Ant Nest Corrosion - Digging the Tunnels

Corbett, Richard A. (Corrosion Testing Laboratories Inc.) | Elliott, Peter (Corrosion and Materials Consultancy Inc.)


Approximately 10% of all preliminary failures of copper tubes used in the HVAC industry are the direct result of ant-nest corrosion on a worldwide scale. This unusual form of localized corrosion has been detected in new tubes from some manufacturers, as well as early in service (i.e., less than one year). Laboratory studies have replicated service failures. The sources of the corrosive species that promote this form of attack are considered.

This is the continuing saga into a very insidious form of corrosion apparently limited to refrigeration grade copper (i.e., DHP copper, UNS C12200) although other forms of attack, similar in appearance, have been observed in other alloys. (1 - 5) This very localized attack is known as "ant-nest corrosion" (a.k.a. formicary corrosion). A previous review of this phenomenon (6) describes the mechanism where the copper base-metal oxidizes to form copper oxides and copper carboxylates (primarily copper formate). The literature provides several attempts to classify the morphology of ant-nest corrosion (6, 7) although no individual pattern has been identified.

Laboratory studies (7, 9 - 14) have been successful in replicating ant-nest corrosion found in the field using carboxylic acids, hydrochloro- solvents, and petroleum- and water-based drawing and finning lubricants. Case studies reported in the past two years suggest that this form of attack continues to be a problem in the heating, ventilation and air conditioning (HVAC) industry. (6,15) It is reiterated that approximately 10% of all preliminary failures of copper tubes used in the HVAC industry are the direct result of ant-nest corrosion on a worldwide scale. (6, 9, 16,17) This paper reviews not only the current understanding of the mechanism of attack but cites new case studies including a thorough investigation into the cause of over 160 crates of new copper condenser and evaporator tubes that were affected by this form of attack.

Residual organic compounds that remain on copper tubes during production and fabrication into chiller units, are able to advance to ant-nest corrosion only with the simultaneous presence of moisture, air and the decomposition of the organics to acids. It is concluded that the mechanism of the ant-nest corrosion is a modified pitting process involving a very small pit (termed a "micro-anode") where the copper base-metal oxidizes and dissolves according to:

Cu ° "-) Cu ++ e- (1)

In the presence of carboxylic acid (e.g., formic acid) the copper ions react to form an unstable copper (I) complex:

Cu + + HCOO" ~ Cu(CHOO) (2)

This species is further oxidized to form a copper (II) carboxylate (e.g., copper (II) formate) and copper (I) oxide (cuprite):

4 Cu(CHOO) + ½ 02 @ 2 Cu(CHOO)2 + Cu20 (3)

Copper (II) formate has a monoclinic crystalline form and is blue in color. Micro-cracks develop and radiate outward within the pit due to the wedging effect of the deposited copper (I) and copper (II) complexes. The micro-cracks expose more surfaces of copper and the process proceeds within the micro-crack to give the copper (I) complex according to:

Cu(CHOO)2 + Cu ° ") 2 Cu(CHOO) (5)

Thereon, reactions 3, 4 and 5 repeat over and over until tunnels are formed leading to ultimate through- wall penetration.

If, however, there is any disruption in the presence of the basic three components (i.e., moisture, air and organic acid), then the micro-cell mechanism shuts down and ant-nest corrosion ceases to propagate. Therefore, a tube may be infected with ant-nest corrosion but failure has not resulted in a leak.

During the last several m