Cavitation Damage

     We recently had an interesting failure analysis project that illustrates the diversity of problems we encounter and gives us the opportunity to introduce our readers to a material degradation phenomenon they may not be familiar with.

     One of our clients reported that he was experiencing through-wall pitting in the ultrasonic generators of his two-year old commercial ultrasonic cleaning tank (see photo above).

     A preliminary review of the failure suggested that something more than simple corrosion was responsible for the failure. First of all, the tank was made of stainless steel, which, of course, is relatively resistant to corrosion. Next, the fluid in the tank was ordinary tap water, which shouldn't be particularly aggressive to the stainless steel, and the material being cleaned, pyrolylzed polypropylene debris, was also not considered to be corrosive. Finally, a closer examination of the damage revealed a curious pattern of attack, with a repeating, symmetrical array of pitted and nonpitted areas.

     We suspected that a cavitation mechanism was probably involved in the damage. But to establish this with certainty, a more detailed laboratory analysis would be required.

A Little Bit About Ultrasonic Cleaning

     Before we continue with our story of the failure analysis, let's take a small detour and discuss just how ultrasonic cleaning works. Ultrasonic cleaning depends upon cavitation, the rapid formation and collapse (implosion) of minute bubbles or cavities in a liquid. As the voids collapse and the liquid surfaces meet, considerable energy is released. The agitation by countless small and intense imploding bubbles creates a highly effective scrubbing action.

     Cavitation energy is introduced into a liquid at very high (ultrasonic) frequencies by means of transducers that convert electrical energy into acoustic energy. The transducers consist of vibrating elements tuned to the specific frequency desired; for ultrasonic cleaning, they generally operate at 25 to 80 kHz, but frequencies up to 400 kHz are used in some applications.

     The properties of the liquid in the tank are important in determining its cavitation characteristics and therefore its effectiveness as a cleaning agent; these properties include the liquid's density, viscosity, surface tension, and vapor pressure. The choice of an aqueous or solvent-based system depends primarily on the particular application and must take into consideration the chemical characteristics of the contamination to be removed from the parts; liquids should be judged for effectiveness of cleaning, safety, simplicity of use, and longevity.

     Aqueous cleaning solutions nearly always require a wetting agent (surfactant, detergent) to effectively transfer the ultrasonic energy into the solution and generally operate more efficiently at elevated temperatures (140 to 160 degF). Using a detergent with a corrosion-inhibited formulation is often an important consideration; contamination of the cleaning solution with time is a factor that also must be considered.

Cavitation Damage

     Although in ultrasonic cleaning cavitation is an intentionally induced phenomenon, there are a variety of situations where cavitation occurs unexpectedly and causes severe damage to equipment. We have seen cavitation damage on ship propellers, water turbine blades in pumps, and in high-velocity flow lines, especially at fittings where the smooth bore of the pipe is interrupted. Basically, turbulence or changes in temperature within the fluid can cause the pressure in local zones to fall below the vapor pressure of the liquid, resulting in localized cavitation.

     The energy imparted to a metal surface as a result of cavitation can cause mechanical damage. The repetition of this process millions of times in a localized area may leave a surface severely roughened and cause significant metal loss over time. (photo at right)

     In addition to causing mechanical damage, the impingement energy can break down the protective passive surface film that forms on stainless steel and provides its characteristic corrosion resistance; additionally, it can remove corrosion products that may be protecting the surface in an aggressive environment; this combined action can cause accelerated material losses. When the mechanical damage of cavitation is aided by corrosion, the term cavitation-corrosion is generally applied.

Back to Our Story of the Failures

     The cleaning tank contained 24 immersible ultrasonic generators; each generator consisted of twelve 40-kHz transducers mounted within a seal-welded stainless steel can (photo at left). Visual inspection of the stainless steel generator housings revealed severe surface pitting, which eventually led to the perforation of the housings and allowed ingress of water from the cleaning tank into the generator housings, resulting in an electrical short.

     When the ultrasonic cleaning tank was first installed, city water had been used in the tank, with a rust inhibitor added to protect against corrosion. The ultrasonic tank had been operated approximately 12 hours per day at a temperature of 150°F, and the liquid in the tank was filtered to remove particles larger than about 1 micron. After about six months, the client had stopped using the rust inhibitor, and for the last eighteen months, the cleaning agent in the tank had been ordinary city water with no additives.

     Faced with replacing these very expensive ultrasonic generators, our client wanted us to determine the mechanism responsible for the failures and provide recommendations to ensure the ultrasonic generators lasted longer the second time around.

     Through our laboratory examination we determined the surface pitting and failure of the stainless steel generator housing was caused by cavitation-corrosion damage. The exterior damage to the housing was most severe on the top “active” surface of the housing. The damage was concentrated between the individual transducers in the array, indicating the presence of a “standing wave” caused by reinforcement of the wave pattern from neighboring transducers (see photo at top right of page). The morphology of the attack as revealed in the scanning electron microscope (photo below) coupled with our metallographic analysis (second photo above at right) suggested that a combination of cavitation and corrosion mechanisms were at work.

     Although the city water was not particularly aggressive, it is likely that progressive concentration of contaminants in the water (after the use of treatment chemicals was ceased) contributed to a disruption of the passive film, thereby accelerating the cavitation damage, to the extent that the failure mechanism could be classified as cavitation-corrosion.

     After being told of the failures, the tank manufacturer recommended that the water be treated with a corrosion-inhibited detergent (surfactant). Based on our analysis of the failures, we concurred with the manufacturer's suggestion and made an additional recommendation for periodic analysis of the tank contents to ensure that they did not become aggressive to the stainless steel generator housings.

     We also offered some suggestions for repairs that could be made to prolong the life of the existing housings. These involved the application of polymer-based putties like those used in the power generation industry to repair cavitation damage on hydroelectric turbines.

Passivation

     One of the tricks used to keep metal from corroding, aside from coating it with paint or another metal, is to fool it into making its own corrosion-resistant surface coating or “passive surface layer.” (Well maybe you can't actually fool Mother Nature, but that doesn't mean you can't take advantage of what she already provides.)

     Let's start with a textbook definition of passivation: Passivation is the changing of the chemically active surface of a metal to a much less reactive state. Passivity is important in the real world because certain metals and alloys, most notably stainless steel, aluminum, chromium, nickel, and titanium corrode at very low rates in environments that would normally cause them to corrode very rapidly. They do this by rapidly forming a barrier (i.e., a passive film) that is impervious to further corrosion. The chemical and structural nature of these films is often very complex, and naturally occurring protective passive films are generally only a few atoms thick (30 angstroms or less).

     Discovery of the passivation phenomenon is not at all new. Around 1900, it was observed that iron reacted rapidly in dilute nitric acid but was not attacked in concentrated nitric acid. Further, iron exposed to concentrated nitric acid and then placed back into dilute nitric acid was immune from attack. Why? A passive film forms upon exposure to the concentrated acid which then protects the iron if it is exposed to dilute acid. Just one little drawback: The protective film is very weak, and a small disruption of the film (e.g., a scratch) will cause rapid and sometimes violent corrosion of the iron.

     To some extent all metals naturally form a surface layer on contact with air, moisture, and other contamination. For example with coinage metals (i.e., money), oxide, hydroxide, and carbonate layers form naturally (if you consider rubbing a coin between greasy fingers natural!) and are strong enough to protect the coins during normal use. For many metals, however, these naturally occurring layers are weak and nonprotective; a stronger and thicker layer is needed. Anodized aluminum is an example of a material which has a thick, artificially formed oxide layer that protects the metal better than the thin naturally formed layer. And, unlike the naturally occurring passive layer, anodized aluminum coatings can be made in some pretty neat colors!

     Alloys used in surgical instruments (stainless steels) and orthopedic implants (stainless steel, titanium and cobalt alloys) all rely on their ability to passivate to withstand autoclaving conditions and/or the chemical and biological attack of the human body (actually a pretty corrosive environment). In these applications, the integrity of the naturally forming passivation layer is enhanced by the use of a chemical passivation process.

     For stainless steels, chemical passivation usually involves removal of contaminants from the surface to allow the stainless steel to spontaneously form its natural passive film upon exposure to air. Standard practices for stainless steels usually involve thorough cleaning and degreasing of the parts, followed by an acid treatment (generally 20-50 volume percent nitric acid or 4-10 weight percent citric acid) to dissolve embedded contaminants; this is followed by neutralization, thorough rinsing in deionized water, and drying. It is important not to handle passivated parts immediately after treatment, as grease and salts transferred to the parts will interfere with the formation of a quality passivation layer.

     Although through Mother Nature's efforts, certain metals and alloys benefit significantly from their own ability to form protective passive surface films, the best passive and protective oxide films can only be produced with a little outside intervention.

     We have had several projects involving the use of stainless steel parts in a variety of industrial applications (for example, gas-handling equipment for the high-tech industry) where passivated components have failed in service. In many of these cases, the quality of the passivation was suspect. Why? Well, as mentioned earlier, passive layers in stainless steel are very thin, and can be somewhat fragile under certain conditions. Thus, starting with a good passive film in the first place is obviously very important.

Materials Properties Tutorial

     We were recently asked by one of our pulp mill clients to conduct a training session for the mill's engineering and maintenance staff to familiarize them with the differences between the materials used in their old recovery boiler and the materials used in their recently completed, brand new recovery boiler.

     The old boiler, which had been built in about 1970, was designed for a maximum allowable working pressure of 500 psi and had been built with the typical carbon steel materials then in common use (ASME SA-178, SA-106, and SA-210-A1). In contrast, the new recovery boiler was designed for an operating pressure of 1,250 psi, which of course also means higher steam and metal temperatures.

     To accommodate the higher temperatures and stresses of the new boiler's higher pressure design, higher alloy materials with improved properties were specified for the tubes, headers, and branch connections (2¼Cr-1Mo and 1¼Cr-½Mo steel alloys). Additionally, to minimize some of the corrosion problems experienced industry wide in the lower firebox of recovery boilers, composite tubes were specified for the lower walls and floor tubes adjacent to the side walls. The composite tubes used in this case consisted of an outer layer of corrosion resistant material (Type 304L stainless steel or high-nickel Alloy 825, depending on tube location in the boiler) clad over the carbon steel core structural tube material.

     The primary focus for the training session was to highlight the differences in metallurgical properties between the old and new materials and the effects of these changes on welding procedures. As you can imagine, the procedures for welding the Cr-Mo steels and composite tubes are a lot different, and more complicated, than welding the carbon steel materials used in the old boiler. Additionally, the training session addressed the changes in the ASME (American Society of Mechanical Engineers) and NBIC (National Board Inspection Code) requirements brought on by the use of the upgraded materials.

MEI-People

John W. Simmons, Ph.D. PE, Senior Metallurgical Engineer received notification from the Oregon State Board of Engineering Examiners that he passed the metallurgical engineering examination given last October and has been registered as a professional engineer. Congratulations, John!