Corrosion, Failure Analysis and Materials Selection Specialists

 

 

 

 

 

Corrosion Testing / Failure Analysis

 

 

 

Corrosion Testing

Failure Analysis

Field Investigations

Litigation

Metallography

Technical Papers

CTL Profile

Pricing & Policies

Contact CTL

Quality Assurance

Return to Failure Analysis Case Histories

Failure Avoidance 

Galvanic Effects In Carbon Bed Absorbers

 

Richard A. Corbett

© Copyrighted by NACE International 

 

Background 

As part of a remediation site project, activated carbon, resin bed adsorption units are used to remove chlorinated hydrocarbons and carbon disulfide from soil. 

The resin bed is inserted, and gasket-sealed inside, in an open top, carbon steel chamber.  The resin is retained with a 3-foot x 4-foot (0.9 m x 12 m) aluminum frame about 6-inches (15-cm) deep.  The frame is bolted with stainless steel nuts and bolts to a type 304 stainless steel (UNS S30400) top plate that provides a platform for externally mounted inlet and outlet nozzles, valves, solenoids, and controllers.  

The open faces of the frame are covered with a layer of fine mesh (80 mesh) type 304 SS wire screen which is, in turn, covered with a 1/16-in. (1.6-mm) thick, perforated type 304 SS plate.  For added stiffness, aluminum bars are bolted with SS nuts and bolts across the perforated plate.  

The bed assembly is covered with aluminum shrouds that form chambers for solvent laden air to uniformly access the face of the bed.  Electric strip heaters are attached to the outside of the aluminum shrouds to prevent condensation on their inner surfaces.  

Before early 1996, the recovery design used recycled nitrogen atmosphere in an attempt to reduce the presence of moisture.  Due to cost considerations and other reasons, in late 1996 the N2 atmosphere was replaced with outside air resulting in an additional nominal moisture content to that already in the solvent.  

 

Problem 

·        The aluminum shrouds had corroded through and were ineffective

·        The stainless steel components had significantly corroded.  

·        Resin escaped from the bed – indicating the wire screens, which retained the carbon-based resin, were breached.  

 

What happened 

The additional moisture in the air atmosphere increased the acidity of the solvent-laden carbon bed.  During regeneration, warm dilute (~8%) hydrochloric acid was formed.  

Type 304 SS and aluminum are not resistant to corrosion by aqueous, acidic, chloride-containing compounds – at near ambient temperatures.  Aluminum can corrode in dilute hydrochloric acid (HCl) at a rate greater than 1/16-inch per year (1.6 mm/y).  Type 300 series SS will pit and stress corrosion crack in chloride solutions.  

Carbon tetrachloride, one of the chlorinated hydrocarbons to be removed, and carbon disulfide, once adsorbed in the carbon resin bed, decompose, react with residual moisture, and acidify during the desorption portion of the bed regeneration cycle.  This acidified vapor corrodes the aluminum and SS surfaces with which it comes in contact with. 

The carbon-based resin bed is in contact with the aluminum frame, and stainless steel screen and support plates, forming a tri-galvanic couple.  Aluminum is the least noble, and carbon is the noblest of the three materials.   Under nitrogen atmosphere there in insufficient moisture to create a continuous electrolyte between these three materials, however under redesign, using outside air, sufficient moisture is present to “wet” the components.  This creates the fourth component of the galvanic corrosion cell, the common electrolyte.  

Gene Liening, of the Dow Chemical Company, reported on a similar case history in ASTM STP 908 (1986), and the special problems associated with carbon bed absorber vessels due to organics, temperature and galvanic effects.  At Corrosion/91 (Paper Number 165), he addressed the use of alternative materials for carbon bed absorbers, but stressed that each application be tested and verified. 

 

Laboratory Studies 

Due to the organics present in this case study, only metallic components were deemed useable.  Therefore, following the guidelines of Liening, electrochemical tests were performed to evaluate several candidate materials of construction to dilute HCl (7.67 g/L) when galvanically coupled to activated carbon.  Initial tests were performed at 32oC (90oF) under process flow conditions and at 90oC (194oF) under regeneration conditions.  Materials evaluated were type 304 as the control; alloys C-276 (UNS N10276), B-3 (UNS N10675), and 825 (UNS N08825); titanium, unalloyed grade 2 (UNS R50400), and zirconium 702 (UNS R60702). 

 

Laboratory Findings  

Galvanically coupled to activated carbon,  

·        Type 304SS exhibits an unacceptably but predictably high corrosion rate of over 25 mpy (0.64 mm/y),

·        Alloys B-3 and 825 and zirconium have high corrosion rates and cannot be recommended for this service, while

·        Alloys C-276 and unalloyed titanium have acceptable rates and can be considered suitable replacement materials for SS in this system. 

 

Lessons Learned

This case continues to illustrate the need to recognize the effect of mixed materials in contact with each other in hostile environments.  It should be obvious that mixed metals must be electrically isolated from each other, but also to recognize that carbon, as graphite, is capable of conducting electricity and can therefore enter into the electrochemical process of corrosion.  If graphite, being on the noble end of the galvanic series, contact with other metals is unavoidable, such as in the carbon bed absorber unit, then these metals must be similarly noble.  Only through testing of compatibility will the effects of galvanic corrosion be known.

Site Index

Site Copyright © 1995 - 2007, All Rights Reserved,

Corrosion Testing Laboratories, Inc.

60 Blue Hen Drive

Newark, Delaware USA 19713

Phone: 1-302-454-8200

Fax: 1-302-454-8204

web@corrosionlab.com