Treatment of Lead Contaminated Soil using Solidification ...
The effects of lead on the solidification and preferred ...
Transcript of 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.
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
•
Figure 24. LOM cross-sectional photomicrographs (500X, Amyl Nitric) of the Fe-Zn alloylayers of Sample 4, top and bottom sides.
71
Figure 25. LOM cross-sectional photomicrographs (500X, Amyl Nitric) of the Fe-Zn alloylayers of Sample 5, top and bottom sides.
72
Figure 26. LOM cross-sectional photomicrographs (500X, Amyl Nitric) of the Fe-Zn alloylayers of Sample 6, top and bottom sides.
73
98 .
C\
Figure 27. LOM cross-sectional photomicrographs (500X, Amyl Nitric) of the Fe-Zn alloylayers of Sample 9, top and bottom sides.
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
Figure 30. LOM cross-sectional photomicrographs (500X, Amyl Nitric) of the Fe-Zn alloylayers of Sample 12, top and bottom sides.
77
o
Figure 31. LOM cross-sectional photomicrographs (500X, Amyl Nitric) of the Fe-Zn alloylayers of Sample 13, top and bottom sides.
78
•
\~?J. ,
Figure 32. LOM cross-sectional photomicrographs (500X, Amyl Nitric) of the Fe-Zn alloylayers of Sample 14, top and bottom sides.
79
. ,~r,"
.. ... ~
~~~~~~~~)~~~'.. -
Figure 33. LOM cross-sectional photomicrographs (500X, Amyl Nitric) of the Fe-Zn alloylayers of Sample 15, top and bottom sides.
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
Figures 38. LOM photomicrographs (50X) of representative spangle morphologies.
85
dimpled
ridged
Figures 39. LOM photomicrographs (50X) of representative spangle morphologies.
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
backscattered electron image
lead map
Figure 42. Planar EPMA photomicrographs and element map (540X) of sample 1.
89
backscattered electron image
lead map
Figure 43. Planar EPMA photomicrographs and element map (540X) of sample 9.
90
backscattered electron image
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 •
backscattered electron image
lead map
Figure 44. Planar EPMA photomicrographs and element map (540X) of sample 15.
91
backscattered electron image
lead map
Figure 44. Planar EPMA photomicrographs and element map (540X) of sample 15.
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
Figure 47. Plot of mean (103) zinc planes times random versus coating lead content(error = 1esd).
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
Figure 48. Plot of mean (102) zinc planes times random versus coating lead content(error = 1esd).
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
\
\
Figure 49. Plot of mean (101) zinc planes times random versus coating lead content(error = 1esd).
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
Figure 50. Plot of mean (100) zinc planes times random versus coating lead content(error = 1esd).
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
Figure 51. Plot of mean (110) zinc planes times random versus coating lead content(error = 1esd).
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)
Figure 54. Average ZIPF results for 0.024 Pb and 0.029 Pb coating.
~
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)
Figure 55. Average ZIPF results for 0.035 Pb and 0.041 Pb coating.
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)
Figure 56. Average ZIPF results for 0.045 Pb and 0.052 Pb coating.
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
Figure 57. Average ZIPF results for 0.074 Pb and 0.088 Pb coating.
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