Click for next page ( 7

The National Academies of Sciences, Engineering, and Medicine
500 Fifth St. N.W. | Washington, D.C. 20001

Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement

Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 6
7 CHAPTER 2 FINDINGS COLLECT AND REVIEW DOMESTIC AND ter particles, faster output, and superior adhesion than do FOREIGN LITERATURE flame spray techniques (2, 612). Various thermal spray coating materials are available for Literature Review use over steel. Table 1 lists and categorizes some of the mate- rials commonly used for metallizing, as well as typical wire A comprehensive information search was conducted to gauges that are available. Zinc, aluminum, and their alloys are obtain and review information relevant to the design, speci- commonly used as TSMCs on steel in water immersion. The fication, and installation of coated steel pilings. Literature sacrificial corrosion protection that they offer, in combination was obtained by using available government, university, and with their relatively low corrosion rates, make them suitable industry databases. The search included the following: for such harsh environments (2, 7, 9, 11, 1319). Thus, zinc, aluminum, and an alloy of the two metals (8515 weight per- The Transportation Information Research Service (TRIS), cent [wt%] zinc/aluminum) were selected for testing under The National Technical Information Systems Database, this program. Table 2 lists some of the properties of these AASHTO Listings of Research in Progress, three TSMCs commonly used on steel in water immersion. The Current Technologies Index, Thermally sprayed zinc and aluminum coatings are com- Engineered Materials Abstracts, monly used without sealers in mild environments. However, Federal Research in Progress, TSMCs inherently contain porosity that has a major effect on International Conference Papers Abstracts, corrosion performance. When exposed to harsh environments, The Federal Highway Administration, such as marine atmospheres and/or freshwater or saltwater The U.S. Army Corps of Engineers, immersion, the application of a sealer on top of the thermally The Soil Conservation Service, sprayed coating is generally recommended (2, 7, 2023). The State DOTs, purpose of the sealer is to mitigate corrosion caused by the The National Institute for Standards and Technology, penetration of moisture and corrosive ions through pores. The American Society for Testing and Materials, Sealers are often used to enhance the appearance of the coated The National Association of Corrosion Engineers (NACE) structure as well as to extend the life of the thermal spray coat- International, ing. Previous studies have involved dozens of different seal- The Society for Protective Coatings (SSPC), ers, and the studies generally agree that an important property The American Welding Society (AWS), of the sealer is to adequately fill the pores in the thermally The Materials Information Society (ASM), and sprayed coating (2, 8, 9, 14, 16, 22, 24). Some of the more The Thermal Spray Society. common sealers tested include vinyls, silicones, epoxies, ure- thanes, phenolic resins (may react with zinc [24]), and alu- This literature included performance data in technical papers minum pigmented silicone (for high temperature) (7, 8, 20). and reports and specifications for the application of TSMCs. Previous studies have indicated that corrosion rates of steel This literature was sorted, and the most relevant reports were in marine environments are lower in the immersed and inter- reviewed. Of primary interest were data related to the per- tidal zones than in the splash zone (25). The lower corrosion formance of TSMCs in natural waters, application factors rates in the immersed and intertidal zones are believed to be, affecting the life of TSMCs, and appropriate quality assur- in part, the result of the attachments of various organisms and ance tests. Literature of significant value was abstracted and marine growth. Marine fouling is not as prominent in the is included in the report appendixes. splash zone. Because of this lack of fouling and the wet/dry Although the literature identified several available thermal cycling in the splash zone, corrosion rates tend to be highest in spray systems, the ones used most frequently were wire-arc this area. Unprotected steel corrosion rates in the splash zone spray and wire-flame spray. Also, these two systems are often generally range from 4 to 10 mils (102 to 254 m) per year. the most practical application methods for TSMC on pilings. Corrosion of steel pilings in a marine environment may also However, wire-arc spray techniques generally allow for hot- be the result of exposure to variable oxygen concentrations.

OCR for page 6
8 TABLE 1 List of materials commonly available as TSMCs Classification Materials/Wire Comments Aluminum and zinc are available Al99.0%--11 and 14 gauge and pre-alloyed or can be pseudo 1/8-in. wire alloyed with the proper metallizing Anodic Zn 99.9%--11 and 14 gauge and system. These coatings offer 1/8-in. wire sacrificial galvanic protection to a steel substrate. Cu 99.8%--14 gauge Cu 9Al 1Fe--14 gauge Fe 13Cr 0.5Si 0.5Ni 0.5Mn 0.35C-- 14 gauge While these coating systems are Fe 18Cr 8.5Mn 5Ni 1Si 0.15C--14 considered corrosion resistant, gauge Corrosion they are all cathodic to steel and, Fe 28Cr 5C 1Mn--14 gauge Resistant thus, offer no sacrificial protection Ni 5Al--14 gauge to a steel substrate at coating Ni 5Mo 5.5A1--14 gauge defects. Ni 5Mo 5.5A1--14 gauge Ni 18Cr 6A1--14 gauge Sn 7.5Sb 3.5Cu 0.25Pb--14 gauge C 276 Ni Alloy--14 gauge These TSMCs are generally used Chromium Hard Coatings in applications where abrasion Tungsten-Carbides resistance is a desired property. NOTE: 1 in. = 2.54 cm. Al = aluminum, Zn = zinc, Cu = copper, Fe = iron, Cr = chromium, Si = silicon, Ni = nickel, Mn = manganese, C = carbon, Sn = tin, Sb = antimony, Pb = lead. Corrosion will be more severe at zones with low oxygen con- thickness (DFT) of a coating over an "I beam" were 10 mils tent and lower at more aerated zones. Highly localized corro- (254 m) and the minimum DFT over an edge of the beam sion known as macrocell corrosion, or oxygen concentration were 5 mils (127 m), then the edge retention would be 50 per- cell corrosion, can result. This phenomenon is believed to be cent. Most liquid coatings tend to "pull back" or flow away the primary cause of corrosion of piles in heterogeneous soils from sharp edges during application. This further exagger- (26). Other studies have also concluded that macrocell corro- ates the low film thickness encountered at the edges of a part. sion is involved in the corrosion of steel in the tidal zone (25). Many coating specifications make up for this lower edge cov- In oxygen-deficient areas, excess metal dissolution occurs, erage by adding "stripe coats" over all edges and nonuniform and the local pH falls; in oxygen-rich areas, oxygen reduction surfaces (e.g., bolts and welds). TSMCs are applied via a of hydroxide ions occurs, and the local alkalinity increases. "line-of-sight process" (very similar to conventional and air- Severe concentrated corrosion can occur just below the low less spray of liquid coatings) yet do not "flow" as many liq- tide zone as a result of oxygen concentration cells. uid coatings do. TSMC deposition only occurs when the metal The edge retention of coating systems is a property that can particles collide with a surface. As a TSMC application gun affect the overall system performance. If the edges of a struc- is turned away from perpendicular to the surface, the deposi- ture are not coated well, the system may not be acceptable for tion efficiency decreases dramatically. Unlike spraying liq- complex shapes or assemblies. Edge retention is defined as uid coatings, in which the "wet" material may have more of the percentage of the flat surface film thickness that covers an a tendency to adhere from a severe angle or long spray dis- edge of the substrate. As an example, if the nominal dry film tance, TSMC relies on impact energy and a short molten TABLE 2 Properties of commonly used TSMC materials TSMC Metal Alloy Alloy Rational 1 99.9% pure Zn Excellent cathodic protection properties, harder than aluminum, life proportional to thickness, not for acidic environments or high temperatures. 2 99% pure Al Low self-corrosion rate, good seawater performance, high-temperature resistance, lightweight, acid (pollution) resistant. 3 85:15 wt% Zn- Harder than aluminum, has shown better atmospheric Al performance than zinc or aluminum. NOTE: Al = aluminum, Zn = zinc.

OCR for page 6
9 phase for adhesion, so a TSMC particle has a greater ten- rates of 0.31 mpy (7.8 m/yr) for 8515 wt% zinc-aluminum. dency to "bounce" from the surface. Over an edge, where the These corrosion rates agree quite well with measurements angle of application will stray from perpendicular, a TSMC obtained as part of this program on samples from both the will deposit with less efficiency than on a flat surface, so a laboratory and the field. These data are discussed below. reduced thickness on the edge is expected. The sharpness of Such low corrosion rates are consistent with the findings of the edge will also affect TSMC deposition on the edge. A field exposure testing listed in the literature (i.e., 12-mil very sharp, 90-degree edge will retain less coating thickness [300-m] coatings lasting for 20 years with little to no base- than a chamfered or "broken" edge. metal deterioration). While many liquid coating systems rely on stripe coatings Several materials, as discussed above, exhibit the findings to build the film thickness at edges, a TSMC that is anodic to of low corrosion rate and extended service life. This extended steel may not need additional edge coverage because of the service life is in the range of 20 years or more. Several sys- sacrificial protection offered by the other areas of TSMC. tems, properly applied, seem to be able to meet this criterion. However, because piles often have several different edge Thus, inherent material corrosion rate is not a key issue for configurations, edge retention was considered during the lab- the better performing materials; the key issue becomes the oratory phase of the project. coating process needed to obtain a coating that meets this Although the literature identifies various coating systems performance expectation. This is a specification and quality for potential application in natural waters (predominately sea- assurance issue as opposed to a material selection issue. water), there are three principal systems that have received the The most common defect cited in the deterioration of most consideration: commercially pure aluminum, commer- TSMCs is inter-coat "blistering" or delamination. In this cially pure zinc, and 8515 wt% zinc-aluminum. Table 3 process, significant section loss of the TSMC is observed. shows the TSMC materials of interest tested most frequently Extensive propagation of such delamination can impact the within the literature. useful service life of the material. It appears that delamination Of these primary TSMC materials, the material most often may be related to inter-coat corrosion occurring at selected suggested as having superior performance in seawater and pores or defects in the coating. This may occur along oxide freshwater environments is aluminum. Aluminum is listed as boundaries or, in the case of mixed-metal TSMC, at differ- the outstanding candidate by the National Materials Advi- ing alloy phases/compositions. sory Board at the National Academy of Sciences in their On a visit to test piles at the North Carolina DOT test site report, Metallized Coatings for Corrosion Control of Naval at Ocracoke, North Carolina, defects in thermally sprayed Ship Structures and Components, in the American Welding aluminum (99.5%) coating were visible, as shown in Figure 1, Society (AWS) 19-year report on metallized coatings, and in after 2 years of service in a seawater piling application. The several papers dealing with TSMC used in the offshore defect appears as a split in the coating at the bend in the sheet industry (i.e., used in seawater) (7, 8, 11, 12, 14, 22, 2731). piling. Similar inter-coat defects with a slightly different con- Other papers have also concluded that aluminum is an excel- figuration also appeared on an 8515 wt% zinc/aluminum lent performer in freshwater environments (12, 24, 28). piling treated with TSMC at the same site. A 7030 wt% zinc/ From a material property standpoint, the corrosion rate of aluminum pseudo-alloy TSMC appeared in the best shape at all the TSMCs is quite low. These general corrosion rates appear in the 0.1- to 0.4-mpy (2.5- to 10-m/yr) range. The literature suggests corrosion rates of 0.13 mpy (3.3 m/yr) for aluminum TSMC. The same literature suggests corrosion TABLE 3 Materials and frequency of testing Percent of Time Material Tested Pure Al 75 Pure Zn 60 85:15 wt% Zn-Al 25 65:35 wt% Zn-Al 5 55:45 wt% Zn-Al 5 90:10 wt% Al-Al2O3 10 95:5 wt% Al-Mg 10 60:40 wt% Zn-Fe 5 Al-Zn-In 5 NOTE: Al = aluminum, Zn = zinc, Al2O3 = aluminum oxide, Mg = magnesium, Fe = Figure 1. Thermal spray aluminum (99.5 percent) coating iron, In = indium. in seawater after 2 years exposure.

OCR for page 6
10 the site, free of such delaminations through 2 years of ser- ments include an SSPC-SP-10, "Near-White Blast Clean- vice. Yet, some areas of the piling appeared to exhibit incip- ing," or SSPC-SP-5, "White Metal Blast Cleaning," produc- ient blisters. Similar types of defects have often been the ing a 2- to 4-mil (25- to 102-m) profile with angular grit. deciding factor in the ultimate rating of TSMCs. If the occur- Testing under this program included variations in surface rence of such defects can be related to specific application preparation quality to determine the influence of such factors parameters, then the life of the coating can be extended con- as surface profile and surface contamination. siderably. Historically, appropriate sealer coats are often listed as reducing the tendency for these types of defects. Several thermal spray systems were reviewed for testing Initial Testing and Field Surveys under this program. Some of these systems include wire-arc spray, wire-flame spray, powder flame spray, high-velocity One of the initial laboratory tests was a galvanic current oxygen-fuel (HVOF) spray, and plasma spray. Wire-arc spray test in which polyvinyl chloride (PVC) panels were coated generally provides faster output and superior adhesion when with thermally sprayed aluminum and zinc by wire-arc spray. compared with flame spray techniques (8). Production rates These panels, which were 2 in. (5 cm) in diameter, were elec- of up to 90 and 300 lb (41 and 136 kg) per hour have been trically coupled to bare steel panels (1/32 in. 4 in. 6 in. obtained for wire-arc sprayed aluminum and zinc, respec- [0.079 cm 10.2 cm 15.2 cm]) and immersed in natural tively (15). Various other thermal spray techniques, includ- seawater. Current flow between the coated PVC panels and ing HVOF and plasma spray, are also available. However, the bare steel panels was monitored and plotted as a function these alternative thermal spray systems, when compared with of time, as shown in Figure 2. The preliminary electrochem- wire-arc spray, generally do not offer the same levels of out- ical data indicated that while zinc may be more active ini- put and cost-effectiveness. tially, its corrosion rate gradually decreases to a level closer There are also reports in the literature that using a larger- to that of aluminum. Visual observations indicated that the diameter feed wire can increase productivity (32). How- steel panel coupled to the aluminum coating exhibited sig- ever, other technical papers suggest that as wire diameter is nificantly more corrosion (red rust) than did the panel cou- increased, porosity also increases (9). This may lead to more pled to the zinc coating. Figure 2 shows that the galvanic cur- coating defects. Any such tradeoffs between productivity and rent flows from the aluminum- and zinc-coated panels were at coating performance need to be considered in testing. the same level after 3 weeks of immersion. These data indi- Surface preparation is considered the most important part cate that the corrosion rate of zinc, relative to aluminum, in of thermally sprayed coating application. Typical require- long-term exposure, may not be as high as much of the liter- 7 6 5 4 Zinc 3 Aluminum 2 1 0 0 20 40 60 80 100 120 TIME (days) Figure 2. Galvanic current data for TSMC materials (zinc and aluminum) coupled to steel panels in seawater.

OCR for page 6
11 ature suggests. The conclusions in the literature that zinc will The primary source of coating deterioration was inter- corrode at higher rates may be somewhat misleading because coat cracking and delamination between the base metal of the lack of data from long-term exposures. However, and coating. according to the AWS "Corrosion Test of Flame-Sprayed Deterioration occurred at rolled corners. Coated Steel--19 Year Report," unlike aluminum, the ser- Electrochemical data confirmed that the seal coats vice life of zinc thermal spray coatings is dependent on coat- were not effective electrical barriers. Seal coats are not ing thickness (28). intended to behave as full barrier coatings. A test piling was set up outside of Corrpro's Ocean City lab- A 2-year exposure period is inadequate to extrapolate oratory to evaluate corrosion mechanisms associated with a the coating performance over a 20-year period. seawater environment. Seven steel segments were exposed to various zones in seawater--mud, immersion, tidal, and splash Interviews with Coating Applicators zones. Five of the segments were coated with zinc applied by wire-arc spray. Two segments, one below the mudline and Six thermal spray coating applicators were contacted. Of one in seawater immersion, were bare steel. The five zinc- these, three were willing to discuss the TSMC procedures. coated segments were electrically connected with the bare Structural Coatings, Inc., was visited in July 1999, at the steel segment below the mudline. Current flow between the beginning of the project, to view some pipe piles being coated segments was monitored periodically. Current data indicated and to obtain some initial information. Structural Coatings, that the majority of the sacrificial protection to the bare steel Inc., was visited again on February 20, 2001, and question- segment came from the zinc-coated segment below the mud- naires were submitted to the other two applicators. Informa- line. One of the zinc-coated segments in the seawater immer- tion gathered included types of abrasive used, use of mixed sion zone was consistently receiving protective current. This abrasives, treatment of flame-cut edges, spray techniques used, lack of uniform galvanic current flow is indicative of macro- types of application equipment, types of sealers used, qual- cell corrosion effects. These results imply that, as a design ity control procedures, and qualifications. The three applica- agent, long-term local corrosion driven by the macrocell cor- tors provided useful information, which helped to focus and rosion, as well as the general corrosion rate of the material, reduce the amount of laboratory testing needed. The thermal are corrosion mechanisms that need to be considered. The spray applicators contacted were the following: majority of long-term piling tests have focused on the visible corrosion at or above the water line. The segmented piling Jupiter Painting Structural Coatings, Inc. data imply that corrosion below the water line, via macrocell 1500 River Rd. P.O. Box 334, Highway 70 E corrosion, may be a factor in determining the service life of Croydon, PA 19020 Clayton, NC 25720 a coated piling. These results lead to the question of whether 215-785-6920 919-553-3034 the piling should be coated below the mudline to limit gal- Contact: Paul Tsourous Contact: Ray Hails vanic interaction. Corrpro personnel performed a field survey of existing test CSI Coatings pilings installed in late 1997 by the North Carolina Depart- 2102 5 Street ment of Transportation (NCDOT) at the South Side Cape Nisku, Alberta T9E7X3 Canada Hatteras Ferry Dock on Ocracoke Island in North Carolina. 780-955-2856 Five steel test pilings had been exposed for about 2 years at Contact: Wayne Duncan the time of the field survey. The test piles consisted of the fol- (CSI Coatings is a wholly lowing TSMCs: owned subsidiary of Corrpro Companies, Inc.) Arc-sprayed aluminum and sealer, Arc-sprayed 7030 wt% aluminum-zinc and sealer, As a result of the literature search and field and initial labo- Arc-sprayed 8515 wt% zinc/aluminum and sealer, ratory studies, the following focus was determined for follow- Inorganic zinc-rich paint, and on research: Bare steel. Further testing should be limited to the most commonly In addition, the propeller wash piling wall is coated with used alloys for corrosion control, that is, the anodic coat- coal tar. The results of the field survey are summarized as ings of "pure" aluminum and zinc and the 8515 wt% follows: zinc/aluminum alloy. Properly applied, these materials show every indication of being able to meet the intent of Electrochemical data indicated that the uniform corrosion the program (i.e., to significantly extend the life of steel rates of the thermally sprayed coating materials were rel- piling materials in immersion in natural waters). atively low; however, visual observations indicated local- As opposed to testing an increased variety of TSMC base ized areas of deterioration. materials, a significant testing focus should be appropriate