Improved Microstructure Evolution and Corrosion Resistance in Friction-Welded Dissimilar AISI 1010/D3 Steel Joints Through Post-Weld Heat Treatment

Abstract

To achieve the desired material properties of automotive components made by friction welding, post-weld heat treatment is critical. The high temperatures encountered during the friction welding of steels can lead to changes in the microstructure, especially in the heat-affected zones. In the present work, a D3 tool steel and an AISI1010 structural steel are friction welded by varying the rotational speed, and this is followed by post-weld heat treatment. Microstructural evaluation was performed on the friction-welded joints and those produced after heat treatment. Micrographs taken by scanning electron microscope show the formation of distinct zones with ultrafine grains at the interface. Zone measurements at the interfaces of the joints provide information on the proportions of the various zones formed during friction welding. Depending on the rotation speed, the width of the heat-affected zone (HAZ) can range from 10.8 to 19.5 mm, and the width of the total deformed zone varies from 700 to 1070 µm. The width of the fully plasticized zone is between 48 and 380 microns. The region of the friction-welded joint at 1600 rpm shows fine ferrite grains with a width of 48 µm FPDZ, which increase the strength of the joint according to the Hall–Petch equation. Primary carbides are dissolved in the ferrite matrix, and secondary carbides are formed due to the effects of alloying elements such as chromium in particular. Although the formation of secondary carbides cannot be prevented, at higher speeds the primary carbides are dissolved and the tendency to form secondary carbides is reduced. Post-weld heat treatment helps to redistribute these phases and leads to a more homogeneous material structure. The results show that post-weld heat treatment greatly improved the corrosion resistance of dissimilar AISI 1010/D3 steel joints produced by means of friction welding. Coarse grains have been eliminated, and thus the galvanic corrosion at the weld interface is alleviated and reduced. Post-weld heat treatment reduces the corrosion rate and weight loss significantly, by 54.8% and 60%.

1. Introduction

Friction welding is a solid-state joining process that produces the coalescence of materials in contact while rotating or rubbing each other under pressure [1]. In friction welding, mechanical work is converted into thermal energy at the interface of the workpieces [2]. Sufficiently high forging pressure at elevated temperatures makes the mating surfaces form a sound metallurgical bond, and diffusion and mechanical mixing could be the joining mechanisms in friction welding [3]. Though friction welding is suited for both similar and dissimilar components, it has been especially resorted to for cases of joining dissimilar metals that cannot be joined by conventional fusion welding processes [4]. Fusion welding is unsuccessful for dissimilar materials due to incompatibility in terms of metallurgical aspects and differences in melting point, thermal properties, etc. [5]. As there is no bulk melting of materials in friction welding, many defects associated with fusion welding are reduced or even avoided. Thermal degradation is low as there is no fusion. The heat-affected zone is very narrow in friction welding [6]. There is no production of poisonous waste gas [7]. Its reproducibility is high, and there are high material savings and short production times. High energy savings are another advantage. One of the major process parameters for friction welding is the rotational speed [8]. There are varied experiences with the effect of rotational speed, and hence it has become a major study among researchers. In the detailed review by Li et al. [9], while highlighting the effect of process variables, the effect of rotational speed is discussed. The speeds tried were 3800, 4300, and 4500 rpm for the friction welding of the superalloy GH4169. The reported tensile strength was a maximum of around 1350 MPa for the joint made at 4300 rpm, and the strengths were approximately 1250 MPa for the lower and higher speeds. However, they also reported another case of friction welding of Ti17 with which there is no improvement in tensile strength by varying the speed from 700 to 1500 rpm. The strength remained constant at about 1100 MPa.

Selvamani et al. [10] carried out a sensitivity analysis with process parameters using friction-welded AISI 1035 steel rods. While studying the effect of rotational speed in the range 1100–1700 rpm, they observed considerable improvement in tensile strength under a specific combination of friction pressure, forging pressure, friction time, and forging time. However, this trend of improvement in tensile strength was not maintained for other combinations of pressure and time. From the sensitivity analysis, they inferred that small variations in friction pressure and friction time affect the tensile strength to a greater extent. They found that tensile strength is less sensitive to rotational speed. Ozdemir et al. [4] performed studies on the effect of rotational speed on the friction-welded joints of AISI 304L and AISI 4340. They varied rotational speeds from 1500 rpm to 2500 rpm, keeping other process parameters constant. They observed that the width of the fully plasticized and deformed zone (FPDZ) decreased, and that the constituents are a mixture of parent metals. Moreover, they observed that both the yield and tensile strengths increased with increasing rotational speed. They warned that the width of the FPDZ should not be allowed to become thinner, as this might be detrimental to joint strength. Benkherbache et al. [11], in their studies on the friction welding of steel to steel, tried to find the effect of rotational speed in the range of 900–1800 rpm. They reported that an increase in speed reduced the elastic limit, which is undesirable.

Hascalik et al. [12] observed that an increase in rotational speed increased the hardness of the HAZ and FPDZ but tensile strength, as well as percentage elongation, were reduced. Kirik and Ozdemir [13] and Mercan et al. [14] confirmed adverse effects on mechanical strength due to a decrease in FPDZ with an increase in rotational speed, combined with higher friction pressure. Kirik and Ozdemir [15] recommend that higher rotational speeds should be accompanied by a lower friction time. Bati et al. [16] and Mercan and Ozdemir [17] comment that friction time alone has no effect other than increasing flash. It can be understood that not one or two but a combination of all variables together results in a specific change, which cannot be predicted so easily. This study attempts to verify the quality of weldments formed with the judiciously selected parameters and to study the effects of rotational speed. The study analyses the effect of rotational speed through microscopic studies which form a base for the mechanical behavior.

2. Materials and Methods

The overall process of friction welding comprises two important phases, namely the friction phase and the forging phase. During the friction phase, speed remains constant along with a constant friction pressure through the prescribed time. Torque is also almost constant. By the end of the friction phase, as forging has to take place, forging pressure and hence torque are increased. Torque has a peak at the end of the friction stage, presumably due to a decrease in heat generation and a resulting increase in material resistance [18]. Forging pressure is kept high for the completion of the welding process. As the prescribed time lapses, the machine is released, and the product is ready. The friction welding machine developed in-house is shown in Figure 1.

D3 is a high-carbon, high-chromium steel that possesses higher resistance to abrasion and good dimensional stability under heat treatment. It is also resistant to decarburizing and, hence, can be effectively nitrided. Other uses of this material are for the making of blast nozzles and spindles. When this material is effectively used for such applications, there is a possibility to make it functionally graded to reduce costs. AISI 1010 is a structural steel with a lower cost. Hence, friction welding a piece of D3 steel to AISI 1010 steel is preferred and is studied in this work. The suppliers’ catalog presents the chemical composition of these materials, as shown in

Table 1. Chemical composition (in mass %) of workpieces.

Material	C%	P%	Si%	S %	Mn%	Cr%	Fe%
Steel D3	2.10	0.03	0.30	0.03	0.40	11.50	Remaining
AISI 1010	0.1	0.04	-	0.05	0.45	-	Remaining

, with the balance being iron. An in-house-constructed friction welding machine with a programmable logic controller has been utilized here. Regarding the observations of Sahin [8] and Sathya et al. [19], process variables have been selected judiciously. To study the effect of rotational speed, a forging pressure of 0.6 MPa and a forging time of about 20 sec were chosen.

Table 2. Process parameters used in the friction welding of specimens.

Specimen	Rotational Speed, rpm	Axial Pressure, MPa	Forging Time, s
S1	1600	0.6	24.6
S2	1400	0.6	9.2
S3	1200	0.6	10.6

presents the process parameters used for the analysis. The last column is the forging time, different values of which were observed. The friction-welded specimen is shown in Figure 2. There is an axial shortening of workpieces due to the formation of flash [20].

Concerning Emre and Kacar [21] and Ma et al. [22], to relieve residual stresses and achieve a uniform microstructure in the welded interface, heat treatment was carried out on the welded specimens. As per the vendor’s recommendations, the specimens were gradually heated to 870 °C within 30 min, held at this temperature for 30 min, and then furnace-cooled. As these heat-treated specimens performed better than those that had not been heat-treated [23], they were preferred for the study of the effects of rotational speed, since it is considered to be a major factor [24,25,26,27]. For the microscopic examinations, the welded and heat-treated pieces were halved longitudinally. Mechanical polishing and chemical etching with a solution of ferric chloride were performed on the surfaces to be examined. Microscopic examinations and SEM studies were carried out on the ZEISS EVO 18 SEM equipment.

3. Results and Discussion

The examination included panoramic analysis, micrographs, SEM, EDX, and XRD. The specimens are named S1, S2, and S3, which correspond to the speeds of 1600, 1400, and 1200 rpm, respectively. Each of these methods offers distinct insights into friction-welded joint interfaces that are essential to assessing the joints’ quality, structure, and composition. SEM offers a thorough visual representation of the surface and microstructure, whereas EDX supplies information on the elemental composition. On the other hand, XRD offers crystallographic data and can be applied to the determination of residual stress. By combining these methods, a thorough understanding of the joints’ characteristics and quality is acquired.

The panoramic view provides an elongated field of view greater than that of the human eye, extending up to about 160°. It helps to visualize the specific spot of interest mapped onto the overall geometry, which is normally impossible. Figure 3a–c shows panoramic views of the three specimens that correspond to rotational speeds of 1600 rpm, 1400 rpm, and 1200 rpm, respectively.

Table 3. Measurements of the widths of different zones.

Zones	S1 1600 rpm		S2 1400 rpm		S3 1200 rpm
HAZ (in mm)	10.8		14.8		19.5
PDZ (in µm)	D3 Steel	AISI 1010	D3 Steel	AISI 1010	D3 Steel	AISI 1010
	200	180	380	220	520	360
DZ (in µm)	D3 Steel	AISI 1010	D3 Steel	AISI 1010	D3 Steel	AISI 1010
	150	550	200	680	320	750
Total DZ (in µm)	700		880		1070
FPDZ (in µm)	48		220		380

shows a reduction in the plastically deformed zone as the rotational speed increases from 1200 rpm to 1600 rpm. Table 3 details the sizes of various zones created during welding for two types of steel—D3 Steel and AISI 1010—at various rotational speeds. The breadth of the heat-affected zone (HAZ) varies from 10.8 to 19.5 mm. For D3 Steel and AISI 1010, the width of the plastically deformed zone (PDZ) ranges from 150 to 380 µm and 180 to 360 µm, respectively. For D3 Steel, the width of the deformed zone (DZ) spans from 200 to 520 µm, while for AISI 1010, it extends from 200 to 750 µm. The width of the entire deformed zone varies from 700 to 1070 µm. The fully plasticized deformed zone (FPDZ) is 48 to 380 µm wide. The decreasing order of the fully plasticized deformation zone (FPDZ) in particular can be clearly observed with the increase in rotational speed.

3.1. Microstructural Evaluation

Observations on micrographs provide insight into the effect of rotational speed on friction-welded components made of D3 steel and AISI 1010. Figure 4a–c are the SEM micrographs of specimens S1, S2, and S3 (after Nital etching) at the welded zone, which correspond to rotational speeds of 1600 rpm, 1400 rpm, and 1200 rpm, respectively. When the speed of rotation increases, the plasticized zone becomes narrower, and its width decreases. Primary carbides become dissolved in the ferrite matrix, and the formation of secondary carbides takes place. Secondary carbides are formed due to the effects of alloying elements. Chromium, in particular, forms secondary carbides. At the rotational speed of 1600 rpm, the plasticized zone is thin and narrow, and it is more uniform in S1. It can also be interpreted as a fully plasticized region, as shown in the micrograph. Recrystallization is evident, and the grain flow is more visible. SEM micrographs reveal three kinds of zones in the friction-welded joint at 1200 rpm after post-weld heat treatment. Distinct zones such as the partially deformed zone (PDZ), the deformed zone (DZ), and the fully plasticized deformed zone (FPDZ) are clearly seen. The plasticized zone appears wider and thicker because of the greater amount of mass transferred at the weld interface. The presence of primary carbides, martensite, and cementite could be attributed to the decrease in joint strength. Figure 5a shows how the primary carbides and cementite networks are broken. Carbon becomes dissolved, and finer grains of ferrite appear. A fine grain structure leads to higher strength as per the Hall–Petch relationship [28,29]. At a rotational speed of 1400 rpm, the plasticized zone contains a small amount of coarse pearlite and martensite. There is an intermixed zone, and the primary carbides are not fully dissolved. In some areas, alternate layers of ferrite and cementite are seen. A larger quantity of carbides from chromium and a small amount of molybdenum are also present as secondary carbides. Figure 5b shows spots of martensite, patches of alternate layers of ferrite and cementite, and the dispersion of spheres of the interstitial solution of ferrite. There is a partial diffusion of carbides, and the transferred materials are not fully dissolved in the interfacial region. A smaller mixture of pearlite and cementite is seen along with clusters of martensite, as in Figure 5c. Diffusion is not proper, and the primary carbides are not dissolved. The intermixed region is characterized by alternate layers of pearlite and cementite. Martensite and austinite lead to the formation of intermetallic compounds. Since this intermetallic compound is thick, hard, and brittle, failure usually occurs at the joint interface. The particle analysis of the SEM micrographs of the weld interface at 1600 rpm, 1400 rpm, and 1200 rpm reveals particle densities of 0.946 particles/µm², 0.854 particles/µm², and 0.881 particles/µm², respectively.

3.2. EDX and XRD Analysis

The results of the EDX analysis are shown in Figure 6a–c, respectively, for the specimens S1, S2, and S3. The figures show the concentration profiles of Fe and Cr across the interfacial region of the joint. Post-weld heat treatment was found to be beneficial, and it was found to enhance joint strength. Post-weld heat treatment often reduces the tendency for phase transformations and residual stresses. Figure 6b shows the bonding mechanism of dissimilar steel joints where Fe can mix with Cr constituents and form intermetallic phases such as M₂₃ and C₆. Figure 6c clearly shows the wide gap between the concentration profiles of Fe and Cr. In the interfacial region, there is less of a tendency for the formation of intermetallic compounds. But, instead of a brittle intermetallic compound, ferrite is present as refined, recrystallized, finer grains. However, at lower rotational speeds, as shown in Figure 6c, there is a greater tendency for the formation of metastable, deformed phases. This is revealed in the concentration profiles of Fe and Cr and how they are ready to intermix and form intermetallic brittle compounds such as M₂C, which is hard and brittle. The XRD result at the interface, as shown in Figure 7, confirms this strong tendency.

Figure 7 shows the XRD for specimens S1, S2, and S3, along with that of a non-heat-treated specimen as a reference. The non-heat-treated specimen was friction-welded with the same process parameters but at a rotational speed of 1600 rpm. The XRD reveals the presence of a greater amount of M₂₃C₆ at the interface of the joint made at 1400 rpm. More and more quantities of finer grains of ferrite are found at the interface as the rotational speed is increased. There is a wide gap between the concentration profiles of Fe and Cr. At lower rotational speeds, there is a greater tendency for the formation of metastable, deformed phases of M₂C (chromium carbide). This is brittle in nature. Hence, at higher rotational speeds, the tendency for the formation of M₂C is less and results in better mechanical properties.

4. Corrosion Effects

A galvanic couple is formed between two materials welded together, especially when they are exposed to aqueous environments. This galvanic couple is a matter of concern as it promotes corrosion. Contents of Cr, Mo, and N reduce corrosion cracking caused by the presence of ferrite. A higher ferrite content and coarse grains are the major factors which decrease both the corrosion resistance and the mechanical properties of welded joints. Corrosion is found to be a threat to the welded specimens. This section is concerned with the assessment of the corrosion behavior of non-heat-treated and post-weld heat-treated friction-welded specimens made of the steels AISI D3 and AISI 1010.

Friction welding between dissimilar metals is highly prone to galvanic corrosion, especially at the interface. Depending on the position of the metals in the galvanic series, the potential difference will be set up. The higher the potential difference, the faster the less noble metal will be corroded. Hence, one of the methods by which galvanic corrosion could be controlled is by the use of metals or alloys as close as possible to the galvanic series. The galvanic series of metals ranges from the most noble, which are classified as cathodic, and the least noble, which are anodic. The combination of plain carbon steel and high-chromium steel falls very marginally within the area of galvanic corrosion risk. Figure 8 shows a heat-treated component welded at a rotational speed of 1600 rpm before and after corrosion.

The corrosion rate is calculated as per the formula given in Equation (1). Wb − Wa is the weight loss in mg, A is the exposed area in sq.cm, T is the time of immersion in the medium in hours, and d is the density of the sample in g/cc. Since two dissimilar steels are involved, density is considered to be the average of the density of AISI 1010 (7.87 g/cc) and D3 steel (7.7 g/cc).

$$CR={\displaystyle \frac{k· (Wb-Wa)}{ A· T· d}}$$

(1)

$$d=0.5 (d1+d2)$$

(2)

$$CR={\displaystyle \frac{k· (Wb-Wa)}{0.5 · A· T· (d1+d2)}}$$

(3)

Initially, the corrosion rate was very high because of the fully unprotected surface of the AISI 1010/D3 steel joint. More oxygen was dissolved in the water. So, the formation of rust, which is iron oxide, began. As more and more iron oxide covered the surface, it acted as a barrier for corrosion. So, the corrosion rate started to decrease along with the weight loss, as shown in Figure 9. Pitting corrosion started at this stage in the localized region. After 30 h, the rate of corrosion decreased, and corrosion continued in the localized spots.

Heat treatment modifies the microstructure of steel, relieving internal stresses and refining the grain. As reflected in the lower amount of weight loss of heat-treated joints, this results in better corrosion resistance. In untreated steel, this rusting occurs immediately because of the absence of a protective oxide film. Heat-treated steel forms a more even and protective oxide layer which retards corrosion. Due to the post-weld heat treatment of the dissimilar steel joint, there is a significant reduction in weight loss of 50 to 60%.

In corrosive environments, heat-treated AISI 1010/D3 steel joints show a comparatively lower corrosion rate. Heat treatment correspondingly decreases the corrosion rate of AISI1010/D3 steel weldments. The corrosion rate of both untreated and heat-treated steel joints decreases with time. This reduction indicates the creation of a protective oxide layer on the surface, inhibiting further corrosion. The untreated steel clearly demonstrates a far higher rate of corrosion throughout and suggests that the oxide layer is less effective or more porous (Figure 10). When it comes to long-term exposure to acidic conditions this makes heat-treated steel a better option as it is more resistant to corrosion. Overall, the corrosion rate of a post-weld heat-treated AISI1010/D3 steel joint was more effectively reduced to about 54.8%.

Figure 11 and Figure 12 show corrosion at the interface of friction-welded AISI 1010/D3 steel joints without heat treatment and with heat treatment. Figure 11a shows the friction-welded joint prior to heat treatment. The D3 tool steel showed clear grain boundaries and distinct precipitated carbides at the interface. A segmented micrograph (Figure 11b) shows the deposition of grain-boundary carbide, localized corrosion, and reduced ductility. Slice analysis (Figure 11c) reveals that irregular eutectic phases promote corrosion. A two-dimensional profile (Figure 11d) shows surface damage and imperfections and pitting corrosion. Figure 11e indicates significant corrosion and the roughness of the initial untreated joint. Figure 12a shows the fine distribution of ferrite and pearlite and that the D3 steel hosts fine carbide phases. A segmented micrograph (Figure 12b) shows fewer pits and corrosion sites when compared to the second segmented micrograph (Figure 11b), indicating better corrosion resistance. Slice analysis (Figure 12c), the 2D profile (Figure 12d), the roughness profile (Figure 12e), and the 3D profile (Figure 12f) indicate less corrosion damage and a smoother surface. These figures indicate the relative microstructural changes, surface topology, and corrosion damage.

Severe pitting corrosion and a rough surface morphology are found in the untreated joint (Figure 11). Nitric acid aggressively penetrates without any protective oxide film, resulting in a massive loss of material. The corrosion resistance of heat-treated steel (Figure 12) was enhanced significantly by microstructural changes. The presence of a stable oxide layer is more protective to corrosion in nitric acid and avoids pitting and material degradation. The heat treatment improves the corrosion resistance of friction-welded AISI 1010/D3 steel joints due to the refinement of the microstructure, the reduced segregation of carbides, and the formation of an oxide layer. These findings validate post-weld heat treatment as an essential procedure to enhance the durability and dependability of dissimilar metal joints in aggressive service conditions.

Primary carbides and secondary carbides are seen as dark and grey spots, respectively (Figure 11a). During the heat treatment, carbides are precipitated as eutectic phases. They are distributed all around without visible of grain boundaries. Accordingly, Figure 12a reveals the uniform distribution of eutectic phases all over the ferrite matrix. The presence of carbides at the grain boundaries results in better ductility, since the sliding of grain boundaries is prevented in most situations. Prominent carbides, such as M₇ C₃, M₂₃ C₆, M₂ C, etc., are found in the heat-treated joint. This is further confirmed in the XRD analysis, as explained before in Figure 7. As the post-weld heat treatment temperature is increased, the primary carbides and secondary carbides do not dissolve, and they eventually increase. The austenite content keeps on decreasing when the post-weld heat treatment temperature is increased. Hence, it is evident that the heat treatment procedure changed the microstructural constituents. The morphological features were studied and pits were seen in many places on the surface of the D3 tool steel specimen. It is evident that a post-weld heat-treated joint exhibits less damage than an untreated joint. Corrosion resistance obviously increased in the heat-treated joint due to the oxidizing nature of nitric acid. A reduction of nitric acid in water takes place, and the corrosion product iron tri-nitrogen oxide is formed. The reaction can be shown as indicated below. This reaction leads to the evolution of nitrogen (II) oxide and the production of Fe (NO₃)₂, which lead to further coloration of the medium.

Fe + 4HNO₃ Fe → (NO₃)₂ + 2H₂O+ 2NO₂

(4)

5. Conclusions

In general, with an increase in rotational speed, the integrity of a joint increases, but with a reduction in forging time. It is understood that if the forging time is not properly controlled, a very thin FPDZ will form, which would be detrimental to the joint [30,31]. Post-weld heat treatment was found to be significant in terms of an improvement in the favorable microstructural evolution of dissimilar steel joints.

• Zone measurements at the interface of the joints shed light on the proportions of the various zones formed during friction welding. Depending on the rotating speed, the heat-affected zone (HAZ)’s width can range from 10.8 to 19.5 mm. With values ranging from 180 to 750 µm, the plastically deformed zone width varies with different kinds of steel and different rotational rates. The width of the entire deformed zone varies from 700 to 1070 µm. The fully plasticized deformed zone is between 48 and 380 microns wide.
• The interface of a joint friction-welded at 1600 rpm shows fine grains of ferrite, which enhance the joint’s strength as per the Hall–Petch equation. SEM micrographs also reveal that the joint fabricated at 1600 rpm has a thin plasticized region (48 microns width) with ultra-fine grains at the interface.
• At lower rotational speeds, there is a greater tendency for the formation of metastable, deformed phases such as M₂C, which is hard and brittle.
• Though the formation of secondary carbides cannot be prevented, at higher rotational speeds primary carbides become dissolved. This reduces the tendency for the formation of an intermetallic compound and increases joint strength.
• Post-weld heat treatment results in controlled cooling, which can refine the microstructure and improve the overall material properties of dissimilar steel joints.
• The heat treatment of joints provided significant improvements in corrosion resistance. The untreated joint revealed coarser grains, segregated carbides, and deep pitting corrosion owing to aggressive nitric acid attack on the surface that resulted in a higher loss of material; whereas, a homogeneous distribution of the eutectic phases was observed on the surface of the heat-treated joint, which exhibited a more stable protective oxide layer reducing the probability of pitting behavior and improving the joint’s corrosion resistance.
• Overall, the corrosion rate of post-weld heat-treated AISI1010/D3 steel joints was more effectively reduced by about 54.8% and the weight loss by 60%. Post-weld heat-treated AISI1010/D3 steel joints are preferable in applications where failure due to corrosion occurs.