Long-Term Economics

Economics of Thermal Spraying for the Long-Term
Corrosion Protection of Steel Structuresby: Hugh Morrow, III
Zinc Institute Inc.

Long-term corrosion protection of bridges may be achieved by a coating system that requires little maintenance during the design life structure, often 50 or more years, or by a system that requires several major maintenance operations during the structure's life.

Typically, longer lasting systems cost more initially, both in materials and in the labor required to apply them. The higher labor cost are related to more stringent cleaning requirements for long- lasting materials, and to special methods, equipment, or techniques of application.

Two examples of coatings systems used for long-term corrosion protection of bridges are zinc or zinc-aluminum metalizing, and a three-coat high performing paint system consisting of an inorganic zinc-rich primer, an epoxy intermediate coat, and a urethane topcoat.

While these materials have a higher initial cost than, say, a more conventional three-coat alkyd system, this disadvantage must be weighed against three important benefits. The first is the long- term costs, or costs per sq ft per year of service, are going to be lower for long lasting systems if maintenance efforts are substantially reduced. The second is that, because maintenance efforts on bridges tend to be disruptive, a longer lasting system will minimize higher quality of transportation service. Finally, three-coat paint systems, both conventional and high performing ones, require a vigilant, well designed program of maintenance to achieve their maximum service life. A good maintenance painting program, itself, is an expensive item, because it requires specialized engineering expertise and a good deal of work, as recent Journal articles have shown.1,2,3

All three of these benefits accrue to the economic advantage of metalizing or thermal spraying. While it is impossible to be exact about life cycle costs, given the variables of specific service environments and their influence on deterioration, general trends as reported in the literature confirm that metalizing can offer a substantial economic advantage over other coating systems.

METALIZING TECHNIQUES

Metalizing may be described very simply as spraying with small droplets of a molten material. It is similar to paint spraying, with, of course, some important differences. Instead of feeding paint through a hose to the spray gun, solid wire or powder is fed to the gun where it is melted and subsequently sprayed by gases onto the prepared steel surface. The powder may be housed on the spraying gun or fed through it to flow smoothly. Metalizing wire is fed directly from coils. Processes have also been proposed that feed liquid zinc directly to a spraying nozzle for in-shop production work.

Melting of zinc or zinc-aluminum alloy wire or powder is accomplished by a gas combustion technique or an electric arc method. A schematic diagram of gas flame wire gun I shown in Figure 1a (all figures available upon request). The wire is drawn, typically by an air motor, through knurled feed rolls, into an oxygen-fuel gas flame where the wire is melted. The molten metal is then projected onto the prepared steel surface by the gas combustion products, augmented by an additional compressed air blast. The wire is normally 1/8-in. or 3/16-in. diameter. A gas flame powder gun is shown in figure 1b (all figures available upon request). It is similar to the wire gun except that the zinc is fed as a fine powder. The electric arc gun is shown in Figure 1c (all figures available upon request). Here two wires are fed such that they arc and melt at the tips. A compressed air jet then impels the molten zinc droplets to the steel surface, and the wire gap is continuously adjusted to maintain arcing as the zinc is blown away from the tips of the wires. The metal wire used for the electric arc gun normally ranges from about 1/16-in. to 1/8-in.

Table 1 - Coating Thickness of Principal Types of Zinc Coatings

THICKNESS OF ZINC COATINGS

Metalized coatings may be applied in thickness ranging from 3 mils to 12 mils, or even more in some cases. Thermally sprayed zinc and zinc-aluminum alloys utilized for corrosion protection applications are normally applied in thickness from 4 mils to 10 mils, depending on the environment and expected coating life. The service life of zinc-coated steel has been found to depend, in part, on the thickness of the coating and the exposure environment, as shown by the data in Figure 2, which is based on field experience and reported by the American Hot Dip Galvanizers Association. The heavier the coating, the longer its service life for a given environment.

The typical thickness of the six principal types of zinc coatings are summarized in Table 1. Hot dip coatings, as thick as 5 mils to 6 mils, can be produced in certain cases. Thermally sprayed zinc is normally applied in thickness greater than 6 mils. Zinc dust primers typically have an effective zinc film thickness in the 0.5 mil to 3.5 mil range, taking into account the relative amounts of zinc binder in the dried film. Continuous galvanizing, zinc electroplating, and mechanical zinc plating are normally techniques reserved for applying thinner zinc layers to small parts or sheet steel.

In addition to the actual thickness of the zinc film, there is also the factor of the form in which the zinc is present. Pure zinc, iron-zinc alloy layers, or zinc pigments dispersed in a wide variety of binders do not all exhibit the same corrosion behavior. For pure galvanic protection, one would assume that the more zinc the better.

Metalized zinc and zinc-aluminum are normally sealed with a thin coating of a low viscosity sealer, usually based on vinyl, phenolic, modified epoxy-phenolic, or polyurethane resins. Metalized coatings have been used, however, in the unsealed condition, in some environments with good success. They are, furthermore, excellent substrates for subsequent paint topcoats, and are probably the best zinc coatings with regard to paint adhesion, owing to the presence of some porosity in the coating, which is shown in Figure 3.

EXAMPLES OF APPLICATIONS

Hundreds of steel structures have been metalized to provide long-term corrosion protection, several of them dating back to the 1930's. A short listing of major metalized bridges in the US, Canada, and Europe is shown in Table 2. The year that thermal spraying of the bridge began and the latest date at which it was inspected and the coating found intact are noted. The complete coating systems are not known for some of these bridges, and in one case, the initial sprayed zinc thickness has been lost in the records somewhere. Based on the long life of the Ridge Avenue Bridge in Philadelphia, however, and the apparent spraying thickness practices in the US at that time, the original applied coating thickness on that bridge was probably 10 mils. Figure 4 shows this bridge being coated in 1938.

Several of these bridges are quite large, with significant metalized surface areas. The Pierre-Laporte Bridge, spanning the St. Lawrence River near Quebec City, is the largest on-site metalized structure in the world. The 1.8 million sq ft of the bridge's steel surface are protected from corrosion by thermally sprayed zinc. Some of the other metalized bridges are also quite sizable. The Vilsund Bridge in Denmark, for instance, involves 129,000 sq ft of zinc metalized steel surface, and the Forth Road Bridge in Scotland is more than a mile and a half long. On the Forth Road Bridge, one million pounds of zinc were thermally sprayed on some 20,000 tons of structural steel. In addition to the bridges listed in Table 2, there are also many smaller bridges in Norway, Sweden, Denmark, France, the United Kingdom, and Canada that have been thermally sprayed.

Table 2 - Metalized Zinc Bridges

THE ECONOMICS

Most of the studies of the economics of the zinc thermal spraying process have been carried out in Europe, where it is more heavily used than in North America. However, there have been cost surveys and analyses performed here as well as in Europe to establish the relative economic merits of metalizing compared to other zinc coating methods such as galvanizing or painting with a zinc- rich paint system.

In 1972 in London, Porter and Payne4 presented a paper entitled "Economic Aspects of Metal Spraying" at the symposium on "The Protection of Steel Structures by Metal Spraying". In this analysis, they examined the cost quotations from eight organizations in England and Wales for eleven different coating systems on six different kinds of steelwork that varied widely in their surface area-to-weight ratio. Only their findings on plate, girderwork, and box girders, which are the structural appropriate to bridges, will be presented here. These structures are characterized by a surface area-to-weight ratio of 90 sq ft per ton to 100 sq ft per ton, and, in the Porter and Payne survey, were quoted in 500-ton lots with the maximum weight of any one item being 10 tons.

The data for two zinc metalizing systems and the longest-lived paint system are compared as they are the most appropriate systems for bridges. The conditions for cost comparison of metalizing with either 4 mils or 10 mils of unalloyed and unsealed zinc are as follows:     

• shop coated in January 1972; 
• quotes are costs to customer;   
• costs do not include transport, marking, sorting, cleaning, and other special costs;   
• white metal grit blast surface preparation;   
• Zn 4 = 4 mils zinc, unsealed; lifetime = 8.5 years  
• Zn 10 = 10 mils zinc, unsealed; lifetime = 21 years; and   
• moderate industrial environment.

The lifetimes for the zinc metalizing systems above are based on long-term tests by the British Iron and Steel Research Association as reported by Porter and Payne.

The conditions for the cost analysis of an organic zinc-rich primer plus two epoxy topcoats are as follows: 

• shop coated in January 1972;
• quotes are costs to customer;
• costs do not include transport, marking, sorting, cleaning, and other special costs;    
• white metal grit blast surface preparation;   
• organic zinc (OZ) rich primer, epoxy based, 1.6 mils minimum DFT;   
• intermediate coat, two-pack epoxy, 2 mils minimum DFT, applied in shop;
• topcoat, two-pack epoxy, 2 mils minimum DFT, applied at job site;
• moderate industrial environment; and    
• life of three-coat paint system = 11 years.

The 11-year lifetime of this painting system is the estimate of Porter and Payne, based on experience. It does not differ substantially from service life estimates of similar systems reported in the Steel Structures Painting Manual.5

The cost quotations for the two metalized coating systems and the three-coat paint system on plate, girderwork, and box girders are summarized on Table 3 in terms of the life-cycle costs of the particular coating per unit area per year over the lifetime of the coating.

Table 3 - Life Cycle Costs* for Typical Bridge Structurals

*Life cycle costs in pence per sq. ft. per year (1972 British currency); shown for relative comparison only.

The 4-mil zinc metalizing system was the least expensive to apply initially. However, it proves to be the most expensive when considered in terms of life-cycle costs. Conversely, the 10-mil metalized zinc system was the most expensive to apply initially but the most cost-effective in terms of life-cycle costs. The three-coat paint system was intermediate in both initial and life-cycle costs.

In 1978, Porter6 extended the survey and analysis to include galvanized products, sealed metalized coatings, and more painting systems. In terms of life-cycle costs, galvanizing was shown to be very cost-effective for high surface area-to-weight ratio work. In the area of bridge girderwork, however, metalizing with either zinc or aluminum was more cost-effective.

In 1985, Stoneman7 re-examined the results of the 1978 survey. He updated the cost information to reflect current conditions. These results are summarized in Figure 6. The costs for the following three coating systems are shown:

• pickle and hot dip galvanized (3.3 mils minimum zinc coating thickness);    
• grit blast and zinc thermal spray (4 mils nominal zinc thickness); and   
• grit blast and airless spray, either (a) three coats of an organic zinc-rich paint system (total of 6 mils nominal thickness), (b) three coats of a high build chlorinated rubber (12 mils total), or (c) three coats of a two-pack chemical resistant paint (11 mils total).

Stoneman's surprising data on initial costs agrees with the data of Porter and Payne. Both show a 4-mil coat of metalizing to be less expensive than three-coat paint systems, even in initial costs. This finding is explained, in part, by conditions in the UK, where metalizing is a more common practice than in the US. Greater volumes of work and more competitive bidding drive down the costs of metalizing.

What about bridge metalizing costs in North America? The costs of metalizing the Pierre-Laporte Bridge over the St. Lawrence River in Quebec is an excellent example of the cost-effective economics of this process. Because of the high cost of maintenance painting which had previously been required on this bridge, and the favorable experience found with metalizing other smaller overpasses in Quebec Province, cost analysis was performed and the decision made to metalize the structural girderwork, as see in Figure 7, of the Pierre-Laporte Bridge. The two end spans of the bridge were metalized beginning in 1977, and the center span was thermal sprayed beginning in 1979. The 1979 unit costs for the entire project were $29.26 per square meter (Canadian currency), which converts to the US equivalent of $2.17 per sq ft.

It is estimated by Quebec Ministry of Transport that the coating will need no maintenance for at least 25 years.8 This estimate is based upon extrapolations from the Ministry of Transport's experience in the UK. Thus, the annual cost of protection for the Pierre-Laporte Bridge is expected to be $0.087 per sq ft per year expressed in 1979 dollars. These costs, of course, reflect the economies of scale realized when a structure as large as the Pierre-Laporte Bridge is protected from corrosion by metalizing. However, similar economies were reflected in US studies when the costs for metalizing were examined.

In the Summer of 1984, the Zinc Metalizers Task Group of the Zinc Institute conducted a survey of six organizations that perform both thermal spraying and painting operations. These organizations included companies in the eastern, midwestern, and southwestern US. The conditions of the cost analysis were as follows:  

• shop coating;
• eight I-beams; 
• dimensions: 60 ft length, 6 ft web, 14 in. flanges; 
• quoted in July 1984; 
• life based on five percent red rust; 
• moderate industrial environment; and 
• quotes from six vendors.

The I-beams under cost examination here are typical of those that might be used on overpass bridges in the US. The metalizing system consisted of a white metal blast surface preparation, 8 mils of metalized zinc sealed with a thinned vinyl, and two coats of unthinned vinyl.

The costs for the metalized coating were compared with those for two paint systems. Both systems were suggested by the Federal Highway Administration as a good basis for comparison. Paint system number one, a three coat system, consisted of near white blast surface preparation, 3 mils of an inorganic zinc-rich primer, an epoxy polymide intermediate coat, and 2 mils of a urethane topcoat. Paint system number two, a two-coat system, employed a near-white blast surface preparation, 3 mils of an inorganic zinc-rich primer, a vinyl butyrate wash primer, and 3 mils of a high build vinyl topcoat.

The initial and long-term costs of corrosion protection using these three systems are presented in Table 4. The striking difference between the metalized system and the two paint systems studied is their comparative service lives. The 8-mil metalized and sealed zinc coating topped with two coats of vinyl will last more that 25 years, according to survey results, while the two paint systems will require maintenance in less than ten years. These life estimates are based on information from the Metalizers themselves, and are generally consistent with those listed in the Steel Structures Painting Manual5 and with those presented by A.H. Roebuck9 in numerous papers. It must be noted, however, that the life cycle costs calculated here are very sensitive to the lifetimes of the coating selected. If paint system number one had a lifetime of about 17 years and paint system number two had a lifetime of about 15 years under these same conditions, then the life cycle costs of the paint systems would be about the same as those of the metalized system, with its life of about 25 years. The large differences in coating lives means that, even though a metalizing may be more expensive to apply initially, it will be more cost effective when considered on the basis of life-cycle costs.

Table 4 - Comparative Coating Costs* from 1984 Zinc Metalizers Task Group Survey

*Average coating costs in dollars.
**Figures in parentheses are estimated lifetimes reported in the Steel Structures Painting Manual

Dividing the initial coating cost, however, by the coating lifetime yields only an approximate life cycle cost. In reality, a steel structure normally lasts far longer than its protective coating, and thus the coating must be maintained over the entire lifetime of the structure. Therefore, in comparing corrosion protection systems, the costs of maintenance as some time in the future must also be included.

The basic principles of coating economics may be illustrated using a hypothetical example (Table 5). Imagine that a steel structure requires corrosion protection for 20 years. Coating A will provide this protection by one application at a cost of $0.75 per sq ft. Coating B will provide this protection by an initial application at a cost of $0.35 per sq ft, and regularly scheduled maintenance after 5, 10, and 15 years at a cost of $0.25 per sq ft each time. The total cost of corrosion protection for 20 years using Coating A is $0.75 per sq ft, while that for Coating B is $1.10 per sq ft. If we had compared the two systems based on the equation that life-cycle cost = initial coating cost divided by lifetime, the life-cycle cost for Coating A would be $0.75 divided by 20 years of $3.75 per sq ft per year. The life-cycle cost for Coating B without maintenance would be $0.35 divided by 5 years or $0.07 per sq ft per year. the life-cycle costs of Coating B including maintenance at 5, 10, and 15 years would be $1.10 divided by 20 years or $0.055 per sq ft per year. Both methods of cost comparison clearly indicate that Coating A is the more cost-effective.

Table 5 - Hypothetical Coating Cost* Comparison

*Costs in dollars per square foot.

However, the analysis for Coating B is oversimplified because the costs for maintenance at 5, 10 and 15 years will rise due to inflation (Table 6). If four percent inflation is assumed, maintenance in 5 years will cost $0.30 per sq ft, in 10 years $0.37, and in 15 years $0.45. But then, if the lower cost Coating B is initially chosen, the savings realized from not using the expensive Coating A can be invested to earn interest. At the required maintenance intervals, the savings plus the interest, its present value is greater than its value in the future. Thus, a projected cost at some time in the future must be discounted to its present value.

The discounted costs for the future inflated maintenance costs are shown in the third line in Table 6 assuming a discount or interest rate of eight percent per year. Thus, the total cost for Coating B is $0.86 per sq ft, while that of Coating A is $0.75 per sq ft. Coating A is still more cost effective corrosion protection system, but by less than suspected at first glance.

Table 6 - Present Worth of Coating Cost* for Coating B.

*Cost in dollars per square foot.

The data in Table 4 are now re-examined using discounted cash flow to compare the 27-year costs of metalizing and paint systems numbers one and two. For paint system number one, we assume that repainting will be required after 9 and 18 years, and for paint systems number two, after 7, 14, and 21 years. For this consumption, the repaint costs are assumed to be the same as the initial costs (in real dollars). Normally, however, a system with five percent surface rust would not require a complete reblast and repriming. The methodology was that described by Appleman.1

The revised costs of paint systems numbers one and two are $0.277 per sq ft per year and $0.256 per sq ft per year, respectively. These are still higher than the costs for metalizing ($0.157 per sq ft per year), but by a smaller amount. If one assumes a repaint interval of 15 years for the two paint systems, the costs are reduced to $0.159 per sq ft per year for paint system number one and $0.141 per sq ft per year for paint system number two compared to $0.157 for metalizing. Thus, the conclusions to be drawn depend not only on the initial paint costs and the average life time, but also on the cost to repaint, the number of years repainting can be deferred, and the inflation and discount rate.

FACTORS INFLUENCING SELECTION

For the corrosion protection of steel structures in private industry, the factors of depreciation and taxes must also be considered. These two cost items tend to offset each other as do the inflation and interest factors. However, it is the specific rates of each of these factors that are so important in determining the most economical coating system. If interest rates are high and inflation is low, it is better to invest in a system with a lower initial cost. If the inflation rates are high and the interest rates low, it is more prudent to choose a higher cost, high quality system such as metalizing that will require no further maintenance. In general, high tax rates favor use of lower cost system while high depreciation factors favor the high quality system. The problem with this analysis is that for a long- lived steel structure, these factors are all likely to vary considerably and uncontrollably over the long term.

Beyond these considerations, there are other factors of great importance. If the overall cost of the project is high and the corrosion protection system but a very small part of that cost, then the tendency would be to use a high quality coating system, regardless of initial cost, to protect that large investment. Likewise, high shutdown or service interruption costs dictate the use of a coating system that will not require maintenance for long intervals. Difficulty of access for maintenance is also likely to favor a maintenance-free coating system such as metalizing. The surface area-to- weight ratio of the work being coated, as previously discussed, is clearly important, too. Factors favoring the use of metalizing are listed below. 

• High Inflation Rate
• Low Interest Rate
• Low Tax Rates
• High Depreciation Allowance
• High Initial Cost of Projects
• High Shutdown Costs
• Difficult Maintenance Access

Surface Area-to-Weight Ratio Below 120 sq ft per ton

On the basis of some of the information presented herein, there appear to be specific areas where each of the three methods for coating structural steel with zinc - galvanizing, metalizing, and painting - are most cost-effective. For high surface area-to-weight ratio structurals, galvanizing appears the most cost-effective for service lives up to 40 years. For low surface area-to-weight ratio materials and corrosion protection required beyond 25 years, metalizing should be considered the most cost-effective on a life-cycle basis. For low surface area-to-weight ratio steel and design lives of less than about 25 years, painting with zinc-rich paints is attractive, cost-effective alternative. These general guidelines also must be considered in the light of some of the factors mentioned in the list above.

It is important to remember, however, that regardless of these general considerations, there are circumstances in which one or more of these methods simply cannot be used. Galvanizing, for example, cannot be employed for the recoating of a bridge that is already in place or for structural members where heating and cooling leads to distortion or changes in mechanical properties. Metalizing and painting cannot be performed on inaccessible surfaces such as the insides of tubes and pipes.

In summary, metalizing is one of the most cost-effective techniques available today for the long-term corrosion protection of large steel structures such as bridges. Structures in Europe and Canada have demonstrated that thermal spraying life-cycle costs are equal to or less than those of three-coat paint systems. Surveys in the UK and the US clearly indicate the life-cycle cost effectiveness of the process. In North America today, where so much of our infrastructure is crumbling from corrosion and where maintenance of bridges seems to be a never ending problem, there is a clear-cut, cost- effective solution, metalizing.

REFERENCES

Appleman, B.R., "Economics of Corrosion Protection by Coatings," Journal of Protective Coatings & Linings, March 1985. Appleman, B.R., "Maintenance Repainting of Structural Steel: Chemistry and Criteria," Journal of Protective Coatings & Linings, August 1984. Delahunt, J.F., "Maintenance Paint Surveys of Refineries and Petrochemical Plants," Journal of Protective Coatings & Linings, December 1984. Porter, F.C. and Payne, J.H., "Economic Aspects of Metal Spraying," presented at "The Protection of Steel Structures by Metal Spraying," London, 1972. Sline, M.R., et al., "Comparative Painting Costs," Steel Structures Painting Manual, Vol. 1, Good Painting Practice, Steel Structures Painting Council, Pittsburgh, PA 1982. Porter, F.C., "Comparative Costs of Protecting Steel," Zinc Development Association, London 1978. Stoneman, A., "Protective Coatings for Steel: Initial Costs in Perspective," Zinc Development Association, London, 1978. Jodoin, N. and Nadeau, M., "The Biggest On-Site Metalizing Project: Pierre LaPorte Bridge," International Thermal Spraying Conference, The Hague, May 19-23, 1980. Roebuck, A.H. and Brevoort, G.H., "Coating Works Costs: Computer Application and Inspection." CORROSION/86, National Association of Corrosion Engineers, Houston, TX, 1986.