1. INTRODUCTION
Ductile iron is one of the most commonly used pipe materials in our modern society. The first ductile iron pipe was cast experimentally in 1948 and was introduced into the marketplace in 1955. By the late 1970's, ductile iron pipe had replaced gray cast iron pipe in the marketplace. Cast iron is an old principal pipe material that was used particularly for water and wastewater service in North America for the past 150 years. Today, ductile iron pipe is widely used in the transportation of raw and potable water, sewage, digester gas, slurries, and process chemicals.
Deterioration of underground pipes due to corrosion or mechanical failure is of increasing concern to users in both the private and the public sectors in North America and other countries. The primary concern relates to the increasingly large number of leaks that have occurred in potable water and sewage systems. The demand for privatization within the water industry to yield cost savings, as well as more aggressive pressures from governments to reduce wastage of resources, has meant that methods of preventing the rate of deterioration of buried pipes has become an important issue.
As the number of ductile and gray cast iron buried piping systems far exceeds the number of other types of water distribution systems, it would be expected that corrosion of both ductile and gray cast iron would have received major attention from engineers, municipals, water companies,
and pipe producers. It is interesting to note, therefore, that a notable lack of consistent opinion appears to exist on issues such as failure mechanisms, corrosion resistance, and optimal
corrosion control methodologies for buried ductile and gray cast iron pipelines. Such controversies can be for example seen from the 1992 NACE International Report from its Task Group T-10A-21 on Corrosion Control of Ductile and Cast Iron Pipe 1. The most obvious example of this lack of consensus is the relative service life expectancy of ductile and gray cast iron when exposed to underground conditions. A second controversy is that relating to the relative effectiveness of loose polyethylene encasement system, a corrosion control method that is widely used for ductile and gray cast iron pipe but rarely used as a corrosion prevention approach for pipes of other types of materials.
The present paper focuses on corrosion and protection of buried ductile iron pipe. External corrosion mechanisms and corrosion resistance of ductile iron pipe are reviewed. In addition to considering the relative merits of uncoated cast and ductile iron pipes, the value of corrosion control measures such as bituminous coatings, sacrificial metals, polyethylene encasement, tape wrapping, bonded coatings, cathodic protection, the impact on cathodic protection of electrical insulation at pipe joints, and the value of improvements in trenching methods are evaluated.
2. THE COMPARATIVE CORROSION RESISTANCE
OF DUCTILE IRON,
GRAY IRON, AND STEEL
Ductile iron is a high carbon, cast ferrous material with a composition that is generally similar to gray iron. There is little difference in the chemically analyses of ductile iron and gray iron. Both contain a similar amount of carbon, which affects the machineability and the corrosion resistance of both materials. In gray cast iron, most of the carbon is present in the form of continuous network of flake graphite platelets, which are dispersed throughout the metal matrix. In gray iron this matrix is the major factor controlling its mechanical properties and is responsible for its relative weakness and lack of ductility. In ductile iron, on the other hand, an additional procedure during the casting process causes the graphite to ball-up in the form of spheroids or nodules. When properly processed cast ductile iron essentially consists of a near single-phase ferritic material with only minor discontinuities due to the presence of the graphitic spheroids. In consequence, the mechanical strength and ductility of the material is much close to that of steel. Ductile iron therefore is a material that offers the low-cost foundry manufacturing characteristics of a gray iron but with the mechanical properties similar to a steel.
It might be assumed on the basis that there is little difference between the chemical analyses of ductile and gray cast irons that the corrosion resistance of the two materials also would be similar. However, there is disagreement as to whether or not ductile iron should have better corrosion resistance than has gray cast iron because of the spheroidal morphology of the graphite nodules. LaQue2 considered earlier that the interconnected and overlapping flakes of graphite in gray iron (as opposed to the well-dispersed and separate graphite nodules found in ductile iron) could tend to promote a greater depth of penetration of corrosion along the margins of the graphite flakes. This suggestion was supported by Fuller3 based on his analysis of data from ductile pipe that had been exposed in field installations. It was suggested that ductile iron would be less susceptible to deep localized pitting than was gray iron because its spheroidal graphite structure might encourage a given amount of corrosion attack to spread out over the surface of the metal. Ferguson and Nicholas 14 reported from their tests on adjacent ductile and gray iron mains that the durability of ductile iron pipe was better than that of gray iron pipe due to lesser pitting rate, greater strength, and greater ductility. And they concluded that these three factors more than compensated for the reduced wall thickness of the ductile iron material.
An alternative view expressed by Cox 13 and others is that the flake graphite matrix of the gray cast irons was a highly effective diffusion barrier that served to impede both the access of aggressive species to the corrosion interface of the ferrite phase, and to retain the corrosion products within its matrix, thereby stifling the subsequent corrosion activity. Experience in British water distribution systems revealed that in old cast iron water mains which had been laid in the latter part of the last century and the early part of the present century, graphitization could result in a substantial reduction in residual pipe wall thickness over large areas of piping without loss of product containment. It was considered that the presence of the graphite matrix has extended considerably the service life of the pipeline, compared to the service life uncoated steel or ductile iron pipes exposed in the same environment.
Other research concluded that the corrosion behavior and corrosion resistance of ductile and gray cast irons would have no significant differences 1, 4-10. Some studies indicated that ductile iron might indeed corrode faster than gray cast iron 10-13. When corrosion of ductile iron occurs, the graphite that is present will be left is an integral part of the corrosion by-products, which may serve as a preferential cathode, thereby promoting the corrosion mechanism. By contrast, in addition to their protective value, the flake graphite-containing corrosion products have considerable mechanical strength, which is believed to contribute to the long service life of some unprotected gray cast iron pipelines in corrosive environments. The strength and adhesion of the flake graphite-containing corrosion products of gray cast iron are believed to be greater than is obtained in the case of corrosion products formed by ductile iron, due to the microstructural and minor composition difference between the two materials 10,12. The iron corrosion products on gray iron are tightly bound together as well as to the pipe metal substrate by the residual flake graphite structure and the remaining eutectic network. The eutectic network in the case of phosphorus-rich gray iron is made of the more corrosion resistant phosphide eutectic. In the case of ductile iron, the graphite is in the form of discrete nodules, which are readily detached. There is negligible phosphide eutectic present in ductile iron due to the lower levels of phosphorus necessary to achieve the essential spheroidal graphite structure during the casting process. Consequently, neither of these features that confer strength to gray iron corrosion products could be present in the case of ductile iron 10,12. LaQue 2 reported from tests on specimens buried in an abnormally corrosive environment of clay soil at two beaches in Europe that ductile iron resisted attack to about the same extent as gray cast iron.
The compositions and microstructures of ductile and gray iron differ significantly to those of carbon steel, which is commonly used for steel pipes. Steel has a much lower carbon content and all of the carbon is present in a combined form and usually exists as pearlite, whereas in the cast irons a large portion of the carbon exists as graphite flakes or spheroids. The ferrite phase of steel is subject to the same corrosion processes as is ductile iron or gray iron, except that graphitization does not occur. Therefore, if the graphite containing corrosion products of ductile iron or gray iron can act as a diffusion barrier, as discussed above, the steel will be less corrosion resistant than because no similar diffusion barrier is formed. On the other hand, if the graphite containing corrosion products remain permeable to the further penetration of corrosive liquids, the attack on the underlying iron will not only continue but may even proceed at a rate accelerated by the galvanic effect of the graphite-containing corrosion products, which act as an enlarged cathode to supplement the other cathodes within the structure of the underlying iron 2.
It is understandable that corrosion behavior in any given case will depend upon the balance between the pipe material, the service environment, and the various effects of the corrosion products. Hence, it is not possible to make any generalized statements as to the relative corrosion resistance of steel and cast irons. In particular, it is difficult to substantiate claims, as made by Horn 15, that steel has less inherent corrosion resistance than has ductile iron in buried pipeline applications, when the claimed beneficial effects of the graphite corrosion products of ductile iron are under question as above. However, it is generally agreed that the cast irons are greatly superior to carbon steel and copper steel in their resistance to atmospheric corrosion 2.
It is clear that there are many different mechanisms and parameters, as will be discussed later, involved in the corrosion and failure of buried piping materials. This degree of complexity accounts for the controversy surrounding the claimed corrosion performance of ductile iron. However, the main failure mode of ductile iron pipe is not uniform corrosion or graphitization, but pitting attack. The arguments based on protection by graphite containing corrosion products are irrelevant. More research would be necessary before the claimed merits of ductile iron corrosion resistance could be justified.
From the practical point of view, results of research already completed have, at a minimum, indicated beyond doubt that gray cast or ductile iron does not possess superior corrosion resistance to other materials. While reviewing the results of two extensive comparative corrosion tests on different ferrous materials conducted by the U.S. National Bureau of Standards (NBS) 44 and by a British research sub-committee of the Institute of Civil Engineers in collaborating with the Corrosion Committee of the British Iron and Steel Research Association, Ulick R. Evans wrote:
The most conspicuous results of both tests is that - Whilst the rate of penetration varies greatly from one soil to another - there is little difference between the behavior of one ferrous material and another, unless expensive stainless steels are used. Small amounts of chromium, copper or nickel produce no marked effect on corrosion rates. Steel and cast iron seem to corrode at about the same rate; if the thickness were the same, the steel would probably outlive the iron 41.
William A. Pennington, a scientist at the Bureau of Reclamation of the U.S. Department of the Interior who worked with steel, gray cast-iron, and ductile iron pipe, also reviewed the tests of NBS and of the Bureau of Reclamation. He concluded of the three pipe materials, steel and gray and ductile cast irons, steel is the best for a given thickness where buried bare in soil, although gray cast-iron in larger commercial sizes is the only one not requiring an external coating for a service life of 50 years. Ductile cast iron, made according to manufacturers schedule, has about the same service life as steel in severely corrosive soil, even though its pitting rate is greater than either steel or gray cast iron. He also concluded that ductile iron "should have an exterior coating for 50-yr life" based on a 12 percent allowance in pipe thickness, and that if the 12 percent allowance had not been made for ductile iron, the necessity for coating would appear to be all the stronger" 45.
It is also important to note that the wall thickness of ductile iron pipe is as much as 50% thinner than gray cast iron pipe for equivalent nominal diameter and is typically only slightly thicker than or equal to the thickness of steel pipe that would be used for similar service 9. These factors imply that ductile iron pipe should be protected from external corrosion at the same level as would be the case for pipes made of materials such as steel.
3. MECHANISMS OF EXTERNAL CORROSION ON BURIED DUCTILE IRON PIPES
It is generally believed that the characteristics and mechanisms of external corrosion of ductile iron are similar to those of steel. However, ductile iron pipe does not fail in the same way or at the same rate as pipe made of other materials 11, 16.
Corrosion as graphitization is a major factor influencing gray cast iron failure. Early laboratory and field studies tended to treat it the same as for ductile iron 2, 7. Later research and case histories have indicated ductile iron pipe cannot graphitize as originally thought.
Graphitization of gray cast irons can be expected when soil conditions favor anaerobic bacterial growth, with the appropriate conditions of pH, dissolved salts, and organic content. The result is a matrix consisting of a mass of residual graphite flakes interspersed with oxides of iron, which have been referred to in this paper as the graphite-containing corrosion products.
As previously discussed in this paper, uncertainty exists over the contribution of the graphite-containing corrosion products in acting as a protective barrier to continued corrosion, especially in the case of corrosion products formed on ductile iron pipe. Subsequent pipe failures frequently come from a mechanical stress or a hydraulic shock (roadwork, transport damage, or ground movement). The pipe would likely not have failed had it not been weakened by corrosion-induced loss of wall thickness.
2)
Pitting Corrosion
Pitting corrosion is a concentration of corrosion in one particular area whereby the metal goes into solution preferentially at that spot, rather than at other adjacent areas.
Pitting corrosion has been reported to be the primary mode of failure for ductile iron pipes. The National Research Council of Canada has recently presented a report of data collected on water main breaks during 1992 and 1993 from 21 Canadian cities 17. The report shows that ductile iron pipe constitutes 24% of the water distribution network and the break rates for this material in 1992 and 1993 were 9.3 breaks/100 km/year and 9.8 breaks/100 km/year, respectively. Between 76% and 78% of the ductile iron pipes failures were reported to have failed as a result of holes or pits. In contrast, only approximately 20% of gray cast iron failures were reported to have failed due to pits or holes.
A survey of 359 corrosion failures on 118 ductile iron mains in the UK Water Industry was conducted by WRc in 1984 10. The survey results suggested that pitting corrosion was a primary mode of failure and the average maximum pitting corrosion rate for unprotected ductile iron was typically in the range of 0.50 to 1.5 mm/year (20-60 mils/year), with values of up to 4.0 mm/year (160 mils/year) in some instances.
It is generally believed that the rate of external pitting attack on unprotected ferrous materials is governed primarily by the corrosivity of the environment, and the material type has no pronounced influence.
Originally developed and recommended by the Cast Iron Pipe Research Association (the former name of the Ductile Iron Pipe Research Association DIPRA), a 10-point system is commonly used in soil analysis by the ductile iron pipe industry. The 10-point system and its interpretation are incorporated in Appendix A of ASNI/AWWA C105/A21.5, Standard for Polyethylene Encasement for Ductile Iron Pipe Systems 18. This soil evaluation system is applied only to the gray and ductile cast iron industry but not typically used by industries of other materials.
Another commonly used method for soil analysis is shown in Table 1, which uses soil resistivity only to give out rough indications of the soil corrosivity. However, Table 1 should be carefully used to evaluate the nature of corrosivity of an environment. Although it is generally true that the most rapid corrosion takes place in soils of the lowest resistivity and the least rapid corrosion takes place in soils of the highest resistivity, so many other factors affect the corrosion rate that certain soils of low resistivity may be found to be no more corrosive than certain soils of high resistivity.
Table
1
Rough Indications of Soil Corrosivity vs. Resistivity 19
Resistivity (Ohm-cm) |
Soil Corrosivity Description |
Below 500 |
Very corrosive |
500 1,000 |
Corrosive |
1,000 2,000 |
Moderately corrosive |
2,000 10,000 |
Mildly corrosive |
Above 10,000 |
Progressively less corrosive |
In 1956 and 1957, the Cast Iron Pipe Research Association conducted two surveys on soil corrosion in eight cities located in areas representing six of the seven major soil classifications in the U.S. 20 The major soil groups and the locations of the cities studied are shown in Figure 1.
The soil conditions of the eight cities are shown in Table 2.
Figure
1 Major Soil Groups in the United States and Location of
Eight Cities 20
Table
2 Soil Conditions of Eight Typical U.S. Cities Reported by CIPRA
in 1956 and 1957 20
City |
Great soil group |
Resistivity (ohm-cm) |
Type of soil |
pH |
Relative corrosivity |
Des Moines, Iowa |
Prairie | 1,100-1,500 |
Gary clay with sand inclusions, wet |
6.5-6.9 |
Corrosive |
Denver, Colo. |
Dark brown and brown | 3,000 5,000 |
Sandy loam, dry |
6.5-6.9 |
Mildly corrosive |
St. Paul, Minn. |
Podsol and gray-brown podsolic | 500 1,100 |
Clay mixed with organic matter, wet |
6.2 |
Very corrosive |
Detroit, Mich. |
Gray brown podsolic |
350 - 500 |
Dense blue and brown clays, moist |
6.8-6.9 |
Very corrosive |
Trenton, N.J. |
Gray-brown podsolic |
5,000 9,000 |
Sand loam, moist |
6.1-8.1 |
Mildly corrosive |
Amarillo, Tex. |
Chernozem |
2,380 2,500 |
Loam with over lay of white clay, dry |
7.8-7.4 |
Mildly corrosive |
Meridian, Miss. |
Red and yellow |
6,070 24,500 |
Loam and clays, wet |
3.6-5.2 |
Moderately corrosive |
Greenville, S.C. |
Red and yellow |
6,225 17,300 |
Heterogeneous red clay with sand, moist |
6.3-7.2 |
Moderately corrosive |
Soils of lower resistivity are likely to cause more rapid pitting
attack to ductile iron at rates that increase as the resistivity
decreases. The shaded area in Figure 2 gives a typical range of
the average maximum pitting rates of ductile iron pipe vs. lower
soil resistivity. It is compiled from data found by various
surveys or studies done in the U.S., Canada and Europe 1-2,
12, 14, 21, 44-45.
It should be noted that pitting rates tend to decrease with time, and the data compiled in Figure 2 are based on maximum pitting rates actually measured from ductile iron pipes 15 years or younger. These maximum pitting rates for ductile iron pipes are considerably higher than those obtained from gray cast iron pipes of much greater age 25. These data may explain to some extent the apparent discrepancy between the pitting rates of ductile and gray cast iron pipes. Bearing in mind the fact that the wall thickness of ductile iron pipe is as much as 50% less than that of gray cast iron pipe, these results may also explain the manifestation of corrosion failures on ductile iron pipe after only a few years service in corrosive environments. This supports reports by Gummow 9 and De Rosa and Parkinson 10 on more rapid failure of ductile iron pipe, and also failure case histories of ductile iron pipes in Calgary 21, in Bayside, Wisconsin 22, and in Scarborough, Ontario 23. In the latest case, ductile iron constituted 31% of the total piping in the water distribution system of the City of Scarborough, and was first installed in 1965. The first corrosion-related failure on the ductile iron piping occurred in 1972, just seven years after its introduction, and the pitting rate and failure rate continued to increase until the problem was recognized and brought under control. It was found that the diameter of a ductile iron pipe had a significant influence on its in-service failure rate. For example, 6-in. (150-mm) water mains accounted for 50% of the ductile iron system but are responsible for 67% of the failures. The reason for this was observed to be the thinner wall dimension of the 6-in pipe as compared to that for 8-in. (200 mm), 12-in. (300 mm), and 16-in. (400-mm) water mains of the same class.
It should also be noted that pitting corrosion of ductile iron pipe may be influenced by a number of factors, and it can occur on ductile iron in combination with other corrosion mechanisms. This makes the failure analysis of ductile iron pipes and the determination of the above maximum pitting rates a complicated task.
As noted previously, the initiation of localized corrosion of ductile iron pipe may be influenced by a number of factors, not least the presence of surface oxides that have been formed during the manufacturing process 10. The susceptibility of spun pipe (both ductile and gray cast iron) to external corrosion can be increased by damage to the annealing oxide scale, which inevitably occurs during normal handling and installation. In corrosive environments, localized attack also tends to initiate at sites corresponding to the reverse peen marks on the external surface of ductile iron pipe where the annealing oxide scale is generally thinner and more exposed to mechanical damage. Such damage to the electrochemically more noble thermal oxide scale, which exposes the bare metal substrate can lead to the formation of a galvanic cell between the relatively small area of bare iron (anode) and the reminder of oxide-coated pipe, which is more cathodic. The large ratio of cathodic to anodic areas, together with the fixed position of the damaged site, can thus provide the necessary conditions for rapid, localized corrosion attack of the pipe. The above mechanism for residual oxide scale also applies to the case for works-applied bituminous coating on ductile iron pipe 10.
3)
Galvanic Corrosion
Ductile Iron Pipe may also corrode because of galvanic reaction of dissimilar metals with copper services. The service piping used in North America is almost exclusively copper, with a small amount of lead and galvanized steel pipe being used in the older areas. When ductile iron and copper pipes are connected, the mixed metal system may accelerates corrosion where it is connected to iron piping which acts as the anode of a galvanic corrosion cell in which the copper acts as the cathode. While describing corrosion of municipal water mains in Detroit, Michigan, Gummow 9 has suggested that copper has an undeserved good reputation for corrosion resistance in soil environments because in many instances copper receives current produced by corroding iron, thus having a reduced corrosion rate.
For example, it has been reported that in 1980, 6.2% of the 42 water main failures (per 100 km) in Calgary resulted from attack to service saddles that were joined to copper service lines 21. Stetler has reported that about 80% of the 23 failures of the ductile iron mains in Bayside, Wisconsin during 1972 and 1976 occurred within 1 m (3 feet) of a copper service pipe or a copper bond strap, and about 33% were reported to have occurred within 20 cm (8 inches) of a copper item. 22 Nevertheless, in overall terms the number of failures from this cause is relatively low, and the proportion confirmed to have been due solely to a galvanic corrosion mechanism is even few.
4)
Microbiologically Influenced Corrosion
Buried ductile iron pipe can also be subject to microbiologically enhanced attack. Biological organisms fall under two groups based on the type of corrosion they engender: (a) anaerobic corrosion and (b) aerobic corrosion. Sulfate reducing bacteria (SRB) from the genera desulforvibrio are a typical example of anaerobic bacteria. If sulfides are found in the corrosion products, the presence of sulfate-reducing bacteria is possible, or even probable. The sulfide test is qualitative and is accomplished by introducing a solution of 3-percent sodium azide in 0.1N iodine into a test tube containing a small quantity of soil removed from pipe depth, preferably from the surface of the pipe. If sulfides are present, they catalyze a reaction between sodium azide and iodine with the release of nitrogen 24, and this is indicative of the presence of SRB. Electrochemical techniques such as Zero Resistance Ammetry, Electrochemical Impedance Measurement and corrosion potential logging have been used to study the nature and mechanisms of microbiological influenced corrosion of ductile iron pipes in soils.
Evans ascribed biological attack as resulting from anaerobes ability to render the oxygen present in sulfates, nitrides, and carbonates available for the acceleration of the cathodic reaction 26. This means that corrosion can proceed even in the absence of dissolved oxygen. He also indicated that the corrosion rate of iron under anaerobic conditions is as much as 19.5 times greater than that under sterile conditions. Microbacterial action can also promote local anodic attack.
King and his associates 5 tested blast-cleaned specimens of cast irons and concluded that both ductile iron and gray cast iron pipe can suffer extensive corrosion by sulfate reducing bacteria. Both ductile and gray iron pipes were observed to corrode at similarly high corrosion rates (50-60 mils/year) in SRB containing soils. They also reported that specimens that did not have the asphaltic coating and annealing oxide removed did not corrode at a high rate. However, during their investigation at WRc, De Rosa and Parkinson 10 found that the deepest corrosion pit was produced on mill scale coated (non-blasted) ductile iron pipe after 250 days in SRB soil test exposure. The pit depth was approximately 0.7 mm (28 mils), equivalent to a rate of corrosion of 1mm/year (40 mils/year). This rate is similar to that reported for premature external corrosion failures on unprotected ductile iron mains in the U.K. under service exposure conditions in aggressive soils. According to a study in Australia, it was also estimated that 50% of all failures of buried metal were due to microbiological causes 14.
5)
Corrosion Due to Dissimilar Electrolytes
Corrosion cells develop on a piece of metal exposed to different electrolytes and it is a particularly common problem on underground structures 27. Potential differences develop, for example, on a long continuous pipeline that passes through different types of soils. One portion of the line might be laid in sandy loam while another lie in clay. Substantial natural pipeline currents (long-line currents) may occur, which leads to corrosion cells as called long line cells. In soils of low resistivity where such currents exit from the pipeline, causing the metal at the exit points is lost by anodic dissolution (corrosion). Anodes and cathodes may be miles apart.
Similarly, mixtures of soils in the backfill will cause corrosion. In the ground there are usually areas of varying nature which might form geological batteries, i.e. galvanic cells. Clods of clay, for example, mixed into a sand backfill will lead to sever corrosion where the clay contacts the pipe. The same phenomenon causes corrosion on pipe exposed to soil and concrete or other highly alkaline backfill. The problem often develops where a pipe passes through a concrete wall or floor. The resultant cells lead to corrosion of the portion of pipe exposed to the soil.
Corrosion cells may be developed as a result of different electrical (magnetic) ground currents the changes in the ground currents due to the earths magnetic field or caused by lighting discharges.
6)
Stray Current Corrosion
Stray current corrosion is caused by current flow through paths other than the intended circuit or by any extraneous current in the earth. Metal structures buried in the ground, like pipelines, can often provide a better conducting path than the soil for earth-return currents from electric rail and tramway systems, electrical installations, and cathodic protection systems on nearby pipes. Sometimes such routes exhibit higher conductivity than a sheathed earthling cable. Accelerated corrosion of the pipeline may occur at the point where the positive current flow leaves the pipe and enters into the earth.
The term stray current corrosion differs from other forms of corrosion in that the current, which causes the corrosion, has a source that is external to the affected structure. It may include the following different types of currents on buried or submerged metallic structures:
· Stray currents from direct current (DC) systems such as railways, trolley bus systems, cathodic or anodic corrosion protection systems, welding equipment in shipyards, and household appliances, etc.
· Interference currents such as HVDC (high-voltage direct current) power lines with full or partial ground return.
· Stray currents from alternating current (AC) systems such as AC currents from certain household appliances.
Modern ductile iron pipe are manufactured in 18- and 20-ft (5.5- and 6.1 m) nominal lengths, and a rubber-gasketed jointing system may be employed to join successive lengths into a continuous pipeline 24. Joints gasketed in this matter offer resistance that may vary from a fraction of an ohm to several ohms but, nevertheless, of sufficient magnitude that ductile iron pipelines are considered to be electrically discontinuous (and are therefore unsuitable for cathodic protection without substantial modification). The rubber-gasket joints limit attack of ductile iron by long-line stray currents, but not necessarily by local currents 1.
The question of whether underground pipes are attacked to a significant degree by stray AC currents is controversial and Horton 28 has reviewed this subject. Stray AC current densities in excess of 1 mA/cm2 were reported to be required for significant AC corrosion. It was reported that stray AC current could initiate and/or accelerate corrosion of unprotected metals by exaggerating the potential of existing anodes and cathodes on the surface, and/or by depolarization of existing bimetallic or galvanic cells. There was also concern that stray currents could be introduced in a cast iron pipeline if it ran parallel to high-voltage cables, the alternating current apparently being partly rectified by residual oxide films on the pipe 1.
4. EXTERNAL CORROSION PROTECTION OF
DUCTILE IRON PIPE
The impact of deterioration and failure of buried ductile iron pipelines due to external corrosion is very significant:
· It has been reported that, in 1992 and 1993, the total number of water main breaks in 21 Canadian cities was 3601 and 3773, respectively 17. Assuming an estimated cost for repair/break of $2500, the average annual total cost of repairs would be $9.2 million. The total number of breaks for 1992 and 1993 occurred in 21 cities representing 11% (3.14 million) of Canadas population. Extrapolating on a population basis, the average annual total national cost for water main repair is $82 million. Ductile iron pipe constitutes 24% of the water distribution network in the 21 cities. Breaks in ductile iron pipe accounted for 10.89% of the total number of water main breaks in 1992 and 10.97% in 1993. Ductile iron pipe for water main use was first manufactured in Canada around 1961 though significant use of ductile iron pipes probably began between 1963 and 1968.
· Ductile iron pipe was introduced in the UK in the mid 1960s and by the mid 1970s had displaced the use of gray cast iron pipe. Water companies were left with hundreds of poorly protected ductile iron mains installed between 1965 and 1984, facing more than 359 premature failures and the need for early replacement. Meanwhile increasing problems from leakage, repairs and interruptions to customer supplies also occurred 10, 29. As much as 40% of the water entering the UK distribution system was lost by leakage before it got to the end user 13. Similar rates of water loss were reported more recently in Mexico 13.
· It was reported that as much as 190 million liters (50 million gallons) a day were leaking from the 10,000 km (6,200 miles) of water pipes under New York City. It was also reported that 114 million liters (30 million gallons) of water per day were leaking at a daily leaking rate of about 220 from the City of Houstons pipes in 1985. A quantity report in the June 1985 issue of Forbes magazine outlined the extensive impact of replacing 12% of the 1,324,000 km (823,000 miles) of sewer pipe in the United States 1.
It should be noted, however, that much of the pipe failures mentioned above are not only on ductile or gray cast iron but also on other pipe materials, and that not all these failures resulted from corrosion. On the ductile and gray iron pipe failures which were experienced, installation of unprotected pipe in corrosive soils, improper application of protection measures, improper installation of the pipe, stray current from grounding services, bi-metallic corrosion from stainless saddles, and other documented problems were all the add-ons to corrosion.
Although perhaps more relevant to internal corrosion than external corrosion, results of recent research have associated the corrosion of cast iron pipes with human behavioral disturbances, including loss of impulse control and outbursts of violent behavior under stress 30. Neuroscientists believe that manganese, coming out from water supplies due to the aging or rusting of cast iron pipes and conduits, may inflict its damage by lowering the amount of serotonin in the brain. Low levels of serotonin in the brain are associated with mood disturbances, poor impulse control and increases in aggressive behavior. The contribution of manganese from corroded cast iron pipes to the chemistry of violence is perhaps valid because typical ductile and gray cast iron pipes normally contain 0.4 - 0.6% manganese. Details of the discussion of this subject can be seen from a 1998 report on the chemistry of violence in Popular Mechanics magazine 30.
Uncertainty does exist as to whether or not corrosion protection is required to protect ductile or gray cast iron pipes. DIPRA (Ductile Iron Pipe Research Association) wrote in its Fall/Winter 1997 semi-annual publication Ductile Iron Pipe News that 31: The majority of soils found in North America are not considered corrosive to Ductile or Cast Iron. Therefore, in these soils, corrosion protection of any nature is not required. DIPRA has suggested that those soils, described early by themselves in their 1956 and 1957 surveys as corrosive in Table 2, are not considered as corrosive anymore when analyzed in accordance with the 10-point soil evaluation procedure. In addition, despite the fact that gray cast iron and ductile iron have possessed significantly different modes of failure mechanisms as well as different wall thickness, case histories of long service life associated with some gray cast iron pipes are often mixed with case histories of ductile iron pipes which have been in service of much shorter times.
Nevertheless, the following methods are commonly used for corrosion control of underground ductile iron pipes:
1)
Polyethylene Encasement
Although not normally used for the protection of other pipe materials, loose polyethylene jacket (encasement) is a standard corrosion control method specifically and favorably recommended by DIPRA for the protection of ductile iron pipes in soils at landfill sites and in similar corrosive environments. Since its first use in operating water systems in Lafourche Parish, Louisiana, and Philadelphia by the late 1950s, polyethylene encasement has been taken in various standards in the U.S. and in other countries such as Japan, UK, Germany, Australia, and an international ISO standard. Following this control method, the ductile iron pipe is encased with either loose 8-mil (200 microns) low-density polyethylene or loose 4-mil (100 microns) high-density cross-laminated polyethylene.
There are strong disagreements about the benefits of polyethylene encasement. Advantages inferred by DIPRA on polyethylene encasement include that it is: relatively inexpensive, easy to install, does not require maintenance or monitoring, and is easy to repair if damaged. A 1972 paper, Corrosion Prevention with Loose Polyethylene Encasement by Smith 32 claimed: After almost 20 years of experience, including research and application in the field, there has been no failure of pipe so protected. Most case history reports by DIPRA indicated minimum attack to ductile iron pipes installed in the United States because of polyethylene encasement 33-35. Most of these reports, however, mixed case histories for ductile iron pipes with those for gray cast iron pipes. The case histories for ductile iron pipes mostly covered time periods of 6 to 21 years only in soils with resistivities of 310 to 4,000 ohm-cm.
However, there are indeed reports that describe failures of polyethylene wrapped pipes ever since the first kind of tests done by the Cast Iron Pipe Research Association (CIPRA) in the late 1950s and early 1960s. During these first tests, bolts made of 0.5% copper-content cast iron from polyethylene-wrapped joints that were buried by CIPRA in the Atlantic City tidal marsh were reported to have lost an average of 28.9 to 33.1 grams per year. This indicated a significant corrosion rate that might have resulted from the tidal action which was reported to force water to migrate into the void between the polyethylene and the pipe 43. A study conducted for Calgary in 1975 also concluded that loose polyethylene was not protective and wrapped pipes and fitting could be severely corroded 36. Results of a study by Vrable on buried pressurized steel drums also indicated that polyethylene encasement was not a reliable corrosion barrier 37.
A further objection to the use of polyethylene encasement is that the polyethylene films restrict the subsequent use of cathodic protection 1. Another is that the encasement is easily damaged, resulting in holidays, rips, or tears. These defects are easily created during handling, and during pipe laying and backfilling operations, in an industry where a high standard of workmanship was not norm. Such defects may admit environmental water into the interface between the films and pipe surface, leading accelerated attack to the pipe in the vicinity of these defects. Polyethylene encasement is not recommended by DIPRA as the sole protection method where high-density stray currents may be present 35. It was also suggested that polyethylene wrapping might not provide enough protection in continuously-saturated soils, although it might be used in conjunction with cathodic protection systems 43. Polyethylene also exhibits significant softening at temperatures over 82oC(180oF) and will melt around 104 to 110oC (220 to 230oF). 31
It is therefore very critical to evaluate the soil and application conditions before recommending and using the polyethylene encasement method for corrosion protection for ductile iron pipe. It is improper to make such a general statement that this method is the most effective corrosion protective method for ductile iron, without talking about the actual and detailed soil and application conditions in a specific pipeline project.
2)
Cathodic Protection
For the past 25 years, the U.S. Department of Transportation has required all new underground metallic piping which typically is steel conveying petroleum and natural gas to be cathodically protected as a secure measure to reduce the risk of catastrophic corrosion-related failures. Cathodic protection can be adapted for ductile iron water and wastewater pipe applications to slow the overall corrosion rate or in fact to prevent corrosion entirely. Two cathodic protection methods can be used, namely, sacrificial anode and impressed current. The sacrificial anode system is the more appropriate one for corrosion prevention on cast or ductile iron piping systems because the pipes seldom have electrical continuity and are therefore susceptible to stray current corrosion if an impressed current system were to be installed.
The sacrificial anode method utilizes galvanic anodes of zinc or magnesium that are packed in a low resistivity backfill, the chemical composition of which is selected such that anode polarization will be inhibited. Zinc has a lower corrosion potential and a lower protective current output so that it is practical use zinc anodes only in low resistivity soils, or where only a small cathodic protection current is required. Magnesium anodes have a larger protection current output and hence they are applicable over a wider range of soil resistivities, and can be used to protect larger pipe sizes. The basic criterion for adequate cathodic protection of water mains is generally taken as the application of a protection current from the anodes equivalent to 10 mA per square meter of water main pipe surface 23.
Cathodic protection can be a powerful method of preventing leakage from existing ductile iron pipes, and it may be less expensive than the use of other combinations of corrosion control. However, cathodic protection does not work well on extensively corroded gray cast iron, even assuming that there is adequate electrical conductivity across the joints, which is frequently not the case. Corroded gray cast iron is highly sensitive to physical disturbance and therefore will continue to fracture and leak in proportion to the mechanical force applied, regardless as to whether corrosion has been halted. It can be cost effective for new ductile iron water main systems to be cathodically protected and, where appropriate, the pipes can be coated and electrically isolated from stray current effects 9.
Case histories have proven that cathodic protection is an effective corrosion method for ductile iron pipe 21-23, 38. The first failures of ductile iron pipe in the City of Scarborough occurred after the pipe system had been only in service for seven years, and the rate of failure continued to increase until a cathodic protection program was implemented 23. Cathodic protection was also used to curtail further leakage in ductile iron water mains in Bayside, Wisconsin 22. Sacrificial anode cathodic protection is mandatory for all metallic pipes in current procurement specifications in Calgary 21.
The main objections to use cathodic protection for new ductile iron water pipelines are the additional capital cost and the need for continual monitoring. Craft39 has recently made a cost comparison between cathodic protection and polyethylene encasement method of corrosion prevention for this type of application. He assumed that polyethylene encasement could be sufficiently undamaged after installation that it would provide effective corrosion prevention as alternatively would cathodic protection, for a 100 year service life of 1 mile of a 30 in. ductile iron pipe. The initial acquisition and installation cost of the cathodic protection system was 18 times the cost of purchasing and installing loose-film polyethylene encasement. He estimated that the operating costs of a cathodic protection system would be 370 times as much as the polyethylene encasement and 6 times the initial purchasing cost of the ductile iron pipe.
3)
Bonded Coatings
Bonded coatings control corrosion by creating a physical barrier that isolates the ductile iron pipe from the surrounding corrosive soil environment. Over the years, significant advances have been made in coatings technology specifically pertaining to ductile iron pipe.
Cold-applied tape coatings have been used for steel pipes but are strongly discouraged by DIPRA for protecting ductile iron pipe from corrosion because:
a) Tape coatings have organic nutrients inherent in the adhesive/primer used to bond the tape to the pipe. These nutrients combined with cathodic hydrogen from the pipe surface, when and if the pipe is also cathodically protected, enabling sulfate-reducing bacteria to generate metabolic energy and reproduce rapidly;
b) It is difficult to apply tape coatings to ductile iron joints and fittings;
c) Tape coatings frequently are damaged during shipping, handling, storage, and installation;
d) Tape coatings area expensive to install and to maintain.
Some arguments have been made to show that the quality of new tape coatings may be improved for ductile iron pipe through careful engineering design. Noonan and Bradish 40 have recently reviewed the critical performance issues in adapting the use of tape coatings and cathodic protection on non-steel water pipelines, and highlighted several important engineering design specification issues.
Various experimental studies and case histories, however, have indicated that the thin (about 50-70 microns or 2-3 mils) sprayed zinc/bitumastic coating approach offers at best only a marginal enhancement of short-term corrosion protection for ductile iron pipes. Corrosion pits on ductile iron tend to initiate at sites corresponding to the positions of the reverse peen marks (casting pips) on the external surface of the pipe. Because of the surface profile, works-applied bitumastic coating is thinner on the tops of the pips and corrosion initiates quite quickly at these poorly-protected locations. Pits also form at sites of mechanical damage on the external surface of the pipe 10. This type of damage is unavoidable for the thin sprayed-zinc/bitumastic coating, especially when scratching can remove it almost completely at these locations during handling and installation. Over the period 1956 to 1980, 43% of the total 1821 failures in Calgary on ductile iron pipe occurred where a coating was present 21, and most of the coatings involved in these failures were of thin zinc/bitumastic type, both with and without polyethylene encasement.
100% solids polyurethane coatings More recently there has been a movement towards the development of proper high-performance coatings for protection of ductile iron pipes and carbon steel pipes. The specially designed plural component and zero VOC polyurethane (PU) coatings have been demonstrated to be by far the most successful protective coating systems used for both exterior and interior application. These polyurethane coatings have excellent adhesion to ductile iron or steel, combined with excellent chemical resistance, impact resistance, resistance to cathodic disbondment, and abrasion resistance. The fast setting (drying) nature of the 100% solids polyurethane coatings also means they are very suitable for rapid application lines during pipe production, and they can be applied at normal ambient temperatures. The success of the 100% solids polyurethane coatings on ductile iron and steel pipes has been confirmed in water distribution systems already service in North America, the Middle East, several European countries (in particular, in France 41), and very recently, in a number of Asian countries. The polyurethane coatings also have successfully been used in conjunction with cathodic protection systems for protecting ductile iron and steel pipes, where the function of the coatings is both to improve degree of protection and also to reduce the capacity of the cathodic protection installation required to achieve complete immunity from corrosion attack.
In June of 1993 a San Diego corrosion engineering firm tested the corrosion protection system installed in 1991 on the 12", 6 mile, ductile iron pipeline known as Fiesta Island Replacement Project (Phase 1) in San Diego, California. The system uses 25 mils (625 microns) of a 100% solids polyurethane coating in combination with sacrificial magnesium anode cathodic protection system. The consultants analysis showed that the coating system had an installed efficiency of 99.66% and the pipe had an actual current requirement (for corrosion protection) three times less than the design value.
Other coatings such as cold applied laminated tape and epoxy coating, both fusion bonded and 100% solids thermosetting liquid epoxy, are used mainly for protection of the valves, tee-pieces, and other fittings for ductile iron pipes.
4) Other Corrosion Control Methods
Insulated joints are used to break the metallic electrical connection between two pipelines, pipeline components or structures, thereby preventing the flow of electrical current between them. This method has limited applications in ductile iron pipes that use a rubber-gasket jointing system and are considered to be electrically discontinuous. Insulated joints often are used however, to isolate ductile iron pipelines from other underground structures that are cathodically protected or are part of an electrical grounding system. They may also be used at the junction of ductile iron pipe and copper service pipe to prevent galvanic current flow.
Trench improvement of corrosive soils to reduce their corrosive tendencies may provide only short-term protection to ductile iron pipeline. It has been observed, in many instances, that the substitute less corrosive backfilling material eventually takes on the characteristics of the surrounding soil 24.
The use of sacrificial metal (i.e. a corrosion allowance) by increasing wall thickness of ductile iron pipe to allow metal loss is neither cost-effective nor reliable, particularly when pitting is the primary failure mode for ductile iron pipe.
5. SUMMARY
The results of both experimental researches and practical case histories have concluded that ductile iron does not possess superior corrosion resistance to other pipe materials such as gray cast iron or carbon steel when exposed in corrosive underground service environments. This, coupled with its thinner wall thickness compared with that of gray cast iron, implies that ductile iron should be protected from external corrosion to the same degree as for pipes made of steel.
The primary failure mode for buried ductile iron pipe is localized attack (pitting corrosion). Pitting rates obtained from ductile iron pipes of age less than 15 years are much higher than those obtained from gray cast iron of much older age. Pitting corrosion of ductile iron pipe may be exacerbated by a number of things such as residual oxide scale, casting marks, and poor quality zinc/bitumastic coatings.
External corrosion of ductile iron pipe not only makes a significant impact on the loss of water resources but also the cost for repair or replacement.
Loose polyethylene jacket (encasement) is a standard corrosion control method specifically and favorably recommended by DIPRA for the protection of ductile iron pipes in soils at landfill sites and in similar corrosive environments. However, it is improper to make such a general statement that this method is the most effective corrosion protective method for ductile iron, without talking about the actual and detailed soil and application conditions in a specific pipeline project.
In recent water industry installations, carbon steel and ductile iron pipelines coated with 100% solids polyurethane systems have exhibited excellent corrosion prevention performance. When used in combination with a well-designed cathodic protection system, such coatings provide the most cost-effective corrosion control measure for the protection of ductile iron pipes in corrosive soils. Cathodic protection also may be the most powerful and cost-effective control method for existing ductile iron at locations where it is at risk from external corrosion.
References
1. NACE Task Group T-10A-21, "Corrosion Control of Ductile and Cast Iron Pipe", NACE publication 10A292, Item No. 54293, NACE, 1992
2. F.L. LaQue, "The
Corrosion Resistance of Ductile Iron", in Corrosion of
Ductile Corrosion of Ductile Iron Piping, Edited by
Michael J. Szeliga, NACE International, 1995
3. A.G. Fuller, "Corrosion resistance of Ductile Iron Pipe", BCIRA Report 1442, 1981
4. Cast Iron Pipe Research Association, "Soil Corrosion Test Report: Ductile Iron Pipe", 1964
5. R.A. King, et al, "Corrosion Behavior of Ductile and Gray Iron Pipes in Environments Containing Sulphate-Reducing Bacteria," NACE-8, Biologically Induced Corrosion (Houston, TX: NACE, 1986), p83
6. E.C. Sears, "Comparison of the Soil Corrosion Resistance of Ductile Iron Pipe and Gray Cast Iron", Materials Protection, Vol. 7, No. 10, 1964, p33
7. M. Romanoff, "Performance of Ductile Iron Pipe in Soils", J. AWWA, Vol. 60, No. 6, 1968, p645
8. W.H. Smith, "A Report on Corrosion Resistance of Cast Iron and Ductile Iron Pipe", Cast Iron Pipe News, Vol. 35, No. 3, 1968, p16
9. R.A. Gummow, "The Corrosion of Municipal Iron Watermains", Materials Performance, Vol. 23, No. 3, 1984, p39
10. P.J. De Rose and R.W. Parkinson, "Corrosion and Protection of Ductile Iron Pipe", Proceeding of 6th International Conference on the Internal and External Protection of Pipes, Nice, France, November, 1985, p249
11. J.H. Fitzgerald III, "Corrosion of Various Type of Pipe", PSG Corrosion Engineering, Inc., Detroit, Michigan, 1984
12. P. Nicholson, "Corrosion of Municipal Water Systems", NACE Corrosion'91 Conference, Houston, TX, NACE, 1991, Paper No. 592
13. W.M. Cox, Corrosion Management Ltd., UK, Private Communications, 1998
14. P. Ferguson and D. Nicholas, Corrosion Australasia, April, 1984, p12
15. L.G. Horn, Product Advisory: Tape Coat, Ductile Iron Pipe Research Association, 1993
16. J.H. Fitzgerald III, PSG Corrosion Engineering, Inc., Detroit, Michigan, Private communications, 1998
17. National Research Council Canada, Water Mains Breaks Data on Different Pipe Materials for 1992 and 1993, 1995
18. A.W. Peabody, Principles of Cathodic Protection (Houston, TX, National Association of Corrosion Engineers, 1970), p5
19. American National Standard for Polyethylene Encasement for Ductile Iron Pipe Systems 1993 ANSI/AWWA C105/A21.5, Denver, Colo.: American Water Works Association.
20. H.L. Hamilton, Effects of Soil Corrosion on Cast-Iron Pipe, J. AWWA, May 1960, p639
21. Underground Corrosion of Water Pipes in Canadian Cities, Case: The City of Calgary, Caproco Corrosion Prevention Ltd., CANMET Contract Report No. 0SQ81-0096, Catalogue No. M38-17/2-1-1985E, May, 1985
22. F.E. Stetler, Accelerating Leak Rate in Ductile Cast Iron Water Mains Yields to Cathodic Protection, Materials Performance, Vol. 19, No. 10, 1980, p.15
23. B.J. Doherty, Controlling Ductile-Iron Water Main Corrosion, NACE Corrosion/89, New Orleans, Louisiana, 1989, Paper no. 588.
24. AWWA Manual M41, Ductile-Iron Pipe and Fittings, American Water Works Association, 1996, p173
25. L.L. Shreir, Corrosion, John Wiley & Sons, Inc., New York, 1963, p3.93
26. U.R. Evans, Metallic Corrosion, Passivity and Protection, New York, NY: Longmans, Green and Co., 1948, p301
27. J.H. Fitzgerald, Fundamentals of Corrosion, Basic Text for the Twenty Ninth Annual Appalachian Underground Corrosion Short Course, West Virginia University, College of Minerals and Resources, 1984
28. A.M. Horton, Corrosion Effects of Electrical Grounding on Water Pipe, NACE Corrosion/91, Cincinnati, Ohio, March, 1991, Paper No. 519
29. R.F. Yalden, In Situ Cathodic Protection of Ductile Iron Pipeline, Proceedings of the 11th International Conference on Pipeline Protection (Florence, Italy, October 9-11, 1995), Published by Mechanical Engineering Publications Ltd., Suffolk, UK, 1995, p197
30. J. Wilson, The Chemistry of Violence, Popular Mechanics, April 1998, p42.
31. Ductile Iron Pipe Research Association, Ductile Iron Solutions Ductile Iron Pipe News, Fall/Winter, 1997
32. W.H. Smith, Corrosion Prevention With Loose Polyethylene Encasement, Water & Sewage Works, May, 1997
33. A.M. Horton, Protecting Pipe With Polyethylene Encasement, 1951-1988, Waterworld News, May/June, 1988, p26
34. T.F. Stroud, Corrosion Control Measures For Ductile Iron Pipe, NACE Corrosion/1989, New Orleans, Louisiana, April, 1989, Paper No. 585
35. DIPRA, Polyethylene Encasement, 1997
36. D.E. Hawn, J.R. Davis, Special Corrosion Investigation, Report for the City of Calgary, Water Transmission and Distribution System, Caproco Corrosion Prevention Ltd., Edmonton, Alberta, 1975.
37. J.B. Vrabs, Materials Protection and Performance, Vol. 11, No.3, 1972, p26
38. B.M. Green, P.J. De Rosa, Retrocat and Retrovac: In-situ Cathodic Protection of Existing Ductile Iron Pipe, Proceedings of the 19th International Water Supply Conference, Budapest, 1994
39. G. Craft, Corrosion Protection A Cost Comparison, U.S. Piper, 65, (2), Fall-Winter, 1995-1996, p14
40. J.R. Noonan and B.M. Bradish, New Bonded Tape Coating Systems and Cathodic protection Applied to Non-steel Water Pipelines: Quality Through Proper Design Specifications, Proceedings of the 2nd International Conference on Underground Pipeline Engineering, Bellevue, Washington, 1995, p765
41. J. Mailliard, "Polyurethane Resin Base External Coating for the Protection of Buried Ductile Iron Mains", Proceedings of the 6th International Conference on the Internal and External protection of Pipes, Nice, France, November 1985, Paper F1
42. Ulick R. Evans, The Corrosion and Oxidation of Metals: Scientific Principles and Practical Applications, Edward Arnold, London, 1960, p272
43. Ian Lisk, The Use of Coatings and Polyethylene for Corrosion Protection, Water Online, 1997
44. M. Romanoff, Underground Corrosion, National Bureau of Standards Circular No. 579, U.S. Government Printing Office, Washington, D.C., 1957
45. William A. Pennington, Corrosion of Steel and Two Types of Cast iron in Soil, in Highway Research Record No. 140 Corrosion and Protection of Metals, Division of Engineering, National Research Council National Academy of Sciences National Academy of Engineering, Washington, D.C. 1966.