Investigation of Corrosion of Stainless Steel Piping
Corrosion of stainless steel piping at the water treatment plant has developed at the welds after less than one year of operation. A metallurgical examination was requested to determine the cause of the corrosion, which is manifested as discoloration and pitting in the welded areas. The questions at hand are to determine if the welding has shortened the operating life of the water treatment plant due to corrosion and if the discoloration (rust) affects the aesthetics of stainless steel.
The OD welding degraded the corrosion resistance on the ID since no backing gas was employed, allowing pitting and crevice corrosion to occur.
Discussion / Results
There has been some debate regarding corrosion of stainless steel on the inside of a weld; can the weld parameters affect the microstructure such that corrosion will readily occur? Recent work mentioned the diffusion of Cr from the grains to the grain boundaries, which effectively negates the corrosive resistant properties of this material [1], which is commonly known as desensitization. Corrosion associated with welding stainless steel could be eliminated by using a backing gas in the pipe during welding [2-4] or the use of alternate filler metal that is high in Si [5, 6]. Future welding could also be performed using an aluminized backing tape that would protect the ID of an OD weld [7]. The use of a backing gas is effective in eliminating the presence of oxygen, to the extent that AWS has published a standard documenting the level of discoloration as a function of oxygen present during welding [8], a portion of which is shown in Figure 1. Standards published by Stainless Steel World [9] and Force Institute [2] describe acceptable levels of oxidation on the welded surface that include shiny weld beads with minimal discoloration on HAZ.

Figure 1. Portion of AWS color chart showing color levels of halo as a function of oxygen [8].
The leaks in the joint are due to the lack of fit-up and allowing the liquid to enter a crevice as shown in Figure 2. There are several types of corrosion that can occur : crevice corrosion and accelerated corrosion due to Cr degradation within the alloy due to HAZ of welding without a backing gas. Based on the color chart the level of oxygen appears to be at the highest level.

Figure 2. Pin-hole leaks on T-joint due to lack of fit-up and lack of weld penetration. Halo is present between arrows.

Figure 3 shows a pit that developed in the halo region of the joint and some surface roughness due to oxidation. The pit depth is approximately 0.004" deep. Some of the Cr in the discolored region has diffused to the grain boundaries as shown in Figure 4, making the base metal susceptible to corrosion.
Figure 3. Pit in halo region on joint and surface oxidation. Ignore "1364", a measurement tool was used to draw a reference line.

Figure 4 is an SEM "dot map" that shows elemental concentrations. This Figure shows the HAZ beneath the discolored region. The Cr is no longer evenly distributed; the Cr near the grain boundary has partially diffused to grain boundaries.
Figure 4. SEM "dot map" of Cr. Orange is FexCry at grain boundary (shown as sketched line). The central region is the center of the grain and maintains the proper Cr level.
To see the resume of the expert associated with this case study, see the link below.