Plasma boriding of high strength alloy steel with nanostructured surface layer at low temperature assisted by air blast shot peening

Prévia do material em texto

Surface & Coatings Technology 228 (2013) 229–233
Contents lists available at SciVerse ScienceDirect
Surface & Coatings Technology
j ourna l homepage: www.e lsev ie r .com/ locate /sur fcoat
Plasma boriding of high strength alloy steel with nanostructured surface
layer at low temperature assisted by air blast shot peening
H.P. Yang ⁎, X.C. Wu, Y.A. Min, T.R. Wu, J.Z. Gui
School of Material Science and Engineering, Shanghai University, Shanghai 200072, People's Republic of China
⁎ Corresponding author. Tel.: +86 21 56331153; fax:
E-mail address: hpyang1993@163.com (H.P. Yang).
0257-8972/$ – see front matter © 2013 Elsevier B.V. All
http://dx.doi.org/10.1016/j.surfcoat.2013.04.033
a b s t r a c t
a r t i c l e i n f o
Article history:
Received 26 February 2013
Accepted in revised form 15 April 2013
Available online 23 April 2013
Keywords:
H13 steels
Nanostructured surface layer
Air blast shot peening
Plasma boriding
Plasma boriding of high strength alloy steel (AISI H13) with nanostructured surface layer fabricated by air
blast shot peening(ABSP) was performed at 580 °C for 4 h. A continuous coating layer with thickness of
about 4 μm was composed with two phases of Fe2B and FeB, and the nanohardness of borided layer was as
high as 20 GPa. These results indicate that cyclic deformation and the angle between shot jet and sample sur-
face play a critical role in the process to produce nanocrystalline in the surface layer. By using ABSP as
pre-treatment, boron diffusion and the nucleation rate of boride can be remarkably enhanced in nanostruc-
tured surface layer. In addition, the weight loss rates of borided samples in molten aluminum alloy were only
as much as ~40% of that of untreated ones, and the main reason was that the coating layer can effectively pre-
vent the substrate directly contacting with molten aluminum alloy.
© 2013 Elsevier B.V. All rights reserved.
1. Introduction
The AISI H13 steels are used extensively for extrusion dies as well
as die casting tool steels for aluminum alloy and many other metal
workpieces. They are alloy steels with high strength and toughness.
However, this kinds of die steels is commonly used in aggressive
environments, and it is necessary to adopt surface treatment tech-
nique to improve their surface properties, such as wear resistance,
hardness and corrosion resistance. Boriding technique has attracted
extensive attention in thermochemical surface treatment, owning
to the borided layers possess excellent adhesion to the substrate
when compared to prevalent physical coating process. The boriding
process can be carried out in solid, liquid or gaseous medium [1].
Many previous literatures have reported the boriding techniques
which are used to prepare borided layers on the surface of steels
[2–5]. Yet, most of them have the disadvantages of requiring relative
high processing temperature (750–950 °C) or time consuming, which
are harmful to the mechanical properties of the matrix and create
the deformation of workpiece. For H13 steels, a critical requirement
for the boriding process is that it must be conducted at temperatures
below 650 °C in order to prevent grain growth and carbide precipi-
tation and hence to preserve their high strength and toughness.
Therefore, boriding at lower temperatures and in shorter time have
attracted extensive significant interest over the past decades. Plasma
boriding was expected to be able to accelerate boron atoms diffusion
into the steels, due to the nature of this method. It was regarded to be
carried out at lower temperatures [6,7]. However, to date, reports on
+86 21 56331461.
rights reserved.
the plasma boriding that is used to synthesize borided layers on the
surface of steels below 650 °C are scarce. In addition, little work has
been focused on the study of dynamic erosion resistance properties
of H13 die steels with plasma boriding treatment.
It is well known that borided layers generated in thermochemical
treatment depends on boriding condition and on the properties of
the materials itself. Both factors are strongly affected by grain bound-
aries and defect densities. The diffusion dynamics of boron atoms can
be sufficiently enhanced in materials with nanostructured surface
layer, due to the large grain boundaries and defect densities. Various
severe plastic deformation methods have been proposed to produce
nanocrystalline materials. Among these techniques, shot peening
treatment can not only lead to residual compressive stress, but also
fabricate ultrafine grain on the surface layer. Nanostructured surface
layer can be obtained on stainless steel [8], low alloy steel [9], and
silicon steel [10] by using air blast shot peening(ABSP), which has
been applied extensively in material processing. But until now, high
strength alloy steels with nanostructured surface layer through this
method has not been reported. In this study, plasma boriding was car-
ried out at low temperature for AISI H13 steels assisted by ABSP.
Moreover, dynamic erosion resistance properties of them in molten
aluminum alloy were investigated.
2. Experimental
2.1. Materials and shot peening treatment
The chemical compositions of AISI H13 steel used in the experiments
contain (wt.%) 0.42C, 4.93Cr, 1.40Mo, 0.98Si, 0.87 V, 0.30Mn, 0.018P,
0.005S and balance Fe. The ferritic steel was pretreated by quenching
8
6
5
230 H.P. Yang et al. / Surface & Coatings Technology 228 (2013) 229–233
at 1030 °C and tempering twice at 610 °C to obtain amartensiticmicro-
structure. The hardness of sample was 46–47HRC. Before ABSP, the
samplewith the shape (60 mm × 60 mm × 4 mm)wasmirror polished.
Then, the sample was processed by a flow of cast steel balls with diam-
eter of 0.8 mm at 0.5 MPa for six cyclic deformation, and the time of
each cycle was 5 min.The angle between the shot jet and the sample
surface is in the range of 70–90°. The small samples with dimensions
of 15 mm × 15 mm × 4 mm for plasma boriding treatments were
machined from the bulk sample as mentioned above. The specimens
were ultrasonically cleaned with alcohol and acetone before plasma
treatment.
9 1
2
3
7
4
Fig. 1. Schematic diagram of experimental set up used for dynamic erosion test system:
1 borided sample, 2 untreated sample, 3 graphite sleeve, 4 ADC12 aluminum melt, 5
thermocouple, 6 alumina crucible, 7 salt bath furnace, 8 rotational speed control sys-
tem, 9 electromotor.
2.2. Plasma treatment
The plasma boriding system mainly consisted of vacuum chamber
equippedwith electrical power and gas flow control units. After loading
the small sample, it was cleaned by sputtering with a direct electrical
power of 520 V in a gasmixture of Ar andH2. At the same time, the tem-
perature of vacuum chamber increased to 580 °C with a heating rate of
2 °C/min. BF3was fed into the vacuumchamberwith a certain flow rate.
The gas consisted of BF3 (3% vol.%), Ar (37% vol.%) and H2 (60% vol.%).
During our experiments, the total pressure had been maintained be-
tween 650 and 700 Pa for 4 h. The voltage applied on the samples
was about 550 V.
2.3. Dynamic erosion measurements
Plasma boriding treatments were carried out for cylindrical sam-
ples with dimensions of 10 mm in diameter and 60 mm in length,
which were used in the dynamic erosion tests. These samples were
compared with the untreated ones. As the H13 steels are used exten-
sively in die-cast formation for aluminum alloy, their dynamic erosion
resistance to a molten aluminum alloy is very important. ADC12 alu-
minum alloy is a kind of die-casting aluminum which is widely used
in industrial production. In order to study the performance of dynam-
ic erosion resistance of borided samples and untreated ones to a mol-
ten aluminum alloy, ADC12 aluminum alloy was used in the dynamic
erosion measurements. The chemical compositions of ADC12 alumi-
num alloy contain (wt.%): 2.13Cu, 0.87Fe, 10.72Si, 0.31Mg, 0.36Mn,
0.85Zn, 0.054Ti, 0.054Ni, 0.056P, 0.037Sn and balance Al. The proper
quantities of ADC12 aluminum alloy was put into an alumina crucibleand heated. When the melt temperature was stabilized at around
700 °C, borided sample and untreated one were held symmetrically
on a rotating shaft driven by a DC electric motor and dipped into
the molten aluminum alloy. The rotating shaft rotated at a speed of
120RPM. The schematic diagram of experimental set up used for dy-
namic erosion test system is illustrated in Fig. 1. The erosion studies
were carried out for 5, 10, 15 and 30 min, respectively. Before measur-
ing the weight loss rate, the aluminum which existed on the surface of
specimens was removed with a NaOH solution and thoroughly rinsed
with water, and the specimens were dried in an oven at 60 °C for 4 h.
2.4. Characterization
The deformation layer and nanostructured features obtained by
ABSP were observed using a JSM-6301F scanning electron microscope
(SEM)and a field emission transmission electron microscopy(TEM,
JEM-2010F), respectively. The crystalline structure of samples was
characterized with a RigakuD/Max-RBX-ray diffractometer by using
CuKa radiation and a secondary beam graphite monochromator. The
hardness gradient of nanoindentation test by using a Berkovich dia-
mond indenter with a tip radius of about 200 nm was performed on
Triboindenter In-Situ Nanomechanical Test System.
3. Results and discussion
3.1. Characterization of the nanostructured surface layer
The SEM image of the cross section of the sample treated by ABSP
is shown in Fig. 2(a). It can be seen that a plastic deformation layer
(~10 μm) is present in the treated surface layer, which is indicated
by a red–red line, where the microstructural morphology differs
from that in the matrix. The boundary between deformation layer
with dark contrast region and substrate is not distinctive. In sub-
surface zone, it is work-hardened region, which is characterized
with the elongated grain boundaries. Fig. 2(b) displays a typical
bright-field TEM image of the top layer and the statistical distribution
of grain size(inset). The size of vast majority of grains is at 5–15 nm.
The selected-area electron diffraction(SAED) pattern is shown in
Fig. 2(c). It reveals that the microstructure is characterized by
ultrafine equiaxed grains with random crystallographic orientation.
The strong diffraction rings are assigned to polycrystalline α-Fe.
There is also one reflection that could be interpreted as austenite re-
flection(311). It may reveal that α to γ transformation in the nano-
structured surface layer at about room temperature during severe
plastic deformation, due to the formation of supersaturated nanocrys-
talline ferrite and the compression stress at the boundaries of nano-
size grains [11]. There is no carbides diffraction ring that can be
detected, which may be attributed that the carbides might be broken
into ultrafine particle and they are too small to be detected. Alterna-
tively, they dissolved into the ultrafine ferrite matrix on the top sur-
face layer where very large strain and strain rate were applied
during the repetitive impacts of a lot of balls [12]. Fig. 2(d) shows
the HRTEM image of the top surface layer. There is one elongated
grain (indicated by dashed line) with the short axis of ~20 nm,
which contains a subgrain boundary. It is composed of several edge
dislocations. The elongated grain could be further refined with defor-
mation increment.
The production of nanocrystalline surface layer by ABSP has consider-
able importance since ABSP is a popular process in industries. According
to the previous report [13], there are some necessary and favorable de-
formation conditions to produce nanostructured surface layer on metal
materials. Firstly, the minimum amount of strain necessary to produce
nanocrystalline structure is considered to be around 7–8, depending
on the deformation techniques and materials employed. Secondly, the
nanostructured surface layer could be produced in the range of strain
Fig. 2. (a) Typical SEM image of the cross section of the sample treated by ABSP,(b) bright field TEM image showing the microstructure of the top surface layer in the ABSP sample and
corresponding the graph of statistical distribution of grain size(inset), (c)selected-area electron diffraction(SAED) pattern, and (d)HRTEM image of the nanocrystallization layer.
231H.P. Yang et al. / Surface & Coatings Technology 228 (2013) 229–233
rate from 0.5 to 104S−1. High strain rates are favorable to obtain ran-
domly oriented fine grains with large misorientation. Other important
factors include low temperature deformation and multidirectional de-
formation. Low deformation temperature slows down self-diffusion,
anddelay grain recovery kinetics.Multidirectional deformation activates
multi slip systems and increases dislocation interaction frequencies
which lead to the development of fine cells. Based on the discussion
above, in accordance with our ABSP experiment parameters, the nano-
crystalline surface layer can be fabricated on high strength alloy steel.
It should be noted that cyclic deformation and the angle between the
shot jet and the sample surface play a critical role in the process to pro-
duce nanocrystalline. The former suppresses recovery since specimens
are cooled during each strain interval, and the latter is needed in order
to facilitate the grain refinement process.
3.2. XRD measurements
The XRD patterns of top surface layer of samples with and without
ABSP are presented in Fig. 3(a). The diffraction peaks of α-Fe are ob-
viously observed due to the martensitic microstructure. It shows that
the diffraction peaks become a little broader for the sample treated by
ABSP than that of untreated one, indicating that the grain refinement
and microstrain exist in the surface layer. The XRD pattern of the
sample with plasma boriding treatment is given in Fig. 3(b). It is ob-
vious that two phases of FeB and Fe2B are detected. Additionally,
three diffraction peaks of α-Fe appear in the pattern, due to the thick-
ness of boriding coating, which is not enough thick to completely
mask the diffraction peaks of matrix, because the results of XRD mea-
surements may obtain the structure information of a surface layer of
about 10 μm thick [14]. Moreover, the intensity of (002) diffraction
peak of FeB is much stronger than other characteristic peak of its
own, which indicated that the growth of FeB phase exhibits a (002)
preferred orientation [15].
3.3. Microstructure and hardness of the borided layer
The SEMmicrostructure of the cross section of the samplewith plas-
ma boriding treatment is showed in Fig. 4(a). It can be seen that a con-
tinuous borided layer is on the surface, which is about 4 μm thick.
Usually, The morphology of coating layer exhibits tooth-like growth
into substrate, which attributes that neither FeB(orthorhombic) nor
Fe2B(tetragonal) has cubic crystal symmetry [15]. Diffusion in these
phases is of anisotropic nature. However, in our experimental results,
the tooth-like growth of borided layer is not obvious. Actually, the
chemical composition of substrate plays an important role on the
growth of borided layer. The alloying elements, such as Mo, V, Cr and
Ni, inhibit the growth kinetic of the coating layer. They tend to concen-
trate at the tips of the borided columns and considerably reduce the ac-
tive boron flux in this zone [16]. Owning to H13 steel with relatively
high amount of alloying elements, the borided layer on it does not pos-
sess the obvious tooth-like morphology. There is a work-hardened re-
gion just adjacent to borided layer(indicated by dashed line). The
microstructure of it shows no obvious difference with the sample that
is treated by ABSP, indicating that recrystallization does not clearly
occur in the work-hardened zone during plasma boriding. When com-
pared to the process of pack-boriding at high temperatures [2], it is in-
teresting to discover that there is no sharp boundary between FeB and
Fe2B. It might compose of mixed-phases of the two borides. Fig. 4(b)
Fig. 3. XRD patterns of (a) the original sample(A) and thesample treated by ABSP(B),
and (b) the sample with nanocrystallization layer after plasma boriding treatment at
580 °C for 4 h.
Fig. 4. (a)Typical SEM image of the cross section of the sample with plasma boriding
treatment, and (b)the corresponding hardness gradient of nanoindentation test.
232 H.P. Yang et al. / Surface & Coatings Technology 228 (2013) 229–233
shows a variation of hardness along the depth from the treated surface,
which is determined with nanoindentation tests. The graph indicates
that the two compound layers exhibitmuch greater hardness(~20 GPa)
than that of the substrate(~5 GPa), and so does the hardness(~7 GPa)
of transition zone, owning to work-hardened region, which is an effec-
tive supporting under-layer for the borided layer. The evolution of hard-
ness along depth agrees well with the cross section structural analyses
results.
3.4. Formation mechanism of the borided layer
In the process of plasma boriding, near the cathode, the high-energy
collision creates a series of induced or dissociated states of various
chemical components [17]. It can be considered that the reduction of
BF3 is performed by molecular and atomic hydrogen. The active boron
can be absorbed on the surface of sample and diffuse into the matrix.
At the initial stage, the boron atoms could dissolve in solid solution
with martensite matrix. When the adsorbed boron concentration
Cads
B on the sample surface is sufficiently high for FeB formation, both
FeB and Fe2B phases are formed and proceed to grow [2]. If the boron
concentration is in the range of CupFe2B b CadsB b ClowFeB (CupFe2B is the upper
limit of boron concentration in Fe2B phase, and ClowFeB is the lower limit
of boron concentration in FeB phase), only Fe2B phase is formed. By
reacting with active boron atoms, the Fe2B phase could transform to FeB
phase. The whole process of plasma boriding could be presented as fol-
lows: 1. BF3 + 3/2H2 → B (active) + 3HF and BF3 + 3H (active) → B
(active) + 3HF; 2. If CadsB > 0.5(at%), B(adsorption) + Fe → FeB; if
0.33 b CadsB b 0.5(at%), B(adsorption) + 2Fe → Fe2B; 3. B atoms jump
through FeB or Fe2B lattice driven by chemical potential; 4. If FeB phase
appears on the surface of sample firstly, 2FeB → Fe2B + B; if Fe2B
phase appears on the surface of sample firstly, Fe2B + B → 2FeB; 5. At
Fe2B/Fe interface, 2Fe + B → Fe2B.
Assisted by ABSP treatment, the temperature of plasma boriding for
H13 steel can be decreased to 580 °C, which is about 300 to 400 °C
lower than the conventional treatment temperatures. This phenome-
non is similar to some previous study results [18–20]. It should be
noted that this strategy may provide a general route for the surface
chemical treatments to improve the performance of engineering mate-
rials. The beneficial result can be ascribed to a large number of grain
boundaries and lattice defects such as dislocations in the deformation
layer induced by ABSP treatment, which can significantly reduce the ac-
tivation energy for boron diffusion in iron lattice. Our experimental re-
sults show that a greatly enhanced diffusivity of boron in the H13 steel
occurs when ABSP is taken as pre-treatment. Moreover, the enhanced
formation rate of borided compounds may be further understood in
terms of the reaction kinetics of nucleation and growth [21]. Numerous
defective non-equilibrium grain boundaries existing in the nanostruc-
tured surface layer are expected to be favorable nucleation sites with
a heterogeneous nucleation rate of [22,23]
NB ¼ ωCB exp −ΔGm=kTð Þ exp −ΔG�B=kT
� �
nuclei m−3s−1; ð1Þ
where ω is a factor including the vibration frequency of the atoms and
the surface area of the critical nucleus, T is the temperature, ΔGm and
ΔGB⁎ are the activation energies for atomicmigration and barrier against
nucleation on the grain boundary, respectively. The concentration of
nucleation sites on the grain boundary (CB) can be given by:
CB ¼ 3δ=dð ÞCv ; ð2Þ
Fig. 5. The graph of weight loss rate of untreated samples and borided samples in dy-
namic erosion tests at various time.
233H.P. Yang et al. / Surface & Coatings Technology 228 (2013) 229–233
where Cv denotes the number of atom sites per unit volume, δ and d are
the grain boundary width(~1 nm) and the grain size, respectively. It
shows that the nucleation rate increases significantly while the grain
size reduces into the nanometer scale because of the apparent incre-
ment of CB and the boron diffusivity, e.g. the nucleation rate may in-
crease by an order of about 106 when the grain size reduces from
40 μm to 40 nm [23]. It is apparent that the precipitate will grow at a
much higher rate with nanocrystalline.
3.5. Dynamic erosion measurements in molten aluminum alloy
The weight loss rate of borided samples and untreated ones in dy-
namic erosion tests are shown in Fig. 5. It can be seen from the graph
that the borided samples possess superior dynamic erosion resistance
to a molten aluminum alloy. When the time of dynamic erosion tests
is within 30 min, the weight loss rate of borided samples is only as
much as ~40% of that of untreated ones. Whether two metals can
form intermetallic compound depends on some factors, such as less
than 15% difference in atomic radii, same electronegativities and sim-
ilar crystal structures. Although the size difference between alumi-
num and iron atoms is less than 15% and their electronegativities
are also close, they have different crystal structures, which limit the
solubility of iron in molten aluminum [24]. However, owning to the
Al–Fe–Si affinity, some of intermetallic compounds could form, such
as Al8Fe2Si, Al5FeSi and Al12Fe5Si [25]. Therefore, the dynamic erosion
rate of H13 steels without surface treatment in molten aluminum
alloy is quick. With plasma boriding treatment, the borided layer is
very useful to prevent the matrix directly contacting with molten alu-
minum alloy. Another reason may be that the diffusion coefficient of
Al or Fe atom in Fe2B and FeB phases is small, the similar study results
had been obtained by Nazari et al. [25]. Their experimental results
showed that H13 steel with nitride-coated layer had good erosion
resistance, due to the formation of iron nitrides which prevented
the surface of the steel from multi-element diffusion conditions.
4. Conclusions
In conclusion, nanostructured surface layer on high strength alloy
steel (AISI H13) has been successfully fabricated by ABSP, which
has been applied extensively in industrial production. By using it as
pre-treatment, plasma boriding treatment was carried out for H13
steels at 580 °C for 4 h. A continuous coating layer with thickness of
about 4 μm is composed with two phases of Fe2B and FeB, and the
nanohardness of borided layer is as high as 20 GPa. It shows that
boron diffusion and the nucleation rate of boride can be remarkably
enhanced in nanostructured surface layer. When the time of dynamic
erosion tests is within 30 min, the weight loss rates of borided sam-
ples in molten aluminum alloy are only as much as ~40% of that of
untreated ones. The main reason is that the coating layer can effec-
tively prevent the substrate directly contacting with molten alumi-
num alloy.
Acknowledgments
The authors acknowledge the “11th Five” National Science and Tech-
nology Support Project of China(Project Number: 2007BAE51B04) and
Shanghai LeadingAcademicDiscipline Project forfinance support(Project
Number: S30107). The authors would like to thank J.C. Peng from Instru-
mental Analysis and Research Center of Shanghai University for the help
with the TEMmeasurements.
References
[1] V.V. Uglov, D.P. Rusalsky, V.V. Khodasevich, A.L. Kholmetskii, R. Wei, J.J. Vajo, I.N.
Rumyanceva, P.J. Wilbur, Surf. Coat. Technol. 103 (1998) 317.
[2] L.G. Yu, X.J. Chen, K.A. Khor, G. Sundararajan, Acta Mater. 53 (2005) 2361.
[3] K. Genel, Vacuum 80 (2006) 451.
[4] I. Gunes, S. Ulker, S. Taktak, Mater. Des. 32 (2011) 2380.
[5] J.H. Yoon, Y.K. Jee, S.Y. Lee, Surf. Coat. Technol. 112 (1999) 71.
[6] S. Bourithis, S. Papaefthymiou, G.D.Papadimitriou, Appl. Surf. Sci. 200 (2002) 203.
[7] K. Bartsch, A. Leonhardt, Surf. Coat. Technol. 116 (1993) 86.
[8] T.S. Wang, J.K. Yu, B.F. Dong, Surf. Coat. Technol. 200 (2006) 4777.
[9] S. Bagherifard, M. Guagliano, Eng. Fract. Mech. 81 (2012) 56.
[10] J.L. Liu, M. Umemoto, Y. Todaka, K. Tsuchiya, J. Mater. Sci. 42 (2007) 7716.
[11] N. Min, W. Li, X.J. Jin, Scripta Mater. 59 (2008) 806.
[12] Z.B. Wang, N.R. Tao, W.P. Tong, J. Lu, K. Lu, Acta Mater. 51 (2003) 4319.
[13] M. Umemoto, Y. Todaka, K. Tsuchiya, Mater. Trans. 44 (2003) 1488.
[14] N.R. Tao, M.L. Sui, J. Lu, K. Lu, Nanostruct. Mater. 11 (1999) 433.
[15] C.M. Brakman, A.W.J. Gommers, E.J. Mittemeijer, J. Mater. Res. 6 (1989) 1354.
[16] M. Carbucicchio, G. Palombarini, J. Mater. Sci. Lett. 6 (1987) 1147.
[17] E. Filep, S. Farkas, Surf. Coat. Technol. 199 (2005) 1.
[18] Z.B. Wang, K. Lu, G. Wilde, S. Divinski, Appl. Phys. Lett. 93 (2008) 131904.
[19] Y.M. Lin, J. Lu, L.P. Wang, T. Xu, Q.J. Xue, Acta Mater 54 (2006) 5599.
[20] W.P. Tong, N.R. Tao, Z.B. Wang, J. Lu, K. Lu, Science 299 (2003) 686.
[21] R.D. Doherty, Diffusive phase transformations in the solid state Physical metallurgy,
Elsevier Science, The Netherlands, 1996. 1363.
[22] D.A. Porter, K.E. Easterling, Phase transformations in metals and alloys, Chapman
and Hall, London, 1992. 263.
[23] Z.B. Wang, J. Lu, K. Lu, Acta Mater. 53 (2005) 2081.
[24] D. Pirzada, E.G. Baburaj, R. Govindaraju, F.H. Froes, Surf. Eng. 16 (2000) 164.
[25] K.A. Nazari, S.G. Shabestari, J. Alloys Compd. 478 (2009) 523.