The spangle size results are consistent with measurements by Kim et al. [3]

which were shown in Figure 10a and replotted in Figure 41 with the results of this

experiment along with results from Fasoyinu et al. [4]. The apparent agreement with

Kim et al. is expected because the cooling rates in the two experiments were

comparable. Kim et al.'s data was for a cooling rate range of 25-60C/sec versus the

cooling rate for this ~xperiment which was 45C/sec. The cooling rate in the Fasoyinu et

al. experiment was 4C/sec and the spangle sizes ranged from 1 to 13mm, thus

demonstrating that lower cooling rates yield larger spangles because lower

undercoolings produce less nuclei. While the spangle sizes measured by Fasoyinu et al.

were larger due to the slower cooling rate, the transition to large spangles still occurred

at 0.04%Pb which is consistent with the results of this investigation.

Lead could affect the solidification of zinc coatings in several ways. One effect of lead

29

on the solidification of zinc coatings is that lead lowers the interfacial energy of liquid

zinc and therefore inhibits the heterogeneous~ucleation of zinc crystals [8,9]. Lowering

the interfacial energy of liquid zinc causes an increase in the contact angle, e, between

crystalline zinc and the substrate and raises the energy barrier, boG *i' to heterogeneous

nucleation (equations 4-6). Zinc crystals grow outward from each crystal nucleus until

they meet adjacent dendrites from neighboring nuclei. Lead, by inhibiting

heterogeneous nucleation, increases the spacing between neighboring nuclei, thus

possibly coarsening the spangle size. A second effect of lead on the solidification of

zinc coatings is that lead causes constitutional supercooling of liquid zinc which

promotes dendritic solidification. Lead is soluble in liquid zinc and insoluble in solid zinc,

therefore, as liquid zinc crystallizes, lead is rejected by solid zinc and concentrated in

/"liquid zinc. When the meltin.g point of the lead enriched liquid zinc is greater than the

prevailing temperature gradient, the liquid is constitutionally supercooled and dendritic

growth will occur. As the lead content of the original liquid zinc alloy increases, solute

redistribution (equation 8) and the constitutional supercooling driving force for dendritic

growth also increase. As solidification progresses, lead is concentrated between the

zinc dendrite arms as shown by the EPMA photomicrographs in Figures 42-44. These

results show again that lead solubility in zinc is nil but this excess lead has no effect on

spangle size because the lead particles form primarily interdendritically and not

excessively at the spangle boundary.

A relationship between coating iron content and ,spangle size was also observed (Figure

45). When the coating iron content was in excess of 1% Fe, the spangle size was high.

Because coating iron content is a measure of Fe-Zn alloy layer development, it appears

that the presence of the Fe-Zn alloy layer somehow hinders nucleation. Zinc crystals

30

nucleate at the zinc/substrate surface and any factors which alter the zinc/substrate

surface should accordingly affect the spangle size of the zinc coating. The formation of

Fe-Zn intermetallic compounds significantly alters the substrate surface and tne

chemical reaction is exothermic. This source of heat could be a factor in zinc· crystal

nucleation. The heat of transformation for zinc solidification is -7 KJ/mole versus the

heat of formation of the Fe-Zn phases which is -11 KJ/mole [35,36]. Therefore, the heat

released by Fe-Zn formation could remelt the zinc nuclei and result in a decrease in

nuclei density and coarsened spangle size as observed in this study. Another possibility

is that the Fe-Zn alloy phases compete with zinc for heterogeneous nucleation sites

along the zinc/substrate surface. The formation of Fe-Zn phases may consume

nucleation sites that would have otherwise been available for zinc crystal nucleation

and, accordingly, coarsen the spangle size.

Zinc Inverse Pole Figure Results (ZIPF)

Individual ZIPF results for the eighty eight samples are listed in Table 12 and plotted

against coating lead by zinc plane in Figures 46-51. The error bars are equal to one

estimated standard deviation (e~d) and are typical of ZIPF's for coarse grained materials.

Averaging the individual results by coating lead content yields ZIPF's which are more

representative of the true bulk zinc coating orientation. Table 13 summarizes the

averaged normalized intensity ratios from each coil and Figures 52-57 show the average

ZIPF's by coating lead content. The plot of average basal orientation versus coating lead

(Figure 46) shows that there is no clear relationship between the two variables. Some

basal orientation was observed in all of the samples as shown by Figure 58, which plots

basal orientation versus spangle size and could be interpreted to show that spangle size

is independent of 'basal plane orientation. This is consistent with the findings of

31

previous investigators [5-7] who also found that the preferred orientation of hot dipped

zinc coatings was basal and that there was no apparent dependeflce on spangle size

over the range of spangle sizes considered in this experiment.

Plotting basal orientation against coating iron content (Figure 59) revealed that the

samples with the most developed Fe-Zn alloy layers (Samples 13, 14, and 15) nad low

basal plane orientations and that the Fe-Zn alloy layers of all of the highly oriented

samples (Samples 1,4,10,11, and 12) were underdeveloped. This suggests that a

coating iron content threshold of approximately 1%Fe may be critical to whether or not

a strong basal orientation developed in the study material. Basal orientations in zinc

coatings develop due to the unrestricted two dimensional growth of zinc dendrites in the

< 100> direction and factors which interfere with free growth should tend to randomize

the preferred orientation. The chemical reaction between iron and zinc is an exothermic

source of heat which may be disruptive to the prevailing temperature gradients and

dendrite growth. Additionally, the alloy layer is significantly rougher than the base steel

surface which may also tend to randomize the orientation of the zinc coating by

introducing growth defects into the zinc dendrites.

In summary, the production of hot dip zinc coatings can be optimized for either high

spangle size or high basal preferred orientation. In order to produce zinc coatings with

well developed spangles, it is necessary to alloy the molten zinc with an element such

as lead which either lowers the interfacial energy of liquid zinc or precipitates at the

substrate/coating interface, affecting inhibition, and consequently inhibits the

heterogeneous nucleation of zinc crystals. This study also showed that enhancing the

growth of the Fe-Zn alloy layer can also contribute to large spangle size. Large spangles

32

are characteristic of products such as guard rails and lamp posts which are coated with

zinc by batch dip processe~ which involve long immersion times and result in very heavy

Fe-Zn alloy layers. Factors which promote the formation of the Fe-Zn alloy layer include

lowering the, soluble aluminum content of the pot, increasing the operating temperature

(pot and strip) of the line, slowing the line to increase immersion time, switching to

more reactive substrates, and increasing the lead content of the pot. In order to

produce zinc coatings with a highly developed basal preferred orientation, this study

showed that it is necessary to minimize the Fe-Zn alloy layer growth. Factors which

minimize the 'development of the Fe-Zn alloy layer include increasing the soluble

aluminum content of the pot, decreasing the operating temperature (pot and strip) of the

line, increasing'the line speed to decrease immersion time, switching to less reactive

substrates, and decreasing the lead content of the pot. Lead was shown to be a very

potent alloying addition in the zinc pot because of its duel influence on both spangle

development and alloy layer growth.

33

Conclusions

The following can be concluded from this study of the effect of lead additions on the

zinc coating of continuously hot dipped galvanized sheet steel:.

1. Lead levels greater than 0.04% Pb resulted in larger spangle sizes. Lead additions to

liquid zinc significantly reduce the surface energy of liquid zinc resulting in an increase in

the contact angle, 8, between the zinc crystal and substrate. Increasing 8 raises the

energy barrier, llG*j, to heterogeneous nucleation which increases the distance between

nuclei, thus coarsening the spangle size. The presence of lead in liquid zinc also

promotes constitutional supercooling and dendritic growth.

2. Lead in liquid zinc promotes the formation of the Fe-Zn alloy layer, and consequently,

the amount of aluminum required to control the Fe-Zn alloy layer may be higher in

leaded zinc baths. The coating iron content was constant up to a level of 0.04% Pb,

beyond which the coating iron content increased significantly with lead content. The

absence of a significant homogeneous distribution of lead in the Fe-Zn alloy layer as

revealed by EPMA analysis indicates that lead does not affect the diffusion of zinc

through the Fe-AI inhibition layer. However, lead may hinder the formation of the Fe-AI

inhibition layer by the formation of discrete lead particles at the steel/coating interface

which results in enhanced Fe-Zn alloy layer growth.

2. The Fe-Zn alloy layer enhances spangle size by altering the zinc/substrate surface

""'-which affects heterogeneous zinc crystal nucleation. When the coating iron content

was greater than-1-$1o-Ee.-the-spangle size-was-high.Jt isproposed_thaLthe_heat _

34

.,

./

released by the formation of Fe-Zn intermetallics may remelt some zinc nuclei and/or the

Fe-Zn phases may be in competition with liquid zinc for he.terogeneous nucleation sites.

4. The presence of the Fe-Zn alloy layer tends to randomize the preferred orientation of

the zinc coating.

5. There is no relationship between the preferred orientation and the spangle size over

the range of spangle sizes investigated. It appears that both spangle size and preferred

orientation are influenced by the coating iron content. In addition, there is nO

relationship between the pr~ferred orientation and the lead content of the coating.

------------ ---

35

References

1. Sebisty, J. J. and Edwards, J. 0., Proceedings of the 5th Int'I Conf. Hot DipGalvanizing, 1958, Zinc Development Association, Sidney Press, London (1959),p.213.

2. Camero~, D. I., Harvey, G. J., Ormay, M. K., Journal of the Australian Institute ofMetals, V10 N03, August 1965, p. 255.

3. Kim, Y. W. and Patil, R. S., 1st International Conference on Zinc Coated SteelSheet, Munich, 1985, p. 0/1.

4. ~soyinu, F. A. and Weinberg, F., Met. Trans B, 21 B, June 1990, p. 550.

5. Wall, N. J., Spittle, J. A., Jones, R. D., Proc. 1st International Conference on ZincCoated Steel Sheet, Munich, 1985, p. C/1.

6. Leidheiser, H. and Kim, D. K., Journal Of Metals, 1976, 28, p. 19.

-7. Klang, H. and Kiusalaas, R., Proc. 3rd International Conference on Solidification

Processing, 1988.

8. Thresh, H. R., White, D. W. G., Edwards J. 0., and Meier, J. W., "Properties ofMolten Zinc Alloys, Part III : Surface Tension," Canadian Department of Mines andTechnical Surveys, Ottawa. ILZRO Pr?ject ZM-54, 1964.

9. Fleming~, M. C., Solidification Processing, McGraw Hill Publishing, 1974,p. 36,60,290-7.

10. Reed-Hill, R. E., Physical Metallurgy Principles, D. Van Nostrand Publishing Co.,1964, p. 47.

11. Renaux, B., and Sylvestre, A., Intergalva 1988, p. sb3/1.

12. Mathewson, C. H., Zinc: the Science and Technology of the Metal, Its Alloys andCompounds, Reinhold Publishing, 1959.

13. Chalmers B., Trans AIME, May 1954, p. 519.

14. Cameron, D. I. and Harvey G. J., 8th International Galvanizing Conference, 1967,p.86.

15. Bablik, H., Galvanizing (Hot-Dip), John Wiley & Sons, 1950.

16. Ghuman, A. R. P. and Goldstein, J. I., Metall. Trans., 2 , 1971, p. 2903.

17. Urednicek, M. and Kirkaldy, J. S., Z. Metallkde., 64, 1973, p. 899.

18. Tang, N., and Adams,G. R., The Physical Metallurgy of ZinG Coated Steels;nProcessing Structure. and Properties, 1994, p. 41.

36

19. Gagne, M., Guttman, H., L'Ecuyer, J., Brummitt, G. G., Adams, G. L.,Kieimmmeyer, D., 82nd Galvanizers Association Meeting, 1990.

20. Belisle, S., The Physical Metallurgy of Zinc Coated Steels: Processing, Structure,and Properties, 1994, p. 65.

21. Fasoyinu, F. A., "The Solidification of Hot Dipped Galvanized Coatings on Steel,"PhD Thesis, University of British Columbia, May 1989.

22. Lindenmayer, C. S., Orrok, <3. T., Jackson, K. A., Chalmers, B., Journal ofChemical Physics, V27, No3, September 1957, p. 822.

23. Nash, G. E. and Glicksman, M. E., ActaMetallurgical, 22, October 1974, p. 1283.

24. JCPDS-ICDD Card 4-831 .

25. Cullity, B. D., Elements of X-Ray Diffraction (2nd Ed.), Addison-Wesley PublishingCo., 1978.

26. Edmunds, G., Trans AIME, V143, 1941, p. 114.

27. Edmunds, G., Trans AIME, V161, 1945, p. 183.

28. Jaffrey D., Browne, J. D., and Howard T. J., Met. Trans B, liB, Dec 1980, p. 631.

29. Leidheiser, H. and Kim, D. K., Met. Trans B, 9B, December 1978, p. 581.

30. Spittle, J. A., Brown S. G. R., 3rd Intenational Conference on Zinc Coated SteelSheet, Barcelona, June 1991, p. S4K/1.

31. ASTM Method E112.

32. Bramfitt, B. L. and Longenbach, M. H., IMS Proceedings, 1969, p. 45.

....

33. Schaffer, S. J., Morris, J. W., and Wenk, H. R., Zinc Based Steel Coating SystemsMetallurgy and Performance, TMMMs, 1990, p. 129.

34. van der Heiden, A., Hoogovens Groep, The Netherlands, PrivateCommunication,1994.

35. Kubaschewski, 0., Evans, E. LL., Alcock, C. B., Metallurgical Thermodynamics(4th ed.), Pergamon Press, p. 388.

36. Dauphin, J. Y., Perrot, P., and Tchissambot, U. G., Memoires et EtudesScientifiques Revue de Metallurgie, 1987, p. 329.

37

Table 1. Processing Parameters for the Eleven Samples

Code Furnace Snout Temp Bath Line Speed ImmersionTemp (F) (F) Temp (F) (FPM) Time (SEC)

#1 987 831 858 102 4

#4 1010 869 864 102 4,

#5 1037 884 882 103 4

#6 1009 876 864 102 4

#9 1005 845 865 102 4

#10 1032 854 866 103 4

#11 1012 847 859 102 4

#12 1031 855 866 103 4

#13 1013 887 861 103 4

#14 1028 893 858 102 4

#15 1019 890 864 102 4

Table 2. X-ray Diffraction Instrument Parameters

Instrument Diano XRD 8535 Diffractometer

Source Chromium (Kalpha = 2.291 A) CA8l X-Ray Tube;20mA, 45 kV , Vanadium Beta filter

Slits Beam - 0.2 degrees , Recieving - MR

Scan Speed 2 degrees per minute

Time Constant 2.5'to

38

Table 3. Calculated Intensities for Zinc with Chromium Radiation, Kalpha = 2.291angstroms.

l;"lt h'<-IS;,.. G

( ~r'\ If/ z. f fe //1,+ Rel1ttte s;'" c9 """£- - - -

-~---

1 (002) 27.6 0.463 0.202 22.31 1991 2 6.969 27751 38

2 (100) 29.7 0.495 0.216 21.78 474 6 5.905 16794 23

3 (101 ) 33.4 0.548 0.239 20.91 1312 12 4.625 72816 100

4 (102) 42.7 0.678 0.296 18.74 351 12 2.978 12543 ,17

5 (103) 58.6 0.854 0.373 16.56 823 12 3.185 31453 43

6 (110) 59.3 0.860 0.376 16.49 1088 6 3.256 21255 29

Table 4. Sample Calculation of Normalized Intensity Ratio for a Random Zinc PowderSample.

Zinc Measured Measured Calculated Measuredl Normali Normalized

Planes Integrated Relative Relative Calculated zation IntensityIntensity Intensity Intensity ReI. Int. Factor Ratio

(002) 12750 51 38 1.34--- 0.875 1.17

(100) 5868 23 23 1.00 0.875 0.87

(101 ) 25020 100 100 1.00 0.875 0.87

(102) 5016 20 17 1.18 0.875 1.03

(103) 13356 53 43 1.23 0.875 1.08

(110) 8059 32 29 1.11 0.875 0.97

Sum 6.86

NormalizationFactor = 0.875

6/sum

39

Table 5. 'Results of ZIPF Verification with Random Zinc Powder

Run Zinc Planes

(002) (103) (102) _ (101) (100) (110)...

.,

First 1.15 1.08 1.03 0.88 0.89 0.99

Second 1.17 1.08 1.03 0.87 0.89 0.97

Third 1.11 1.06 1.12 0.88 0.85 0.97

AVG 1.14 1.07 1.06 0.88 0.88 0.98

Table 6. Chemical Analysis of the Primary Zinc

Zn AI Si Fe Pb Cu Cd Sn Sb (ppm)

99.1 <0.05 <:0.05 <0.05 <0.05 <0.05 <0.05 <0.05 <1.0

Table 7. Chemical Analysis of Primary LeatJ/

Pb Sb As Bi Cd Cu Fe Sn Ag

99.9 0.02 0.07 0.07 <0.005 <0.005 <0.005 . <0.005 <0.005

40

Table 8. Chemical Analysis of Steel Substrate Coils

ICoil ISample Code I C I Mn I p I S I Si I Ni I AI I

#1 1,4 . 0.032 0.29 0.009 0.011 <0.01 0.022 0.039

#25,6,9,10,11,

0.049 0.29 0.010 0.019 <0.01 0.01 0 0.05112,13,14,15 ,

~-

Cu N Cr Ti Mo Sn V

#1 1,4 0.083 0.0065 0.024 <0.002 <0.02 0.003 <.003

#25,6,9,10,11, 0.016 0.0053 0.018 <0.002 <0.02 0.005 <.00312,13,14,15

,

41

Table 9. Zinc Coating Weights

Coating

CodeWeight

(g/m 2 zincper side)

#1 117

#4 115

#5 115

#6 113

#9 118

#10 109

#11 108

#12 111 "-

#13 119

#14 105

#15 123

42

Table 10. Chemical Analysis of the Zinc Pot and Coating

Code Lead Aluminum Iron

Pot ., CoatingPot Pot Coating Pot Coating(total) (soluble) (average)

#1 0.002 0.002 0.14 0.14* 0.21 0.03 0.70

#4 0.010 0.012 0.14 0.14* 0.23 0.04 0.67

#5 0.024 0.018 0.14 0.14* 0.20 0.03 0.81

#6 0.026 0.024 0.14 0.14* 0.19 0.04 1.06

#9 0.030 0.029 0.13 0.13* 0.18 0.04 0.88

#10 0.038 0.035 0.13 0.13* 0.19 0.03 0.69

#11 0.044 0.041 0.14 0.14* 0.20 0.05 0.66

#12 0.049 0.045 0.14 0.14* 0.20 0.04 0.87

#13 0.060 0.052 0.12 0.12* 0.23 0.04 2.31

#14 0.082 0.074 0.15 0.14 0.21 0.04 1.l-7

#15 0.109 0.088 0.11 0.11 * 0.22 0.04 2.24

*soluble aluminum equals total aluminum

43

Table 11. Results of Spangle Size Measurements

Code Coating LeadSpangles per Spangle Diameter

Square Millimeter (mm)

#1 0.002 3.92 0.57

#4 0.012 3.01 0.65 -,

#5 0.018 4.53 0.53

#6 0.024 2.26 0.75

#9 0.029 0.75 1.30

#10 0.035 0.27 2.19

#11 0.041 0.23 2.34

#12 0.045 0.26 2.20....... ~~

#13 0.052 0.24 2.31

#14 0.074 0..22 , ~.39."#15 - 0.088 0.20 2.51

\

44

HOMER RESEARCH LABORATORIES 4BETHLEHEM STEEL CORPORATION

TABLE 12. RESULTS OF X-RAY DIFFRACTION ZIPF ANALYSISFOR THE ELEVEN COILS

COATINGLEAD ID SIDE (002) (103) (102) (101) (100) (110)

0.002 1-1 TOP 4.20 0.72 0.24 0.32 0.25 0.280.002 1-4 TOP 2.12 1.64 ~.09 0.49 0.48 0.190.002 1-2 TOP 3.58 0.61 0.34 0.82 0.31 0.340.002 1-3 TOP 3.86 0.77 0.44 0.47 0.16 0.290.002 1-1 BOT 4.10 0.81 0.64 0.21 0.13 0.110.002 1-4 BOT 2.60 1.44 1.03 0.30 0.15 0.480.002 1-2 BOT 3.96 0.58 0.46 0.26 0.23 0.520.002 1-3 BOT 2.85 1.37 1.13 0.30 0.01 0.330.002

/ 0.012 4-2 TOP 3.82 1.14 0.78 0.14 0.03 0.100.012 4-1 TOP 4.12 0.35 0.28 0.31 0.21 0.730.012 4-3 TOP 2.84 1.22 0.56 0.54 0.68 0.150.012 4-4 TOP 4.04 0.31 0.74 0.35 0.21 0.340.012 4-2 BOT 2.77 0.70 1.10 0.78 0.25 0.400.012 4-1 BOT 3.62 0.50 0.45 0.42 0.63 0.360.012 4-4 BOT 3.65 0.42 0.72 0.79 0.17 0.250.012 4-3 BOT 2.84 1.04 0.86 0.18 0.46 0.620.0120.018 5-1 TOP 0.72 1.60 1. 75 0.93 0.69 0.310.018 5-4 TOP 2.94 0.93 0.95 0.58 0.23 0.36·0.018 5-2 TOP 2.68 1.24 0.51 0.66 0.23 0.680.018 5-3 TOP 2.84 0.83 0.49 0.53 0.27 1.050.018 5-1 BOT 1.81 1.09 0.90 1.39 0.42 0.390.018 5-3 BOT 2.10 1.33 0.78 0.55 0.48 0.760.018 5-4 BOT 2.49 0.91 0.94 0.81 0.39 0.460.018 5-2 BOT 3.20 0.96 0.86 0.34 0.22 0.420.0180.024 6-1 TOP 3.03 0.52 0.53 0.60 0.27 1.050.024 6-2 TOP 0.99 1.46 1.55 0.86 " 0.01 1.• 130.024 6-3 TOP 2.16 1.65 0.88 1.01 0.2'4 0.060.024 6-4 TOP 1.43 0.34 1.48 0.85 0.15 1.740.024 6-1 BOT 3.25 0.37 0.37 1.00 0.68 0.330.024 6-2 BOT 1.64 L82 1.35 0.82 0.19 0.170.024 6-2 BOT 2.47 1.19 0.91 0.97 0.23 0.220.024 6-3 BOT 1.98 0.77 2.04 0.47 0.45 0.280.0240.029 9-1 TOP 3.21 1.15 0.27 0.64 0.49 0.230.029 9-4 TOP 1.62 0.76 1.54 0.16 0.56 1.360.029 9-3 TOP 0.53 0.81 1.26 1.05 0.70 1.640.029 9-2 TOP 2.19 1.64 0.46 1.02 0.05 0.64

~ 0.029 9-1 BOT 0.44 2.27 0.44 0.37 1. 46 1.030.029 9-2 BOT 0.09 2.52 1.28 1.86 0.21 0.050.029 9-3 BOT 2.32 1.32 0.39 0.45 1. 32 0.200.029 9-4 BOT 3.23 1.33 0.55 0.44 0.34 0.120.0290.035 10-1 TOP 5.04 0.16 0.28 0.19 0.32 0.020.035 10-2 TOP 4.28 0.27 0.12 0.12 0.01 1.190.035 10-4 TOP 3.36 1.28 0.37 0.11 0.00 O.BB0.035 10-3 TOP 3.42 0.02 2.14 0.31 0.11 0.000.035 10-2 BOT 3.58 0.36 0.36 0.70 1.00 0.000.035 10-1 BOT 3.60 1.17 0.30 0.16 0.72 0.05

45

HOMER RESEARCH LABORATORIES 5BETHLEHEM STEEL CORPORATION

TABLE 12. RESULTS OF X-RAY DIFFRACTION ZIPF ANALYSISFOR THE ELEVEN COILS

COATINGLEAD ID SIDE (002) (103) (102) (101) (100) (110)

0.035 10-4 BOT 4.07 0.44 1.07 0.24..- 0.16 0.020.035 10-3 BOT 0.3B 1.40 0.60 0.50 0.10 3.030.0350.041 11-1 TOP 5.1B 0.23 . 0.33 0.27 0.00 0.000.041 11-4 TOP 5.B4 0.11 0.04 0.01 0.00 0.000.041 11-3 TOP 4.48 0.05 O.BB 0.54 0.02 0.020.041 11-2 TOP 4.07 0.29 0.27 1.34 0.03 0.000.041 11-1 BOT 3.51 0.90 0.96 0.59 0.04 0.010.041 11-4 BOT 2.65 1.24 0.78 0.89 0.36 0.090.041 11-2 BOT 5.05 0.22 0.24 0.22 0.01 0.250.041 11-3 BOT 3.9B 0.75 0.35 0.84 0.04 0.020.041 .0.045 12-1 TOP 3.46 1. 25 0.50 0.17 0.05 0.570.045 12-3 TOP 3.01 1.19 0.64 0.32 0.72 0.130.045 12-2 TOP 5.09 0.40 0.40 0.05 0.05 0.000.045 12-4 TOP 1.B6 1. 32 0.49 0.23 0.20 1.890.045 12-1 BOT 3.82 1.14 0.7B 0.14 0.03 0.100.045 12-2 BOT 4.15 0.94 0.01 0.83 0.06 0.010.045 12-3 BOT 4.15 0.56 1.06 0.09 0.00 0.140.045 12-4 BOT 3.BB 0.13 0.72 0.46 0.3B 0.420.0450.052 13-1 TOP 0.97 1.14 0.32 0.02 0.00 3.550.052 13-2 TOP 0.B7 2.23 1.29 0.B2 0.11 0.670.052 13-3 TOP 0.02 0.63 2.68 2.02 0.64 0.010.052 13-4 TOP 0.79 0.05 LBO 2.25 0.11 1.000.052 13-1 BOT 5.59 0.00 0.22 0.10 0.07 0.020.052 13-2 BOT 5.00 0.33 0.41 0.10 O.OB 0.080.052 13-3 BOT 0.43 3.57 1. 40 0.36 0.16 0.080.052 13-4 BOT 4.44 0.56 0.38 0.49 " 0.12 0.00

. 0.0520.074 14-1 TOP 3.51 1.34 0.77 0.16 0.19 0.030.074 14-3 TOP 1.24 3.50 0.95 0.26 0.01 0.050.074 14-2 TOP 0.32 2.63 0.55 2.31 0.12 0.060.074 14-4 TOP 2.21 1.71 1.23 0.13 0.12 0.590.074 14-1 BOT 1.52 0.94 1.61 0.60 1.22 0.100.074 14-4 BOT 0.44 1.37 0.71 0.70 0.00 2.780.074 14-3 BOT 0.03 2.03 _ 3.67 0.23 0.00 0.040.074 14-2 BOT 4.01 0.77 1.15 0.06 0.00 0.020.074O.OBB 15-1 TOP 4.03 0.99 0.51 0.46 0.00 0.00O.OBB 15-2 TOP 0.20 2.29 2.92 0.42 0.11 0.07O.OBB 15-3 TOP 4.74 0.50 0.22 0.53 0.00 0.01O.OBB 15-4 TOP O.BB 0.60 3.17 0.34 0.00 1.00O.OBB 15-1 BOT 0.19 1.90 3.09 0.52 0.31 0.00O.OBB 15-3 BOT 1.93 1.90 1. B3 0.25 0.01 '. 0.09O.OBB 15-4 BOT 0.94 1. 63 2.76 0.59 O.OB 0.01O.OBB 15-2 BOT 0.79 0.97 1. 73 1. 29 0.14 1. 08O.OBB

46

Table 13. Average Normalized Intensity Ratios for the-Eleven Coils

Code Coating Zinc PlanesLead

(002) (103) (102) (101 ) .--- (100) (110)

#1 0.002 3.41 0.99 0.67 0.40 0.22 0.32

#4 0.012 3.46 0.71 0.69 0.44 0.33 0.37

#5 0.018 2.35 1.11 0.90 0.72 0.37 0.55

#6 0.024 2.12 1.02 1.14 0.82 0.28 0.62

#9 0.029 1.70 1.48 0.77 0.75 0.64 0.66

#10 0.035 3.47 0.64 0.66 0.29 0.30 0.65

#11 0.041 4.35 0.47 0.48 0.59 0.06 0.05

#12 0.045 3.68 0.87 0.58 0.29 0.19 0.41

#13 0.052 2.26 1.06 1.06 0.77 0.16 0.68

#14 0.074 1.66 1.79 1.33 0.56 0.21 0.46

#15 0.088 1.71 1.35 2.03 0.55 0.08 0.28

47

740

720

700

Eu'680I/)Q) gc>-~660c0

'ii)c~640

Q)u0't620='en

600

580

560

540

The Effect of Leadon the Surface Tension of Zincat 435 OC with Correction forV~ TrCllsport Effect

• Corrlctldo \Jnc:orrlclld

Figure 1. The effect of lead on the surface tension of liquid zinc [8],

48

o

o

~

v,C\I

NN'

0,N

~,

~,

"0C

!' Q)...J~0

~--.J::OlQ)

33

cod

<Dd

otd

<'t°0

otr<)

q

o

oIIIq

Figure 2. The effect of lead on the viscosity of liquid zinc [8].

49

+~ CLUSTERST NUCLEI t--

<.!)

U~.

>-<.!)a:: llG7wzwww 0a::lL. rtt

RADIUS ---+

Figure 3. Plot of crystal nucleation total free energy [9].

50

)0:

O"SL

Figure 4. Crystal cap and wetting angle, e, for heterogeneous nucleation [9].

51

Fl

780

Tn ••-----e------------- ----------

The Effect of Alrniunon the SlI'fOC4l Tension of zncat 43~"C with Correction fOf

Va,xu TrCllSpol't Effe~t

7110

7411

Sn, 0i --0--0---0--- 0 ---0------~T6!1 0 ~

~5 T6.-5

.!! TIlII

~

74

Figure 5. The effect of aluminum on the surface tellsion of liquid zinc [8].

52

"0 0

.:J • •"CI 0 0uN CD=• •.. .. ..0 0 0

" " "• • •:J :J :J'0 a a> > >>0 >0'>0......- .- .-" " "0 0 0Co) Co) Co)

.!! " " 0> :; :;

~0 • 0

0-............ ~o A Co)..............

E0 :J

/ .=.0 E

/ :J-/0 ..: <t

0

0

0 <0 cD q- C\I 0 <0 CD •~ If) rIl If) rIl rIl N N Nq q q q q q q q 0.

as!od c ~I!SOOS!I\

Figure 6. The effect of aluminum on the viscosity of liquid zinc [8].

53

At.%Pb/0 20 JO 40 $0607090

1008040 60Wt.%Pb .

"k:::i<t< mit::::K/

Liq A +Liq B\

, ..

o.

I

CI

417.8 0

0·9% '98·0%I

Zn .,. Liq B 1327.4 0

\. 99.f"%\,c Zn+ Pb Sol. SOI..,~.. Zn Sol. Sol. in Pb,\

~ ~

oc700

600

SOO

400

JI8'2

JOO o

4/9·4

Figure 7. The zinc-lead phase diagram [1 21.

54

(0)

DISTANCE. x" ~(c)

tz c.o L.-(j')oa.~ouo CO):::::>o...J

(b) SOLUTE ENRICHED. LAYER IN FRONT OFLIQUID - SOLIDINTERFACE

DISTANCE. x" ~(b)

CONSTITUTIONALLYSUPERCOOLEDREGION

DISTANCE. x"~(d)

Figure 8. Schematic of constitutional supercooling [9]. a) Typical binary equilibriumphase diagram. b) Solute redistribution. c) Relationship between the actual temperaturegradient and the melting point of the alloy when the solid-liquid interface is st(ible. d)Relationship between the actual temperature gradient and the melting point of the alloywhen the solid-liquid interface is unstable resulting in dendritic solidification.

, k~

55

At. % Zn0c ,0 O">0tP ,>0 tP ,,0 '60900 I

2~1 •BOO ;c ___ 554'S~/ ~,.. ;SO"700 -- . --- 8/ ~--4---P-

50 2·7 . 410o 5 08 100

600 -Wt. %Zn - --·Wt. %Zn -J-----'2>-l-i-q·4--lI---1

I500 liq

400

Figure 9. The zinc-bismuth phase diagram [1 2).

56

.. ,,0."

•

'.0. ..01. • 01

Lud Conlenl 1",1

•001•

•

II

'F' ,!III

But: 0.2% AI . ZnCooling rile: n.lur,1 conveclive cooling

,n H2 . N2 'Imolpher. •

•

II..;;

A Effect of lead at a constant aluminum level (0.2%) on~anQle size.

I.' e.~: pur. ZinC

Cooling r'le: nlturel convectlv, coohng in H2 . N2 'Imolpher.

ij "!•..';r,!l •.•

j

• • o. .01 0" .,.

Aluminum contenl ('101oro ...

B Effect of aluminum on spangle size.

Figure 10. The effect of aluminum and lead on span-gle size [3].

57

800"-480~o"""C--------------"'",700

470°C,..... 600 -,"ItI0,..>C ~ Actual analysis of

~unfiltered sample:

0

)!400r.:-..J

- Correction line-m 300 < s= %Fe = 0.78 %AI::J..J0en 200(1)

LL Intersection giving thedissolved Fe and AI

100 level at 480 °C.

"\,

0 0.1 0.2 0.3 0.4 0.5 0.6

AI CONCENTRATION (wt 0/0)

Figure 11. Correction line estimation of soluble aluminum [191.

58

Figure 12. Effect of soluble bath aluminum on coating aluminum [20].

59

14.0

12..0

10.0

8.0

6.0

4.0

2..0

Zn-Sb

•

0--

.j

0.00.00 0.04 0.03 0.12 0.16

SOLUTE CONCENTRATION (wt.%)

0.20 0.24

Figure 13. Plot of grain size versus alloy concentration for zinc-lead and zinc-binaryalloys [4].

60

<001> <001> <001>

<100> <100>~4H+~~<1OO>(--+-+-.....I

(002),0 degrees (103). 35.5 degrees [102). 1j7 degrees

{001> <001> <001>

{l OO>(--~!+!+.....I

(101 I. 65 degrees

<100> <100>~~~~

(100). 90 degrees (11 O~ 90 degrees

Figure 14. The first six zinc planes and the angle that the plane makes with the basalplane. The <001> and < 100> directions are also shown.

61

,. -Scannina eleclronmicrogl1lph (SEM) of lhe "mirror·like"lype of spanale. Their surfaces generally appear relalively nat Indfeatureless. This specimen hu been lilted to Iccenluate shallowscralches and surflce imperfections. Magnification SOO times. TiltIngle 4' dca.

-SEM of the surface of I "Feathery" spangle. The groovesin the bollom left hind corner are due to I marking pen.Magnification 44 limes. Till angle 4' deg.

-SEM of the surface of I "dimpled" spangle. Magnification350 times. Tilt anale 4' dca.

. -Scannina electron microeraph of I "Ridged" spangle wilh Is\Rgle system of ridges. Mignilication 350 limes. TIll angle." dea.

Figure 15. Mirror-like, dimpled, feathery, and ridged spangle morphologies [28].

62

100 ,90

80 -• • •e! 70 • •

~ •Ql •-a 60.D

1550Ql

01~ •cB 40..Ql0.eu •e! 30c(

20•

10

0I I I I I I I , .;

I

0 2 3 4 5 6 7 8 9 10 11 12 13Grain size (mm)

Figure 16. Plot of area percentage basal texture versus grain size [51 ....;

63

Brushing and Alkal1ne Cleaning: Induction furnaceIIIIIIIIIIII

~

..............

Zinc Pol

I1.

./

Nitrogen Wipers: Air Blowers

~ :::~:j::!!::!~

Figure 17. Schematic of 6" Continuous Hot Dip Galvanizing Pilot Line at BethlehemSteel's Homer Research Laboratories.

64

Many Small Cryst2llsContinuous Cones

Few Llnge Crystals·Discontinuous Cones

An XRO sC2In mayor may not intersectIntllnslty 0106 discontinuous cone froma statlon8ry sample -however. spinning the sample rot2lte'sthe cone and assures that all of theIntensity Is slImDled.

Figure 18. Formation of continuous and discontnuous cones of diffracted intensity [25]cmd the effects of sample spinning.

65

IIII

T

\\\

'\

'\\

\--··-··'·-r·-'

/

-- - - -,\---+- ",]"--- - - - j

1: . j'_...lI.- _

Figure 19. Schematic of sample spinning XRD stage.

66

."cO'c~

Cll

N0

X Random Zinc Powder Pattern::D0

I' enC1 2000OJ:J

0-.. 1800OJ~

OJ:J 16000.03N 1400 r I <. \0\):J I till..)C1

"0 "C

0 § 1200~ 0

C1l0\ 0. en-....l Cll

~ 1000'.~

enOJ III

1\ . l \"'!> \3 -c"0 :J

800 t ("r0- ou

II 1\ I LI'o)600

(100) II ~ llo'1. ') IIIl lbO"i)

400

200

0

50 60 70 80 90 100 110 120 130 140 150 160

Degrees 2-theta

~oozc(a:(,/)w~i=en(")

II

...ae.zN

· 1o<0

oLtl

oM

0N

0 III...... eu...~'"eueu

0 ..Cl

0 eu0

0en

oco

,..;; 0

r--

~ 10,.. t 0(, -.J <0

0....--~

J 0-' Ltl

0 0 0 0 0 0 0 0 0 0 00 0 0 0 0 0 0 0 0 00 co <0 '<t N 0 CXl <0 '<t NN

pUO:las Jad Sluno:>

Figure 21. XRD scan of a sample 'with approximately four times random basal planes.

68

Bath Lead Content versus Coating Iron Content

0.12 .-

.,.'/

0.1 f-

• "/

0.08 /

.,/"I... "c: ,.

.<

4)

/r...

c:0

(.)

"0 0.06III4)

..J

.c:)~...

III •m

•0.04

"-I---' •

•••

0.02 l-

•

oo

I

0.5

I I

1.5

II

2

1

2.5

Coating Iron Content

Figure 22. Plot of coating iron versus bath lead.

69

" .~~~~~,.-v:_

\1:> . . .,

Figure 23. LOM cross-sectional photomicrographs (500X, Amyl Nitric) of the Fe-Zn alloylayers of Sample 1, top and bottom sides.

70

~·B '

"F 9

•


71


72


73

98 .

C\


74

..,-lO \

. .

()

i

~Q~~~j(~~

Figure 28. LOM cross-sectional photomicrographs (500X, Amyl Nitric) of the Fe-Zn alloylayers of Sample 10, top and bottom sides. t '

75

!

Figure 29. LOM cross-sectional photomicrographs (500X, Amyl Nitric) of the Fe-Zn alloylayers of Sample 11. top and bottom sides.

76


77

o


78

•

\~?J. ,


79

. ,~r,"

.. ... ~

~~~~~~~~)~~~'.. -


80

T'

LEHIGH, 2·22·94

A

·26 ·24 ·22 ·20 ·16 ·16 ·14 ·12 ·10 ·6 -e -4 ·2 0 2 4 6 6 10 12 14 16

MICRONS

LEHIGH, 2·22·94

0.5

0.4

0.3

0.2I-Pbl

/

·26 ·24 ·22 ·20 ·16 ·16 ·14 ·12 ·10 -a -e -4 ·2 0 2 4 6 8 10 12 14 16

MICRONS

HAL,2·8·94

• • • •

~~

B:,0 ·8 -6 -4 ·2 0 2 4 6 8 10 12 14 18 18 20 22 24

MICRONS

HAL,2·8·94

M. O. L. Pb • 0.07wt'J(,

·10 ·8 ·8 -4 ·2 0

~~

6 8 10 12 14 18 .18 20 22 24

MICRONS

Figure 34. EPMA line scans of the zinc coating of Sample 15.

81

Figure 35. EPMA backscattered electron image and lead map of a typical lead particle inthe coating of Sample 15.

82

Figure 36. EPMA iron (top) and lead (bottom) area maps of Sample 15 coating.

83

Figure 36. EPMA iron (top) and lead (bottom) area maps of Sample 15 coating.

83

(

Figure 37. EPMA aluminum area map and backscattered electron image of Sample 15coating.

84

Figure 37. EPMA aluminum area map and backscattered electron image of Sample 15

coating.

84

mirror-like

feather

Figures 38. LOM photomicrographs (50X) of representative spangle morphologies.

85

feather


85

dimpled

ridged


86

dimpled

ridged

\Figures 39. LOM photomicrographs (50X) of representative spangle morphologies.

86

(

Spangle Size (Dmm) vs Coating Lead Content

3

2.5

\/2

EE0

ClN 1.5ii)Cl"0cIIIQ,

CIl

0.5

a

a 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09

Coating lead Content

Figure 40. Plot of spangle size versus coating lead content

87

Spangle Size versus Bath Lead Content

10

9

8

• Fasoyinu at al. [4)

II This Study7

---0- Kim at al. [3)

6E§GlN

en 5Gl0.ccoc.

CI)

4

3

2

o

o 0.02 0.04 0.06 0.08 0.1 0.12

Bath lead Content (wt%)

Figure 41. Spangle size versus lead content according to Kim et. al. [3], Fasoyinu et. al.[4], and this investigation.

88

backscattered electron image

lead map

Figure 42. Planar EPMA photomicrographs and element map (540X) of sample 1.

89


lead map


90


lead map

Figure 43. Planar EPMA photomicrographs and element map (54QX) of sample 9.

90

. I'.. ~ •. ..'

.. . ", .:.:.:-::: .:~":\..'..:\\ ..~ .<:<>.{::~:::;.:.:' ','.' .:::...".. :~.. ;:.':.:: ::;::..::;....:.. :.:.::~.: .' . :;:-:: .:: I: ;. .•.. .' . .. ;:.:•..: ::· :~~·~.:;.·;·..:· :.:.~:.I.;::..:.:..:.; : :.:~.~,:..~::: :..:.:.~..;.::..:'.;.·:.·..:: :·..l~:.:~>:: ::~:~.:.:: ::• •.::'~'. '.::~~ • '. :;".' ,0 •• • .. :.~.~'.: :.:••:•••~:.:•.•: :••:~.•: •.•.••:••.:.~. •. :' ~. ~ .,:" .,' ' . .'::.:.::- ;.. ...::':~~:"" ::.. ~>:"': ;.....:~\:.: ):':":':~:"'~:" .;.... :?:.::'.:. .;·.::..f;~.~~·..:...... , ','..' .' ..... ..:::;.~:. '.:.'::~: ':' .' '.':' .:'" ." ....•.•

: ~:: ::;':~:::;.r f:~ ..":;. i~·.·~.:..~·;··.: ···.:~·t~~.:·~.·.·..·.·...~:·.·:....:.:.:.....~..:...:..;.~.: ~.:.:::.;.: :;.:. r:;~~t\;?:.:...··..:.... :....:.:••:••.• ,.~.•,.,;, ...... .: ,t. ,0 • ;:..... .: ". ".:. 'I. . • ... ~ ~~'.. .:::. '.:..~; •. :':

,0 'I' ':~-::,::.~<~": .. 0••". • . • 0' .:~::':.:".,'., ."':.',: ." ••••• ,:.,,:. ":.~~'"....••: ..:: :::~:.\. : ': ;':.'! .•..: .. .' . . ..' .' . :~.,,: ":;;:.':.' . ;:;:.: •.::'~.':;'; .

..•:~:' 0° -:.::.:' ~. '0. :.:.:.~~:. ..':." ::~.;" .• :~: .••••••••:••.••.•: .. ~~ •••::•.••••:••.••.:•••::.:::.•••::..•.••• '":.::~".:' :::~.:~ • ::. "., •. e,••!.. ..:., ..., ....! •. : ~.:~:.:-.: .•~.::" . . .... :: ~' ...' .." .. ,,' :. .:'::"',' '. .. ,.'.' ...• :.:' \. . :.. :.,": " :;. :.~"...:...... ';.::.:..:5.:··.··~:: ':.':.' '.':',' ':~:~ ..•..\~:;::~::~~.;.~:.:.:.:;

: .: '. . . .'" . • • . . .! .. :<.~: ' ..X:.~ ;·.:··~:.:·:.··:· .:.~. .'. ~ .; -:..~ ::. ..•..... .. .. :.:... . .

. .•.• ." : ••..• ' . " :,.:; .:;••:•••• '" ..e .: •..... :...:..... ': :... ; ' ..~.;. '. .. .... .;.....;.;... :..~ .:., .. '

: . ". ..' ..:':.: . . ... ' .•..•~.:.:: . .-••,.•••:.:. c.' • '. '. .:.' ., e' :.' • ••• •• ~:."".•' .=:.:.:..:;. .' .' . .:..'. •... .' ;.: :~.; .. f :.' ~.~.~ .:.: ..... e •


lead map


91


lead map


91

2.5 r

2 t--

Coating Iron vs Spangle Size

,/{.

'#.! 1.5 f-...c:

C1l...c:o

(J

c:o.=Cl

.~ 1 f-lOo(J

0.5

•• •

•------_.-8--._.--- _. __

~.

• .-.

oo

II

0.5

II

1I

1.5

Spangle Size (Dmm)

II

2

1I

2.5

II

3

Figure 45. Plot of coating iron versus spangle size.

92

6 tIIII

II5 +

IIIII

4 tM IE IA I

)N II

( 3 +0 I0 I2 I) I

I ,j

2 t

1 +---------------------- ---------------IIIII

o t

I---t-------------t-------------t-------------+-------------+-------------+--

0.00 0.02 0.04 0.06 0.08 0.10

COATING LEAD

Figure 46. Plot of mean (002) zinc planes times random versus coating lead content(error = 1esd).

93

6 +IIIII

5 +IIIII

4 +M IE IA IN I

I( 3 +1 I0 I3 I) I

I I2 +IIIII

1 +---IIIII

o +I---+-------------+-------------+-------------+-------------+-------------+--

0.00 0.02 0.04 0.06 0.08 0.10

COATING LEAD


94

II,;

t

t------ ---- - -[ -

II

6 +IIIII

'"5 +IIIII

4 1"IIIII

3 +IIIII

2 1"IIIII

1 +---

IIIII

o 1"I---+-------------+-------------+-------------+-------------1"-------------+--

0.00 0.02 0.04 0.06 0.08 0.10

(

1o2)

M

E

AN

COATING LEAD


95

II

6 tIIIII

'5 tIIIII

4 t

M IE- IA IN I

I( 3 t

1 I0 I1 I) I

I"---2 t

IIIII

1 j---------------r---

i H-l"o t

r

I---t-------------t-------------t-------------t-------------t-------------t--

0.00 0.02 0.04 0.06 0.08 0.10

COATING LEAD

\

\


96

M

E

A

N

I

I6 +

I,.. I

III

5 +

IIIII

4 +

IIIII

( 3 +

1 Io Io I) I

I2 +

IIII _ ~

1 1----------------------+----------------------------------------------------IIIII

o +

I---+-------------+-------------+-------------+-------------+-------------+--

0.00 0.02 0.04 0.06 0.08 0.10

COATING LEAD


97

I/

6 t

5 t

IIIII

4 t

M' IE IA IN I

I( 3 t

1 I1 I0 I) I

I

rIII

r

It

t ,, I-------j------ --- ---------------

l--+o t

IIIII

1 t-------------------IIIII

2 t

I---t-------------t-------------t-------------t--------- ---t-------------t--

0.00 0.02 0.04 0.06 0.08 0.10

COATING LEAD


98

RANDOM ZINC POWDER ZIPF

6:!:g 5z~ 4enw:!:t= 3enw

~ 2Q.

o 1zN

oZN(002) ZN(103) ZN(102)

hid

ZN(101) ZN(100) ZN(110)

0.002 Pb ZIPF

6 ,;

:;E

g 5z~ 4enw:;Et= 3enwz 2~Q.

01zN

0ZN(002) ZN(103) ZN(102) ZN(101) ZN(100) ZN(1101

hid

Figure 52. Average ZIPF results for random powder and 0.002 Pb coating.

99

0.012 Pb ZIPF

6~

8 5z~ 4t/)w~i= 3t/)wz 2SQ.

(.) 1zN

oZN(002) ZN(103) ZNl1021

hid

ZN(101) ZN(100) ZN(110)

6~

8 5z~ 4t/)w~i= 3t/)wz 2SQ.

(.) 1zN

o

0.018 Pb ZIPF

ZN(002) ZN(103) ZN(102)

hid

ZNl1 01) ZN(100) ZN(110)

Figure 53. Average ZIPF results for 0.012 Pb and 0.018 Pb coating.

100

0.024 Pb ZIPF

6~

g 5z~ 4(/)w~i= 3(/)wz 2:3Q"

() 1zN

oZN(002) ZN(103) ZN(1021

hid

ZNOO1) ZN(100) ZN(1101

6~

g 5z~ 4(/)w~i= 3(/)wz2:3Q"

() 1zN

o

0.029 Pb ZIPF

ZN(002) ZN(103) ZN(102)

hid

ZN(101) ZN(100) ZN(110)


~

101

.; 0.035Pb ZIPF

6~0

50z~ 4enw~i= 3enwz 2:3Q"

~ 1N

oZN(002) ZN(103) ZN(102)

hid

ZN(101) ZN(100) ZN(110)

6~oo 5z~ 4enw~i= 3enwz 2:3Q"

~ 1N

o

0.041 Pb ZIPF

ZN1002) ZN(103) ZN(102)

hkl

ZN(101) ZN(1001 ZN(110)


102

0.045 Pb ZIPF

6~

g 5z~ 4(/jw~i= 3(/jwz 2:30.

o 1zN,

oZN(0021 ZN(103) ZN(1021

hid

ZN(101) ZN(100) ZN(1101

6~

g 5z~ 4(/jw~i= 3(/jwz 2:30.

01zN

o

0.052 Pb ZIPF

ZN(0021 ZN(1031 ZN(102)

hid

ZN(1011 ZN(1001 ZN(110)


103

0.074 Pb ZIPF

ZN(002) ZN(103) ZN(102)

hkl

ZN(101) ZN(100) ZN(110)

0.088 Pb ZIPF

6.,

:E0

50z~ 4(/)w:Ei= 3(/)wz 2~a.0ZN

0ZN(002) ZN(103) ZN(102) ZN(101) ZN(100) ZN(110)

hid


104

(002) Planes Times Random

6 -

5 -

•4 +-

E0

"C •t:m •a: .-UIQ)

Ei= 3 -I-

UIQ)

t:m

c::N •0 •0 •- 2 +

....

• • •

random = ,

oo

II

0.5

II

I,1.5

I,

2

II

2.5

II

3

II

3.5

Spangle Size IDmm)

Figure 58. Plot of basal planes versus spangle size. All coils showed some (001)preferred orientation.

105

(002) Planes Times Random vs Coating Iron Content

4.5 -r

~!

4 -f-

•3.5 ..

3E0"0r::ltla:III 2.5

\Fe-Zn Alloy Layer Inhibition

Q)

E •i= •III •Q)

r:: 2ltlii:N • •0 •0- 1.5

0.5

oo

I

0.5

I

1

II

1.5

II

2

I

2.5

Coating Iron Content

Figure 59. Plot of basal planes versus coating iron content.

106

Name-

Place and Date of Birth -

Parents -

Education -

Professional Experience -

Biography

Richard Eugene Fraley, Jr

Winchester, VAOctober 6, 1958

Mr. and Mrs. Richard Eugene Fraley

Millersville State University, BA Physics, 1980

Bethlehem Steel Corporation, 1981-present

107

The effects of lead on the solidification and preferred ...

Documents

Transcript of The effects of lead on the solidification and preferred ...