Microstructure, Physical-Mechanical, and Magnetic Characteristics of a Butt-Welded Joint Obtained by Rotary Friction Welding Technology of Bimetallic Pipe

Abstract

The development of technology, including in the oil and gas industry, necessitates the creation of materials with special sets of properties, such as high strength characteristics combined with corrosion resistance. One such material is bimetallic pipe, but we are faced with the problem of creating extended structures and obtaining high-quality butt-welded joints of such industrial bimetallic pipes. The microstructure in different parts of the thermomechanically influenced zone of a butt-welded joint of a bimetallic pipe obtained by rotary friction welding (RFW) was investigated by optical and electron microscopy methods. It was established that during rotary friction welding of the bimetallic pipe in standard mode, one metal flowed into the zone of another. This could be explained by the different plastic properties of the steels that made up the bimetal, which must be taken into account in future welding. Standard RFW mode did not result in the formation of a high-quality weld; defects and discontinuities were observed in the joint area. The maximum hardness values were observed directly in the weld joint. It is concluded that rotary friction welding can be used as a welding technology for bimetallic pipes, but the most attention should be paid to the welding mode to obtain a high-quality butt-welded joint.

1. Introduction

Modern trends in industrial production development are characterized by increasing requirements for quality and operational properties. The development of technology, including in the oil and gas industry, necessitates the creation of materials with special sets of properties, such as high strength characteristics combined with corrosion resistance. Layered metal compositions have found wide application. Such materials include bimetals—materials consisting of two or more dissimilar metals and/or alloys.

Bimetallic pipes are widely used in the oil and gas industry [1,2]. They are used in pipelines for transporting oil and gas, as casing pipes for wells, and in various equipment for the production and processing of oil and gas. The operating conditions in the oil and gas industry often include high-pressure and aggressive chemical environments, such as salt water, acids, and other corrosive substances. Bimetallic pipes, due to their construction from two or more different metal materials, are able to withstand such operating conditions and ensure the durability and reliability of the equipment [3].

However, to create extended structures, it is necessary to obtain butt-welded joints of industrial pipes.

It is known that fusion welding technologies, characterized by complexity and labor intensiveness in their implementation, are being developed for welding bimetallic pipes used in the oil and gas industry. However, as shown in [4,5], these methods, in some cases, contribute to the formation of defects in the weld seam, which lead to premature failures of welded joints during delamination, resulting in significant economic losses due to the shutdown of oil and gas production operations.

To address these issues, solid-state welding technologies can be employed. These technologies prevent the formation of crystallization cracks as well as the formation of segregations and undesirable compounds in the weld joints [6,7]. One method of solid-state welding of bodies of revolution is rotary friction welding (RFW). However, during friction welding, it is necessary to carefully select the welding mode and parameters to ensure the desired properties. These parameters determine the nature of plastic flow in the bimetal material and the temperature regimes of the process. In Russia and abroad, research on joining various single-layer materials and establishing the optimal rotary friction welding parameters to ensure the required joint properties is actively being conducted. All RFW parameters, such as the heating pressure, heating rotation frequency, and forging pressure, are interconnected and influence the formation process and properties of the welded joint. Primarily, the optimal parameters of the RFW process are predominantly determined empirically. In this case, tabular data on already established regimes [8,9,10] for different steels in various countries can be used as guidelines.

And if the issue of developing technologies for ensuring high-quality welded joints of single-layer pipes made of dissimilar materials is being dealt with very intensively in various countries [11,12,13,14,15,16], then joining bimetallic pipes together, including by means of rotary friction welding (RFW) [17], is a very promising direction of development, wherein there are still many blank spots that are attractive for study. The complexity of welding bimetallic materials means that the established modes for steels are not always acceptable for bimetals due to the differences in the mechanical and thermophysical properties of the individual layers.

Statistical methods of planning and processing experiments are widely used to establish optimal parameters [18,19]. Thus, in [18], the parameters of the process of tensile testing of homogeneous joints made of medium-carbon steel AISI 1035, according to the criterion of achieving the maximum tensile strength of solid samples and samples with a stress concentrator, were optimized. The paper [19] presents the results of optimizing the parameters of RFW of dissimilar joints made of austenitic stainless steel AISI 304 and copper. In [20], joints of dissimilar steels AISI 304L and AISI 1040 are described. In [21], the effects of rotation speed and friction time on the properties of a joint of dissimilar steels AISI 4340 and 2205 were investigated. The effects of friction pressure and friction time on the properties of joints of armor steel 500 and DSS 2205 were investigated in [22]. In [23], the effect of friction pressure on the microstructure and mechanical properties of DSS 2205 and AISI 1045 joints was assessed and it was found that, with an increase in friction pressure, the tensile strength and ductility increased due to the suppression of diffusion processes at the interface of the steels, which contributed to the embrittlement of the joint.

Rotary friction welding has now proven itself as a reliable and effective method for joining tubular or cylindrical blanks in the oil and gas industry [24,25,26,27,28].

In [29,30,31], attempts were also made to create a thermomechanical model of the process of friction rotation welding to analyze changes in temperature and plastic flow in the welded materials as well as the distribution of temperature and deformation fields depending on the preparation of the edges of the welded bimetals.

However, the interaction features of various materials that make up the bimetal and their properties and structures during welding require detailed study. This is important not only to ensure the quality and strength of the butt-welded joint but also to prevent potential defects and damage that may occur during operation. Thus, an in-depth study of the welding process of bimetallic pipe-like products becomes necessary for the development of effective assembly methods and technologies aimed at increasing the service life and improving the reliability of the infrastructure, including in the oil and gas industry.

To begin with, it is necessary to have a general understanding of the capabilities and applicability of the developed and applied modes of RFW for single-layer cylindrical structures on bimetallic ones. This is a huge area of further research. The objectives of this particular work are to study the structure and mechanical and magnetic characteristics of the butt-welded joint obtained by the technology of rotary friction welding using standard parameters for welding bimetallic pipes for oil and gas purposes in order to determine the capabilities and prospects of using RFW for bimetallic pipes.

2. Materials and Methods

Structural pipe steel and corrosion-resistant high-alloy steels were used as the materials for manufacturing the bimetallic pipe. A SPECTROMAXx emission optical spectrometer (SPECTRO Analytical Instruments GmbH, Kleve, Germany) was used to determine the chemical composition of the materials in the study. The chemical composition of the studied steels is given in

Table 1. Chemical composition of investigated steels.

Alloying Elements, wt. %
Steel	C	Si	Mn	S	Cr	Ni	Mo	Cu	Ti
Constructional	0.35	0.50	1.55	0.01	0.14	0.17	0.03	0.21	-
Corrosion-resistant	0.08	0.76	1.09	0.02	16.58	9.82	0.2	0.23	0.59

.

The bimetallic pipe was manufactured by the lining method. Lining technology [32] can be used, for example, as a means of reconditioning pump-compressor tubing. Use of this technology makes it possible to make tubes composed of composite materials with high performance characteristics. In this investigation, lining consisted of the joint distribution of a structural steel pipe acting as the outer shell and a thin-walled stainless steel pipe acting as the inner shell [33]. During the lining process, the joining of pipes is primarily achieved due to compressive residual stresses at the interface between the layers. The main advantages of such a joining method compared to other methods of producing bimetal pipes are lower production costs and less time required to master the production of new products. In the oil and gas industry, there is positive experience with pilot-industrial batches of layered composite pipes in onshore (the Netherlands, North America, and Indonesia) and offshore (North Sea, Indian Ocean, New Zealand, Gulf of Mexico (USA), Alaska (USA), Norway, the UK, Philippines, and Malaysia) oil wells. Positive experience in using bimetal pipes in aggressive environments is also reported by companies such as Kawasaki Heavy Industries, Ltd. (Tokyo, Japan), in close collaboration with NAM (Assen, the Netherlands), Shell Oil (Houston, Texas, USA), and Kuroki Tube & Pipe Co., Ltd. (Tokyo, Japan) [34,35,36]. A schematic diagram of the lining process of joining two pipes is presented in Figure 1a: 1—outer and inner layers (pipes), 2—specially produced mandrel.

The butt weld of two parts of bimetallic pipe was carried out on a Thompson-60 automated friction welding machine manufactured in Great Britain (Thompson Friction Welding Company, Halesowen, UK). Continuous drive friction welding (CDFW) was used with a friction time of approximately 3.3 s. The welded joint schema of a bimetallic pipe and a real photo of it are shown in Figure 1b. The parameters of the rotary friction welding (RFW) process, which were used in this work, were recommended for a group of medium-carbon steels by the manufacturer of welding equipment (Thompson-60), namely: lapping pressure of 30 MPa, friction pressure (heating) of 60 MPa, forging pressure of 130 MPa, rotational speed at heating of 800 rpm, and specified value of precipitation of 8 mm.

Longitudinal samples for metallographic analysis were made from welded joints with a weld in the center. Microstructural studies were carried out using a NEOPHOT-21 optical microscope (Asma-pribor, LLC, Moscow, Russia) and scanning electron microscopy using a Tescan VEGA II XMU scanning microscope (TESCAN GROUP, a.s., Brno, Czech Republic) equipped with an Advanced AZtecHKL diffraction (EBSD) analysis system. The specimens were polished using diamond pastes of different dispersion degrees. The final polishing was carried out using colloidal silicon suspension and finishing ion polishing on a Linda machine.

Measurements of microhardness (HV 0.025) were carried out using the method of reconstructed indentation on a Shimadzu HMV-G21DT (Shimadzu Corporation, Kyoto, Japan) device at an indenter load of 0.25 N. Measurements were carried out along the specimen under the study at a distance from the contact line in both directions to the base metal zone.

Kinetic microindentation with recording of the loading diagram was carried out on a Fischerscope HM2000 XYm measuring system using a Vickers indenter (Helmut Fischer Group, Sindelfingen, Germany) at a maximum load of 0.245 N, loading time of 5 s, load dwell time of 20 s, and unloading time of 5 s. The Martens HM hardness (in which elastic deformation is taken into account in addition to plastic deformation) was determined. In addition, the plasticity ratio δ_A = W_r/W_t, where W_r—work of residual molding, W_t—total mechanical work of indentation, was calculated based on the results of the data obtained during instrumental microindentation.

Using the MICROSCAN 600 magnetic noise analyzer (MNA) (Stresstech Group, Vaajakoski, Finland), we determined the MNA parameters, namely the RMS value of the MNA voltage (U) and the number of Barkhausen jumps (N) at a remagnetization frequency of 95 Hz in a package of 10 cycles. The remagnetization frequency was selected by the maximum signal-to-noise ratio. When measuring the MNA parameters, the sample was magnetized along and across the long axis using the analyzer transducer. An attached transducer with a pole cross-section of 3 × 8 mm and a pole spacing of 4 mm served as the primary transducer.

The Barkhausen noise parameters were measured by placing the sensors from the center of the weld toward the base metal of the samples on both sides of the structural and corrosion-resistant steels. For each point, a series of five measurements was performed with sensor repositioning, after which the obtained results were averaged. The maximum deviation of the measured values from the average value did not exceed 5%.

3. Results and Discussion

This article presents the results of the radiographic inspection of the obtained butt-welded joint. This inspection was carried out on a 3D Tomograph. The results are presented in Figure 2.

The radiographic image of the pipe after RFW shows edge non-melting—elongated parallel lines scattered along the weld axis. There is no joining between the welded parts in some places. The appearance of areas of non-melting can be caused by several factors: insufficient pressure, incorrect selection of welding parameters (rotational speed, clamping force), contamination of surfaces, insufficient preparation of edges, or incorrect assembly of the joint. Non-fusion significantly reduces the strength and tightness of the welded joint.

The microstructure of the butt-welded joint and contact zones in the studied sample after RFW, obtained by optical microscopic examination, is presented in Figure 3. During the RFW process, the welded materials in the near-contact areas are subjected to thermal deformation, which in turn leads to the emergence of a thermomechanically affected zone (TMAZ) characterized by a gradient structure.

According to literature sources [37,38,39], near the welded surface after RFW, the metal suffers a local thermo-deformation affect. So we can observe the simultaneous development of phenomena such as heat generation, friction wear of surfaces, continuous formation and destruction of metal bonds between mating surfaces in the process of their relative motion, immediate heating and rapid cooling of local volumes of metal in the presence of high specific pressures, elastic-plastic deformations of the near-surface layers, etc. Under the influence of these processes, several areas with different structures can be identified in the TMAZ.

The first zone is the area of exposure to strains and temperatures exceeding the critical point AC3. In this zone, on the side of constructional steel, a mixed martensitic-bainitic structure is formed (Figure 3a,b). The formation of the hardening structure in the TMAZ of structural steel is associated with local heating to temperatures above the temperature of polymorphic transformation and further rapid cooling due to the realization of rapid heat exchange between the narrow zone heated in the process of friction and the surrounding areas of the metal.

The second zone is the area of strain and temperature exposure in the range from AC1 to AC3. In this zone, the appearance of a fibrous structure (Figure 4) of the steel is observed, consisting of alternating bands of hardening structure and finely dispersed ferrite-cementite mixture (Figure 3b). The direction of this structure smoothly changes from parallel to perpendicular to the welded joint.

And one more zone is the recrystallization area, in which the temperature is close to AC1. This area is characterized by the absence of visible signs of plastic deformation; also, we can talk about the formation of equiaxed small grains. The recrystallization area is followed by the base metal (Figure 3c,d).

Figure 4 shows a contrast map and misorientation map based on the EBSD analysis results of the butt-welded joint specimen. The appearance of a fibrous structure of the steel, changing direction from parallel to perpendicular, is observed after RFW. The fibrous or directional structure in friction rotation welding is not a defect. These are the features of the plastic flow of the metals during welding. The fibrous structure is the processed textures of the original workpiece during heating. Texture formation is observed; in a large number of grains, the direction [001] corresponds to the direction of lining. It is also seen that there is an influx of one material into the zone of another.

It is known [40,41,42] that the process of forming an unbreakable joint in the contact zone during welding in the solid phase proceeds in three stages: the formation of physical contact, activation of the contact surfaces, and volumetric interaction with the formation of strong chemical bonds both in the plane and in the contact zone volume. However, to ensure the required strength of the joint, it is necessary to develop relaxation processes such as recrystallization or heterodiffusion.

In the EBSD maps presented on Figure 5, it is clear that an oriented boundary between the two materials is not observed. This allows us to consider that, during the rotary welding of bimetallic pipe, interaction in the joint zone in some areas is not limited to the formation of interatomic bonds and there is mutual volume recrystallization.

Figure 6 shows the results of the micro-X-ray spectral analysis of the near-contact zones of the welded joint. Scanning along the line perpendicular to the contact zone does not reveal a transition zone with a smooth change in the content of the main alloying elements (chrome, nickel, manganese), indicating their diffusion movement into each other. The contents of chrome, nickel, and manganese in the contact areas of the welded materials are in accordance with their chemical composition in the initial state. Thus, the role of the recrystallization process in the final stage of welded joint formation is predominant, which was also confirmed in [43].

Figure 7 shows the recrystallization map of the investigated welded joint. In Figure 7, grains within which the crystal lattice has a misorientation angle of less than 2° are indicated in blue. Such grains are considered as recrystallized. The yellow color indicates grains with substructures, within which the angle of lattice misorientation does not exceed 2°, and between subgrains—more than 2° but less than 15°. Grains with an angle of misorientation of atomic planes greater than 15° are highlighted in red; they are considered as deformed.

For the welded joint obtained by RFW technology, near the fusion zone is characterized by the presence of mainly deformed grains (red color on the recrystallization map), which is explained by the intense mechanical impact during welding. As the distance from the weld increases, their amount decreases.

The character of microhardness changes in the fusion zone after RFW of the investigated butt-welded joint, as presented in Figure 8. The hardness of the corrosion-resistant steel is 330 HV 0.025, while the structural steel exhibits a hardness of 390 HV 0.025.

The distribution of microhardness across the welded joint is not uniform and is characterized by a maximum value (658 HV 0.025) at the contact line and minimum values (280 HV 0.025 and 209 HV 0.025) at the boundary zone of the two different steels. There is an increase in microhardness in the weld zone due to processes occurring during welding, such as the formation of deformed structures with increased dislocation density, crystallization and recrystallization of the metals, martensite formation, and changes in the grain structure. This leads to the formation of harder regions around the butt weld, which is reflected in the microhardness distribution.

The microhardness values in the near-contact areas are slightly lower (316–380 HV 0.025) due to the transition from a martensitic structure to a ferrite-carbide mixture, particularly in constructional pipe steel.

At a distance from the contact line, the values of microhardness sharply decrease and become approximately equal to the hardness values for the initial state of the steels investigated in the work. For corrosion-resistant steel, hardness is characterized by more uniform values 332–342 HV 0.025. For constructional pipe steel, the spread is 372–421 HV 0.025, which is 13%, respectively.

The distribution of values of micromechanical characteristics (Figure 9a,b) agrees with the obtained metallographic studies. In the welded joint zone, there is an increase in microhardness HM. The plasticity parameter δA characterizes the plasticity of materials by the share of plastic deformation in the total elastic-plastic deformation. The higher the value of this parameter, the higher the plasticity of the material. It is known that for tensile plastic materials, this parameter is δA ≥ 0.9. The plasticity δA decreases in the TMAZ, which increases the propensity of this zone to cracking.

Thus, the data of kinetic microindentation agree with the results of the metallographic and durometric analyses and indicate hardening of the welded joint.

Figure 9c shows the distribution of Barkhausen magnetic noise parameters across the width of the welded joint. The use of attached magnetic devices has proven itself well, including for assessing changes in welded seams of steel products [44,45,46,47]. So scanning and measurement of MNB parameters was performed using an overhead sensor pointwise along the length of the sample. Parameters such as the RMS value of Barkhausen magnetic noise and the number of Barkhausen jumps were investigated (Figure 9c). The distribution of magnetic parameters is characterized by a local increase in the butt weld zone due to the formation of quenched and hardened structures. The magnetic parameters can be used to estimate the changes in weld joints and help to detect the formation of hardening structures [48].

4. Conclusions

The results of a complex study of the structure and mechanical and magnetic characteristics of a butt-welded joint obtained by rotary friction welding technology allowed us to draw the following conclusions.

The results of radiographic inspection showed the presence of serious defects in the product after the widely used regime of rotary friction welding. This may indicate that the chosen mode used for conventional (single layer) pipes is not applicable for bimetallic pipes. To improve the quality of welded joints, it is necessary to optimize the welding parameters as well as to improve the preparation of edges of welded objects.

It is established that, during rotary friction welding, the influx of one metal into the zone of another is a problem. This could be explained by differences in the plastic properties of the materials.

The microstructural investigation showed that the formed TMAZ was characterized by a gradient structure. Three microstructural zones could be distinguished quite clearly. The distribution of microhardness across the welded joint was inhomogeneous and was characterized by the maximum value in the weld zone and minimum values in the two different steels.

The possibility of using the Barkhausen magnetic noise scanning method for determining and controlling the width of the TMAZ as well as the hardness level of the resulting welded joint is shown.

The obtained results showed which way to go next. Further experiments are being conducted to vary both the force during heating and forging and the edge preparation method, taking into account the different plastic properties of the materials comprising the bimetallic pipe. Thus, optimization of the welding parameters is required, which will further allow us to obtain a reliable butt-welded joint.