Metallurgical and Materials Transactions A

, Volume 43, Issue 13, pp 5115–5121

Carburizing of Duplex Stainless Steel (DSS) Under Compression Superplastic Deformation


    • Department of Mechanical Engineering, Faculty of EngineeringUniversity of Malaya
  • Iswadi Jauhari
    • Department of Mechanical Engineering, Faculty of EngineeringUniversity of Malaya

DOI: 10.1007/s11661-012-1357-4

Cite this article as:
Ahamad, N.W. & Jauhari, I. Metall and Mat Trans A (2012) 43: 5115. doi:10.1007/s11661-012-1357-4


A new surface carburizing technique which combines superplastic deformation with superplastic carburizing (SPC) is introduced. SPC was conducted on duplex stainless steel under compression mode at a fixed 0.5 height reduction strain rates ranging from 6.25 × 10−5 to 1 × 10−3 s−1 and temperature ranging from 1173 K to 1248 K (900 °C to 975 °C). The results are compared with those from conventional and non-superplastic carburizing. The results show that thick hard carburized layers are formed at a much faster rate compared with the other two processes. A more gradual hardness transition from the surface to the substrate is also obtained. The highest carburized layer thickness and surface hardness are attained under SPC process at 1248 K (975 °C) and 6.25 × 10−5 s−1 with a value of (218.3 ± 0.5) μm and (1581.0 ± 5.0) HV respectively. Other than that, SPC also has the highest scratch resistance.

1 Introduction

Conventional surface hardening via diffusion methods such as carburizing, boronizing and nitriding (regardless under solid, liquid or gas medium) can be considered as a static process. This is due to the fact that there is no force element introduced into the system during the process which will affect the shape and dimension of the substrate. Hard layer surface is basically via through thermo-chemical reaction, whereby heat is required to enhance the diffusion of hardening atoms. Generally, the layer thickness exhibits time-temperature dependence.

Studies have shown that superplastic boronizing, whereby boronizing is conducted concurrently with superplastic deformation of the substrate, could provide faster rate and superior mechanical properties compared with conventional methods.[1,2] The introduction of a force leading to substrate plastic deformation is considered to be the novelty of the process. However, previous studies are carried out under tensile and it is felt that a more practical study under compression mode is warranted.[3] On the other hand, mechanics of superplasticity of metals and alloys has been studied mainly by means of a tensile test in which a true constant strain or a constant displacement rate is imposed, and the steady-state flow stress is recorded. The optimal superplastic deformation temperature, strain rate, maximum elongation, and flow stress, as well as strain rate sensitivity of the material are obtained through experiment.[3] However, in practical applications, there are few cases similar to the simple tension test but many involve compressively formed products. In this experiment, a different approach of SPC under compression method was studied. This method was selected because it is beneficial to find the optimal parameters of superplastic deformation compare with tension tests. In addition, compression tests have the advantage of avoiding the problem of necking particularly at high strain rates.

The surface properties and kinetics of SPC of DSS under compression mode have been studied, and the results show that the carburizing process is highly accelerated by the concurrent superplastic deformation.[4] This process is conducted by SPC under a fixed strain rate condition.

In the present work, carburizing of DSS is carried out in order to understand the carburizing process behaviour under superplastic deformation. The effects of deformation temperature and strain rate on the carburizing process are investigated. The results are compared with those from conventional carburizing (CC) and non-superplastic carburizing (NSPC). It is expected that this study will provide a fresh insight on carburizing process development and the application of DSS in industries.

2 Materials and Experimental Procedures

Duplex stainless steel (DSS) was used as the substrate material, whereas Wilcarbo powder was used as the deposit material for all processes. Table I shows the chemical composition of the substrate. To obtain a fine-grained microstructure, the as-received DSS was solution-treated at 1573 K (1300 °C) for 1 hour, quenched in water and then cold-rolled to a plate through a reduction area of 75 pct. SPC and CC specimens were cut from the thermo-mechanically treated plate to dimensions of 15 × 10 × 8 mm and 10 × 10 × 8 mm respectively. NSPC specimens were cut from the as-received DSS with the same dimensions as SPC specimens.
Table I

Chemical Composition of Duplex Stainless Steel (JIS SUS329J1) in Weight Percent



















Prior to all processes, the specimens were ground by emery paper and cleaned by alcohol to remove oxide layers and irregularities at the outer surface in order to enhance carbon uniformity.[4,5]

SPC process was conducted using compression test machine (Instron), which is equipped with high-temperature furnace under a controlled gas atmosphere. Schematic diagrams of the experimental apparatus and compression process procedure are similar to the previous study.[4] The thermo-mechanically treated DSS was heated to a carburizing temperature and held for 1 h before compression up to 0.5 strain. SPC was performed at different temperatures ranging from 1173 K to 1248 K (900 °C to 975 °C), and strain rates (\( \dot{\varepsilon } \)) of 6.25 × 10−5, 1 × 10−4, 2 × 10−4, and 1 × 10−3 s−1. Table II shows the SPC conditions in this study.
Table II

The Carburized Layer Thickness and the Surface Hardness Values for SPC Process

Compression Strain

Temperature [K (°C)]

Strain Rate (s−1)

Carburized Layer Thickness (μm)

Surface Hardness (HV)


1173 (900)





1 × 10−4 



1223 (950)





1 × 10−3 



2 × 10−4 



1248 (975)

1 × 10−4 



6.25 × 10−5 



NSPC was conducted using the same procedure as SPC at 1248 K (975 °C) and 6.25 × 10−5 s−1.

For the CC process, the specimen was packed in a sealed stainless steel container filled with Wilcarbo powder (carburizing agent), as shown in Figure 1. The carburizing process was performed in a furnace under a controlled atmospheric condition at 1248 K (975 °C) for 133 minutes.
Fig. 1

Schematic diagram of conventional carburizing container

In order to study the flow behaviour of the material, the flow stress was plotted against the strain rate, as represented by the following relation:
$$ \sigma = {\text K}\dot{\varepsilon }^{m} $$
where σ represents the stress, K is the material constant, \( \dot{\varepsilon } \) is the strain rate and m is an exponent generally known as the strain rate sensitivity.

Several techniques were used to characterise the samples. To investigate the microstructure of the carburizing layer, optical and scanning electron microscopy tests were conducted. X-ray diffraction (XRD) of the specimens before and after SPC was carried out. Phase identification was made by X-ray diffraction using Philips X’Pert MPD PW3040 with Cu-kα radiation at 1.54056 Å X-ray wavelength. The surface and layer hardness were determined using Vickers hardness tests at a load of 25 gf. Scratch test was also performed on the samples using microscratch tester with 1000 mN load at 300 μm constant distance.

3 Results and Discussion

3.1 Flow Stress Behaviour

The stress-strain curve of the compressed specimen at different strain rates is shown in Figure 2. The curves exhibit a typical continuous dynamic recrystallization (DRX) characteristic, whereby the stress increases to a peak, followed by softening, and then remains constant. The DRX characteristic is caused by microstructural evolution of work hardening, DRX occurrence and steady state. The flow stress decreases with decreasing strain rate at a fixed temperature. The lowest flow stress is obtained at the slowest strain rate, with a value of 17.4 MPa.
Fig. 2

Stress-strain relationship of the DSS at different strain rates under 0.5 strain (7.5 mm) and 1248 K (975 °C)

Figure 3 shows the relationship between threshold stress and strain rate using the data extracted from Figure 2. The natural logarithmic plots of σ vs\( \dot{\varepsilon } \) indicate that the strain rate sensitivity parameter, m, is associated with superplastic behaviour with a value of 0.52. Most superplastic materials have m values typically between 0.3 and 0.8.[6] The fine grains size of less than 10 μm in the substrate is also a factor for this superplastic behaviour, which will be shown in the following sections.
Fig. 3

Stress and strain rate relationship at 1248 K (975 °C)

3.2 XRD Analyses

XRD analyses were performed on the samples to detect the existence of carbide phases. Figure 4 represents the XRD pattern for the DSS before and after SPC process at 1223 K (950 °C) for 0.5 strain. This figure is taken from previous study.[4] From the relative peak intensity in the XRD pattern, the presence of carbide phases of Fe3C and Cr3C2 were detected after SPC, proving that carburizing had taken place. The details about these carbide phases are however, beyond the scope of this study.
Fig. 4

X-ray diffraction pattern of superplastically carburized DSS surface

3.3 Structural Observations

3.3.1 Microstructure analyses

The typical microstructure of the specimens before and after deformation is taken from previous study.[4] Figure 5a shows the microstructure before deformation. The large and elongated grains are attributed to the thermomechanical treatment of the as-received DSS, which contains solely of α-ferrite (white phase). Figure 5b shows the microstructure of a specimen after compression for 0.2 strain, which yields fine grains with an average size of approximately 3 μm.
Fig. 5

Optical micrograph and SEM image of the specimen microstructure (a) before deformation and (b) after compression at a strain rate of 1 × 10−4 s−1 for 0.2 strain at 1223 K (950 °C)

Figure 6 shows the appearance of specimen before and after SPC at 1 × 10−4 s−1 strain rate and 1223 K (950 °C). The specimen is superplastically compressed without cracks.
Fig. 6

Side view of specimen before (left) and after (right) superplastic carburizing at 1 × 10−4 s−1 strain rate and 1223 K (950 °C)

3.3.2 Carburized microstructure and layer thickness

Figure 7 shows the cross-sectional views of the substrate carburized at 1 × 10−4 s−1 and different temperatures. From the figure, a thick, uniform, smooth and dense morphology of carburized layer was formed on the surface of the carburized specimen. The thickness of the carburized layer ranges from (62.4 ± 0.5) μm to (159.1 ± 0.5) μm, depending on the temperature. Higher temperatures give a thicker layer thickness.
Fig. 7

Cross-section images of SPC specimens carburized at 1 × 10−4 s−1 and different temperatures; (a) T1 = 1173 K (900 °C) (b) T2 = 1198 K (925 °C) (c) T3 = 1223 K (950 °C) and (d) T4 = 1248 K (975 °C)

Figure 8 shows the carburized layer at different strain rates. The carburized layer thickness is found to be within the range of (36.2 ± 0.5) μm to (218.3 ± 0.5) μm. Table II summarizes the carburized layer thickness attained at each condition and the time required. From Table II, it is noteworthy that a carburized layer thickness of (63.0 ± 0.5) μm is achieved although the specimen was carburized for as short as 8 min under 1 × 10−3 s−1 strain rate. The value is higher than that of CC at longer periods. It can be seen that the carburized layer increases with carburizing temperature and time. Slower strain rate means longer carburizing periods, and therefore, a thicker carburized layer is produced at slower strain rates.
Fig. 8

Cross-section images of SPC specimens at 4 different strain rates; (a) \( \dot{\varepsilon }_{1} \) = 1 × 10−3 s−1. (b) \( \dot{\varepsilon }_{2} \) = 2 × 10−4 s−1. (c) \( \dot{\varepsilon }_{3} \) = 1 × 10−4 s−1 and (d) \( \dot{\varepsilon }_{4} \) = 6.25 × 10−5 s−1 

Figure 9 shows the carburized layer for CC and NSPC specimens. Both specimens are carburized at the same temperature and time. A carburized layer thickness of (28.0 ± 0.5) and (96.4 ± 0.5) μm are obtained for the CC and NSPC specimens, respectively. The values are obviously low compared with the value obtained for SPC specimen, which is (218.3 ± 0.5) μm.
Fig. 9

Cross-section images of the specimens for different processes (1248 K (975 °C), 133 min/6.25 × 10−5 s−1); (a) SPC (b) CC (c) NSPC

From these results, it shows how superplastic deformation has managed to produce a carburized layer with thickness much higher than that of the other methods. The carburized layer thickness has differences although carburized at the same temperature and time condition. It gives the interpretation that a different controlling the diffusion process of the carbon atoms in the DSS at each carburizing method. In the CC method, the self and grain boundary diffusion is expected to be the control mechanism. On the other hand, under the NSPC, the conventional creep deformation mechanism where deformation occurs through stress-induced diffusional flow that elongates the grains. It is expected to take place since the microstructure of the as-received DSS is considered coarse about 10 μm. Meanwhile, it is well understood that superplastic deformation occurs mainly from the grain boundary sliding and slipping of the fine grains. Thus, it is expected here that the same grains were to have been transporting the carbon atoms from the surface to the inner part of the substrate through the said grain boundary sliding and slipping. As a result, a much faster diffusion rate of carbon was obtained through the SPC method. The NSPC shows thicker carburized layer as compared to the CC owing to the stress-induced diffusional flow. It is a well known fact that simultaneous deformation could enhance diffusion coefficient through recrystallization.[1113] However the studies diffusion is restricted to the elements originally exits in the material. In this study we have been able to show that the diffusion of outsider element (element which is originally not exits in the material) can also be enhanced through simultaneous deformation, and through superplastic deformation the diffusion process can be further enhanced.

3.3.3 Hardness profile

Table II tabulates the surface hardness of SPC specimens. The surface hardness is found to be within the range of (550.5 ± 5.0) to (1581.0 ± 5.0) HV. The highest hardness is achieved at 1248 K (975 °C) and 6.25 × 10−5 s−1 . The cross-section hardness of the SPC specimen is compared with CC and NSPC specimens, which are carburized at the same carburizing time. Figures 10 and 11 exhibit the hardness indentation images and the hardness gradient from the surface to the core of the carburized specimens, respectively. It is confirmed that the SPC specimen shows the highest hardness from the surface to the inner part of the substrate.
Fig. 10

Optical micrographs showing the variation in hardness indentation from the surface to the core under different processes (1248 K (975 °C), 133 min/6.25 × 10−5 s−1); (a) SPC (b) CC (c) NSPC
Fig. 11

Cross-section hardness profile from the surface to the core of different processes

Although showing thicker carburized layer than CC which as explained previously, NSPC gives lower surface hardness. The amount of carbon exist on the surface can be reflected by the surface hardness. This means that the amount of carbon atoms diffused into the DSS layer under NSPC is lesser than CC. It is again indicates that the grain size of the material does have an effect on the surface hardness which is similar as the formation of carburized layer. Therefore, the smaller the grain size, the harder the carburized layer formed. This is due to the higher value of grain boundaries available on the surface for fine size grains compared to the coarse size grains which promotes more diffusion of carbon atoms at the surface area. As a result, more carbon atoms are available at the surface area of the fine grain microstructure DSS resulting in a high amount of carbon atoms which in return increases the surface hardness. Although NSPC has coarse grain size, it could be diffused further the carbon atoms into the substrate however the diffused amount is lesser.

Meanwhile, under SPC, the carbon atoms are diffused further into the substrate and the diffused amount is considerably higher than the other two carburizing processes. This clearly indicates that the superplastic deformation effect under the compression process not only increase the carbon layer thickness but also improves the surface hardness of DSS as well.

3.3.4 Scratch test

To further evaluate the surface property of the carburized samples, scratch test is performed. Scratch test is one test that is performed on the material to prove the hardness and durability of the materials prior to use for specific applications. Figure 12 indicated the results of scratch test for SPC, CC and NSPC specimens. The scratch test results are identical to the surface hardness results where the SPC sample shows the highest scratch resistance followed by CC and NSPC.
Fig. 12

Depth (nm) vs distance (μm) for scratch test result (a) SPC (b) CC and (c) NSPC

4 Conclusions

In this study, a new method for carburizing DSS, called superplastic carburizing under compression mode has been carried out successfully. In this method, the SPC is strongly affected by temperature and strain rate. The layer thickness and the surface hardness for the SPC process are (281.3 ± 0.5) μm and (1581.0 ± 5.0) HV respectively at 1248 K (975 °C) and 133 min/6 × 10−5 s−1 strain rate. This shows that a thicker and harder carburized layer is attained on the surface of the substrate and also the highest scratch resistance compared with the conventional and non-superplastic carburizing method. The diffusion of carbon atoms into the substrate is highly accelerated via grain boundary sliding and slipping mechanisms. This finding will certainly enhance the potential of SPC process for more explorations.


The authors graciously acknowledge University Malaya for funding this research under the High Impact Research (HIR) (Grant No: J-16001-73804).

Copyright information

© The Minerals, Metals & Materials Society and ASM International 2012