Problems with reliability and safety of hot dip galvanized steel structures

(Problemas com a confiabilidade e segurana de estruturas de ao galvanizadas a quente)

L. Mraz1; J. Lesay2
Welding Research Institute Industrial Institute SR, Racianska 71, 832 59 Bratislava, Slovakia
2
ZIN, s.r.o., apkova 2, 811 04 Bratislava, Slovakia.
E-mail: [email protected]

Abstract
Hot dip galvanizing is very effective means of protection against corrosion. Some recommendation concerning the steel quality are
generally known and accepted. The process consists of cleaning (pickling or sand blasting) and dipping the structures or pieces into
liquid zinc bath. The case study of hot dip galvanized steels is presented. Some recent failures of hot dip galvanized welded structures
and hot dip galvanized high strength steel screws are presented. Structures were made of S355 grade steel and MIG/MAG process
was applied for welding. Large cracks were observed in the vicinity of welds after hot dip galvanizing process. The presence of both
hydrogen and liquid metal embrittlement was identified and associated mainly with higher hardness of HAZ or the quenched and
tempered steels. Possible cracking mechanisms are discussed. The influence of chemical composition and production process (welding,
heat treatment) was analyzed according to data published in literature. The solutions and recommendations for avoiding the failure in
hot dip galvanized structures are proposed.
Keywords: steel, welding, cracking, hot dip galvanizing
Resumo: Galvanizao a quente um meio muito efetivo de proteo contra a corroso. Recomendaes relativas a qualidade do ao
so geralmente conhecidas e aceitas. O processo consiste de limpar (decapagem ou jateamento) e mergulhar as estruturas ou partes
destas em um banho de zinco lquido. O presente trabalho apresenta casos de falhas recentes em estruturas soldadas e em parafusos
de aos de alta resistncia galvanizados a quente. As estruturas foram fabricadas com ao do grau S355 e o processo MIG/MAG foi
usado para a soldagem. Os parafusos foram fabricados com ao de alto carbono. Grandes trincas foram observadas nas proximidades
das soldas aps o processo de galvanizao a quente. A ocorrncia de fragilizao tanto por hidrognio como por metal lquido foi
identificada e associada com a dureza elevada tanto da ZTA como dos aos temperados e revenidos. Os possveis mecanismos de
fissurao so discutidos. A influncia da composio qumica e processo de fabricao (soldagem e tratamento trmico) analisada
de acordo com dados publicados na literatura. Solues e recomendaes para evitar a falha em estruturas galvanizadas a quente so
propostas.
Palavras-Chave: Ao, Soldagem, fissurao, galvanizao a quente

1.Introduction
Hot dip galvanizing is very important production process
used for steel protection against corrosion. This process seems to
be even more economical and effective compared to protection
by painting. Hot dip galvanizing is used for protection of
steel products for more than 100 years. During application of
this process the recommendations have been developed for
steel specification. These recommendations should assure
the technological reliability and also the quality of zinc layer.
Recommendations for constructions to be galvanized (such as

(Received in 02/09/2009; Final Text in 03/06/2009).

Based on paper published on the Welding & Material Testing
3/2008 (Romania). Re-printed under the consent of Welding &
Material Testing.
184

ventilation etc.) are the basic requirement for the continuous

galvanized process. The quality of hot dip galvanizing
also influence not only the technological conditions (bath
chemical composition, cleaning process etc.) but also chemical
composition of the steels.
The quality requirements and the procedures for quality
assessment of hot dip galvanized process are specified in EN
ISO 1461 [1] and the construction requirements are given in EN
ISO 14713 [2].
The influence of Si and P in the steel to be galvanized
on the thickness and quality of zinc layer is generally known.
This requirement for the content of both Si a P is specified in
the standards of structural steels EN10025-1 to 6 [3] for the
normalized and also for the quenched and tempered but also for
TMCP steels.
The intermetallic phases containing different Fe and Zn
content are formed on steel surface during hot dip galvanizing.
Soldagem Insp. So Paulo, Vol. 14, No. 2, p.184-190, Abr/Jun 2009

Problems with reliability and safety of hot dip galvanized steel structures

The stresses and also the effect of strain level on the quality of
steel products were studied. It is generally known that there is no
remarkable problem with galvanizing mild steels.
Cracking of steels has been observed in steels with the yield
strength over 1000 MPa [4]. Specific problem was observed
with high strength steel fasteners [4]. Fracture was not observed
for the fasteners without zinc layer. Some failures in welded
structures of hot dip galvanized steel are reported in [69].
The main reasons for this failure are indicated the following
parameters such as high residual stresses, local plastic strain and
local hardening.
The subject of this contribution is to attain an overview on
possible fracture mechanisms which were observed during hot
dip galvanizing of steel constructions.

475C than the killed and also not killed steel.

Steel welded structures are hot dip galvanized very often.
Mainly mild steel are used for these structures. Liquid metal
embrittlement caused by zinc has also been observed in welded
steel structures galvanized after welding. This problem, related
with steel bridge structures was studied by Abe H. et al. [9].

2.Failures of hot dip galvanized steel products

There were three basic embrittlement mechanisms observed
according to [4] when the hot dip galvanized steel structure
failed. It is:
- liquid metal embrittlement
- hydrogen embrittlement and
- strain ageing embrittlement.
McDonald [5] supposed also the influence of Fe - Zn
intermetallic layer on the fracture. The role of all mechanisms is
complex and one of them may miss in the actual case.
The liquid metal embrittlement caused by zinc and
cadmium was studied by Arrata at el [6] in liquid Cd, Zn and
Cd + 5 % Zn, Cd + 50 % Zn alloys. The authors identified that
liquid metal influences the plasticity of steels. The ductility
decreases with increasing Zn content (Figure 1). The change
of plasticity was associated with the actual fracture mechanism
from transcrystalline to intercrystalline. They also observed
that the fracture associated with the liquid metal embrittlement
takes place when the minimum stress level and or deformation
is applied (Figure 2).

Figure 2. Influence of load on liquid metal embrittlement [6].

The influence of chemical composition was identified for
hot dip galvanized steel failure. The following equation was
developed by the regression analysis [10]:
Steels with C < 0.12 %
S LM-400 = 201-370C22Si51Mn35P+33S-28Cu22Ni87Cr
123Mo275V-182Nb-82Ti-24Al-1700N-155000B
(1)
Steels withC > 0.12 %
S LM-400 = 227-320C10Si76Mn50Cu30Ni92Cr88Mo
220V-200Nb+200Ti
(2)
where S LM-400is the parameter which identifies the influence
of liquid metal at 470C on the notched strength Rm in zinc /R m
x 100 %, and also zinc carbon equivalent (CEZ):
without zinc
CEZ = C+Si/17+Mn/7.5+Cu/13+Ni/17+Cr/4.5+Mo/3+V/1.5+
Nb/2+Ti/4.5+420B
(3)

Figure 1. Influence of Cd and Zn on the steel elongation [6].

Pavlidis et al. [7] found out, that the pure zinc does not
change the ductility of steels within the temperature range from
450 to 500C. The liquid metal embrittlement is remarkably
influenced by the presence of Pb in the liquid metal.
Friehe a Hankel [8] also identified that iron is more
susceptible to intercrystalline liquid metal embrittlement at
Soldagem Insp. So Paulo, Vol. 14, No. 2, p.184-190, Abr/Jun 2009

Japanese authors also supposed that liquid zinc

embrittlement is caused by diffusion of zinc along the primary
austentic grain boundaries due to thermal and residual stresses.
They also concluded that embrittlement can be reduced by the
presence of ferrite on the HAZ grain boundaries.
There are no serious problems when the steels of yield
strength in the range from 235 to 460 MPa are galvanized
according to [12]. It is also recommended that the killed steels
resistant to strain ageing or fine grained steel grades should be
used for steel structures containing bent parts. For the steels
with yield strength higher than 460 MPa it is recommended
to use shorter pickling time in order to avoid the hydrogen
embrittlement. The influence of elements such as Sn, Pb and
also Bi on liquid metal embrittlement is reported. The content of
185

Mraz, L.; Lesay, J.

both Sn and Pb should not be higher than 1.3 % and the content
of Bi should be 1.0% max.
The cracking phenomenon is expected according to [13]
when hardness would exceed the level of 34HRC or 340HV
due to hydrogen. Unfortunately, these criteria are not mentioned
in the standard EN14713.
Hydrogen embrittlement during hot dip galvanizing process
is discussed also in [14], [15]. The guideline [14] considers the
hydrogen embrittlement as one of the fracture mechanism of
galvanized structures. The steel production process, welding and
pickling are possible sources of diffusible hydrogen. No criteria
such as maximum hardness can be found in this document. The
EN 1011-2 standard is recommended to follow, in order to avoid
the hydrogen induced cracking. Usually inhibitors are added to
liquid metal for avoiding this problem. Anyway, liquid metal
embrittlement is expected when steels of the yield strength
higher than 355MPa are used for welded structures.
According to the literature related to cracking of welded
structures after hot dip galvanizing, it is supposed that the liquid
metal embrittlement is associated with local strain or stress
seems to be the main mechanism for the failure. The influence
of hydrogen is expected but there are no criteria given in order
to reduce the hydrogen induced cracking.

Figure 4. Fracture surface in L profile.

3.Failed steel structures after hot dip galvanizing

Several failed welded structures of zinc coated steels and
steel products have been analyzed in the last few years. The
welded structures were fabricated from structural steel grade
355MPa and welded by using MIG/MAG process.
First example of failed steel structure is a weldment in 6m
length fabricated from L profiles in dimension 6x80x80mm
and 5 x 50 x 50 mm. The cracks sometimes attaining almost
70mm in length were observed after hot dip galvanizing (Figure
3). These cracks initiated close to the fillet welds and propagated
perpendicularly to the profile surface (Figure 4). No cracks were
identified by visual testing just after welding.

Figure 3. Fracture in a welded steel structure close to L profile

weld.
186

Figure 5. Fracture of circumferential weld between the tube

and plate.

Figure 6. Crack morphology in a circumferential weld between

the plate and tube.

Soldagem Insp. So Paulo, Vol. 14, No. 2, p.184-190, Abr/Jun 2009

Problems with reliability and safety of hot dip galvanized steel structures

Second failed steel structure consisted of a tube welded to

10 mm steel plate. The plate 10 mm in thickness was made of
S355 steel grade. Several cracks were identified perpendicularly
to the circumferential weld (Figure 5). A number of cracks has
been identified by using dye penetrant test when the zinc layer
was removed by grinding. The crack surface was covered by
zinc layer and it propagated both, in plate and tube as well
(Figure 6).
Third case of hot dip galvanized steel structure was the
fracture of a sizable steel structure containing multi-pass welds.
The structure failed in the location of stress concentration
(Figure 7).

3.1 Investigation of the failed welded steel structures

Fracture mechanism of steel structures was investigated
using optical microscope METAVAL. The specimens for
investigation were cut from the broken welds. These specimens
were prepared for microscopical analysis by grinding, polishing
and etching. Also Vickers hardness measurements of base
metal and HAZ were performed at the load of 49.05 N (HV 5).
Chemical composition of steels used for failed steel structures is
given in Table 1.
Macrostructure of the L profile fillet weld is shown
in Figure 8. The cracks initiated in the notch at the interface
between the weld and base metals, e. g. in the HAZ. The crack
length is shown in Figures 9 and 10. The cracks propagated
perpendicularly to the L-profile length from the HAZ into base
metal. The cracks were trans- and intergranular in the HAZ and
intergranular in the base metal. The crack surface is covered by
zinc layer (Figure 11).

Figure 7. Crack in zinc coated welded steel structure.

Table 1. Chemical composition of steels used for steel structures.
Parent material
tube
plate
profile 80 x 6 mm
S355J2G3

C
0.080
0.207
0.181

Mn
1.41
1.30

Max. 0.27 Max. 1.70

Composition (weight. %)
Si
P
S
0.01
0.014
0.006
0.41
0.010
0.020
0.37
0.019
0.011
Max.
Max.0.60
Max.0.055
0.055

Al
0.034
0.045

N
0.005
-

CE*
0.440
0.397

max.0.55

* CE = C + Mn/6 + (Cr+Mo)/5 + (Ni+Cu)/6

Macrostructure of the tube to plate circumferential weld is
shown in Figure 12. The cracks were oriented perpendicularly to
the circumferential weld (Figure 13) and propagated into plate
and also into the tube. In some areas the cracks oriented parallel
to weld seam were observed. Fine trans- and intergranular
microcracks were observed in the HAZ close to the main
circumferential crack (Figure 14). The intergranular cracks were
observed in the base metal.
The presence of cracks was identified in a large weldment
in the vicinity of multi-pass welded joint located in the stress
concentration area. Cracks initiating from the weld surface
and weld root were identified in one cross-section (Figure 15).
The crack initiation area could not be identified. The cracks of
branched type were observed with intergranular propagation in
the base metal and filled also by zinc layer (Figure 16).
Hardness measurements have shown that the HAZ hardness
in the tube to plate weldment attained 274 HV5 and the HAZ
hardness in L profile of structure attained 360 HV5.

Figure 8. Location of crack initiation (arrow).

Soldagem Insp. So Paulo, Vol. 14, No. 2, p.184-190, Abr/Jun 2009

187

Mraz, L.; Lesay, J.

Figure 11. Crack propagation in the base metal.

Figure 9. Crack initiation and propagation.

Figure 12. Cross-section of the circumferential weld.

Figure 10. Crack initiation in the HAZ.

Figure 13. Crack in the circumferential weld.

188

Soldagem Insp. So Paulo, Vol. 14, No. 2, p.184-190, Abr/Jun 2009

Problems with reliability and safety of hot dip galvanized steel structures

Figure 14. Crack in the circumferential weld.

Figure 15. Crack location in the steel welded structure.

initiation has always been observed in the vicinity of welds.

Hardness level higher than 270 HV 5 has been measured in
coarse grained HAZ. All failed structures were made of S355
steel grade. The crack surface was always covered by a zinc layer.
Trans- and intergranular fracture close to the fracture initiation
and intergranular fracture in the crack propagation direction
were identified by microscopic examination. The presence of
transgranular fracture suggested that most probably hydrogen
embrittlement of the HAZ region is responsible for crack
initiation and liquid metal embrittlement (intergranular fracture)
and for crack propagation in the base metal. Nevertheless,
the local strain during hot dip galvanizing cannot be excluded
as a possible factor responsible for the fracture initiation and
propagation.
Failures of welded structures have appeared in zinc coated
applications. It seems that coating process unanimously affects
the reliability of products. In case of welded structures, failures
appeared before putting the structure into service and were
associated with higher hardness, diffusible hydrogen, liquid zinc
and also local residual stresses as the consequences of welding.
Based on the presented case studies, we assume that the
local stresses, high hardness and diffusible hydrogen are the
main factors responsible for failures of zinc coated products in
general. The presence of active, diffusible, hydrogen may be
affected by the steel production, welding process but we assume
that in case of steel structures mainly by the pickling. This
process is used for cleaning the structures prior to coating.
Anyway, according to literature survey also local strains
may increase the risk of failure due to non-uniform heating
during hot dip galvanizing. This local strain can increase the risk
that LME could also appear as the main fracture mechanism.
We assume that in presented cases the LME is active in the
crack propagation range but not in the crack initiation range.
The failures have shown that there is a need for criteria for hot
dip galvanized steel structures welded prior to galvanizing, in
order to avoid the cracking occurrence. We suppose that the
HAZ hardness limit or choice of suitable steel with lower CE
can be helpful to avoid the crack occurrence. Of course such
recommendations are highly dependent on other factors such as
structure rigidity which is responsible for local strains during
hot dip galvanizing. More data are necessary in order to estimate
such criteria.
5.Conclusions

Figure 16. Crack propagation in the base metal.

4.Discussion

Several failures of zinc coated steel products have been

presented in this paper. The investigation of the failed steel
structures have shown that the main factor responsible for
fracture initiation seems to be the hydrogen embrittlement due
to presence of high-hardness microstructure and residual or
external stresses. The active diffusible hydrogen appears mainly
from the coating process pickling, which is used for cleaning
prior to galvanizing. The LME is supposed as the mechanism for
crack propagation in welded steel structures.

The analysis of failed steel structures after hot dip

galvanizing has shown some common features. The fracture
Soldagem Insp. So Paulo, Vol. 14, No. 2, p.184-190, Abr/Jun 2009

189

Mraz, L.; Lesay, J.

6. Acknowledgement
The paper has been prepared according to the literature
survey and to the analysis of practical cases within the project
APVT-P06705 financed by APVV.
7. References
[1] EN ISO 1461 Hot dip galvanized coatings on fabricated
iron and steel articles. Specifications and test methods (ISO
1461:1999)
[2] EN ISO 14713 Protection against corrosion of iron and steel
in structures. Zinc and aluminium coatings (ISO 14713:1999)
[3] EN 10025 - 1 to 6 Hot rolled products of structural steels,
Part 1 to Part 6
[4] GALVAINFO
[5] McDonald R. D.: Steel embrittlement problems associated
with hot dip galvanizing causes, mechanisms, controls, and
selected references, Materials Performance, January 1975
[6] Arata Y., Ohmori A., Okamoto I., Ogawa H.: Study on liquid
metal embrittlement of carbon steels, Transactions of JWRI,
Vol. 11, No. 1982
[7] Pavlidis Chr., Schulz R., Fricke M.: Zum Problem der
Rissentstehnung an Feuerverzinkungskesseln, Maschinenbau
52, (1979) Heft 1
[8] Friehe W., Hankel A.: Stand der Kentnisse ber die
Entstehnung interkrystalliner Risse an Verzinkungskesseln,
Stahl und Eisen 94 (1974) Nr. 7
[9] Abe H., Ieazawa T., Kanaza K., Zamashita T., Aihara S.,
Kanazawa S.: Study of HAZ cracking of hot-dip galvanizing
steel bridges, IIW Doc. IX-1795-94
[10] Interpretation zinc assisted cracking on big scale steel
structures and preventive methods, 2001, in ILZRO Project ZC21-2
[11] Katzung W., Schulz W.-D.: Zum Feuerverzinken von
Stahlkonstruktionen Ursachen and Lsungsvorschlge zum
Problem der Rissbildung, Stahlbau 74 (2005), Heft 4, Bericht Nr.
152, Gemeinschaftausschuss Verzinken e. V., GAV-Nr. FC22/1,
[12] Empfehlung zur Vermeidung der Rissbildung an
feuerverzinkten Stahlkonstruktionen, Deutscher Stahlbauverband
DSTV
[13] Schmidt J.: Bewertung zu den Entwrfen DIN EN ISO
1460 aDIN EN ISO 14713, Schweisstechnik and mehr
[14] Guidance note: The design, specification and fabrication
of structural steelwork that is to be galvanized, Guidance note,
British constructional steelwork association, May 2003
http://www.steelconstruction.org/static/assets/source/
Guidance%20Note%20-%20Final%20-%20Issued%2030-0503.pdf
[15] Liquid metal penetration during hot dip galvanizing, TWI
Guide to LMAC

190

Soldagem Insp. So Paulo, Vol. 14, No. 2, p.184-190, Abr/Jun 2009