A Study of Post Plating Heat Treatment in

Automotive Fastener Steels

B. Lonyuk1, R. Hop2, D.N. Hanlon1, S. van der Zwaag3, J. Zuidema3

and A. Bakker3
1
Netherlands Institute for Metals Research, Rotterdamseweg 137, 2628 AL
Delft, the Netherlands
2
Koninklijke Nedschroef Holding N.V., Techno Centre, Kanaaldijk N.W.
71, 5707 LC Helmond, The Netherlands
3
Laboratory of Materials Science, Delft University of Technology,
Rotterdamseweg 137, 2628 AL Delft, The Netherlands

ABSTRACT: In the automotive industry many high strength steel fasteners are zinc
electroplated. The plating process is thought to be a principal cause of hydrogen
embrittlement: fasteners can fail unpredictably at applied stress levels well below the
fracture stress. To avoid this a hydrogen relief heat treatment after electroplating is
commonly applied. In this study the effect of plating and post-plating treatments on the
slow strain rate tensile fracture behaviour of two commercial steels has been investigated.
Testing was conducted on fatigue pre-cracked cylindrical specimens in air. Results
describing the effect of alloy selection, metallurgical processing conditions and heat
treatment on the susceptibility to hydrogen embrittlement are presented. The principal
conclusion drawn from this study is that post-plating hydrogen relief annealing, as
specified by international standards, is not always of benefit.

INTRODUCTION

Automotive high-strength steels for fastener applications are often protected

against corrosion by electrochemical plating. This may result in the
introduction of hydrogen and consequently service failure due to hydrogen
embrittlement [1]. In the presence of hydrogen catastrophic failure of
fasteners may occur unpredictably at applied stress levels far below the
fracture stress. Hydrogen tends to accumulate at areas of high stress
concentration and reduces the stress required for the initiation of fracture
and the energy barrier to crack propagation.
In the automotive industry the most commonly applied method for
reducing the risk of hydrogen embrittlement is a post-coating heat treatment,
which is often referred as baking treatment. The severity of the
embrittlement is strongly dependent on parameters such as the strength level
and microstructure as well as on the amount of hydrogen introduced into the
steel. The baking treatment is assumed to decrease the hydrogen
concentration and consequently increase the fracture stress. According to
established practice the required post-coating baking conditions (time and
temperature) are dependent on the strength level of the steel. According to
specifications, electroplated fasteners do not require heat treatment if made
from steel with a hardness level less than 31 HRC [2]. However, Raymond et
al. [3] report that a standard baking heat treatment had no beneficial effect on
the fracture toughness of zinc plated specimens at a hardness level as high as
52 HRC. In this paper the results of a study of the effect of hydrogen relief
treatment on two representative commercial fastener steels are reported.

EXPERIMENTAL

Circumferentially pre-cracked cylindrical bar specimens (Figure 1) were

produced from two low alloy high strength steels. One is a Boron containing
steel referred to as Steel A and is favoured by the european automotive
fastener industry for 10.9 grade fasteners and the other, 5038 steel, is used here
for the purpose of comparison since it is more typical of the steels used in the
north american industry. The chemical compositions and mechanical
properties of the steels are given in Tables 1 and 2 correspondingly.

Figure 1: Specimen geometry used for fracture toughness testing.

All samples were used in the quenched and tempered condition

(austenitisation at 890 °C, quenching in oil at 60 °C and tempering for 1
hour at 450 °C). In the temper used for these experiments both steels exhibit
a hardness of approximately 32 HRC and a yield stress of the order of 1100
MPa. Consequently according to standard practice a baking step should be
applied prior to service (DIN 50 969). This consists of holding the plated
material at 195 °C for approximately 4 hours. The microstructure of both
steels was observed to be a well-tempered martensite.
 TABLE 1: Chemical compositions of the steels, Wt %

Material C Si Mn P S Cr Ti B Nb
Steel A 0.35 0.05 0.76 0.011 0.008 0.2 0.028 0.003 
5038 steel 0.41 0.23 0.75 0.008 0.008 0.64   0.035

TABLE 2: Mechanical properties of the steels

Material Yield stress σ0.2 (MPa) Ultimate stress σUTS (MPa)

Steel A 1100 1190

5038 steel 1098 1162

To evaluate the effect of the plating and baking treatments the materials
were passed through the hardening, tempering, plating and baking cycles of
a commercial process line. As a result samples in the quenched and
tempered but unplated (designated QT), quenched, tempered and plated
(designated QTP) and quenched, tempered, plated and baked (designated
QTPB) were obtained. A few plated samples were baked for the extended
times of 16 and 24 hours in order to evaluate the effect of the baking time.
The tests on fatigue pre-cracked samples were performed on a tensile
machine in air at room temperature. For the pre-cracking procedure a
rotating-bending configuration was used. The fracture toughness (K1C) of
the steel was determined according to the recommendations of ASTM E 399-
90 at an applied strain rate of 1 mm/min. Susceptibility to hydrogen
embrittlement was determined by loading pre-cracked samples at a lower
rate of 0.001 mm/min. The threshold stress intensity factor (K1H) was
calculated for the load at which the onset of the stable crack growth was
detected. This was done by a method based on the compliance of the
sample: the onset of crack propagation from the fatigue pre-crack tip was
accompanied by an inflection from the initial linear slope of the “load –
displacement” diagram. All sample were tested to failure and the fracture
surfaces were observed in the scanning electron microscope (SEM).
Some of the plated samples were used for measuring the hydrogen
content. Samples were sectioned to make short 12 mm diameter rods of 20
mm in length, washed thoroughly with distilled water and acetone and then
dried. The hydrogen content was determined by a hot extraction technique
with a H-mat 251 microprocessor-controlled analyser.
RESULTS

The results of the fracture toughness tests reveal comparable values of K1C
for all processing conditions (Table 3). This was of the order of 105 MPa√m
for both the Steel A and the 5038 steel and it would appear that any effects
of process condition on the fracture characteristics couldn’t be resolved by
testing at the higher applied strain rate (1 mm/min).
The K1H values observed for the zinc-plated materials (Table 3) are
considerably lower than the observed K1C values. This indicates a degree of
susceptibility to embrittlement due to internal hydrogen introduced by the
plating process. The K1H observed for the 5038 steel was significantly lower
than that observed for the Steel A indicating that this steel is more
susceptible to hydrogen cracking. This might be a result of differences in
chemical compositions since the mechanical properties of both steels are
similar.

TABLE 3: Effect of plating process on hydrogen embrittlement

Fracture toughness Threshold stress intensity

Material Conditions
K1C, MPa√m factor K1H, MPa√m
QT 104.3
Steel A QTP 54.6

QTPB 55.6

QT 107.1

5038 steel QTP 46.2

QTPB 46.9

The K1H values for materials subjected to a post-plating hydrogen relief

anneal (Table 3) are comparable to the values for unbaked materials,
indicating that the baking treatment is ineffective for the steels and tempers
considered here. This might be due to either the application of an
insufficiently long hydrogen relief treatment or may indicate that the zinc
layer presents a strong barrier to hydrogen escaping during baking.
Attempts to reduce the embrittlement by extending the time of baking
have been made. The results of tests conducted on samples from the Steel A
indicate that extended baking up to the 24 hours yields little effect on
threshold stress intensity factor (Figure 2).
 80

60
K 1H, MPa √ m

40

20
Vload = 0.001 mm/min

0
0 4 8 12 16 20 24
baking time, hours

Figure 2: Effect of the baking time on the threshold stress intensity factor
for hydrogen embrittlement in Steel A.

The operation of hydrogen induced embrittlement mechanisms in the

materials tested has been confirmed by SEM investigations. For the
materials tested at 1 mm/min the fracture surfaces comprised two distinct
regions. The region of fatigue fracture was succeeded in all cases by a
region of fast fracture. In this overload region the failure mode was
essentially ductile. In all cases fracture is seen to occur by microvoid
nucleation and coalescence as evidenced by dimple formation associated
with carbides (Figure 3).

a b

Figure 3: Macrofractography (a) and microfractography of dimple fracture

started at fatigue pre-crack (b) taken from samples tested at 1.0 mm/min.
For all materials with zinc coating the fracture surfaces comprised three
distinct regions. In addition to the fatigue and overload regions areas of
stable crack propagation were observed (Figure 4a). The stable crack
growth mode can be identified as quasicleavage with the appearance of a
degree of intergranular character (intergranular fracture occurred along prior
austenite grain boundaries as shown in Figures 4b-d). Many secondary
cracks were also evident on the fracture surface.

a b

c d

Figure 4: Stable crack propagation in zinc-plated materials broken at 0.001

mm/min; a – macrofractographical indication of three zones: A – fatigue
pre-cracking; B – stable crack; C – overload; b – transition from fatigue to
stable crack in unbaked Steel A; c and d – quasicleavage in baked for 24
hours Steel A and baked for 4 hours 5038 steel samples correspondingly.

Figure 5 shows the hydrogen content measured for Steel A samples in the
unplated, plated and plated and baked (4 hours on a commercial process line)
conditions. From these results it can be seen that the hydrogen content is
increased after plating and that baking has little effect on the hydrogen content.
 2.5

Hydrogen content, ppm

2

1.5

0.5

0
QT QTP QTPB

Figure 5: Effect of plating and baking on hydrogen content in Steel A

samples.

DISCUSSION

Both of the steels under investigation were found to be sensitive to

embrittlement by internal hydrogen introduced during plating with zinc. The
unplated fracture toughness could not be restored by the application of a
hydrogen relief anneal applied according to the DIN 50 969 standard
recommendations for high strength steels. Previously published work by
Grobin [4] reported that, in many cases, annealing for up to 23 hours or
more is required in order to reduce the risk of embrittlement due to
hydrogen. However, in this work a baking treatment of 24-hours duration
revealed no beneficial effect for sample of the Steel A. This implies that the
Zinc layer may act as an effective barrier to the release of hydrogen from the
material in the temperature range up to 200 °C. Rebak et al. [5] found that
baking treatment released only 30% of hydrogen from zinc-plated bolts. A
comparable result was obtained in the present work. The hydrogen content
in measured samples reduced from 2.1 ppm for plated material to 1.8 ppm
for that baked for 4 hours.
In contrast, observations of a beneficial effect of baking have been
reported. The positive effect of baking treatment observed by Townsend [6]
was explained in terms of the trap theory. Townsend suggested that
hydrogen, introduced during plating, may be driven into deep trap sites
where it loses mobility and is not available to affect embrittlement. It could
be assumed that the absence of an effect the baking treatment observed in
present work is also related to the hydrogen distribution in the materials.
However, from the fact that hydrogen in metal lattice is always in
equilibrium with the trapped hydrogen [7], it is possible that the
concentration of mobile hydrogen is still sufficient to produce embrittlement.
Trapping theory may also be applied to explain the differences in fracture
behaviour observed between the two steels investigated here. The higher
threshold stress intensity values observed for Steel A may be considered to
arise as a consequence of a higher density of strong (deep) traps compared
to the 5038 steel and therefore a lower mobile hydrogen content.

CONCLUSIONS

Cracking due to hydrogen introduced during the zinc electroplating process

occur in both Steel A and 5038 steel. This results in a threshold stress
intensity factor which is significantly lower than the fracture toughness of
the steels in question (i.e. K1H is significantly lower than K1C).
However, whilst hydrogen embrittlement effects were observed in both
steels the 5038 steel was observed to be more susceptible than the Steel A.
The Steel A exhibited a significantly higher threshold stress intensity than
the 5038 steel for all conditions considered.
The baking treatment applied according to the standard recommendations
for high strength fastener steels is ineffective in reducing the hydrogen
embrittlement. Extending the baking time up to 24 hours produced no effect
on the embrittlement of the steels investigated.

REFERENCES

1. Bickford, J.H., Nassar, S. (1998) Handbook of Bolts and Bolted Joints.

Marsel Dekker, Inc., New York.
2. ASTM B 850-94 (1994). In: 1996 Annual book of ASTM standards.
Section 2. Nonferrous metal products. Vol. 02.05, pp. 589 – 591.
American Society for Testing and Materials, Philadelphia.
3. Raymond, L. (1998) In: Handbook of Bolts and Bolted Joints, pp.723 –
756, eds. J.H. Bickford and S. Nassar. Marsel Dekker, Inc., New York.
4. Grobin, A.W.Ir., (1990) Materials Performance 7, 71.
5. Rebak, R.B., Muchjin, L. and Szklarska-Smialowska, Z. (1997)
Corrosion 53, 481.
6. Townsend, H.E. (1975) Metallurgical Transactions A 6A, 877.
7. Oriani, R.A. (1970) Acta Metallurgica 18, 147.