Effect of welding parameters on pitting behavior of GTAW of DSS and super DSS weldments

Abstract

This work focuses on the effect of welding parameters on corrosion behavior of welded duplex stainless steel (DSS) and super duplex stainless steel (SDSS). The effect of welding parameters, such as heat input, inter-pass temperature, cooling rate, shielding/back purging gas, on corrosion behavior was studied. DSS and SDSS pipes were welded with Gas Tungsten Arc Welding (GTAW) process. After welding, the test samples were non-destructively tested to ensure no defects and test samples were prepared for microstructural examinations and ferrite content measurements. The root region had complex microstructure because of the repetitive heating of the zone during different weld layers. It was observed that at low heat input desirable microstructure was formed. The test samples were subjected to corrosion tests, i.e. ASTM G48 test for the determination of pitting corrosion rate, potentiodynamic polarization tests, and potentiostatic tests to verify susceptibility of the alloys to corrosion attack. DSS weldments had CPT in between 23 °C to 27 °C and SDSS weldments had CPT between 37 °C to 41 °C in potentiostatic measurements. The corrosion test results were correlated to the microstructures of the weldments. The pitting resistance of individual phases was studied and the effect of secondary austenite on corrosion attack was also observed.

Keywords

Duplex stainless steels ; Super duplex stainless steels ; GTAW ; Pitting corrosion ; Heat input ; Secondary austenite ; Inter-metallic phases

1. Introduction

Duplex stainless steels (DSS) [22% Cr] and super duplex stainless steels (SDSS) [25% Cr] are composed of unique ferrite/austenite microstructure which makes them superior material than conventional AISI 316 austenitic stainless steels [1]  and [2] . Super duplex stainless steels are upgraded versions of DSS. They exhibit a higher pitting resistance [Pitting Resistance Equivalent Number (PREN) > 40]. The combination of mechanical properties and the higher corrosion resistance make DSS and SDSS attractive materials in aggressive corrosive seawater environments.

GTAW uses a non-consumable tungsten electrode (EWTh2) to produce arc and filler wire to join the material and it is shielded by an inert gas like helium or argon to protect the molten weld pool from the atmospheric contaminants [3] . GTAW is a major fabrication process for DSS and SDSS materials being used in offshore and marine industries. Despite these advantages, mechanical and corrosion properties could be deteriorated if the weld parameters are not controlled during a welding operation. Rapid heating and cooling cycles may lead to ferritization and precipitation of hazardous inter-metallic phases like sigma, Chi, and secondary austenite. In order to take full benefits of mechanical and corrosion properties of DSS and SDSS, the welding thermal cycle should be controlled carefully.

The local breakdown of the passive protective layer is the cause of pitting in DSS and SDSS [4] . The susceptibility of pitting corrosion is measured by various tests namely, (a) gravimetric test (ASTM G48 ), (b) potentiodynamic polarization techniques (ASTM G5 ), and (c) critical pitting temperature (CPT) measurements (ASTM G150 ) [5] . Gravimetric tests measure the weight loss in the specimen after immersion of the sample in chloride solution. The polarization techniques analyze pitting potential (E_pit ) and corrosion potential (E_corr ) and corrosion current density (i_corr ). The higher positive values of (E_pit ), (E_corr ) and (E_pit – E_corr ) indicate better pitting corrosion resistance of the material. The CPT evaluation gives maximum allowable working temperature for the specimen by potentiostatic measurements.

The pitting resistance depends on number of variables like the austenite/ferrite ratio, the presence of inter-metallic phases, elemental partitioning between both phases, and PREN value. The PREN is a measure of pitting behavior of DSS and SDSS. It is given by the equation PREN = %Cr + 3.3%Mo + (16–30)%N. Due to different element partitioning and volume fraction, both phases have different PREN values. The pitting resistance of DSS and SDSS is controlled by PREN value of the weaker phase. It has been observed that the best corrosion resistance is achieved when both phases attain equal PREN value [6] .

Recent studies have found that cooling rate is one of the important parameters to achieve a desired microstructure [7] . A slow cooling promotes a diffusion phenomenon, which leads to an efficient partitioning of phases. At the same time with slow cooling rates, there are chances of precipitation of unwanted inter-metallic phases such as sigma phase. A rapid cooling leads to equal partitioning of phases (i.e. partitioning coefficient = 1 for all elements), and it may also form hazardous chromium nitrides. Cooling rate depends on various parameters like the heat input ([current × potential]/travel speed), the inter-pass temperature, the material thickness, the thermal properties of material, etc. A cooling time between 1200 °C and 400 °C is more critical than a cooling rate in the lower temperature region, because in this temperature region, major austenite reformation and secondary phase precipitation might take place [8] .

Shin et al. [9] studied the effect of the heat input on the pitting behavior of DSS welds. It was found that at the low heat input, insufficient reheating effect during subsequent passes caused formation of acicular type secondary austenite, which led to the reduction of the pitting resistance. Kordatos et al. [7] suggested that a continuous network of grain boundary austenite formed in the fusion zone after a faster cooling in the ferrite region restricts the corrosion propagation. Kobayashi et al. [10] quoted that secondary austenite is the reason for the loss of chemical balance and of the resistance of the passive layer. There have been a lot of researches on effect of post-weld heat treatments (solution annealing) of duplex stainless steels [11]  and [12] . However, practically, it is very difficult from industry point of view to carry out solution annealing heat treatments for larger products. Hence, it is very important and necessary to control weld parameters, such as heating/cooling rates to get the best corrosion properties of the material.

From the literature review, it has been found that there have been limited studies on the corrosion behavior of welded DSS and SDSS. The main purpose in this work is to study the (a) effect of weld parameters like heat input ([current × potential]/travel speed), inter-pass temperature, cooling rate, and shielding/backing gases on the corrosion resistance of the weld. The following studies were undertaken in this work – (a) pitting resistance of welded pipe joints, (b) the CPT for DSS and SDSS weldments, (c) the correlation between pitting behavior and microstructure of the weldments, (d) the effect of phase balance on corrosion properties of the weld.

2. Experimental details

2.1. Materials

The materials used in this study were 50.8 mm (2 inch) pipes of DSS and SDSS with 5.54 mm thickness and 150 mm length. The materials were selected with varied composition in order to get low and high PREN values for our studies. Table 1 gives the chemical composition of materials used for welding experiments.

Table 1. Base material chemical composition (% wt).

Material grade	Cr	Mo	Ni	N	C	PREN	Remarks
UNS S31803	22.9	3.03	7.92	0.15	0.017	35.15	DSS – Low PREN
UNS S31803	22.9	3.04	7.63	0.17	0.019	36.30	DSS – High PREN
UNS S32750	25.1	3.75	8.86	0.21	0.028	41.40	SDSS – High PREN
UNS S32750	25.1	3.71	8.9	0.2	0.016	40.36	SDSS – Low PREN

A typical microstructure of base material is shown in Fig. 1 . The islands of austenite in the ferrite matrix are clearly observed.

Fig. 1. Typical base material microstructure.

The welding consumable filler-wire composition is given in Table 2 . Filler wires used for our welding trials are manufactured by Sandvik, and 2 mm diameter wires were used for all welding trials.

Table 2. Filler-metal chemical composition (% wt).

Base metal	Filler Grade	C	Si	Mn	P	S	Cr	Ni	Mo	N
DSS	22.8.3.L	≤0.02	0.5	1.6	≤0.02	≤0.015	23	9	3.2	0.16
SDSS	25.10.4.L	≤0.02	0.3	0.4	≤0.02	≤0.015	25	9.5	4	0.25

2.2. Welding process

Gas Tungsten Arc Welding (GTAW) with Direct Current Electrode Negative (DCEN) polarity was used to weld the pipes. Welding was carried out (a) by varying welding heat input and (b) by varying shielding gas/back-purging gas composition and inter-pass temperature. The general welding specifications for first part of the study are given in Table 3 .

Table 3. Welding specifications.

Welding position	5G (Pipe fixed in horizontal position)
Groove design	Single V groove 70° groove angle 1 mm root face, 2.5 mm to 4 mm root gap
Welding current (A)	80–150
Arc voltage (V)	10–12
Welding speed (mm/min)	40–80
Number of weld passes	4–5
Inter-pass temperature (°C)	100–140
Gas flow rate (L/min)	13–18
Heat input (kJ/mm)	0.75–1.25

In the second part of the work, welding experiments were carried out to study the effect of shielding gas, the purging gas, and the inter-pass temperature on the corrosion properties of DSS and SDSS by keeping other parameters constant. During experiments, one of the above parameters was varied and others were kept constant. Table 4  and Table 5 show welding variable details of our studies. A mixture of argon and nitrogen (Ar + N) was used as shielding/purging gas.

Table 4. Welding parameters to the study effect of shielding/purging gas and inter-pass temperature for DSS weldments.

Exp. no.	Shielding gas	Purging gas	Inter-pass temperature (°C)
1	Ar + 2% N	Ar + 2% N	120
2	Ar + 5% N	Ar + 2% N	120
3	Ar + 2% N	Ar + 5% N	120
4	Ar + 2% N	Ar + 2% N	160

Table 5. Welding parameters to study the effect of shielding/purging gas and inter-pass temperature for SDSS weldments.

Exp. no.	Shielding gas	Purging gas	Inter-pass temperature (°C)
1	Ar + 2% N	Ar + 2% N	120
2	Ar + 5% N	Ar + 2% N	120
3	Ar + 2% N	Ar + 5% N	120
4	Ar + 2% N	Ar + 2% N	160

In total, 24 joints were investigated in this study which can be summarized as: (a) four DSS – low PREN joints by varying the heat input; (b) four DSS – high PREN joints by varying the heat input; (c) four SDSS – low PREN joints by varying the heat input; (d) four SDSS – high PREN joints by varying the heat input; (e) four DSS joints by varying the shielding/purging gas composition and the inter-pass temperature; (f) four SDSS joints by varying the shielding/purging gas composition and the inter-pass temperature.

2.3. Metallography

After successful completion of the welding, the welded specimens were polished up to 1200 grit fineness which was followed by a cloth polishing with 0.05 µm alumina powder. Later, the specimens were etched with 20% sodium hydroxide to reveal the microstructures. Microstructures were viewed under optical microscopes.

Ferrite content was determined on base metal, on the heat affected zone (HAZ), and on the weld region through point count method in accordance with ASTM E562 standard [13] . The elemental composition of each phase was checked by Energy Dispersive Spectroscopy (EDS) method.

2.4. Pitting corrosion test

Pitting behavior of the welded samples was studied using ASTM G48 gravimetric test [14] , which is a common method used for CPT of DSS in ferric chloride solution. After welding, specimens were cut into 50 mm × 25 mm size for testing, and the initial weight was measured by digital weighing machine. The specimens were immersed in 6% of ferric chloride solution for a period of 24 h at a constant test temperature of (Failed to parse (MathML with SVG or PNG fallback (recommended for modern browsers and accessibility tools): Invalid response ("Math extension cannot connect to Restbase.") from server "https://test.scipedia.com:8081/localhost/v1/":): {\textstyle 22\pm 1}

)°C and (Failed to parse (MathML with SVG or PNG fallback (recommended for modern browsers and accessibility tools): Invalid response ("Math extension cannot connect to Restbase.") from server "https://test.scipedia.com:8081/localhost/v1/":): {\textstyle 28\pm 1}
)°C for DSS and (Failed to parse (MathML with SVG or PNG fallback (recommended for modern browsers and accessibility tools): Invalid response ("Math extension cannot connect to Restbase.") from server "https://test.scipedia.com:8081/localhost/v1/":): {\textstyle 35\pm 1}
)°C and (Failed to parse (MathML with SVG or PNG fallback (recommended for modern browsers and accessibility tools): Invalid response ("Math extension cannot connect to Restbase.") from server "https://test.scipedia.com:8081/localhost/v1/":): {\textstyle 40\pm 1}
)°C for SDSS. After 24 h period of immersion, specimens were rinsed with water, dipped into acetone in an ultrasonic cleaner, and air-dried. Subsequently, the test specimens were examined for visible pits and weighed to obtain the weight loss due to corrosion attacks.      

Corrosion behavior was also studied through potentiodynamic polarization technique in 1 mol/L NaCl solution at (Failed to parse (MathML with SVG or PNG fallback (recommended for modern browsers and accessibility tools): Invalid response ("Math extension cannot connect to Restbase.") from server "https://test.scipedia.com:8081/localhost/v1/":): {\textstyle 22\pm 1}

)°C and (Failed to parse (MathML with SVG or PNG fallback (recommended for modern browsers and accessibility tools): Invalid response ("Math extension cannot connect to Restbase.") from server "https://test.scipedia.com:8081/localhost/v1/":): {\textstyle 35\pm 1}
)°C temperature for DSS and SDSS, respectively, with a scan rate of 0.5 mV/s from −1200 mV (SCE). The area of the specimen exposed to the solution was 100 mm2 . Some part of the exposed surface was insulated with adhesive tape so that exposed surface would be the cross-sectional area of the pipes. An electrochemical cell was used with specimen as a working electrode, platinum as a counter and calomel as a reference electrode. The potentiostatic measurements were also carried to evaluate pitting corrosion of the weldments as per ASTM G150  standard. An anodic potential applied was about 750 mV (SCE) until the occurrence of stable pitting. The temperature of the solution was increased at the rate of 1 °C/min.      

3. Results and discussion

3.1. Weld macrostructure

The macro section of the weldments of DSS and SDSS joints is shown in Fig. 2 . The weldments are divided in weld cap, root and heat affected zone.

Fig. 2. Macrostructure of welded joints.

3.2. Weld microstructure

The typical microstructures of weld cap and root side are shown in Fig. 3 . The weld root regions were subjected to reheat during weld passes. Hence, intra-granular primary austenite and acicular type secondary austenite are formed at root. The reheating of weld root region was the reason for formation of secondary austenite phase [9] . The amount of austenite formed is higher in root region than that of weld cap. The weld cap region comprises of grain boundary austenite, intra-granular, and Widmänstten austenite formed in a ferrite matrix. The coarse ferrite grains were observed in weld cap region.

Fig. 3. Typical weld microstructures.

When the welding was carried out with higher heat input (heat input ≥ 1.15 kJ/mm) (i.e. slower cooling rate), large grain size and higher contents of austenite were observed for DSS and SDSS weldments. With lower heat input (heat input < 1.15 kJ/mm), (i.e. higher cooling rate), lower austenite content with finer grains were observed for DSS and SDSS weldments as shown in Fig. 4 .

Fig. 4. Microstructural variation with heat input.

The typical HAZ microstructure for DSS and SDSS weldments is shown in Fig. 5 . In this region, an increase in grain size was observed due to re-crystallization, particularly in ferrite. There was an evidence of inter-granular and intra-granular austenite formation in this region.

Fig. 5. Heat affected zones.

The ferrite contents of the welded specimens are tabulated in Table 6 , Table 7 , Table 8  and Table 9 . It was observed that the weld cap contains more ferrite than the root region for the both fusion zone and the heat affected zone. It can also be observed that ferrite volume fraction in HAZ is higher than that of the weld region.

Table 6. Ferrite content (%) measurements DSS weldments with variable heat input.

Specimen	Heat Input (kJ/mm)	Cap	Root	HAZ-Cap	HAZ-Root
DSS – Low PREN	1.05	54 ± 3	38 ± 4	54 ± 5	48 ± 3
	1.10	46 ± 2	43 ± 5	52 ± 3	43 ± 5
	1.15	39 ± 4	36 ± 3	55 ± 2	51 ± 3
	1.20	35 ± 3	34 ± 3	50 ± 1	48 ± 4
DSS – High PREN	1.0	55 ± 2	35 ± 5	65 ± 4	60 ± 4
	1.05	45 ± 2	39 ± 2	51 ± 1	50 ± 2
	1.1	44 ± 1	42 ± 2	50 ± 2	44 ± 4
	1.15	38 ± 3	35 ± 3	44 ± 1	43 ± 3

Table 7. Ferrite content (%) measurements SDSS weldments with variable heat input.

Specimen	Heat input (kJ/mm)	Cap	Root	HAZ – cap	HAZ – root
SDSS – Low PREN	0.95	56 ± 2	39 ± 5	42 ± 3	39 ± 1
	1.05	48 ± 1	46 ± 3	49 ± 2	48 ± 2
	1.15	46 ± 3	43 ± 4	49 ± 2	47 ± 1
	1.25	43 ± 4	38 ± 3	48 ± 3	45 ± 1
SDSS – High PREN	0.75	64 ± 4	46 ± 5	61 ± 3	56 ± 3
	1.0	58 ± 3	49 ± 2	56 ± 1	55 ± 3
	1.1	55 ± 3	51 ± 3	51 ± 2	48 ± 3
	1.2	52 ± 4	45 ± 4	48 ± 2	43 ± 3

Table 8. Ferrite content (%) measurements [study effect of shielding/purging gas and inter-pass temperature for DSS weldments – refer to [[#t0025|Table 4]] ].

Exp. no.	Cap	Root	HAZ cap	HAZ root
1	54 ± 2	38 ± 4	54 ± 5	48 ± 2
2	51 ± 3	37 ± 2	50 ± 3	46 ± 1
3	49 ± 4	34 ± 5	48 ± 3	46 ± 2
4	52 ± 1	36 ± 4	55 ± 4	43 ± 3

Table 9. Ferrite content (%) measurements [study effect of shielding/purging gas and inter-pass temperature for SDSS weldments – refer to .[[#t0030|Table 5]] ].

Exp. no.	Cap	Root	HAZ – cap	HAZ – root
1	56 ± 3	39 ± 4	42 ± 2	39 ± 3
2	51 ± 2	38 ± 5	41 ± 1	37 ± 3
3	49 ± 4	33 ± 3	41 ± 1	39 ± 2
4	55 ± 3	36 ± 4	44 ± 2	34 ± 4

With addition of nitrogen in shielding/purging gas [i.e. Ar + 2% N and Ar + 5% N], the ferrite content was found to decreas as nitrogen is an austenite stabilizer. The increasing N from 2% to 5% has facilitated a better phase balance due to proper elemental partitioning but it is not significant when it is compared with 2% N [15] . Although 5% N promotes better phase balance, 2% N is providing the required phase balance and it is cheaper and readily available in the market when it is compared with 5% N. Thus, based on the results, we concluded that a 2% N containing gas is good enough to promote a phase balance.

An increase in the inter-pass temperature (see Exp. No. 4 in Table 8  and Table 9 ) slightly reduces the ferrite content in the weld region, and the HAZ and secondary austenite were also formed in the weld zone due to continued exposure to sensitive high temperatures by diffusion of ferrite into secondary austenite.

3.3. Pitting corrosion test

3.3.1. ASTM G48 test

The pitting corrosion test results are shown in Table 10  and Table 11 . It can be seen from Fig. 6 that the corrosion rate was found to increase with the increase in the heat input. This could be because of at high heat input the weld region attains sensitive temperature range where formation of inter-metallic such as secondary austenite takes place easily. The secondary austenite contains very low amount Cr and Mo. Hence, these sites are the sites for pitting corrosion attack because of easy breakdown of the passive film. For DSS, at 22 °C, there was no evidence of pitting, i.e. (weight loss < 1 g/(m² .day). But when the temperature was increased to 28 °C, there were pits observed on the weldments. According to ASTM G48 test, the stable pitting is said to be initiated when weight loss is more than 1 g /(m² .day). On all the DSS samples, pits are formed at 28 °C. Similarly for SDSS weldments, a weight loss of more than 1 g/(m² .day) was observed at 40 °C for all conditions.

Table 10. Corrosion test results for DSS weldments.

Specimen	Corrosion rate g /(m2 .day) at 22 °C	Corrosion rate g /(m2 .day) at 28 °C	Heat input (kJ/mm)
DSS – Low PREN	0.250	2.65	1.05
	0.311	3.95	1.1
	0.40	6.11	1.15
	0.591	7.63	1.2
DSS – High PREN	0.099	1.96	1
	0.103	2.39	1.05
	0.111	3.45	1.1
	0.559	4.65	1.15

Table 11. Corrosion test results for SDSS.

Specimen	Corrosion rate g /(m2 .day) at 35 °C	Corrosion rate g /(m2 .day) at 40 °C	Heat input (kJ/mm)
SDSS – Low PREN	0.127	1.148	0.95
	0.139	2.046	1.05
	0.198	2.731	1.15
	0.246	4.842	1.25
SDSS – High PREN	0.091	1.391	0.75
	0.103	1.928	1
	0.178	2.364	1.1
	0.236	3.687	1.2

Fig. 6. Variation of corrosion rate with heat input.

The variation of corrosion rate with heat input is shown in Fig. 6 . The typical samples after ASTM G48 tests are shown in Fig. 7 .

Fig. 7. Typical specimen before and after ASTM G48 test: (a) DSS weldments; (b) SDSS weldments.

In the second part, the shielding/purging gas and the inter-pass temperatures were varied according to Table 4  and Table 5 . The results of the tests are shown in Table 12  and Table 13 . It was found that addition of more nitrogen in the gas mixture resulted improved effects on corrosion properties due to austenite and ferrite phase balance. Pitting resistance of individual phases was also improved due to proper partitioning of elements [15]  and [16] . The increase in inter-pass temperature increased pitting corrosion rate, as noticed by the weight loss results.

Table 12. Corrosion test results for second part [i.e. study effect of shielding/purging gas and inter-pass temperature for DSS weldments].

Exp. no.	Heat input (kJ/mm)	Inter-pass temperature (°C)	Corrosion rate g /(m2 .day) at 22 °C	Corrosion rate g /(m2 .day) at 28 °C
1	1.05	120	0.25	2.65
2	1.05	120	0.124	1.64
3	1.05	120	0.179	1.68
4	1.05	160	0.768	4.49

Table 13. Corrosion test results for second part [i.e. study effect of shielding/purging gas and inter-pass temperature for SDSS weldments].

Exp. no.	Heat input (kJ/mm)	Inter-pass temperature (°C)	Corrosion rate g /(m2 .day) at 35 °C	Corrosion rate g /(m2 .day) at 40 °C
1	0.95	120	0.127	1.148
2	0.95	120	0.0968	0.945
3	0.95	120	0.0984	0.958
4	0.95	160	0.678	5.92

3.3.2. Potentiodynamic polarization tests

The results of potentiodynamic polarization tests for best and worst corrosion behavior for both DSS and SDSS are shown in Fig. 8 . It can be seen that potentials shifted to more positive values for specimen at low heat input.

Fig. 8. Polarization curves for (a) DSS and (b) SDSS.

The corrosion potential (E_corr ) and pitting potential (E_pit ) for DSS and SDSS specimens are shown in Table 14 . The difference between them, i.e. (E_pit – E_corr ) is a measure of resistivity of passive film on the specimen. The larger the difference, better is the corrosion resistance [16] . The least corrosion resistance of few samples was observed because of formation of secondary austenite and improper partitioning of individual phases. The typical corroded specimen images analyzed through SEM are shown in Fig. 9 .

Table 14. Corrosion and pitting potentials for DSS and SDSS after welding (mV SCE).

Specimen	Epit	Ecorr	Epit -Ecorr
Low heat input – DSS	467	−189	656
High heat input – DSS	406	−216	622
Low heat input – SDSS	382	−321	703
High heat input – SDSS	353	−347	700

Fig. 9. Typical SEM images of corroded specimen.

The results of second part of the studies are tabulated in Table 15  and Table 16 . It is clear that with the addition of nitrogen in shielding and back purging gas increased pitting nucleation resistance. With increase in inter-pass temperature, corrosion potential (E_corr ) and pitting potential (E_pit ) shifted to negative side which resulted in decrease of corrosion resistance.

Table 15. Corrosion and pitting potentials of DSS weldments after welding [refer to [[#t0025|Table 4]] ] (mV SCE).

Specimen	Epit	Ecorr	Epit -Ecorr
1	434	−164	598
2	472	−149	621
3	484	−143	627
4	335	−215	550

Table 16. Corrosion and pitting potentials of SDSS weldments after welding [refer to [[#t0030|Table 5]] ] (mV SCE).

Specimen	Epit	Ecorr	Epit -Ecorr
1	392	−418	810
2	430	−367	797
3	445	−328	773
4	375	−336	711

3.3.3. Critical pitting temperature measurements

CPT measurements were done by potentiostatic measurements to confirm maximum working temperatures for weldments. The CPT is the temperature at which corrosion current density reaches 100 µA/cm² . It was found that in all DSS weldments, stable pitting occurred between 23 °C and 27 °C. For SDSS, it was found to be between 37 °C and 41 °C as shown in Fig. 10 .

Fig. 10. Critical pitting temperature measurements.

3.3.4. PREN of individual phases

PREN of ferrite and austenite phases were studied to understand cause and region of pitting attack as shown in Table 17 . The elemental composition of each phase was observed through SEM-EDS. The typical EDS study is shown in Fig. 11 . The pitting resistance equivalent number was calculated by formula: PREN = % Cr + 3.3% Mo + 16% N. Due to different elemental partitioning, each phase was having different PREN value. However, due to rapid cooling cycles, the partitioning ratio tends to unite for Cr, Mo, and Ni. In case of nitrogen, it was assumed that ferrite reaches its saturation level to 0.05%. The rest partitioned in austenite phase. Hence, the nitrogen content in the austenite phase could be calculated based on the content of nitrogen in the whole alloy and in the phase volume fraction. The new austenite phase was found to have reduced Cr content and very low Mo content. Similar observations were made in other literatures [10] . Hence, secondary austenite phase was the region of pitting attack.

Table 17. Alloying element contents (% wt.) and PREN of individual phases.

Specimen	Phase	Cr	Mo	Ni	N	PREN
DSS – Highest corrosion rate	α	22.30	3.32	7.87	0.05	34.05
	γ	21.93	2.81	7.96	0.22	34.72
	γ2	11.52	0.90	8.10	0.20	17.69
DSS – Lowest corrosion rate	α	24.12	3.24	7.35	0.05	35.61
	γ	21.70	2.74	7.81	0.27	35.06
	γ2	12.86	0.96	8.12	0.23	19.70
SDSS – Highest corrosion rate	α	25.24	3.82	8.52	0.05	38.64
	γ	23.12	3.52	8.82	0.39	40.97
	γ2	14.08	1.11	8.76	0.32	22.86
SDSS – Lowest corrosion rate	α	25.15	3.92	8.52	0.05	38.88
	γ	23.46	3.64	8.72	0.35	41.07
	γ2	14.24	1.23	8.74	0.31	23.25

Fig. 11. EDS study of the welded samples.

4. Conclusion

Gas Tungsten Arc Welding was performed successfully on DSS and SDSS pipes. The effect of welding heat input, shielding gases, purging gases, PREN, cooling rate on weld microstructure, and corrosion resistance was studied.

The weld root region was susceptible to pitting attack as compared to weld cap region. This is due to the precipitation of hazardous secondary austenite in weld root due to multiple heating cycles. Lower welding heat input was found to give better corrosion properties than higher heat input. At high heat input, there was a formation of secondary austenite and inter-metallic phases due to repetitive heating cycles.

From potentiodynamic studies, it was found that at lower heat input, specimen tends to push the polarization curve to positive values. From CPT measurements, it was found that critical pitting occurred between 23 °C to 27 °C for DSS and 37 °C to 41 °C for SDSS specimen. From EDS studies, it was confirmed that secondary austenite phase was depleted in Cr and Mo content which leads to corrosion of the weldments. The PREN of secondary austenite was lowered up to 18.61 and 24.14 for DSS and SDSS, respectively, which resulted in pitting corrosion.

With increase in nitrogen content (minimum 2% N) in shielding and back purging gas, the corrosion properties were improved because of balanced microstructure. Higher inter-pass temperatures caused reduction in corrosion resistance due to the formation of secondary austenite and inter-metallic phases.

Based on the experiments and test results, improved corrosion properties were observed at low heat input, low inter-pass temperature, and higher nitrogen content in shielding/back purging gas and faster cooling rate. The following parameters are recommended to obtain better corrosion properties: (a) heat input (0.75–1.1) kJ/mm; (b) shielding/purging gas – Ar + 2% N; (c) stringer bead technique; and (d) inter-pass temperature – 120 °C.

References

1. [1] M.A. M, K.A. Shrikrishana, P. Sathiya; Finite element modelling and characterization of friction welding on UNS S31803 duplex stainless steel joints; Eng. Sci. Technol. Int. J., 18 (2015), pp. 704–712 http://dx.doi.org/10.1016/j.jestch.2015.05.002
2. [2] M.A. M, K.A. Shrikrishna, P. Sathiya, S. Goel; The impact of heat input on the strength, toughness, microhardness, microstructure and corrosion aspects of friction welded duplex stainless steel joints; J. Manuf. Process, 18 (2015), pp. 92–106 http://dx.doi.org/10.1016/j.jmapro.2015.01.004
3. [3] A.K. Srirangan, S. Paulraj; Multi-response optimization of process parameters for TIG welding of Incoloy 800HT by Taguchi grey relational analysis; Eng. Sci. Technol. Int. J. (2015) http://dx.doi.org/10.1016/j.jestch.2015.10.003
4. [4] A. Azizi; Investigating the controllable factors influencing the weight loss of grinding ball using SEM/EDX analysis and RSM model; Eng. Sci. Technol. Int. J., 18 (2015), pp. 1–8 http://dx.doi.org/10.1016/j.jestch.2014.12.007
5. [5] R. Cervo, P. Ferro, A. Tiziani, F. Zucchi; Annealing temperature effects on superduplex stainless steel UNS S32750 welded joints. II: pitting corrosion resistance evaluation; J. Mater. Sci, 45 (2010), pp. 4378–4389 http://dx.doi.org/10.1007/s10853-010-4311-0
6. [6] Y. Yang, Z. Wang, H. Tan, J. Hong, Y. Jiang, L. Jiang, et al.; Effect of a brief post-weld heat treatment on the microstructure evolution and pitting corrosion of laser beam welded UNS S31803 duplex stainless steel; Corros. Sci, 65 (2012), pp. 472–480 http://dx.doi.org/10.1016/j.corsci.2012.08.054
7. [7] J.D. Kordatos, G. Fourlaris, G. Papadimitriou; The effect of cooling rate on the mechanical and corrosion properties of SAF 2205 (UNS 31803) duplex stainless steel welds; Scr. Mater, 44 (2001), pp. 401–408 http://dx.doi.org/10.1016/S1359-6462(00)00613-8
8. [8] H. Sieurin, R. Sandström; Austenite reformation in the heat-affected zone of duplex stainless steel 2205; Mater. Sci. Eng. A, 418 (2006), pp. 250–256 http://dx.doi.org/10.1016/j.msea.2005.11.025
9. [9] Y.T. Shin, H.S. Shin, H.W. Lee; Effects of heat input on pitting corrosion in super duplex stainless steel weld metals; Met. Mater. Int, 18 (2012), pp. 1037–1040 http://dx.doi.org/10.1007/s12540-012-6017-0
10. [10] D. Yuko, S. Wolynec; Evaluation of the low corrosion resistant phase formed during the sigma phase precipitation in duplex stainless steels; Mater. Res, 2 (1999), pp. 239–247
11. [11] Z. Zhang, Z. Wang, Y. Jiang, H. Tan, D. Han, Y. Guo, et al.; Effect of post-weld heat treatment on microstructure evolution and pitting corrosion behavior of UNS S31803 duplex stainless steel welds; Corros. Sci, 62 (2012), pp. 42–50 http://dx.doi.org/10.1016/j.corsci.2012.04.047
12. [12] P. Ferro, A. Tiziani, F. Bonollo; Influence of induction and furnace postweld heat treatment on corrosion properties of SAF 2205 (UNS 31803); Weld. J., 87 (12) (2008), pp. 298S–306S
13. [13] ASTM E562-11; Standard Test Method for Determining Volume Fraction by Systematic Manual Point Count; ASTM International, West Conshohocken, PA (2011) http://dx.doi.org/10.1520/E0562-11
14. [14] ASTM Standard G48-03; Standard Test Methods for Pitting and Crevice Corrosion Resistance of Stainless Steels and Related Alloys by Use of Ferric Chloride Solution; ASTM International, West Conshohocken, PA (2003)
15. [15] R.B. Bhatt, H.S. Kamat, S.K. Ghosal, P.K. De; Influence of nitrogen in the shielding gas on corrosion resistance of duplex stainless steel welds; J. Mater. Eng. Perform, 8 (1999), pp. 591–597 http://dx.doi.org/10.1361/105994999770346648
16. [16] H. Luo, X.G. Li, C.F. Dong, K. Xiao; Effect of solution treatment on pitting behavior of 2205 duplex stainless steel; Arab. J. Chem (2012), pp. 1–5 http://dx.doi.org/10.1016/j.arabjc.2012.06.011