Fracturing in 14MoV6-3 Steel Weld Joints—Including Base Metals—After a Short Time in Service

Abstract

In order to establish the influence of prolonged exposure to high temperatures on 14MoV6-3 steel, three different weld joints were designed. New-to-new material, new-to-used material, and used-to-used material joints were welded using two welding technologies—GTAW and a combination of GTAW + MMA. The weldments were tested by means of microstructure and tensile testing. The results showed that in all weldments, a fracture occurred in the base metal. Also, in the case of the new-to-used welded sample, the fracture always occurred in the used base metal. Since both materials have the same chemical composition, the difference in microstructure was related to long exposure to high temperatures. New steel has a considerably smaller grain size, while the used material underwent grain growth coupled with carbide coarsening, which decreased its strength. The yield strength (YS) of the new material was higher than the YS of the used material, which exhibited similar values in the used base metal and both weldments. It can be assumed that, since deformation starts in the area with the lowest yield point, the used material is the critical place in a given weldment. Therefore, the accurate extent of strength decrease cannot be evaluated based on the testing of new material, i.e., there is a need to reconsider the traditional qualifications of welding technology.

1. Introduction

The main challenges for designers of equipment for long-term high-temperature ser-vice are the minimization of creep and subsequent degradation of mechanical properties [1,2,3,4], surface oxidation, or corrosion. Since no universal solutions are available, in many cases, designers must rely on a limited life, that is, a predefined exploitation period. After this period, components must be substituted. During this refurbishment, welding is used as a common and well-accepted technology. The replaced equipment should be reliable upon further exploitation. On the other hand, reparation in most cases means welding used parts to new parts. This means that weld joints will be formed between different microstructures, even if their chemical compositions are equal or at least similar. The microstructure and properties of the new material are clear, but the microstructure and properties of the used material differ from case to case.

Whether the replacement of the segments is necessary or not depends on the analysis of creep progress. The degradation of microstructures during high-temperature exploitation is followed by the occurrence of individual and spatially independent microcavities (creep) [1,2,3,4]. The intensity of the formation and growth of microcavities largely depends on the working temperature and the stress state in the workpiece [1,2,3,4,5,6]. Therefore, more intensive creep is expected in pipeline elbows because of their complex stress state, which might lead to more intense creep than that in straight pipeline segments. In addition, more intensive creep may cause premature structural failure. To prevent failure and stoppages in production and prolong the lifetime of the equipment, periodic inspections and replacement of worn-out parts should be performed.

The welding of high-temperature and high-pressure components in power plants is used extensively during both construction and maintenance activities. During construction, new materials are used, meaning that all welding evaluations and testing of weldments are performed on the exact material to be used. On the other hand, during long-term service at high temperatures, different types of material degradation can occur, such as grain coarsening, second-phase precipitation, cavitation, oxidation, wear, the occurrence of creep, etc. [4,7,8,9,10]. These changes result in variations (dominantly degradation) in mechanical properties and a higher risk of creep damage and fracture. Therefore, the usual design approach predicts the reparation/change of components after their service life. During this period, periodic evaluation of the structure/properties is continually performed [5,11,12,13,14].

Time-dependent permanent deformation under a constant load or stress at high temperatures is known as creep. Although the material is subjected to a stress lower than the yield strength at that temperature, the material begins to flow slowly. A significant number of failures that occur in constructions/components used at high service temperatures can be solely or partially attributed to creep. At elevated temperatures, material properties are changed: yield strength decreases, vacancy concentration and atomic mobility are increased, diffusional processes are more pronounced, and dislocation glide or climb can occur. Moreover, stresses can lead to cross slip or the cutting of dislocations, introducing new plastic deformation mechanisms such as grain boundary sliding and diffusion creep [4,7,8,10].

In parallel, different processes like recovery, recrystallization, grain growth, precipitation or dissolution of precipitates, overaging, and particle growth (together with surface oxidation and surface depletion of alloying elements) can occur individually or in combination [4,7,8,10].

Development of alloys resistant to creep, on the other hand, must include limitations related to the production line and technological properties like weldability or formability [4,7,8,10].

The rate of deterioration of properties and their level after long-term service at high temperatures is important for the evaluation of residual life and decisions about the further service of damaged pipes [11,12,14]. Several studies have been dedicated to evaluating the level of this deterioration, mainly by comparing the properties and behavior of new and used materials [5,12,13].

Low alloy steel 14MoV6-3 was developed for hot steam pipes and working temperatures over 500 °C. Its predicted service life was designed to be up to 250,000 h depending on the stress in the material [6,15]. At the beginning of service, bainite forms the microstructure of the 14MoV6-3 steel. During long-term high-temperature service, bainite begins to transform into a mixture of ferrite and carbides. This process is accompanied by an increase in grain size. In addition, the precipitation of vanadium carbides can occur. All of these features lead to a decrease in creep resistance at elevated temperatures [3,5,9,12,16,17,18].

Obligatory qualification of welding technology is carried out in accordance with a standardized procedure [19]. This procedure does not take into account changes in the microstructure of the old steel during service. The aim of the qualification of welding technology is to prove that the chosen welding parameters can provide welded joints without unacceptable defects [20,21]. Therefore, precaution should be taken as to how changes in the microstructure will affect the reliability of the obtained results, since the mechanical properties of the used steel are expected to change [4,7,8,9,10].

Finally, working conditions are linked to high temperatures, high pressures, oxidation, and corrosion. Therefore, to satisfy these requirements, a special grade of steel had to be developed, since the material response to degradation mechanisms has the most important role in the prevention of failure and assessment of component service life [22,23,24,25,26,27]. Additionally, requirements for pipes and turbines are different. This leads to large differences in their chemical composition (carbon content and content of Cr, Mo, V, etc.). Steels typically exhibit a martensitic microstructure [23,26,27].

Welding is the one of the most commonly used fabrication technologies. The presence of weld joints is inevitable and is usually considered a “weak point”. Gas tungsten arc welding (GTAW) and manual metal arc welding (MMAW) are two widely used welding technologies in this field.

From the start of service at high temperatures, the microstructure starts to change. The martensitic microstructure undergoes grain growth, decreasing both strength and toughness, together with the coarsening of carbide particles [26,28,29]. This means that the deterioration of mechanical properties starts at the start of service. The change in microstructure is very slow and is usually neglected at this stage.

In the present case, after a short service time, steel was tested in order to understand the changes in the early stages of service. In tension testing of welded joints, as a first step, it is necessary to determine the fracture location.

An assumption is made that the fracture location indicates the critical zone in the weld joint. If fracture occurs in the new steel, it can be assumed that reliability can be evaluated by testing only specimens made of the new steel. On the other hand, fracturing in old steel imposes the need to include it in testing, as the “weakest” spot. Furthermore, knowledge of the fracture position should improve the reliability of the testing/monitoring equipment [5,9,16].

The aim of this study is to establish and estimate the possible difference in testing only weldments made using new material in comparison to the state of weldments that include used parts.

2. Materials and Methods

2.1. Materials

Chemical compositions of the new and used steels tested in this work together with standard requirements are given in

Table 1. Chemical composition of 14MoV6-3 steel and filler materials (mas. %).

	C	Si	Mn	P	S	Cr	Mo	V
EN 10216-2 [15]	0.1–0.18	0.15–0.35	0.30–0.60	0.040 max.	0.040 max.	0.30–0.60	0.50–0.65	0.22–0.32
Used (U) pipe	0.14	0.30	0.56	0.019	0.015	0.39	0.58	0.23
New (N) pipe	0.15	0.22	0.48	0.012	0.010	0.47	0.57	0.24
GTAW-wire [30]	0.08	0.60	0.90	-	-	0.45	0.85	0.35
MMA-electrode [30]	0.065	0.35	1.20	-	-	0.40	1.00	0.50

.

Samples were cut from seamless pipes with an inner diameter of 25.50 mm and wall thickness of 6.3 mm. The used material spent 52,600 h at 540 °C in service.

2.2. Welding

In order to establish the influence of the service deterioration on mechanical properties, three combinations of samples were welded—a new pipe to new pipe (N-N sample), a new pipe to used pipe (N-U sample), and a used pipe to used pipe (U-U sample)—using GTAW and GTAW + MMA welding. Also, two welding procedures commonly used in reparation, GTAW and combination GTAW/MMA, were selected.

Details of the preparation of workpieces for welding and the number of passes are shown in Figure 1.

Preheating. In order to estimate carbon equivalent and preheating temperatures, the following equations were used [31]:

$$CE=C+{\displaystyle \frac{Mn}{6}}+{\displaystyle \frac{Cr+Mo+V}{5}}+{\displaystyle \frac{Ni+Cu}{15}}+{\displaystyle \frac{P}{4}}+0.0024·d$$

(1)

$$T_p=350·{\left(C{E}_u-0.25\right)}^{0.5}$$

(2)

$$C{E}_u=CE·\left(1+0.005·d\right)$$

(3)

where d refers to pipe diametar and T_p refers to preheating temperature.

Preheating temperatures were calcualted to be 185 °C and 175 °C for the new and used pipes, respectively.

Welding parameters. Detailed welding parameters for GTAW and combination GTAW + MMA are given in

Table 2. Welding parameters for GTAW.

Sample	Pass Number	Tungsten Electrode, mm	Filler Material, mm	Current, A	Voltage, V	Protective Gas Ar, L/min
GTAW	1	Ø2.4	Ø2.4	100	11.5	6–8
	2	Ø2.4	Ø2.4	105	12.1	6–8
	3	Ø2.4	Ø2.4	100	12.9	6–8

and

Table 3. Welding parameters for GTAW + MMA.

Sample	Pass Number	Tungsten Electrode, mm	Filler Material, mm	Current, A	Voltage, V	Protective Gas Ar, L/min
GTAW + MMA	1	Ø2.4	Ø2.4	100	11.1	6–8
	2	-	Ø3.25	108	23.5	-

, respectively. All of the welding parameters are selected based on previous experience as well as recommendations from equipment and additional material manufacturers.

Filler Materials. The chemical composition and mechanical properties of filler materials, W MoVSi (EN ISO 21952 [32]), used in GTAW welding and W MoVSi + E MoV B 4 2 H5 and according to producers’ specifications are given in Table 1 and

Table 4. Mechanical properties of used filler materials [30].

	UTS, MPa	YS, MPa	Elongation, A, %
W MoVSi	670	520	24
E Mo V B 4 2 H5	660	510	22

.

Post-Welding Heat Treatment. Stress-relieving post-welding heat treatment (PWHT) was performed according to following procedure: (i) heating to 400 °C in 2 h, (ii) annealing 1 h at 400 °C, (iii) heating to 710 °C in 2 h, (iv) annealing for 2 h at 710 °C, (v) furnace cooling to 200 °C, and (vi) cooling to room temperature in still air, as shown in Figure 2.

2.3. Testing of Welded Joints

Tensile testing was performed using a uniaxial tensile testing machine (INSTRON tensile test machine with a 250 kN capacity, Norwood, MA, USA) at room temperature (RT) and at 450 °C with a testing rate of 5 mm/min. Tensile testing specimens were prepared in accordance with standard SRPS EN ISO 4136 [33]. A technical drawing of the specimen is shown in Figure 3. Three specimens were tested for each set of conditions.

2.4. Microstructure Examination

Microstructure evaluation of weldments was performed in accordance with ASTM E 340-15 [34] and ASTM E 407 [35] standards. After preparation by grinding, polishing, and etching in 2% nital (2% HNO₃ in ethanol), samples were investigated using an OLYMPUS PMG3 light microscope (Tokyo, Japan).

3. Results and Discussion

3.1. Strength

The mechanical properties of new and used materials are given in

Table 5. Mechanical properties of new and used base metal.

	UTS, MPa	YS, MPa	Elongation, A, %	Reduction of Area, % *
EN 10216-2 [15]	490–690	365	20	40
Used pipe (U sample)	567	377	24.1	76
New pipe (N sample)	564	444	26.1	78

* according to standard EN 10028-2 [39], the minimum reduction in the area is 40%.

. Both steels have similar mechanical properties, except for yield strength, which is the consequence of the observed yield point phenomenon in the new steel. Otherwise, it can be assumed that the long exposure to high temperatures, which leads to grain growth, also decreased the yield strength of the used steel [2,7,8,36,37,38].

Typical stress–strain curves obtained in the tensile testing of both new material and used material at room temperature are shown in Figure 4. The stress–strain curve for the new material exhibits a pronounced yield point phenomenon. On the other side, in the stress–strain curve for the used material, a pronounced yield point phenomenon is absent. It is well known that the yield point phenomenon is a consequence of the formation of the atmospheres of solutes at dislocations, known as Cottrell atmospheres [7,8].

The yield point phenomenon can be diminished or removed by introducing free mobile dislocations, or by dilution of Cottrell’s atmospheres at elevated temperatures. Additionally, an increase in average grain size also affects the yield point phenomenon in the same manner. In this case, it can be assumed that after a long period of exploitation at high temperatures, the Cottrell’s atmospheres were diluted. Moreover, the observed grain growth in the used material could also contribute to the absence of the yield point phenomenon [7,8,36]. In other respects, deformation curves for both steels exhibited a similar shape.

The yield strength of the new material (444 MPa) is higher than the value in the new-to-new weldment (396 MPa) (Figure 5 and Figure 6). On the other hand, the yield strength of the used materials exhibited a similar value to the used base metal (377 MPa) and both weldments (used-to-used—373 MPa, new-to-used—376 MPa) (Figure 5 and Figure 6). It can be assumed that, since deformation started in the area with the lowest yield point, the used material is the critical place in the weldment.

Resistance to deformation at high temperatures decreases with an increase in temperature. This behavior is related to more intensive cross-slip of dislocations. Also, the presence of high dislocation density accelerates the diffusion of alloying elements, promoting carbide formation, preferentially at grain boundaries. In this manner, the solid solution becomes depleted. This decreases the solid solution strengthening effect due to easier dislocation movement and lowers the strain-hardening rate, which results in a decrease in ultimate tensile strength. This effect is most pronounced in steels that are exploited for a long time, since the amount of formed carbides is the greatest [5,10,11].

Since differences in the values of mechanical properties obtained during tensile testing (three specimens) are within 2%, average values are given in

Table 6. Mechanical properties of tested welded joints.

Sample	Testing Temperature, °C	YS, MPa	UTS, MPa	Elongation, %	Place of Fracture
GTAW New-to-new	+20	396	519	17.5	Base metal
	+450	310	423	17.4	Base metal
GTAW New-to-used	+20	376	493	9.5	BM—(Used pipe side)
	+450	322	390	3.5	BM—(Used pipe side)
GTAW Used-to-used	+20	373	531	15.6	Base metal
	+450	302	415	15.5	Base metal
GTAW + MMA New-to-new	+20	426	552	17.7	Base metal
	+450	296	423	15.3	Base metal
GTAW + MMA New-to-used	+20	412	533	15.8	BM—(Used pipe side)
	+450	391	415	13.3	BM—(Used pipe side)
GTAW + MMA Used-to-used	+20	347	495	14.2	Base metal
	+450	312	407	13.5	Base metal

. Since the yield stress of the welded joint has no clear physical meaning, this value is apparent and determined as offset stress at 0.2% deformation [7,8,31].

The main feature of all tensile tests performed on welded joints is that fractures always occurred in the base metal. In the case of the new–used combination, fracturing always occurred in the base metal on the used side. These results indicate that the strength of the weld metal (WM) and heat-affected zone (HAZ) is higher than the strength of the base metal (BM). Therefore, the base metal could be evaluated as the critical zone in the welded joint. All the obtained YS values are higher than the minimum proof strength (203 MPa), which is prescribed by the appropriate European standard (EN 10216-2) [15]. This way, a significant amount of strength that would come before plastic deformation may serve as a quantitative rationale for replacing the damaged component rather than the complete structure.

3.2. Microstructure

Typical microstructures of new and used materials are shown in Figure 7 and Figure 8, respectively.

In the new material, the microstructure consists of ferrite with some bainite [17,40], while in the used steel, coagulation of carbides occurred in bainite as well as on ferrite grain boundaries. The most important difference between two microstructures is the grain size. The grains are much larger in the used steel. The used steel was exposed to high temperatures for a long service time, which caused the observed coarsening of the structure. The increase in grain size, in accordance with Hall–Petch equation, led to the observed decrease in yield strength [7,8], as given in Table 6.

In the microstructure of the used steel, individual/isolated microcavities can be seen on the grain boundaries. According to guidelines for rating creep rupture damage [1] and its demarcation criterion, combined with the observed microstructure of the used steel, Figure 8 indicates a creep rating of class 2a. Metallographic analysis of the used steel as an integral part of the welded joint indicates that periodic control of the welded joint, using the metallographic replica method, is necessary. Metallographic results combined with the results of mechanical properties indicate that the remaining lifetime of the repaired structure is another 50,000 working hours, at least.

4. Conclusions

Steel 14MoV6-3 was developed for long-term service at elevated temperatures. It is used in the production of boiler tubes, parts of steam turbines, and boilers. The degradation of such components occurs, among other reasons, due to the complex loading conditions under which they are operating, thus requiring repairment. Since the deterioration of properties starts with the start of service, there was a need to rationalize the influence of 52,600 h/540 °C exposure to elevated temperatures. In that respect, three different combinations of base metals were welded: new-to-new material, new-to-used material, and used-to-used material, together with used material and new material. GTAW welding was applied using filler material W MoVSi (EN ISO 21952), as well as GTAW + MMA using W MoVSi (EN ISO 21952) for the root pass and E MoV B 4 2 H5 for the filling pass. Weldments were tested by means of microstructure and tensile testing.

In all samples, fracturing occurred in the base metal. In the case of the combination of new and used steel, fracturing occurred in the used part. This means that welding technology and parameters were selected correctly, emphasizing that the base metal zone is critical in the welded joint.

The yield strength of the new material (444 MPa) was higher than the value in the new-to-new weldment (396 MPa). On the other hand, the yield strength of the used materials exhibited a similar value in the used base metal (377 MPa) and both weldments (used-to-used—373 MPa, new-to-used—376 MPa). It can be assumed that, since deformation starts in the area with the lowest yield point, the used material is the critical place in the weldment. The decrease of approx. 15% in yield stress (from 444 to 377 MPa) cannot be obtained/registered by testing new-to-new material weldments. Also, it can be assumed that this deterioration of strength will increase in the case of longer service.

New steel has a considerably smaller grain size than used steel. It is assumed that due to exposure to elevated temperatures, the used steel underwent grain growth coupled with carbide coarsening as well as the onset of creep damage. Both changes decreased the effects of grain size and precipitation hardening on strength. Since this effect can be recorded only if used steel samples are tested, used steel should be tested to obtain additional information on the state of construction in order to predict the remaining life of constructions.

Qualification of welding technology requires welding of two parts from new material. On the other hand, since the fracture always occurs in the used base material, the strength of weldments in construction cannot be estimated from testing only new material. This paper clearly identified used metals to be critical for further exploitation. For example, parallel to a common approach to the qualification of welding technology (Standard EN ISO 15612 [41]), testing should be extended with non-standard testing of mechanical properties of new–used welded joints. In this case, the accuracy will be much improved. It is clear that the lowest strength is always in used material, even though both of the applied welding procedures provided adequate strength values. In this respect, the extent of the decrease in strength in the used steel cannot be revealed, i.e., the traditional approach does not provide accurate values of strength. Therefore, reconsideration of the qualification of welding technology (when possible) might be necessary.

The results clearly indicate a decrease in strength, but further investigation should provide a quantitative study of the decrease in the strength of new-to-used weldments as a function of service time and unavoidable used material degradation.