irt No. FHWA/RD-80/059

no.hHWA-wu-ao-uby ALUATION OF HIGHWAY CULVERT

COATING PERFORMANCE

June 1980

Final Report

Document is available to the public through

the National Technical Information Service,

Springfield, Virginia 22161

Prepared for

FEDERAL HIGHWAY ADMINISTRATION

Offices of Research & Development

Structures & Applied Mechanics Di\BsWiED

Washington, D.C. 20590 MO ^ ^RoFHM

FOREWORD

This report presents the results of an evaluation study and recommendsimprovements in current practices and procedures on protective coatingsfor highway culverts. This evaluation consisted of conducting fieldinvestigations in nine States and a comprehensive literature review on thefollowing types of coatings: asphalt, asbestos bonded asphalt, asphaltpaving, coal tar laminate, polyethylene, PVC, vinyl plasticsol, aluminized,aluminum-zinc, and epoxy coated concrete.

Additional coatings and methods of protection are suggested which may be

superior to existing coatings, but have not been sufficiently evaluated to

justify additional cost. An evaluation program is presented for potentialimprovement in existing culvert specifications and recommendations. Thisreport will be of primary interest to supervisors and engineers responsiblefor the design, construction, and maintenance of culverts and pipes usedin highway and road drainage systems.

This report is being distributed under FHWA Bulletin with sufficient copiesof the report to provide a minimum of three copies to each regional office,three copies to each division, and five copies to each State highwaydepartment. Direct distribution is being made to the division offices.

( *£#.*'

Charles F. Sch^feyDirector, Office of ResearchFederal Highway Administration

NOTICE

This document is disseminated under the sponsorship of the Department ofTransportation in the interest of information exchange. The United StatesGovernment assumes no liability for its contents or use thereof.

The contents of this report reflect the views of the author, who is

responsible for the facts and the accuracy of the data presented herein.The contents do not necessarily reflect the official views or policy ofthe Department of Transportation.

This report does not constitute a standard, specification, or regulation.

The United States Government does not endorse products or manufacturers.Trademarks or manufacturers' names appear herein only because they areconsidered essential to the object of this document.

Technical Report Documentation Page

I. Report No.

FHWA/RD-80/059

2. Government Accession No. 3. Recipient's Catalog No.

4. Title and Subtitle

y5. Report Dare

June 19 80EVALUATION OF HIGHWAY CULVERT COATING PERFORMANCE 6. Performing Organization Code

8. Performing Organization Report No.

7. Author's)

W.T. Young

9. Performing Organization Name and Address

PSC Professional Services Group, Inc.

110 Essex AvenueNarberth, PA 19072

10. Work Un.t No. (TRAIS)

3513-224

FEB, -. .

1J1. Contract or Grant No.

DOT-FH-11-9425

12. Sponsoring Agency Name and Address

Offices of Research and DevelopmentFederal Highway AdministrationU.S Department of TransportationWashington, D.C. 20590

13. Type of Report and Period Covered

Final ReportJune 1978 to November 1979

14. Sponsoring Agency Code

15. Supplementary Notes

FHWA Contract Manager: G. W. Ring (HRS-14)

The objective of this study was to evaluate the performance of

coatings for highway culverts and to recommend methods of improving culvert

service life through improved coating systems and other means. A literature

search was conducted through a review of technical literature and interviews

with cognizant state officials throughout the United States. The literature

search revealed durability problems with the most commonly used materials and

coatings. Corrosion and erosion of interior surfaces appear to be the most

serious durability problems in most states. Field investigations conducted in

nine states confirmed the findings of the literature search. Most coatings

exhibit durability problems under conditions of abrasive bed load and extreme

corrosivity. Coating types examined at various test sites and actual drainage

installations include: asphalt, asbestos bonded asphalt, asphalt paving, coal

tar laminate, polyethylene, PVC, vinyl plastisol, aluminized, aluminum - zinc

and epoxy coated concrete. The best overall coating appears to be asphalt coated

and paved asbestos bonded galvanized steel, although significant deterioration

is probable with this coating under certain conditions. Organic coatings applied

to local areas where needed might be more cost effective than present complete

coating methods. Other coatings and methods of protection are suggested which

might be superior to existing coatings. Many of these alternative methods are

more costly than existing coatings and their benefits are not defined yet.

Durability evaluations are needed to deterimine if their use is justified. A

review of existing State culvert specifications and recommendations for improvement

r^ j nolnrlprl . An evaluation program is presented.-axu pre17. Key Words

Culverts CorrosionDrainage Pipe ErosionMaterialsCoating

18. Distribution Statement

No restrictions. This document is avail-able to the public through the NationalTechnical Information Service, Springfield,Virginia 22161

19. Security Classif. (of this report)

Unclassified

20. Security Classif. (of this page)

Unclassified

21- No. of Pages

157

22. Price

Form DOT F 1700.7 (8-72) Reproduction of completed page authorized

OBJECTIVE

The objective of this study was to evaluate theperformance of coatings for highway culverts and to recom-mend methods of improving culvert service life throughimproved coating systems and other means. A literaturesearch was con-ducted through a review of technical litera-ture and interviews with cognizant state officials through-out the United States. The literature search, revealeddurability problems with the most commonly used materialsand coatings. Corrosion and erosion of interior surfacesappear to be the most serious durability problems in moststates. Field investigations conducted in nine statesconfirmed the findings of the literature search. Mostcoatings exhibit durability problems under conditions ofabrasive bed load and extreme corrosivity. Coating typesexamined at various test sites and actual drainage installa-tions include: asphalt, asbestos bonded asphalt, asphaltpaving, coal tar laminate, polyethylene, PVC, vinyl plasti-sol, aluminized, aluminum - zinc and epoxy coated concrete.The best overall coating appears to be asphalt coated andpaved asbestos bonded galvanized steel, although significantdeterioration is probable with this coating under certainconditions. Other coatings and methods of protection aresuggested which might be superior to existing coatings.Many of these alternative methods are more costly thanexisting coatings and their benefits are not defined yet.Durability evaluations are needed to determine if their useis justified. A review of existing state culvert specifica-tions and recommendations for improvement are included. Anevaluation program is presented.

PREFACE

This study is part of a continuing effort toimprove the performance of highway culverts, and in doing so,lower the cost of highway maintenance.

We wish to thank all of those individuals in thevarious State highway departments for their cooperation in

giving the experiences of their States with regard toculvert performance and for supplying specifications andtechnical data. The cooperation and assistance of thoseStates in which we performed the field inspections is mostappreciated. We also thank those coating fabricators andcoating suppliers that we contacted for their cooperation.

11

TABLE OF CONTENTS

PAGE

SUMMARY 1

INTRODUCTION 3

DISCUSSION 5

Literature Review 5

Field Investigation 7

Procedures 7

Results, General 10Results, Individual Coating Types 14

Asphalt 14Asphalt Coated Aluminum 17Asbestos Bonded Asphalt Coated 17Coal Tar Laminate 18Polyethylene 20Polyvinyl Chloride 20Epoxy Coated Concrete 21Others 21Coating Test Sites 23

Asphalt Coating Performance 24Coating Need and Alternative Means ofProtection 29Alternative Coatings 32Galvanizing 35Specification Review 37

CONCLUSIONS 39

RECOMMENDATIONS 41

FIGURES 1-42 42

TABLES I-XI 66

APPENDIX A - Corrosion and Erosion Effects onCulverts 103

APPENDIX B - Literature Review Summary 106

APPENDIX C - References 135

iii

PAGE

APPENDIX D - Corrugated Steel Pipe Fabricators 143Interviewed and Visited, AsphaltRefiners Interviewed

APPENDIX E - Suggested Culvert Coating Specification 145

APPENDIX F - Recommended Test Program 150

IV

List of Tables

Table Title Page

I Literature Review Results 66

II Summary of Culverts Inspected 67

III Culvert Test Sites - Description and 68Location

IV Summary of Field Test Data 79

V Typical Culvert Asphalt Specification Limits 92

VI Variation in Asphalt Composition WithinAASHTO Specification Limits 93

VII Criteria for Using Metallic Coated PipeWithout Supplemental Organic Coating 94

VIII Typical Material Costs for Flame SprayCoatings for Abrasion Resistance 96

IX Alternative Coating Costs Compared toExisting Coating Systems 97

X Primer Adhesion to Galvanized Culvert Sheet 99

XI Asphalt Adhesion to Galvanized Culvert Sheet 101

XII Summary of State Specifications 102

SUMMARY

One of the primary methods of extending highwayculvert life is to apply a protective coating. The purposeof this study is to evaluate the performance of the coatingsused and to determine if better performance can be achievedthrough the use of other methods or coatings. The studyincluded a literature review of the existing state of theart, a field survey of coated culverts in nine states, areview of methods to improve existing coatings, a review ofalternative coating systems and methods of protection and areview of existing specifications.

Protective coatings used by the States includebarrier coatings and sacrificial metallic coatings or acombination of both. Most coatings and materials usedpresent durability drawbacks in many situations, thus re-quiring the use of more costly materials or higher main-tenance cost through a lower life expectancy. Many states,either individually or through the Federal Highway Admini-stration, have conducted surveys to establish coating andmaterial performance under extreme and typical conditions.Coatings evaluated by the states include asphalt, asbestosbonded asphalt, various polymeries, epoxies, zinc coating,aluminum coated steel and aluminum-zinc coated steel.Materials investigated include steel, clad aluminum, stain-less steel and concrete. Fiberglass and polyethylene arebeing evaluated in a few cases.

Most states have criteria to determine whichcoating and material to use in a given exposure situation.One criteria is California Method 643, which utilizes pH andresistivity to establish a life expectancy basis. Becauseof the wide range of conditions throughout the country, thismethod is not universally applicable.

Field inspections at 82 locations in 9 statesindicate that most coatings are effective in situationswhere runoff does not include abrasive debris and the waterdoes not contain a high percentage of soluble salts, par-ticularly chlorides. All organic coatings inspected aresubject to impact and abrasion deterioration and most willdeteriorate under wet alkali or salt conditions. Low pHconditions do not seem to deteriorate the coating as much asattacking the metal substrate at coating defects, causingdisbondment. The best all around existing coating system isasbestos bonded asphalt coated and paved galvanized steel

.

The most commonly used coating, asphalt, suffersfrom poor adhesion under immersion conditions, low impact

strength at low temperatures and water penetration leadingto removal from the substrate. Less than adequate surfacepreparation prior to coating and ill-defined applicationtechniques impair this coating's performance. Possibleimprovements to this coating include the use of organicadditives to improve adhesion and blending with other com-ponents such as inorganic fibers or organic polymers toimprove mechanical properties.

Certain methods might be useful for protectingculverts in erosive conditions instead of organic coatings.Such methods include the use of stilling basins, energydissipators and reinforced paving.

Alternative coating systems might prove beneficialfor protecting culverts. Coating systems include urethanes,epoxies, neoprene, fusion-bonded coatings, ceramics andmetallized coatings. While these coatings are more costlythan any current culvert coating, they have the advantagethat they can be applied only where needed, such as on theinvert. Additional study is desirable to identify the mostcost effective coatings.

Several primers and wash primers were evaluated inlaboratory screening tests to determine if improvements inthe adhesion of asphalt and other coatings to galvanizedsteel can be obtained. The tests show that some adhesionimprovement is possible but that further work is needed toestablish the cost effectiveness of primers under culvertexposure conditions.

State specifications could benefit from someadditions regarding procedures for protecting coated pipe.Suggested additions are included in this report. AASHTOspecifications, notably M190, M243 and M246, are used bymost states to specify culvert coatings. Recommendationsare made for additions to these specifications.

INTRODUCTION

Culvert life expectancy is determined by materialdurability and structural durability. Material durabilityrefers to the ability of rigid or flexible drainage pipe toresist deterioration due to the natural processes of corro-sion, abrasion and erosion. Structural durability is theability of rigid or flexible drainage pipe to carry thestatic and dynamic stresses and strains imposed by the earthand traffic without deterioration. Total durability is afunction of both of these interdependent factors and bothare essential to satisfactory performance and service lifeof any drainage pipe. This study is concerned with materialdurability.

Erosion by runoff debris can gradually abrade thepipe material. Corrosion can deteriorate the pipe materialon both internal and external surfaces. Appendix A presentsa brief description of the corrosion process. Severalmethods can be used to combat corrosion and abrasion. Theseinclude: coatings, linings, cathodic protection, increasedmetal thickness and alternate materials selection.

Coatings used on culverts can be divided into twomajor categories, that is, metallic and barrier. Metalliccoatings are either intended to present a corrosion andabrasion resistant surface to the environment and/or provideprotection to the base metal by acting as the anode in agalvanic cell (cathodic protection). Not all metalliccoatings offer sacrificial protection to the base metal.For example, chromium coatings are cathodic to steel.Sacrificial coatings such as zinc, aluminum or aluminum zincon steel, and aluminum cladding are used on drainage pipe.Barrier coatings are used to prevent contact between thematerial to be protected, metal or concrete, and the en-vironment. Organic coatings are often used effectively inconjunction with sacrificial coatings. Barrier coatingsused on drainage pipe include bituminous coatings (asphaltdips, mastics), epoxies and thin film polymeric coatings.Invert protection against abrasion and corrosion can beprovided by asphalt paving, clay, cement or fiberglasslinings. Other types of coatings such as passivating con-version coatings and inhibited coatings are not used ondrainage pipe

.

Cathodic protection is a method whereby directcurrent is applied to a corroding metal to make the entiresurface a cathode. The direct current is applied by usingsacrificial anodes or impressing current from an externalrectifier through anodes to the protected surface. Cathodic

protection is most usable on the external surfaces of drain-age pipe and will not prevent abrasion or corrosion ofnon-immersed areas. Cathodic protection is seldom used ondrainage pipe.

Increasing metal thickness to compensate for ex-pected corrosion and erosion losses is a method of increas-ing the life expectancy of culverts. The degree of expectedcorrosion and/or erosion must be known for the site in-volved. Severe corrosion or erosion limit the applicabilityof this method making other methods more cost effective fora particular situation.

Materials known to be durable in severe environ-ments can be used in place of coatings or other measures.These materials include: aluminum, concrete, vitrified clay,stainless steel, polymers and fiberglass.

The purpose of this study is to evaluate theeffectiveness of various protective coating systems incontrolling the deterioration of drainage pipe. This pro-ject included seven tasks as follows:

1. Literature review

2. Field investigations

3. Performance review of asphalt coatings

4. Evaluation of coating need and alternatives

5. Coatings other than asphalts

6. Improving coating bond to galvanized steel

7. Specification review

This report summarizes the findings of these tasks andpresents conclusions and recommendations for future action.

DISCUSSION

Literature Review

Table I presents a summary of the literaturereview findings and Appendix B summarizes the interviews andtechnical literature.

Contact with knowledgeable individuals in. each ofthe fifty states provided information on culvert materialsand performance. Appendix C lists the references usedthroughout the study. Additional sources utilized were theNational Association of Corrosion Engineers (NACE) ab-stracts, National Technical Information Service (NTIS),American Association of Highway and Transportation Officials(AASHTO), American Concrete Pipe Association (ACPA), Ameri-can Corrugated Steel Pipe Institute (ACSPI), American Socie-ty for Testing and Materials (ASTM) and American Iron andSteel Institute (AISI).

Most states use essentially the same drainage pipematerials. Culverts are fabricated from galvanized steel,clad aluminum and various types of concrete. Additionalmaterials for under-drains and culverts include vitrifiedclay, bituminized fiber, asbestos cement, polyvinylchloride(PVC) and acrylonitrite-butadiene-styrene (ABS). Coatingsused to protect metallic culverts also are similar fromstate to state. Bituminous coatings to include asphaltdipped and asbestos bonded asphalt are the most common.Polymeric coatings are beginning to be used more frequentlywith U.S. Steel Company Nexon coal tar laminate having thelongest service history of the polymers. Other polymericcoatings being tested are Inland Steel Company Black Klad,Wheeling Steel Company Plasticote, Bethlehem Steel CompanyBeth-Cu-Loy PC and Republic Steel Company Polycote. Metall-ic coatings (excluding zinc) such as aluminum and alumi-num-zinc on steel are being evaluated in several states.Epoxy coatings are used to protect concrete in acidic areasin some states.

Coating deterioration, particularly with asphalt,is common. Asphalt dipped pipe is subject to asphalt dis-bondment both in abrasive and non-abrasive conditions.Reasons cited for poor asphalt performance include waterpenetration, either through the coating or through coatingdefects, alternate freezing and thawing, mechanical abra-sion, poor or variable asphalt quality and inadequate appli-cation techniques.

Asphalt application techniques are similar fromfabricator to fabricator. Asphalt is usually applied togalvanized steel by immersing the completed culvert pipeinto an open liquid asphalt bath for a period of time. Moststates use AASHTO Specification M190 to govern coatingthickness and quality. AASHTO Ml 90 does not specify eitherpipe cleaning procedures or application procedures. Pipe isimmersed in asphalt at a temperature of about 40 °F (204°C)for a time thought to be sufficient to attain asphalt ad-hesion. Immersion times, although critical, are not uniformand mostly depend on experience, or in a few cases, are theresult of testing programs. There does not seem to bequantitative procedure for shop testing asphalt adhesion.Although two studies reported improvement in asphalt ad-hesion to zinc by chemically pre-treating the surface 38 ' 43

,

there is no commonly used pre-treatment of the zinc surfaceother than occasional removal of soil and drying. Asbestosbonding improves asphalt adherence and is usually used wheresevere applications are expected. There is evidence tosuggest that certain chemicals in water runoff can corrodethe zinc substrate and loosen asbestos fibers9

. An effec-tive barrier coating should increase culvert life even ifthe underlying metal is corroded by limiting the surfacearea of metal exposed to the corrodent. There are indica-tions from some states that asphalt quality varies resultingin field performance differences. This is discussed morefully in the section on Asphalt.

Polymer coatings reportedly suffer from pooradhesion in corrosive environments at areas where the coat-ing to metal interface is exposed such as at section joints.Another problem appears to be the abrasion resistance of therelatively thin film coatings. Lack of adhesion betweencoal tar laminate coating and asphalt paving was also re-ported, however, other sources indicate better adhesion ofasphalt to polymeric coatings than to galvanized steel.This investigator has seen proprietary data showing an orderof magnitude increase in adhesion between asphalt and avinyl plastisol coating over that between asphalt to gal-vanized steel and asphalt to asbestos bonding using thereverse impact test method.

External corrosion in most areas does not seem tobe as much of a problem as interior corrosion and erosion.In areas where external corrosion is a problem, non-metallicand metallic coatings are used with varying degrees ofsuccess. Concrete is usually used in severely corrosiveareas. Most concrete is installed uncoated. Cathodicprotection (excluding sacrificial coatings) is not used toprotect exterior culvert surfaces and is not applicable to

interior surfaces unless the culvert is immersed in water.Attempts to use cathodic protection appear to have beenlimited to uncoated galvanized culverts. This would becostly to do and somewhat impractical to protect a sacri-ficial finish. Cathodic protection is best applied tocoated surfaces to minimize protective current requirementsand reduce costs. Metal surfaces intended for cathodicprotection need not be coated with a sacrificial metal.Further discussion of cathodic protection is presented inthe section on Alternate Methods.

Field Investigation

Professional Services Group investigators con-ducted a field survey in nine states to supplement theliterature survey. States were selected from which themaximum amount of information could be obtained and on thebasis of exposure condition types, culvert and coatingmaterials used and existence of test sites where informationcould be gathered for a large number of coatings undersimilar exposure conditions. The states selected for thefield study were: California, Colorado, Kentucky, Louisi-ana, New York, Ohio, Oregon, Pennsylvania and Utah.

Culvert test sites were selected after discussionswith state Department of Transportation engineers. We madean effort to include as many types of coatings and serviceconditions in the study as possible. The preponderance ofasphalt coated culverts inspected reflect the fact thatasphalt is still the major coating material used. We selec-ted test sites to include a broad range of exposure condi-tions so as not to bias the results with extremes of serviceexposures. Culvert locations were selected so as to mini-mize chances of finding high water levels as this wouldprevent inspection of the interior. Generally, the minimumsize culvert examined was 76.2 cm (30 in.) to allow inspec-tion of the interior. Smaller sizes were examined whennecessary. A total of 108 culverts were inspected at 82highway and test sites. Table II presents a breakdown ofthe inspection by state and coating type.

A. Procedures

Information obtained at each site included:

1. General

Location, type, size, constructionmaterial, coatings, installation date, maintenancehistory and prior inspection information.

2. Environmental

Measurements were taken to establish theservice conditions at the culvert tested. Theseincluded: soil and water pH, soil and waterresistivity, sulfides in the soil, sulfates andchlorides in the soil and water, alkalinity andhardness of the water (if present). Resistivity,pH and sulfide tests were performed in the field.Chloride, sulfate, alkalinity and hardness testswere performed in the laboratory using portabletest kits (Hach). Minimum resistivity 'measure-ments were performed on soil samples by saturatingthe soil with deionized water and measuring theresistivity in a soil box. Field resistivityreadings were taken using the Wenner 4 pin methodwith spacings at 5, 10 and sometimes 15 feet,depending on pipe depth. All pH readings weremade with a portable analog pH meter. Sulfideswere measured using a specific ion electrode,however, sulfides were not detected at any loca-tion tested. Tests were not replicated unless aquestion arose as to validity of a particularmeasurement. Questions of this nature were satis-factorily resolved in replicate tests. Measure-ment precision is as follows: chloride (±10%),sulfate (±35 mg/1), alkalinity (±17 mg/1), hard-ness (±17 mg/1), resistivity (±1%), pH (± . 1 pHunit)

.

Approximate flow rates through culvertswere measured where water was present by timingthe passage of a floating ball through the cul-vert. These flow rates, being taken at times oflow flow, do not necessarily represent maximumflow conditions.

3 . Culvert Pipe Condition

This was done visually and by the use ofan ultrasonic thickness instrument, micrometer andpit depth gauge. Detailed corrosion loss measure-ments were not taken since the primary purpose ofthe inspection was to evaluate the coating.

Exterior surfaces were examined bycutting 5 cm (2 in.) diameter samples from theinvert. Samples were examined in detail in thelaboratory. Three samples were usually takenapproximately 10° from the invert at least 3.0 5 m

8

(10 ft.) apart. The first sample was cut at least3 0.5 m (10 ft.) from the inlet or outlet. Weoriginally planned to remove more samples butsample condition was uniform and we felt that noadditional benefit would be gained by cutting moresamples. We used a high speed steel hole saw anda 12 volt D.C. battery powered drill. Holes wererepaired using an expandable rubber stopper (SeeFigures 1 & 2). Samples were not removed from allculverts because of high water level, small sizeor the lack of an exterior coating. Soil samplesfor pH and resistivity measurements were removedfrom the sample holes where applicable.

Electrochemical potential and polariza-tion data20 ' 40 were obtained on most culverts tosupplement the samples. Polarization tests util-ized the E-log I and polarization resistancemethods to measure corrosion current for compari-son between locations. Test instrument accuracyis ±.2 percent full scale or better. Polarizationdata can be used as a means of evaluating therelative coating condition and the protectivecurrent needed if cathodic protection were to beused. Polarization data is currently being usedto evaluate corrosivity in at least three states(Arizona, Utah, Michigan) and the method was

CO

suggested as a means of evaluating coated pipe

Polarization measurements can not,however, be used to determine absolute corrosionrates. This is because test current flows only toareas such as edges, rivets and coating defectswhich allow the electrolyte to reach the pipesurface. Unless the magnitude of these areas areknown by some means, the corrosion current densitycan not be calculated and an accurate corrosionrate can not be determined. A method whichaverages the corrosion current over the entiresurface area, which might be valid for an uncoatedculvert, yields a low and misleading corrosionrate. The corrosion rate of an exposed metalsurface on a coated pipe is likely to be greaterthan the corrosion rate of a similar uncoatedculvert because galvanic corrosion currents, suchas caused by oxygen differential cells are con-centrated on a small area. Resistance and capaci-tance measurements are better methods of evaluat-ing coating performance. They have proven utility

in laboratory tests 84 but have not been estab-lished as being useful tests in field applica-tions.

A pipe corrosion condition rating wasassigned to each culvert based on the condition ofthe interior exposed metal, if any. The ratingsystem is defined in Table IV.

4 . Coating Condition

Interior coatings were examined forcoating disbondment, adhesion, blistering, ero-sion, checking and cracking. A coating conditionrating was assigned as defined in Table IV.Coating thickness was measured on the flat portionof the corrugations in several random areas with anondestructive magnetic-eddy current gauge.Adhesion tests were qualitative in nature made byprying off the coating in several areas. Quanti-tative adhesion data was not obtained as a reli-able test is not available in our opinion.

Random interior culvert surfaces were inspectedfor coating defects which would expose the culvertmetal to water by using a low voltage (67.5 volt)coating fault detector.

Table III lists the culvert sites in-spected and provides information on culvert size,age, type of coating and location. Table IVcontains a summary of the field data for the siteslisted in Table III.

B. Results-General

All of the coatings inspected appear toperform similarly under similar exposure conditions.The data presented in Table IV establishes the environ-mental conditions affecting the culvert and its coat-ing. Comparisons between environmental data and cul-vert condition do not show strong cause and effectbehavior. Relationships examined are as follows:

1. Corrosion Condition vs :

dipped coal tarasphalt laminate

a. water pH r = .07 r = .34b. water resistivity r = .07 r = . 51

10

c. water sulfate content r = -.18 r = -.34d. water chloride content r = -.19 r = .57e. sulfate and chloride r = .18 r = -.67f. pH and resistivity r = .01 r = .20

2. Coating Condition vs :

a. water pH r = - . 18 r=.06b. water resistivity r = -.05 r = .64c. water sulfate content r = .20 r = .17d. water chloride content r = -.17 r = . 51

Figure 3 illustrates the linear regression curve of coatingcondition of coal tar laminate coatings versus water pH.Each of the correlation coefficients were tested against thehypothesis that no dependency exists using the followingrelationship:

zct

= /n-3 In (1 + r)2 (1 - r) (ref. 56)

Where Z = standardized random variable for a normaldistribution

n = number of sample points

r = correlation coefficient (a measure of thecloseness of the correlation, +1 = perfectdependency)

a = 95% confidence level

No significant relationships were found between these vari-ables at a 95 percent significance level.

We also attempted to establish a correlationbetween environmental data and Corrosion Condition using thefollowing relationships: power function (Y = Ax^ ), exponen-tial function (Y = Anx ) and polynomial function (Y =

A+Bx +Cx2 +Dx 3+. . . ) . The data tested in this analysis was

obtained at coal tar laminate coated culvert sites andinclude pH, resistivity, sulfate, chloride and equivalentweights (of sulfate and chloride). The analysis utilized acomputer program using least squares curve fitting tech-niques. We did not establish a reasonable relationshipbetween any of the variables tested. This analysis confirmsthe results of the linear regression analyses so that test-ing of additional data was not considered worthwhile.

11

The ability of water to form a protective scalecan influence the corrosion rate of metal. Two measures ofa water's ability to scale or not to scale are the LangelierIndex and the Ryznar Index 57

. Briefly, the Langelier Indexis defined as the algebraic difference between the actual pHand the pH at which a water is saturated with CaC03 . Posi-tive Index values represent waters with scaling tendenciesand negative values represent nonscaling. The Ryznar Indexis empirical and is defined as 2 x pH at which CaCCU sat-uration occurs minus the actual pH. Values above 7 repre-sent scaling waters and values below 6 indicate corrosivetendencies. Although not on the original test plan/ we feltthat it might be beneficial to determine if there was arelationship between the scaling tendency and corrosioncondition. We chose the Ryznar Index as it presents acontinuous scale which can be used to estimate magnitude andis relatively fast to calculate. Additional data needed tocalculate this index is the hardness and methyl orangealkalinity which were gathered starting with test site 23 ofTables III and IV. The Ryznar Index is calculated as fol-lows :

pH = log 5± - log (Ca++) - log (alk) (ref. 58)S K

2

Where: K2 is the second dissociation constant forcarbonic acid

Ks

is the activity product of calcium car-bonate

Ca++ is expressed in mg/1 Ca

alk (alkalinity) is expressed in mg/1 CaCOo

Figure 4 presents the resulting correlation. Corrosioncondition appears to correlate with the Ryznar Index ata significance level of 95% but not at a 99.5% level.However, the general trend indicates that the morecorroded culverts are found at greater Ryznar Indexvalues. This would not be the predicted relationshipsince the greater Pyznar Index values are associatedwith the lower corrosion tendency. A similar relation-ship was also found to exist between the Ryznar Indexand coating rating. Again, this data is significant atthe 95% level but not at a 99.5% level (See Figure 5).The scatter in this data is too wide and the number ofdata points too small to draw firm conclusions, butperhaps the method deserves further investigation foruse as a design tool.

12

Coating condition appears to be significantlyrelated to corrosion condition as shown on Figure 6.

This appears reasonable because many of the environmen-tal factors which lead to coating deterioration alsocause corrosion of the pipe.

Significant correlations between corrosioncurrent (all coatings) and: average soil resistivity (r= -.12), minimum soil resistivity (r = .12), soil pH (r= -.10), age (r = .24), total culvert surface area (r =

.16) and pipe to soil potential (r = .18) were notfound. This indicates that corrosion current is re-lated to both the corrosivity factors and exposed sur-face area which can not be measured with presentlyknown methods

.

Comparison of the internal coating condition(CtgC) by the three major coating types inspected yields thefollowing mean coating conditions (See also Table IV).

No. CulvertsCoating Unp

MeanCtgC

avedNo. Culverts

PavedMean NoCtgC

Asbestos BondedAsphalt 3.4 9 4.25

Coal Tar Laminate 3.1 20 _

Asphalt 2.3 15 1.73 15

Applying statistical methods to determine the significanceof the differences in means at the 95% level, we find thatthe only difference in coating condition occurs betweem coaltar laminate and asbestos bonded asphalt and that no dif-ferences exist at the 99.5% level. This means the threeunpaved coating types would probably perform similarly undersimilar conditions. The effect of asphalt paving would seemto improve the performance characteristics of asbestosbonded asphalt coated pipe and reduce the performance of theasphalt coated pipe. We think the anomaly with the asphaltpaving is caused by the small number of sample sites in-spected and is probably not a true indication of the effectof asphalt paving.

The lack of statistical correlation and evidentscatter can be traced partially to the limited number ofsites that could be included in this survey. Another factoris variability of the data at a given site. It is likely

13

that the pH, resistivity, scale forming tendencies andabrasive characteristics of the streams flowing through theculverts change considerably depending on seasonal precipi-tation conditions. Most test sites were examined underconditions of normal flow as evidenced by wear patterns onthe culverts and in the streams. Rapid changes do occur,however, as evidenced by the doubling of water volume andvelocity in a culvert in California during a light rainfallas it was being inspected. It is also apparent that themeasured flow rates do not reflect the worst conditions towhich the pipe is exposed.

Coating deterioration on the culverts inspectedwas limited to the invert section and most invert damageseemed to occur in areas of normal stream flow. Abrasionand impact loading by rocks and other debris appear to be amajor cause of deterioration. Small stones appear to be ascapable of causing erosion damage as large rocks based onobservations made at each site. The embrittling effect ofasphalt in cold weather can be a factor in impact and abra-sion deterioration but is not the sole factor as similardeterioration was observed in the warm climate of southernCalifornia. Coating damage above the normal stream flowlevel was not observed at any of the culvert sites in-spected. Coatings above normal stream flow level protectthe pipe from corrosion and adhere well to the galvanizedsteel

.

C. Results-Individual Coating Types

1. Asphalt

a. Field Applied Asphalt

Figures 7 and 8 illustrate two culverts withfield applied mastic which were subject to abrasive flowconditions. Figure 9 shows a culvert with field appliedmastic exposed to alkali water. Generally speaking, thefield applied mastics inspected exhibited poor performance.Coating thickness measurements revealed that coatings werenot uniformly applied. The average thickness being belowthe .127 cm (.050-inch) minimum specified by AASHTO and thestates.

b. Mill Applied Asphalt

Figures 10 and 11 illustrate damage to as-phalt dipped culvert pipe. The damage in Figure 10 wascaused by abrasion and impact while that in Figure 11 was

14

apparently the result of high chloride and sulfate concen-trations in the water. Test samples in Louisiana showextensive damage to asphalt dipped galvanized steel whensubjected to a non-abrasive environment containing chloridesand sulfates (numbers 67, 68). Figure 12 illustrates a pureabrasion failure on a 2 year old steeply sloped culvert inCalifornia. There was no water flow at the time but themaximum flow level was indicated to be one-half the 1.52 m(60-inch) diameter with rocks up to 6-inches and otherdebris visible. We performed a coating thickness profile onthe downstream side of the corrugations and found a definiteerosion loss on the invert. Mean coating thickness on theunaffected portions was .148 cm (.0 58-inch) whereas in theabraded portion, the thickness was .115 cm (.04 5-inch). Thezinc had eroded away on the upstream side of the corruga-tions sufficiently for general corrosion of the steel tooccur. Figure 13 illustrates a debris crib used at onesite. Unfortunately, as Figure 14 shows, this design is un-successful in preventing abrasion damage. Corrosive condi-tions affecting the culverts shown in Figures 10, 11, 13 and14 could not have appreciably influenced the invert coatingdeterioration.

Figure 15 illustrates corrosion perforation of aconnecting band on an asphalt coated asphalt paved culvertotherwise performing well. The coupling bands had appar-ently been field coated. This culvert was in an area of lowresistivity water and rapid flow rate but no abrasion. Thepipe coating was in very good condition, other than atconnecting bands. A few locations were visible at sectionjoints where the paving was recently removed as evidenced bybright galvanizing. This was apparently the result of tur-bulence at the joints. Figures 16 and 17 show two otherculverts where turbulence at section joints caused coatingloss and perforation. The asphalt coating and paving was inotherwise excellent condition.

The exterior performance of asphalt dipped pipecan be considered adequate. Core samples removed from theculverts exhibited fair to good coating adhesion and thecoating seems to protect the pipe surface from seriouscorrosion. Mild corrosion of the zinc under the asphalt wasobserved on three core samples, numbers 18, 58 and 60.However, polarization tests indicate that the average totalcorrosion current existing on the external pipe surface ofasphalt coated pipe is .022 ampere. This amounts to anaverage metal loss of .199 kg (.4 4 pounds) per year forexposed steel and .179 kg (.396 pounds) per year for exposedzinc. The average total culvert area of those sites testedis 155 sq. m (1675 sq. ft), with a range of 26-613 sq. m

15

(282-6597 sq. ft). The metal loss could be spread out overa large area or confined to a localized area which couldresult in a high or low corrosion rate. Pipe to soil poten-tial measurements indicate that the average asphalt coatedculvert has lost the zinc coating on metal exposed to theenvironment, the average pipe to soil potential being -.710volt to a copper sulfate reference. We did not observe pipeperforations from the soil side. The average coating thick-ness on exterior surfaces is .401 cm (.158-inch) based onthe core samples.

We conclude that asphalt coatings applied' directlyto galvanized steel by conventional methods are subject todeterioration at the ends where exposed to infrared lightand in the stream flow channel where abrasion and erosiontaken place. Deterioration of the invert sometimes appearsas removal only at section joints, removal from the upstreamside of the corrugations or complete removal.

Paving does not appear to appreciably improveperformance under severely abrasive conditions. That is,rocks and other debris combined with high water velocitydamage coating. Both conditions must be present. Largerocks were found in culverts with low flow rates not dis-playing coating damage although the flow must have been atleast momentarily high to carry in the debris. Rapidlymoving streams in themselves will not cause coating failureunless turbulence is present at coating discontinuities suchas at section joints. Unfortunately, the field data andculvert service history are not sufficient to correlate flowvelocity and bedload characteristics with asphalt deteriora-tion. This would be a useful project. Water penetrationinto the coating and disbondment occur in the stream flowchannel. Waters and soils containing relatively largequantities of chlorides and sulfates are capable of causingdeterioration of field applied asphalt mastic and asphaltdipped pipe.

Deterioration at ends exposed to sunlight takesthe form of cracking, checking, peeling and eventual removalof the coating by the stream flow.

Asphalt exhibits good performance where abrasionand salts are not a factor even in low pH environments andthe performance above stream level is excellent.

Some improvements in asphalt performance should bepossible. These include increased adhesion to the gal-vanized steel and increased abrasion resistance. The sec-tion on asphalt coating performance discusses these improve-ments in detail.

16

Six States have discontinued the use of asphaltcoating, they are: Hawaii, Kansas, Maryland, Missouri,Pennsylvania and Tennessee. Asphalt is no longer usedbecause it was found to provide an insufficient increase inservice life to justify the cost. Poor service life inthese states was the result of abrasion and impact failure.Two of these states, Missouri and Pennsylvania are using theorganic coatings (polymeries); the others are of the opinionthat organic coatings are unnecessary and that substitutematerials, such as concrete, can be used in corrosive situa-tions. See also Table VII for criteria in using uncoatedpipe.

2. Asphalt Coated Aluminum

Inspections included only four asphalt coatedaluminum culverts (two at test sites). Figure 18 showsa three year old aluminum coated culvert which wasinstalled with a settling basin at the inlet. Somechipping of the coating is seen on the invert. Coat-ings on both aluminum culverts examined in field usewere in good condition with little coating deteriora-tion and no corrosion of the aluminum. Abrasion andrapid stream flows did not appear to be factors in thetwo culverts examined. Asphalt exhibits a poor bond toaluminum, being easily removed by prying. Figure 19shows a test culvert from Louisiana exposed to a mod-erately corrosive environment in which almost no coat-ing remains. While the condition in Figure 19 could bethe result of periodic removals during inspections, itdoes illustrate the poor adhesion between aluminum andasphalt when conventional application techniques areused

.

The sample size examined is too small to makefirm conclusions but it appears that asphalt coatedaluminum is unsuitable for abrasive locations. We donot know how paved aluminum would perform but suspectit would not perform as well as asphalt paved gal-vanized steel.

3. Asbestos Bonded Asphalt Coated Galvanized Steel

Asbestos bonded asphalt coated galvanizedsteel generally exhibited better performance than plainasphalt dipped galvanized steel. This coating is stillsubject to the same modes of deterioration as asphaltcoated pipe, that is: checking and cracking at exposedends and abrasion of the invert.

17

Figure 20 shows an unpaved culvert where thecoating was removed from the upstream side of thecorrugations. Stream velocity was important in coatingdeterioration as the coating was intact where thestream velocity decreased near the outlet. Figure 21shows the interior of a culvert exposed to waters witha high salt concentration. The asphalt has delaminatedfrom the saturated asbestos layer at and below thewater level. Layer type corrosion is occurring on thesteel. In many areas, although the asphalt has beenremoved, the steel is not corroded, being protected bythe asphalt impregnated asbestos and zinc. This obser-vation was made at several other locations. Figure 23illustrates the asphalt delamination in alkali soilconditions at the Utah test site but no corrosion ofthe steel. Figure 2 2 shows an asbestos bonded asphaltcoated test culvert exposed to brackish water fromLouisiana's Hackberry test site. No significant coat-ing deterioration was observed although asphalt delami-nation was observed at the Starks test site (number76). Some of the delamination at the test sites mightbe the result of periodic removal operations but couldalso be the result of freeze thaw cycling. Adhesion ofthe asphalt to the substrate varied, but the mode offailure always consisted of tearing of the asbestoslayer.

Asbestos bonded asphalt coating appears toprotect the exterior pipe surfaces as well as asphalt.Average pipe to soil potential is -.660 volt to coppersulfate indicating depletion of the zinc at exposedareas. Average external coating thickness (includesgalvanizing and asbestos layer) is .467 cm (.184-inch)based on the core samples. There is no significantdifference between average asphalt thickness and thaton plain asphalt dipped pipe.

4. Coal Tar Laminate

Coal tar laminate (U.S. Steel Nexon) exhibitsgood performance except under abrasive stream flows andin low pH and high salt environments. Figure 24 illu-strates the beginnings of abrasive damage at corruga-tion crests. This particular culvert is in the Genes-see Expressway (1-3 90) in New York and was about a yearold when inspected. Most of the damage occurred duringconstruction when angular shale passed through theculvert. Figure 2 5 shows a more advanced stage oferosion on a six year old culvert with a strongly

18

acidic stream flow. In addition to the abrasion da-mage, the coating is disbonding from the substrate,probably as a result of zinc corrosion.

A pitch resin film is laminated onto thegalvanized sheet which is then formed (at some futuretime) into a corrugated culvert. The culverts in-spected in this survey contained a defect at the lockseam which did not seem to impair performance undermildly corrosive conditions but which could seriouslyimpair performance in adverse environments. In thelock seam forming operation, the coating is cut com-pletely through to the base metal by the tooling at thelock seam. Similar damage was observed on polyethylenecoated aluminum test samples being installed at theHackberry, LA test site. In corrosive conditions thisdefect acts as a point where the zinc is attacked anddisbondment occurs. Figure 26 shows an extreme exampleof this where the invert coating is undercut and liftedup by the stream flow and is only restrained by a pipepassing through. Figure 27 shows another extreme casewhere the coating has lifted up and is acting as a dam,collecting silt and other debris. In both cases, noother deterioration of the coating was observed.Figure 28 shows a seven year old culvert in an acidicstream in Ohio. This culvert was reported to haveperforated about one year after installation, probablyas a result of disbondment starting at the lock seam.The condition which led to this deterioration can becorrected in the manufacturing procedure.

Figure 29 shows a coal tar laminate coatedtest culvert exposed to brackish waters at the Louisi-ana Hackberry test site for six years. General blis-tering is occurring over the entire surface although itis more severe at uncoated cut edges and. coating de-fects. Similar blistering and disbondment at edges andthe lock seam were observed in alkali soils at twoColorado test sites (No. 45, 48) but were not observedin the alkali soil of the Utah test site (No. 39). Themajor difference between the Louisiana, Colorado andUtah test sites is the greater chloride content at theColorado and Louisiana test sites (see Tables III-IV).

Coal tar laminate protects the steel fromsoil corrosion (except in soil with a high chloridecontent) based on samples removed from the culvertsexamined and very limited polarization tests. Pipe tosoil potentials indicate that the zinc is essentiallydepleted from exposed metal, the average potentialbeing -.740 volt (Copper Sulfate reference).

19

The yellow coating (called a modified epoxy)used on the opposite side of single side coated Nexonhas little or no protective value. When exposed towater containing high salt concentrations, the coatingallows corrosion attack of the metal substrate (numbers37, 66, 72, 75, 72, Table IV). The coating is appliedthinly, the maximum thickness measured was .0 20 cm(.008-inch), and does not act as an effective barriercoating when tested with a coating fault detector. Theaverage soil side corrosion current of the culvertstested was .06 ampere or 1.08 pound (.49 kg) per yearzinc loss. Average surface area is approximately 137sq. m (1482 sq. ft). Pipe to soil potentials indicatethat zinc was still intact at each of the culvertsusing this coating. The reason that the zinc remainsintact on the modified epoxy coated pipe and not withthe other coatings is that the thin epoxy is not aneffective barrier coat. Most of the zinc is exposed tothe soil so that the corrosion current is spread overthe surface and not concentrated at small areas. Theresulting corrosion rate does not deplete the zinc asquickly at any one point.

5. Polyethylene

Polyethylene (Inland Black Klad) coatedculverts were examined at test sites in Colorado,Kentucky and Louisiana. Figure 30 illustrates a sampleremoved from the brackish Louisiana Hackberry test site(No. 69, 70) and shows extensive general blistering anddisbondment. Minor blistering was observed at coatingdiscontinuities at the mildly corrosive test site atStarks (No. 80, 81). No deterioration was noted at theDillon, Colorado test site (No. 55). Performance inthe acidic waters of Kentucky's Mortons Gap test site(No. 94) is satisfactory except for some disbondment atuncoated edges.

6. Polyvinyl Chloride

Polyvinyl Chloride (PVC) as manufactured byV7heeling Steel Company (Plasti-Cote ) coated culvertswere inspected at test sites in Colorado and Louisiana.PVC coating performance is similar to polyethylene,developing blisters over the surface in areas with highchloride content. Figure 31 shows a PVC coated testculvert removed from Louisiana's Hackberry test site(No. 71). Minor blistering at the lockseam was alsofound at the two Colorado test sites at Fruita andOlathe (No. 46, 49). Significant deterioration was not

20

observed at Louisiana's Starks test site (No. 74) orthe mildly corrosive test site near Dillon, Colorado(No. 54).

Abrasion was not a factor at any of the testsites inspected in Louisiana or Colorado so that we donot know how well the polyethylene or PVC coatingsperforin under abrasive conditions.

7. Epoxy Coated Concrete

Epoxy coatings applied to concrete culvertsseem to be successful in protecting the concrete sub-strate. Figure 32 illustrates a culvert with a coaltar epoxy invert coating. A few areas were disbondedon this invert but the coating was otherwise in goodcondition. Figure 3 3 shows a polyamide epoxy invertcoating which did not exhibit deterioration. Thisparticular coating contains a sand filler which mightimprove its erosion resistance. The stream can erodethe concrete and undercut the coating as in number 8 7

in Table IV if the paving does not cover a sufficientportion of the pipe circumference to shield the con-crete during normal stream flow conditions. The coatedconcrete culverts examined were not exposed to abrasiveflow or rapidly moving water so that performance underthese conditions can not be documented here. An epoxycoated culvert exposed to rapid water movement andabrasion (number 31, Table IV) was given only a cursoryexamination because of rapid water flow and volume.The epoxy appeared to be intact after 3 years.

8. Others

A vinyl plastisol coating (Bethlehem SteelBeth-Cu-Loy-PC) was examined at the Genessee Expresswaysite in New York. Both coal tar laminate and vinylplastisol coated culverts had been installed in sec-tions 7A and 7B of the Genessee Expressway (1-390) nearDansville, New York. The terrain is hilly and abrasiveflow exists, particularly during the period followingconstruction. The two coatings were not necessarilyinstalled under similar exposure conditions, however,the abrasiveness of the bed loads carried by the vinylplastisol appears to be less severe. As previouslymentioned, the coal tar laminate coating is experienc-ing severe abrasive damage. The vinyl plastisol, onthe other hand, has not been damaged. Erosion of thevinyl plastisol has occurred, however, affecting theupstream side of the culvert crests. Evidence of this

21

is seen in a dulling of the coating gloss and a slightthickness loss measured to be .0032 cm (.0013 in.)compared to the unaffected side. Tearing and disbond-ment were not observed. The vinyl plastisol materialwas also inspected at two test sites at a BethlehemSteel plant in Bethlehem, Pennsylvania where pollutedplant runoff flowed through galvanized steel, asphaltcoated steel and the vinyl plastisol. The polymericcoating was resisting deterioration well after 4 monthsexposure where both the uncoated and asphalt testculverts were deteriorating.

Gunite paving was found to be unsatisfactoryin acidic flows. The gunite paving that we examinedappeared to have broken loose from the corrugatedsteel. There was no reinforcing used and this mighthave helped retain the paving. Figure 34 illustratesthe type of deterioration observed on gunite paving.

Figure 3 5 shows a portion of concrete pavingapplied over asphalt paving which was applied overasbestos bonded asphalt dipped pipe. Although thewater pH was neutral at the time of inspection, theconcrete was eroded as much as 12.7 cm (5-inches) inthe normal flow line. In addition, the water wasflowing under the concrete in places.

Figure 36 shows a concrete paved culvertapplied over asbestos bonded asphalt. The concrete isprobably reinforced according to Utah specifications.No deterioration was observed, however, the culvert isexposed to only intermittent flow.

Vitreous clay lined concrete appears toperform satisfactorily, based on inspection of a singlesample. In the culvert examined, there was no erosionof the vitreous clay plates and only some erosion ofthe mortar after 18 years. The unprotected concrete atthe ends showed deterioration from the low pH water.

One aluminized steel test culvert wasexamined in alkali soil at the Utah test site. Figure37 shows a typical view of that pipe. The only deteri-oration noted was at the spiral weld and some corrosionat the soil-air interface.

One test site containing an aluminum-zinc(Bethlehem Steel Galvalum) coated culvert and a gal-vanized culvert was inspected. Figure 38 shows thetest installation with the Galvalum culvert on the

22

right. Figure 39 shows a detail of the galvanizedculvert and Figure 40 shows a detail of the Galvalumculvert. Both culverts are performing similarly on theinterior, namely both coatings have completely deteri-orated. The test installation is near a major road andit is possible that they drain a high concentration ofdeicing salts. The exterior of the Galvalum culvert isdeteriorating more rapidly than that of the galvanizedculvert based on the core samples removed from each,and pipe to soil measurements. Galvalum is practicallydepleted while the galvanizing is intact.

An epoxy coating (number 4 2, Table IV),organic zinc (number 41, Table III), and an inorganiczinc (number 4 0, Table III) were examined at the Utahtest site. All completely deteriorated in the alkalisoil. An unbonded polyethylene wrap similar to thatpermitted in American Water Works Association (AWWA)Specification CIO 5-7 7 for ductile iron water pipe wasinspected at one of the Colorado test sites and foundto be unsatisfactory. Unbonded polyethylene was foundto be unsatisfactory in corrosive soils in anotherstudy also60 .

9. Coating Test Sites

PSC investigators inspected test culverts infour states during the field study. Figures 41-44illustrate the typical appearance of four of thesesites and Table IV provides environmental information(numbers 39, 45, 48, 53, 64 & 74). Only one coatedculvert was examined at Kentucky's Morton's Gap testsite (No. 94) because the site is being abandoned andfew culverts remain. A new test site with similarcharacteristics might be selected nearby. The primarypurpose of the test sites is to expose the test coat-ings to the most corrosive environment in the state.None of the test sites inspected address the problem ofimpact and abrasion which appear to be the common modesof deterioration. Since the laboratory tests used toevaluate abrasion resistance are useful mainly forcomparison of coating types but do not simulate astream flow with an abrasive bedload, considerationshould be given to establishing test sites to evaluateabrasion resistance under actual field conditions.

All of the culverts that we examined were per-forming their intended function, that is, permitting waterflow and keeping the roadbed intact. Several culverts were

23

perforated significantly or had lost their inverts com-pletely but none appeared to be in danger of failure. Inthe case of the few coal tar laminate coatings previouslydiscussed which delaminated extensively, eventual failure bystream blockage is possible.

Coatings appear to be effective for increasingservice life under conditions free from abrasion or highsalt content. Invert pavings appear to be effective exceptunder conditions of severe abrasion. We do not have suffi-cient data to predict the service life of coatings, however,of those we examined the service life can be as low as twoyears or less under severe abrasive conditions and greaterthan twenty years under other conditions. Two methods canbe used to evaluate service life, excluding laboratorymethods which would be useful only in screening tests, thatis, inspection of a large number of culverts or the esta-blishment of test sites which would provide the major ex-posure classifications. Many States have conducted exten-sive field tests on in service culverts. One of the prob-lems affecting this method is that the exposure history ofthe culvert is not always known. The establishment of anexposure site in which both exposure and coating conditioncan be frequently monitored would seem to provide moreconsistent information. The Louisiana program is an exampleof this type of program. A possible test program is dis-cussed in Appendix F.

Asphalt Coating Performance

The literature review raised the question ofasphalt uniformity and whether performance varies because ofthat lack of uniformity. Variation in performance can becaused by application technique or asphalt composition.

PSG investigators interviewed twenty-three culvertfabricators by telephone in order to ascertain if variationsin application procedures exist. We also visited six fabri-cators. Appendix D lists the fabricators interviewed andvisited. Asphalt application procedures are relativelyconsistent among applicators. Pipe is brought in fromstorage, cleaned of loose dirt (not always done), preheatedat some plants and dipped in the liquid asphalt. Asphalt issupplied by refiners to conform to AASHTO SpecificationM190-78. The AASHTO specification is a physical propertyrequirements specification and does not detail applicationprocedures. Guidelines are published by the National Corru-gated Steel Pipe Association (NCSPA) but there is no indi-cation that these are generally followed.

24

Asphalt bath temperature varies between applica-tors, ranging between 182-246°C (360° and 475°?), however,individual applicators seem to maintain a temperature rangeof about 3.9°C (25°F) or less. Some states specify theapplication temperature and range and some specify pipeimmersion times in their specifications.

Pipe immersion times are critical to asphaltadhesion and are established through experience at someplants or as the result of previously performed laboratorytesting at others. Most plants seem to rely on operatorexperience in judging the proper immersion time. Twomethods are used to obtain the minimum required thickness of.127 cm (.050-inch), that is by single or double dipping.Double dipping involves immersion of the pipe for a fewminutes (exact time depends on metal thickness) to bring themetal up to the bath temperature so that the asphalt thor-oughly wets the metal. This is said to provide the bestbond between metal and asphalt. A second dip is made toachieve the minimum thickness. Some states (detailed inTable XII) specify application procedures but most do not,which essentially leaves the entire application procedure upto the fabricator.

It appears that asphalt performance might be im-proved by: more stringent surface preparation techniques,uniform asphalt application temperatures, uniform immersionschedules and by specifying double dipping. We do not knowat this point the optimum application procedures but suggestthat the guidelines published by NCSPA be used until betteimethods are established.

Surface preparation is critical to the adhesion ofa barrier coating such as asphalt. As a minimum, the metalsurface should be thoroughly cleaned of oil, grease, soil,zinc corrosion product, and completely dried. Acid oralkaline etching, phosphating, chromating or the use of aprimer are used to promote adhesion of coatings and might beuseful to improve asphalt performance. These possibilitiesare discussed in more detail in the section on Galvanizing.

All of the asphalt coated aluminum culvertsexamined exhibited less adhesion between the asphalt andaluminum than found between asphalt and galvanized steel

.

Asphalt could be easily pryed loose from the aluminumprobably due to the smooth surface profile of the aluminum.Surface preparation procedures to increase the surfaceprofile such as acid or alkaline etching, light sand blast-ing, anodizing, oxide coatings, phosphating or chromatingcould be used to improve adhesion. Comparative tests are

25

needed to determine which would be cost effective. Asphaltcoated aluminum culverts are generally not used but manystate specifications allow the material. Present specifi-cations call for essentially identical coating procedures tobe used for both steel and aluminum.

The properties of asphalt vary due to the dif-ferent sources of crude oil but it is not clear how theseproperties vary with culvert asphalt or how the variancesaffect asphalt performance. This is the general implicationbased on discussions with applicators, refiners and theliterature search.

Asphalt is a complex substance produced fromnatural deposits, coal tar or petroleum. Culvert asphalt isproduced from crude oil products by air injection (blown) atan elevated temperature. The blowing operation convertsasphalt constituents to heavier oils then to asphalticresins and then to asphaltenes. Asphaltenes briefly de-scribed are organic polycondensates containing aromatic andnaphthenic rings with aliphatic side chains 6

. The effectof blowing is to increase the fusion point and decrease thespecific gravity. Asphaltenes promote hardness and a highsoftening point while asphaltic resins promote ductility andincrease the breaking point. Blown asphalts differ fromresidual asphalts in that residual asphalts are derived fromsteam distillation of semi-asphaltic or asphaltic petro-leums. Blown asphalts are considered more weather resistantthan residual asphalts having the same source, fusing pointand volatile matter 6 3

. Typical constituents of blown as-phalt and composition ranges are: sulfur (trace-7. 5% )

,

paraffins (0-10%), oily constituents (12.2-30.5%), saturatedhydrocarbons (30-7 5%), minerals (0-.5%), carbenes(methylene) (0-8%), asphaltenes (17-42%) and asphalticresins ( 17. 8-41%) 6 k ' 6

7

. As seen, asphalt composition canvary greatly depending on the source of crude (Middle Eas-tern, Venezuelan, Mexican, Gulf Coast, California,etc.) 61*' 6 5 ' 6 6 ' 6 7

. Variations in composition can occur fromthe same source at different times.

The effects of these variations on the performanceof asphalts as applied to culverts do not appear to be wellunderstood. For example, an increase in asphaltene contentwould seem to improve abrasion resistance but would alsolessen impact resistance. The effect of asphaltenes onadhesion and water penetration are not known. Blown petro-leum asphalts are refined to meet the requirements of AASHTOM190 and in-house specifications of the refiner. Table Villustrates some typical specification limits of culvertasphalt as reported by the refiners. Table VI compares

26

AASHTO specifications with some compositions of blown as-phalt obtained from the literature. A narrow range is seenbetween the asphaltene content and penetration whereas otherconstituents vary over wide ranges. Thus, it is possiblefor wide variations in asphalt characteristics to occurwithin AASHTO specification limits.

The acid number of asphalt could also hold animportant clue into its performance behavior. The acidnumber is a measure of the constituents in the asphalt whichhave acidic characteristics such as organic acids, inorganicacids, esters, phenolic compounds, lactones, resins, heavymetal agents and additives. Acid numbers for asphalts canrange between .02-5. 3

6

k

. Two patents were issued in 1943 68

for additives used to neutralize asphaltic acids for betteradhesion to steel. We do not know if significant benefitcan be derived from these methods. The effects of thevarious components of asphalt on adhesion, abrasion andimpact resistance should be investigated to determine ifalterations in specification requirements are advisable.

Additives are used with asphalt to improve ad-hesion to various aggregates for highway paving. Superioradhesion between asphalt produced from Arkansas crude oiland steel was reported in separate correspondences betweenthe PSG investigator and Lion Oil Company and Armco Steel.This asphalt was also reported to have an unknown proprie-tary additive. Additives are not used by culvert fabrica-tors to increase the adhesion of asphalt to steel or alumi-num (except as previously noted) although several manu-facturers, notably Armco Steel and Republic Steel, haveconducted tests to determine feasibility. Some additivesmight be applicable to increasing adhesion to galvanizedsteel. These include: branched polyalkylenepolyamines 6

9

,

f luorocarbons 7 °, polytheylenepolyamine and synthetic fatty

acids (high thermal stability reported) 71, .5% polyamine or

.25% polyamine +.25% stearic acid 72 and stabilized abietyl-amines (Amine D) 73

, petroleum resin+1-pentene (applied toprimed steel) 71

* and petroleum resin+2-methyl-l, 3 butadiene(applied to primed steel) 7 5

. Others, too numerous to listin this report, are reported in the literature to improveasphalt adhesion to aggregates. Reference 63 lists many ofthese additives. Of these additives, we learned of only onewhich was tested and found to improve adhesion significant-ly. Amine D, a high molecular weight amine, derived frompine resin acids, was found to improve adhesion to gal-vanized steel by a factor of 4 in laboratory tests when usedin a concentration of about 1 percent. However, the addi-tive presents difficulties in stability at the elevated

27

asphalt bath temperatures as it decomposes rapidly at as-phalt application temperatures according to manufacturer'sdata. Such problems would require extensive analysis of theasphalt to assure adequate concentration. Coating cost alsoincreased about 10 percent. Laboratory testing is needed toscreen a large number of additives to determine if anyadditive would be cost effective.

Water absorbs into asphalt where immersion condi-tions exist76 '

8 5 and tends to cause disbondment from thezinc coating. This undoubtedly explains the great dif-ference in performance between immersed areas and unimmersedareas. Better adhesion between the asphalt and substratewould improve coating performance. Better surface prepara-tion techniques will improve the bond.

Asphalt coatings are also subject to abrasionforces which cause the asphalt to flow and wear down.Methods to improve abrasion resistance include increasingthe asphaltene content, use of filters and blending. Anincrease in the asphaltene content would increase erosionresistance but would probably decrease impact resistance andweatherability. Penetration resistance is related to as-phaltene content and penetration is greatly affected byexposure temperature 53

. Increasing the asphaltene contentmight make the coating too susceptible to impact damageduring the winter months.

The use of fillers to improve erosion resistanceshould be investigated, particularly for use as invertpaving. Possible filler materials include sand, fiberglass,asbestos or steel. One manufacturer is currently investi-gating the use of reinforced asphalt, using tailings fromanother process. No data is available yet.

A blend of asphalt and coal tar is said to improveadhesion and weathering properties. The blend is alsoreported to have decreased susceptibility to temperaturechanges and increased viscosity at elevated temperaturesCoal tar will evidently not mix well with asphalts in allproportions. The optimum concentration range for improvingasphalt performance is reported to be 15-25 percent coaltar. We were not able to find comparative data. Blendingcoal tar and asphalt appears to be worth investigating.

Abraham presents several other methods of im-proving asphalt weather-resistance 63

. Since we do not knowdetails of the manufacturing processes used in refiningculvert asphalt, some of these might be used now, however,in the interest of completeness, they are, briefly:

28

1. Separate the light and medium lubricating oils from theasphaltenes and resins. Remove the oily constituentsfrom the lubricating oils and reflux the residue withthe asphaltenes and resins - air blow the mixture.

2. Remove resins of low molecular weight from the mixtureof asphaltenes and resins.

3. Separate the oily constituents from the asphaltenes andresins. Treat the oily constituents with a selectivesolvent and recombine with the asphaltenes and resins.

4. Incorporate oxidation inhibitors.

Coating Need and Alternative Means of Protection

Five states (Arkansas, Nebraska, New Mexico, NorthDakota and South Dakota) have never used organic coatingsand four others (Hawaii, Kansas, Maryland, Tennessee) haveceased using them. Most States have criteria which allowthe use of uncoated galvanized steel, aluminum and sometimesother metallic coated pipe. The criteria are usually basedon pH and resistivity data and a certain design life as wellas structural requirements. Other criteria sometimes usedrelate to the type of road and traffic volume. Table VIIlists known criteria used by several states for permittingthe use of corrugated pipe without organic coatings. Manyof the test sites inspected fall into the ranges in thecriteria which permit uncoated pipe but a coating was usedpossibly for abrasion protection. It is also possible thatexposure conditions have significantly changed. An alter-native means of protecting the invert would be economicallyjustifiable in these cases if abrasion resistance weredesired

.

Protective coatings are used to extend culvertservice life where either corrosive conditions exist orabrasive flow is expected, or both. The literature searchand field survey have shown that asphalt type coatings areeffective in controlling culvert deterioration on exteriorsurfaces but are not entirely satisfactory on internalsurfaces. Polymer coatings can be provided on either theinternal or external surface or both, depending on whereincreased durability is needed. Performance is questionablein abrasive or very corrosive environments. Other means ofextending culvert life can be explored.

Exterior deterioration can be controlled throughthe use of cathodic protection either by using sacrificialor impressed current anode systems. Cathodic protection of

29

a large uncoated culvert can involve large numbers of anodesin order to achieve corrosion control. The cost of theanodes, engineering time necessary to determine currentrequirements in the field and inevitable maintenance costs(mandatory on impressed current systems) can easily surpassthe present cost of coating. Cathodic protection could bean economic choice in very severely corrosive environmentsin conjunction with a protective coating. Polarizationtests on externally coated culverts_

[+during the field study

indicate that an average of 3.4 x 10 ampere per sq. m (3.1x 10 ampere per sq. ft.) is needed to project the externa^culvert surface. The range_

7is 3.23 x 10 ^ to 1.81 x 10

ampere per sq. m (2.99 x 10~ to 1.67 x 10~ ampere per sq.ft.). An average of 3.4 17 pound magnesium anodes (range:1-18 ) would be needed for 50 year anode life at an averagematerial cost of $91 (range: $27-$486). The cost of engi-neering tests and installation would raise the total cost ofcathodic protection.

Coating and external cathodic protection might beused in a cost effective manner if ungalvanized culvertmaterial were used. Galvanizing costs $7.42 per hundred-weight ($.163 per kg ) , the average estimated galvanizingcost of the culverts tested being $1026 (range: $64-$2619).Substantial savings might be realized in many installations,however, the total cost of a cathodic protection systemcould exceed the galvanizing cost particularly on smallculverts. Cathodic protection can not be used to protectnon-submerged surfaces and would not be effective in pro-tecting surfaces exposed to alternating wetting and dryingconditions. Anodes would be subject to damage if installedunprotected in the stream flow. An effective coating wouldbe needed to protect internal surfaces.

Abrasive bedloads could be prevented from enteringthe culvert through the use of settlement basins or screen-ing devices. Energy dissipators could be used to lessenstream velocity. Screening devices (i.e. cribs) are used insome cases but are not entirely effective as shown inFigures 13 and 14. Settlement basins might be the bestalternative as they would trap abrasive debris but notaffect the hydraulic efficiency of the culvert. All threemethods require periodic maintenance by highway personnelwhich would be a disadvantage.

Reinforced invert pavings or vitrified clay tileare recommended where abrasive stream flows are encounteredand could be used in place of organic coatings where corro-sive conditions are not a consideration. Reinforced Port-land cement or reinforced asphalt could be used although

30

Portland cement would probably provide the best abrasionresistance. Reinforcing should consist of wire mesh orsimilar steel reinforcing welded to the invert. The cost ofusing reinforced invert protection will probably vary great-ly. Reinforced concrete would be field applied and be laborintensive. Concrete paving would also be limited to cul-verts large enough to allow the applicator to work effi-ciently. Reinforced asphalt could be shop applied usingexisting equipment but would require improved surface prep-aration. The use of fillers, such as fiberglass and metalturnings, mixed with the Portland cement and asphalt, shouldbe investigated.

Other inorganic coatings for invert protectioninclude metallized coatings, clad metal liners, ceramiccoatings, fiberglass lining and clay lining. Mr. H. Johnsof the Bureau of Reclamation is evaluating several metal andceramic coatings for their erosion resistance in drainageapplications 79

. These include stainless steel sheet (whichcould be provided as mill applied metal lurgically bondedcladding on the steel culvert sheet), metallized stainlesssteel, metallized bronze, metallized high chrome stainlesssteel, metallized molybdenum and flame sprayed aluminumoxide. The tests are being conducted using rotating steeldrums coated with the test material and containing water andsilt, sand or gravel. Test results were compared to coaltar enamel. Coal tar enamel has exhibited up to seven timesthe abrasion resistance of asphalt according to tests simi-lar to the Bureau of Reclamation program 80

. Of the metalliccoatings tested, the stainless steel sheet and metallizedstainless steel outperformed the coal tar enamel and all butthe flame sprayed aluminum oxide might be expected to out-perform asphalt based on abrasion resistance. There isinsufficient information to estimate the useful life ofthese other relatively thin coatings compared to the greaterthickness of asphalt pavings. Flame sprayed coatings varygreatly in cost, and are more expensive than presently usedcoatings on a square foot basis. However, since flamesprayed coatings can be applied only on the invert, totalcost could be lower. Table VIII presents typical flamespray material costs. A possible disadvantage to flamespray coatings is that they must be applied to sandblastedsteel which would require the removal of the galvanizing orthe use of ungalvanized culverts sheet.

Fiberglass might also be an effective selectionfor invert protection. The Bureau of Reclamation tests havealso included several glass reinforced resins to includepolyester resins filled with Al2Si03, sand and glass fiber.Both performed as well as or better than the coal tar enamel

31

standard. Industry sources, however, claim that adhesion togalvanized steel is poor and application to corrugated metalcostly. The cost range for fiberglass application is$64.80-$216 per sq. m ($6-$20 per sq. ft.) Fiberglass mightbe better applied to concrete culverts as invert protection.

Although the epoxy coatings appear satisfactoryfor protecting concrete in acidic environments, there arealternative means which might be advantageous, particularlyin environments corrosive to concrete. These methods in-clude sealants such as those being evaluated for- use onbridge decks to prevent chloride penetration. Possiblesealants include wax8 1

, methyl methacrylate, styrene,acrylonitrite and t-butyl styrene82

. Concrete mixtures withpolymer latex, epoxy resin and polyester resin could also beevaluated as invert paving in acidic or abrasive areas 79 ' 82

.

The cost of surface treatments is difficult to predict sincemost of the work done has been experimental in nature. Onesource estimates the cost of impregnating (bridge decks) tobe $5.40 to $10.80 per sq. m ($0.50"to $1.00 per sq. ft.) 82

.

The cost of applying a polyester resin, polymer latex orepoxy resin concrete invert paving will probably not be muchdifferent from that of current epoxy systems.

Polyethylene liners could be used to provideadditional protection to metal or concrete culverts incorrosive or abrasive environments or to increase the ser-vice life of deteriorated culverts. Such liners, which arefield installed, are currently used to rehabilitate sewerpipe. One application to drainage culverts is known inMichigan where galvanized steel deteriorated along 17 5.

Polyethylene liners are available in sizes ranging between1.9 cm (.7 5 in.) to 1.22 m (4 ft.). The thickness of a thinwall liner is based on the diameter with a diameter tothickness ratio of 32. The cost of using polyethylene toline corrugated metal or concrete culvert pipe is in rangesfrom $13.39 per sq. m ($1.24 per sq. ft.) for 15.2 cm (6ft.) to $83.81 per sq. m ($7.76 per sq. ft.) for 1.22 m (4ft.) diameter pipe. The best use of this material wouldseem to be as protection for the culvert against corrosivestreams. The material also might exhibit acceptable abra-sion resistance.

Alternative Coatings

Asphalt and several precoated polymer coatings(coal tar laminate, polyethylene, polyvinyl chloride andvinyl plastisol) are now used for culvert protection. Theliterature 79

'83 suggests several post applied organic coat-

ings which might provide improved performance. Two of the

32

most abrasion resistant are polyurethane and neoprene whichhave abrasion resistances several times that of coal tarenamel . Table IX lists organic coatings which might becandidates as alternatives for existing systems and theirapproximate application costs. In preparing this list, theprimary characteristics were the abrasion and impact resis-tance. Important characteristics for culvert coatingsinclude water penetration, chemical resistance and adhesionas well as impact and abrasion. Comparative test data onthese other properties is not available. The screening ofthese alternative coatings should be the subject of a la-boratory program which would select suitable coatings fortesting under actual field conditions.

Coal tar enamel is included in Table IX since itwould be a low cost alternative to asphalt and could beapplied in the same manner. Plasticized coal tar enamel hassuperior water absorption properties to asphalt 8 5 and has ahigher penetration coefficient which should give it greaterabrasion resistance as indicated in the literature80 . Coaltar is, however, subject to the same adverse weatheringbehavior (on exposure to infrared radiation) as asphalt.Another problem is its possible carcinogenic effect which italso shares with asphalt. Coal tar enamel would almost cer-tainly require a primer to assure adequate adhesion.

The cost figures in Table IX are seen to be con-siderably greater than the conventional coatings. This isbecause the cost figures contain a large application costallowance for cleaning and spray application and multiplecoats are needed in some cases. Costs would decrease ifautomated application procedures could be used. Many ofthese coatings have the distinct advantage of application tospecific areas where they are needed, such as the invert.For example, a polyurethane elastomer could be applied toone-half the interior surface to protect the invert at acost of about $5.14 per ft. compared to $9.57 per ft. forasphalt paving (using the pipe size example given in TableIX).

Alternative coatings can also be applied to alumi-num culverts. Asphalt is the only coating now used onaluminum but at least one manufacturer, Kaiser Aluminum, istesting a polyethylene laminate manufactured by Dow ChemicalCompany.

Two alternatives to galvanizing are inorganic zincand Zincrometal. Zincrometal is a trademark of the DiamondShamrock Company but its use is licensed to many steel

33

companies and is a mill applied process. The coating con-sists of 2.54 ym (.0001 in) of a proprietary zinc-chromiumwater based dispersion with a second coat, 10.16 ym (.0004in) thick, of a zinc rich weldable epoxy. It is applied toautomotive sheet metal prior to fabrication at the presenttime. The cost for coating both sides of a sheet is in theorder of $110 per 908 kg (1 ton) or $5.50 per 45.4 Kg (1cwt) compared to $7.42 per 45.4 Kg (1 cwt) for galvanizing.The manufacturer claims that it is compatible with mosttopcoats and is probably compatible with hot applied as-phalt. Disadvantages include the lack of sacrificial pro-tection and a limit on the thickness sheet that can becoated (up to about .203 cm/. 080 in) at the present time.

Inorganic zinc is a spray applied coating whichdoes offer sacrificial protection. This coating must beapplied to a white metal sandblasted surface (Steel Struc-tures Painting Council SP-5). The material cost wouldamount to $1.19 per sq. m ($0,111 per sq. ft) at a dry filmthickness of .0076 cm (.003 in), sandblast would cost about$10.76 per sq. m ($0.12 per sq. ft) and application about$1.29 per sq. m ($0.12 per sq. ft) bringing the total costto $13.24 per sq. m ($1.23 per sq. ft). The cost to coatboth sides of the culvert example in Table IX would be$33.38 compared to $3.26 for hot dip galvanizing. We do notbelieve that benefits would be derived using inorganic zincsufficient to justify the cost.

Fusion bonded epoxy powder coatings or polyure-thane/polyester powders could offer a significant costbenefit for culverts. Fusion bonded powders exhibit goodabrasion 81 and weathering resistance as well as resistanceto chemicals 87

^88

. A fusion bonded epoxy coating is beingevaluated by Armco Steel Company for use on culverts and oneis approved for potable water pipelines (AWWA SpecificationC213-79). This epoxy is intended for application overnongalvanized steel. Fusion bonded powders can be appliedover galvanized surfaces but blistering and film continuityproblems can be encountered due to trapped moisture andgases. Prebaking the steel prior powder application mightalleviate the application problems. Powder coatings appliedover uncoated steel could save the cost of galvanizing ifproven affective in field tests. Additional protection suchas paving would be needed in abrasive environments. Corro-sive soils might also require additional protection such ascathodic protection as discussed in the section on CoatingNeed and Alternative Means of Protection. Epoxy coatingstend to chalk on outdoor weathering exposure which could bea disadvantage at the ends of the pipe.

34

Adhesion to Galvanized Steel

The adhesion of coatings to galvanized steel isoften less than satisfactory for several reasons including:poor surface profile, silicate coatings applied at the millto prevent white rust, zinc corrosion products and out-gassing of the zinc during high temperature coating opera-tions. The minimum surface preparation normally recommendedfor galvanized surfaces is cleaning and degreasing. Ad-vanced methods include acid or alkaline cleaning, surfaceconversion coatings, wash primers, surface conditioners orsandblasting. Several methods can be used in combination.

The field study indicated the need to improve theadhesion of asphalt to steel, particularly in the invertarea. Several studies have shown that asphalt adhesion issignificantly improved when surface etching or phosphate orchromate conversion coatings are used. Surface treatmentwould add to the coating cost of a culvert. It is possibleto limit the surface preparation to the invert area to lowerthe cost, but unless the culvert is installed correctly, themethod would not be effective.

Typical costs are:

Acid or alkaline cleaning(mill applied) $1.08/m2 ($.10/ft. 2

)

Phosphate conversion coating(mill applied) $1.19/m2 ($.ll/ft. 2

)

Chromate conversion coating(mill applied) $1.94/m2 ($.18/ft. 2

)

Wash primers, conditioners& Primers $1 .49-1.73/m2 ($ . 14-. 16/ft

.

2)

Sandblasting (SP-7) $2.70/m 2 ($.25/ft. 2)

Cleaning & Degreasing $2.70/m2 ($.25/ft. 2)

Test data is not available to determine which, ifany, of these surface preparation methods is cost effectivein a culvert installation. The scope of this study includeda screening test of various wash primers and primers whichmight be used to improve the adhesion of asphalt or othercoatings. Two test methods were used in the screening test,they are: ASTM D2197 Scrape Adhesion test and the Elcometerpull-off test which measures the tensile bond strength of

35

the coating to its substrate. Both tests were performed on18 primers and wash primers and asphalts obtained from 5

sources. Primers were applied according to manufacturersinstructions by brushing onto cleaned and degreased flatgalvanized steel culvert sheet. Asphalt coatings wereapplied by dipping the panels into asphalt at 204°C (400°F)for 3 minutes. Tests temperature was approximately 21 °C(70°F). Asphalt coated panels were also tested at 0°C(32°F). Three of the asphalt primers were tested with oneasphalt topcoat at 21 °C and 0°C.

Table X presents the adhesion data for the as-phalts and primer combinations. The data presented inTables X and XI indicate that there are only a few primerswhich exhibit adhesion strengths to cleaned and degreasedgalvanized steel greater than that of hot dipped asphalt.Some improvement in asphalt adhesion can be obtained byusing one of these primers with the hot applied asphalt. Anexception occurred with the PA- 98 primer in which theadhesion strength actually decreased. This unexpectedresult was replicated and could be caused by a degradationof the primer under heat. Significant changes are observedwith the other two primers exposed to hot asphalt. Thesignificant problem of asphalt adhesion after water hasabsorbed into the coating was not addressed in this program.The ultimate effectiveness of a primer will depend on howwell it maintains its adhesion under immersion conditions.This question should be addressed in future evaluations.

Data indicated on Table XI that the adhesivestrength of the asphalt varies considerably even though allof the asphalts tested are supposed to meet the AASHTO Ml 90specification. Scrape adhesion strength is seen to increaseat the low test temperature most likely because the asphaltitself becomes more resistant to flow. The failure mode at0°C is one of brittle chipping and that at 21 °C is one ofplastic flow.

Scrape adhesion test data and Elcometer test datafollow only a rough correlation with asphalt at 21°C. Thecorrelation coefficient is .82 but the correlation is barelysignificant at a confidence level of 95 percent, perhaps dueto the few data points. The correlation coefficient at 0°Cis only .50, and in the comparison with the other primers,the lack of correlation is evident in the data. The reasonis that the cohesive strength of the coating is sometimesmore dominant than the adhesive strength with the Elcometermethod. An adhesive must also be used with the Elcometertest which could affect the coating in some way. The El-cometer test shows reasonable correlation then only whencomparing coatings of the same type such as asphalts.

36

Specification Review

State and American Association of State Highwayand Transportation Officials (AASHTO) specificationsregarding culvert pipe protection were reviewed. The reviewwas based on the information gathered during the literaturesearch, field survey and asphalt review phases of the study.Each specification was examined for the adequacy of itsprovisions with regard to type of coating, applicationmethods and installation procedures.

Most state specifications for asphalt and polymercoated culvert pipe refer to AASHTO Ml 90, M24 3 and M24 6 anda few supplement the AASHTO specifications with additionalrequirements. There are several subjects that few State orAASHTO specifications address. These are: transportationand handling of coated pipe, surface preparation prior tocoating and backfilling to prevent coating damage. TableXII presents a comparison summary of the various statespecification provisions. Appendix E presents a recommendedspecification for coated culvert pipe which could be used asa guide to revise existing specifications.

AASHTO specification M190-7 8 provides the basicreference for applying asphalt to galvanized and asbestosbonded steel. This is basically a materials specificationwhich calls out the properties of the asphalt used and thetypes and thickness of coating. Since the specification isoften used unaltered and unsupplemented by the states, itshould contain more specific information regarding surfacepreparation, application and handling of culvert pipe.

AASHTO M190 does not specify the means of applica-tion, i.e. spray, dip or brush. Dipping is the method usedbut the specification does not delineate the procedure thatshould be used to obtain satisfactory adhesion. We recom-mend that the practice called out by the Corrugated SteelPipe Association be included in AASHTO M190 unless abetter procedure is found. The NCSPA recommended practiceincludes minimum immersion times, asphalt temperature rangeand requires double dipping.

The physical property requirements should beexpanded to include penetration at 0°C (32°F) - 2 5 min. at200g for 60 sec, penetration at 25°C (77°F) - 25 to 55 atlOOg for 5 sec. by ASTM D5 or AASHTO +49, flash point byASTM D92 or AASHTO T48 - 232°C (450°F) min., specific gravi-ty by ASTM D7 or AASHTO T2 29 - .98 min. and softening pointby ASTM D36 or AASHTO T53 - 93°C (200°F) min./110°C (230°F)max. These suggested values are also referenced from the

37

NCSPA practice with similar limits recommended by the Cana-dian Corrugated Steel Pipe Institute (Specification No.501-78, Revision 11, April, 1978).

Mechanical property requirements in AASHTO M190are limited to shock testing, flow tests and imperviousnesstest. These tests should be augmented with four othertests, notably: abrasion resistance by ASTM G6, D658 or amodified procedure, adhesion test similar to that specifiedin AASHTO M24 3-7 5 by ASTM G9 and freeze thaw resistancesimilar to that specified in AASHTO M246-78. Acceptablevalues for these tests should be determined through a testprogram.

Section 5.1 of ASSHTO M190 calls for asphaltsamples to be taken from strippings removed from the pipe.We recommend that this be changed to specify that the sam-ples be removed from the asphalt bath at the time of coat-ing. This will eliminate the possibility of contaminatingthe asphalt with dirt or anti-sticking compounds. Pipe fromwhich samples are taken are unlikely to be installed withthe uncoated areas always at the top. These areas could beexposed to the stream flow resulting in earlier coatingdisbondment and culvert deterioration.

AASHTO specification M243-75 applies to the ma-terial and application of field applied asphalt mastic andtar based material. Both materials should be subject to thesame performance tests required for asphalt, i.e. shock,flow and imperviousness as well as the additional testsrecommended for asphalt (abrasion, adhesion, water penetra-tion and freeze-thaw) . Currently, the asphalt mastic issubject to flow, acid and alkali resistance, pliability andadhesion tests. The tar base material is not subject to anyperformance test.

AASHTO specification M246-78 provides requirementsfor precoated galvanized steel sheet for culverts and under-drains. We can not recommend any changes to this specifica-tion other than to suggest that the average number of per-missible holidays (Section 7.4 Holidays) be deleted. Eachholiday is a source of blistering and disbondment underexposure conditions. Coatings should be as defect-free aspossible

.

38

CONCLUSIONS

1. Durability problems are encountered with all protectivecoatings now commonly used.

2. Alternate methods are available to protect culvertsother than organic coatings and could have been used toadvantage at many of the locations inspected in thefield study.

3. Organic coatings are, by themselves, not satisfactoryunder abrasive stream flow conditions.

4. The durability of polymer coatings depends on theamount of salts in the soil or water, the continuity ofthe coating, the pH and the abrasiveness of the bed-load. Improvements are needed in production techniquesto prevent damage which adversely affects performance.Polymer coatings are satisfactory where abrasive flowsand high salt conditions are not encountered.

5. Asphalt adhesion to aluminum is poor. This coatingwould not be satisfactory in abrasive or corrosiveenvironments

.

6. Epoxy coatings and vitrified clay liners are effectivewhen used on concrete in acidic streams. They mightalso be useful on corrugated metal under certain severeconditions

.

7. Adhesion between asphalt and galvanized steel can beimproved through the use of surface treatments andprimers. The benefits of improved adhesion should beevaluated.

8

.

Asbestos bonded asphalt coating is somewhat more dur-able than plain asphalt coating but is also subject todeterioration in abrasive or high salt environments.

9. The durability of asphalt coatings is influenced byapplication procedure, adhesion to the substrate,seasonal temperature changes, water asorption, tur-bulence in the stream flow and abrasiveness of thebed load. Asphalt is satisfactory where abrasive flowsand high salt conditions are not encountered.

10. Asphalt mastic is not a durable coating.

11. Asphalt composition varies widely depending on thesource of crude oil. Performance variations of culvert

39

asphalt are attributable to the water absorption andabrasion properties of asphalt and current methods ofapplication.

12. There are several alternative coatings which should beevaluated for use on culverts. These coatings, whilemore expensive than current culvert coatings, could becost effective for selected application such as oninverts.

13. Many state and AASHTO specifications should be mademore specific.

40

RECOMMENDATIONS

1. Evaluate by means of laboratory and field tests, me-thods of improving the bond between asphalt and gal-vanized steel and determine which, if any, are costeffective. Appendix F presents a program for this andother recommended evaluations.

2. Evaluate by means of laboratory and field tests, me-thods of improving the abrasion resistance of asphalt.Appendix F presents a recommended test program.

3. Investigate the variations in asphalt composition andthe effects on asphalt erosion resistance and adhesionto galvanized steel. Appendix F presents a recommendedtest program.

4. Evaluate by means of laboratory and field tests, alter-native organic coatings for use on culverts in corro-sive environments. Appendix F presents a recommendedtest program.

5. Utilize organic coatings only for protection in corro-sive environments. Use additional invert protectionwhere abrasive bed loads or rapid stream flow are anti-cipated. Consider using reinforced concrete as invertpaving until methods of improving asphalt adhesion anderosion resistance are developed. Consider asphaltpaving over polymer coatings.

6. Seal the space at joints between pipe sections toreduce turbulence and lessen the chance of coating andculvert deterioration at these locations. Only theinvert area exposed to normal stream flow conditionsneed be considered for this treatment. A hot pouredasphalt would probably be the most effective sealant.

7. Recommend changes in state specifications to improvecoating uniformity and to detail application, handlingand installation procedures. Appendix E presents a

recommended specification.

41

FIGURE 1 - Removing Core Sample From Culvertto Inspect Exterior Coating

FIGURE 2 - Expandable Rubber Plug Used toSeal Core Hole

42

5BJ Tm n

HI

I

E=E=E=E

:J1:+HO-

-ej-zffit

-Ifl-

:C3:

-61-

=33:

-M- ;p5=

E11=

Sfr-

=aa=

=a£

It:

:t5p:

pn: :lEt -?f :

jjj: je;MSt

:i:c

:jj: mF35 :

:«t tsi::??:

~\\V:9=

:£-m :isp:

-FJ-

-=3= 1 :!!t: 4&i-t3- -£3=1

I mmim 3S43A£-iV- "*i~- -Ed-

ElS3=t35EEI US ;-:E^i !33=23:

ffi~1

ffi=

frft

+H

S tttj

FIGURE 7 Field Applied Asphalt MasticSite No. 7

FIGURE 8 - Field Applied Asphalt MasticSite No. 6

47

FIGURE 9 - Field Applied Asphalt MasticSite No. 29

• i

-*

FIGURE 10 - Asphalt. Site No. 16

48

FIGURE 11 - Asphalt. Site No. 58

FIGURE 12 - Asphalt. Site No. 62

49

FIGURE 13 - Crib to Control large debristhat might plug culvert. SiteNo. 63

FIGURE 14 - Asphalt Coated Pipe in Figure13. Site No. 63

50

FIGURE 15 - Perforation of Field Coated ConnectingBand. Site No. 18

FIGURE 16 - Perforations at Section JointsSite No. 5

51

FIGURE 17 - Perforation at Section JointSite No. 19

FIGURE 18 - Asphalt Coated Aluminum,Site No. 25

52

FIGURE 19 - Asphalt Coated Aluminum,Site No. 7 9

FIGURE 20 - Asbestos Bonded AsphaltSite No. 28

53

FIGURE 21 - Asbestos Bonded AsphaltSite No. 50

FIGURE 22 - Asbestos Bonded AsphaltSite No. 65

54

FIGURE 23 - Asbestos Bonded AsphaltSite No. 43

FIGURE 24 - Coal Tar Laminate. Site No. 12

55

FIGURE 25 - Coal Tar Laminate. Site No

FIGURE 26 - Coal Tar Laminate. Site No. 96

56

FIGURE 27 - Coal Tar Laminate. Site No. 97

FIGURE 28 - Coal Tar Laminate. Site No. 105

57

FIGURE 29 - Coal Tar Laminate. Site No. 6 6

FIGURE 30 - Polyethylene. Site No. 70

58

FIGURE 31 - PVC. Site No. 71

FIGURE 32 - Coal Tar Epoxy on ConcreteSite No. 1

59

FIGURE 33 - Epoxy on Concrete. Site No. 8 6

FIGURE 34 - Gunite on Galvanized SteelSite No. 92

60

FIGURE 35 - Concrete on Asbestos BondedAsphalt. Site No. 27

FIGURE 36 - Concrete on Asbestos BondedAsphalt. Site No. 34

61

FIGURE 37 - Aluminized Steel. Site No. 44

FIGURE 38 - Galvanized and Aluminum-ZincSite No. 21-22

62

FIGURE 39 - Galvanized. Site No. 22

FIGURE 40 - Aluminum-Zinc (Galvalum)Site No. 21

63

^r

'*'

-m '

FIGURE 41 - Utah Test Site near Henderson

FIGURE 42 - Kentucky Test Site at Morton's Gap

64

FIGURE 43 - Colorado Test Site, Dillon

FIGURE 44 - Louisiana Test Site near Hackberry

65

TABLE I

LITERATURE REVIEW RESULTS

(SEE APPENDIX B)

M fr> 0.

s a s &

B

8

g2u

j 0.

o Xu

O U>

2 ga o

s co <2

8

ALABAMA X X X X X X X X X

ALASKA X X X X X X X X X X

ARIZONA X X X X X X X X

ARKANSAS X X X X X X X XCALIFORNIA X X X X X X X X X X X XCOLORADO X X X X X X X X X X X X XCONNECTICUT X X X X X X X X

DELAWARE X X X X X X X

FLORIDA X X X X X X X X X XGEORGIA X X X X X X X XHAWAII X X X X X X X X X

IDAHO X X X X X X X X X X X XILLINOIS X X X X X X X X X X

INDIANA X X X X X X X X X X X X

IOWA X X X X X X X XKANSAS X X X X X X X X

KENTUCKY X X X X X X X X X X X

LOUISIANA X X X X X X X X X X X

MAINE X X X X X X -X X X

MARYLAND X X X X X X X X

MASSACHUSETTS X X X X X X X X X X X

MICHIGAN X X X X X X X XMINNESOTA X X X X X X X X X X X

MISSISSIPPI X X X X X X X X X X

MISSOURI X X X X X X X X X X

MONTANA X X X X X X X X X X X X

NEBRASKA X X X X X X X » X

NEVADA X X X X X X X X X X X

NEW HAMPSHIRE X X X X X X X X X X X XNEW JERSEY X X X X X X X X X XNEW MEXICO X X X X X X X X

NEW YORK X X X X X X X X X X

N. CAROLINA X X X X X XN. DAKOTA X X X X X X X XOHIO X X X X X X X X X X X X X X

OKLAHOMA X X X X X X XOREGON X X X X X X X X X X XPENNSYLVANIA X X X X X X X X X XRHODE ISLAND X X X X X X XS. CAROLINA X X X X X X X X X X

S. DAKOTA X X X X X X

TENNESSEE X X X X X X X X X

TEXAS X X X X X X X X X X

UTAH X X X X X X X X XXX X X

VERMONT X X X X X X X X X X

VIRGINIA X X X X X X X X X X

WASHINGTON X X X X X X X X X X X X

WEST VIRGINIA X X X X X X X X X X X

WISCONSIN X X X X X X X X X

WYOMING X X X X X X X X X X

66

TABLE II

Summary of Culverts Inspected

By State No. Sites

California 7

Colorado 7 (Includes 3 Test Sites)Kentucky 10 (Includes 1 Test Site)Louisiana 5 (Includes 3 Test Sites)New York 11Ohio 14Oregon 6

Pennsylvania 12 (Includes 1 Test Site)Utah 10 (Includes 1 Test Site)

No. CulvertsBy Coating Type Inspected

1. Coal Tar Laminate (U.S. Steel Nexon) 252. Asphalt dipped galvanized 22*3. Asbestos bonded asphalt 16*4. Epoxy coated concrete 9

5. Asphalt Mastic 9

6. Polyethylene (Inland Steel Black Klad) 6

7. PVC (Wheeling Steel Plasticote) 5

8

.

Asphalt dipped aluminum 4

9. Gunite paved galvanized steel 2

10. Vinyl Plastisol (Bethlehem Steel Beth Cu Loy PC) 3

11. Aluminum zinc (Galvalum) 1

12. Epoxy 1

13. Aluminized 1

14. Vitreous clay lined concrete 1

15. Unbonded polyethylene 1

16. Inorganic zinc 1

17. Organic zinc 1

*22 out of the 38 asphalt coated culverts were asphaltpaved

.

67

TO oO -i-

O !->

i#. • >> • >> >> >> >>

i/i c>- >> Q.+J CL4-> -+J *j !->

3 S +? s <_> 5 O • O «o O • >)OO >, (— o t— 1— a. cx+y

+-> T3 -o 2 -D n.-o • •o 2 OJZ C_> 0) TT (/> r— l/l 1

—

— i— 2i— »—u <J i— 3 0) 3 <D 01 h- oi Q- 0) DC TO C 01 TO i- (O -i- C •!- 3 f- C r—TO >— oi •!- .c n- -C i|- 0) 14- i/> it- I— <*- o a>S- <ll S- l»- +J S- (-> V -C s- er) s_ s- +> -i-

CD •- 3 i- s_ re S- 10 un to C7> ro in re m i4_

<4- IC TO (O 0) re 01 O 4) o ai en o> 3 i-• i. —1 0) ^ 1— St .— CB i— CO >— a>i— X re3 ro o o (_> i_> o o 0)

0) •.(_> COi—

1

00 " 00 » C\J « 00 - •—1 (_)

lO (_) to #, ID C\J l£> i—

1

ir> r-~ ro r~. -o «3"

o o o -C o o -C O vo O IT) O r»» CTl ^t .C O -^r—* •< r~ 10 u r~ + (J r-» + r-~ + rv. + VD + (J r^ in <_>

c i—i ro 1-1 + s_ rt CO s- i—c U3 i-i CO i-H in CO CM !_ iHCVJ l.o C 00 =t ro «S i—

t

O UO cr> «c o «cr— OL C or r~- ce >-" OS f-i oi in OS CPl ce «3- o; i-i+J —I (O _J 0) _1 cu _J _j _1 _i ai _i aito .c • -l-> p j-j • +j<-> • ai •> <0 TO » fO 03 m TO « re •> re « re TO * re reo < 3 «=C +J =c +-> =r 4J «c +J < *-> «I +j <C +J i—_j a. or Cu OO O- a. oo D_ D. 00 a. oo O- oo D- U1 o_ o. oo ex

in i— * cai J2o re a>«-^

h- +->

1 <_5 •

i<- j*Wl O • u01 ai •>-

+J -o > ^z•i- c: ct t—O0 Q)

M J-J

oo re01

en

o O i—t i-H

01 Ol 0) OlC LlJ c u-i c cOl— O h- o z: O Ez. o z <_> z <c Z cE

Olcz«t

01

-0ai

fc. ^~is TO

01 4-)>- OO

C'—

'

-68-

T3CD

o<-J>

co r*

•1

—

to+-> cfD ou •1

—

o +->

_J fO•1

—

T3 >c CDfO S-

-Oc -Qo <"r-P S-Q. o•i— <+-

S-

o CDto i

—

CD XIQ fO

I—

M-to OCDP-O*r- cto cu

+-> -t->

to fOCD1— W

CD+-> -MS- OCD -Zl>

i

—

CD13 CDo oo

CD CO * CDCO CO CO • c.

- =>-> * »> CO 3 CD "O 3• +J 4-> •> CD O CJ OT -^ 4-J

Q-O S- >> s- >> CJ in CJ CJ CJ% O -4-J s +-> u < — O 3\- C Q.O t— o < .— ^— or s-

o CO o +jc s- s- t- o c « o • X • CO«.' o> CD CD C (0 >»x: >> >>CL E •c +-> •i- CD Q. l_> a. s- a. •D- it) 3 +J +J ^ X CO X CD x •

i- CJ O O <o u UJ UJ c tej -OC C_> O- CD s: a o orto - ^r CD CD CD (-> CD< cd or CD CO CD r-••CO n cm « i—

i

CO CO CO .—er> + o r» J= CO «tf- .C CO M_ CO >4- CO -i-

o> CO •-H + CJ CT> + o CD O CD O CD 31en «a- s- CO s_ c C C

or «s- DT V£> < cr:co< CD • OJ • CD CT>

_j _i _j Cfl 3 O 3 CD C•» TO •» (O s. * (O s. •> . * . » S-< -l-> C +J < -u >- T3 >- -o >- O-

O- (/I D- CO D_ O- CO c_ z or s or z to

CD -i-

> .C< I—

O COt—i enO O

co enO CNlo o

X)

5

o

to

— en 3

o oO COrH O

en en

O OO enO r~-H O

< < luc_> luc_) «t=c «r «s < <

CD

03•aCJ

s- i

—

m (OOJ *->

>- toc1—1

69-

oO)=5E

•i

—

4->

Coo

co ft

•1— 00+-> EfO Oo •

i

—

o -0J

_l re•1

—

-o >c CDea s-

-Qc -Qo <•r™•M i-a. O•i— 4-s-

u CDCO r

—

CD -QO 03

h-

"4-

00 o<u+J o•r— eCO CD

»-> +Joo f0CL)

1— 00CD

+-> -t-J

i. O<D s:>i— CD3 CDC_) CO

o o • >>10 i~ -C s_ r^ S- 4J

oO u CO co <D • U c>- o • S- 4->

(. CO CO on co oo U Z i_ c 4- cn 3 oE «£ -c <o -c <c </) ct 4- k. O 0J (_>

10 +-> Cu +J Q. ••.c <o 0- o ai o ro o cu.c o- •r- 0. •i- S- CM <_> u cu cuco a c 3 M C+-> o E CU E CU CM l_ o •1- +J CU T3 c10 oo -o 1/0 -a <: (O C -i- 3 •>- 2: i- 1— i. -^ cu—I S- c c _J => O- *-> Eco o •i- O o> ii

CU M- Z3 M- => *-> a> •r- 0) • a. s: m- oc c O O CC O. It- « . E CU in en t-H L. cc cuCU S- r^ r^ O to Cl CO O 4_ »o 4->

15 • CO • 00 (-> CL- 0- E co s- ro cu |-H O t-H J* -1- 4->2 <— z »— IO • cu <o 01 r— O 4-> oj co c+J 00 C S- « +-> • xz 1-i l- 19 OOo> « «oO •>oC ^- CO s- co c vo E 4-) 4J o r— CLr—-0 4-> CM CSJ .—

< CO CM O +J CM >—

i

CM a; 3 • a. to aa01 co . ro • CM (0 CM OJ i- f—

»

»—t » o >> 4-> or-~ -i- D D a. X c c . oo 4-> a: H- * CT)

^— • cu • CC • 0. • i. cu • o • o OO OJ o O i- c>- > +j 4-> 4-> CU +J 3 -t-> +-> 4-> +J oo CO i- •»-Z 0. cc cu Ct CU ce > a o-d cc a> CtL CU => LO >> =3 JC M IO+J

OJ > > o U- c s- 4-> 4-> r-« 4-> « o •s o » 1 3 . Q. . CL «o s- " 3 >,

l-^"io H>- ro >- 1— >- r— ce 4- a s. o CC ro q: ro a: .-i 3 C£. O -MZ 3 z o Z <_) O o OOCQ o 2: o s: O CO 0J O oo oj 3 oo oo

0J+

LD LO

LO t—I r—

<

ro COr^ o O o O© O O o O

o o CO LO<£> CO CO CO CMo o o O i-H

oo

II <c < <t «t <: <c <c <:COCO CO CO CO 00

CD

-O

a0)

s- r-~

m ro

01 4->^ {Si

EC•~

. r_>

.

—

+J3 rooj s;

70-

"Oo>=3C

•i

—

+->

Coo

c:o »\

•i

—

004-J c<T3 oU •r

—

O -t-1

_J fD•r

—

-o >c (Un3 S_

-QE JOO <•I"+J i.Cl o•r— 4-S-<_> <D(/I r—<D JDQ 03

I—

<+-

00 O(U+J T3•I— cGO <U

+-> +->

00 030)\— 00

<D4-> -MS- Ocu 2:>r— QJ3 <UO 0/0

Z <£ -r-

IO •p- •i- dl^- E E D.3 J=

u <*- JH> 0> —I D.s- o co < * *r" r. »

o O •— O 0)CM 0) LU i-H p—

CL >> (0.C 4J J= cO D. O <o O re

00 00 4-J -t-> QJ»—i

- »—

<

= >> => •—+-> S_ 4->

•> CD * trt

1— X 1— 1- E t— IO

o c TD C «-C£ .'-) o a>

• f— • r- *— +->

lu ca a> x IO • roU~l CHLU O0 3 —

IO Q.E 4- +J ^i- <*-

o C O O •i— *CM O E •

4-J

CM LU U- LU LU LO uJC .c T>o (jO CM i_ o a; O i_ o +->

r~~ i-~ < r-» TD r~- «C r-» c•— o

+-» o> l/> ai Ea. Q. r. flj

(— X 1— • I— 1— i_=1 LU Q. => 3 => Cu => LL

cn ai en 0)c »-> c (->

j-> on 4-> l^)

c C3 +-> 3 4->

HI l/l a: on

01 0)>4- 1— 4- 1—o O

« X• o o o

oo •-< oo<0

• -C -C >•i- m (O i—E -(u E *J to

ID =i C5CM CM

*• «Ll_« E C OO

1— O (— O cC3 -o Z3 X! CD

QJ -r-

> sz«S I—

O0 CM

o o

01

CD CD O

O COO -Ho f~~ oro ro cO O rH

O LOCM Oo o

O LO O LOCM O <£> Oo o O O

CD CD CD CDc£ < _JCD CD t—

<_5 CJ •"- •—

i

CD

rt3

I—

-oa>

L. i

—

ro id

a) *->

>- inc

-71-

•»— >> >» c(O (-> (-> 3

1 •-D. 1 « 1 1 • CJ • • -nOl 0) CTi a) a> CD CT) <D •» ** •« •r- •r— T3c 4J • C *J C -M C 4J ID (0 re E cy E <D CC XI

•I- > +» +-> +j CO 00 C-t-> co i— 4-> CO +-> CO *J CO •^ •1— •r— 00 o 00 rec (0 c c c 3 3 3 • s. • s. CM i_3 +J O 3 +-> 3 +> 3 -U S- 0) S- OJ t- OJ CM +J CM +J CD3C CO 3: co 3: on a: m LL. +J U. 4-> U. P c c +J

Q) . a) ro m 0) •r- •n- •r- - o » re osl+- t— a. "4-1—1X1 *- i— <4- I— 00 »l/0 CO o s: ;r

-oCD

O £ O «3" o o vo VO VD LO LO CO CD» o +-> -l-> !-> C

• o o • O o© • o • o co • 00 CO • ai • 0) • 4->

13 CO f—

I

CO .—

1

CO r-H co .—

i

+-> OJ •-> d) +J a> +-> x: +j x: +J re

E CM O ac 1— OZ I— B£ i— C£ <-> ce -t-> ce 1—* -C . XT CO • x: • j= re <D re ID

c 03 co •i- re «s- i- re •i- re >i • >» >v +-> +J4-> o E +J 4-> E -M E -t-> E -u CO 4-> CO !-> CO +j CO o 00 co' CEO

=3 •-> =3 T> =3 =3 <_> • CJ <_5 CO CO • (D+J CM 0) CM t- C\J CM =3 =) =3 =3 l(- =3 |+- =3 OJ13 » c » (O « m (D <D 10 O +-> 4-> 3o O c c * c ai • C » C m *• CO CO CO CO •-+J

e o r— O <D t— o c 1- O t- o O 0) o <u O cv O • tt) • ai O CD •

_l => T3 CD D-DUJ => -o => -a CJ E o 2: <-> e: CJ CO 1— co 1— <-} JD ^o •>

•i— V)+-> efO o Ol

O T-O -M

cCD IV OJ m a> D ai ID CD a)

> C C c c c C c c C c_J fO TO o o o o o o o O

Q. z z z z z z z Z z z"O >E Oln3 S-

X)E -Q -

—

O < * C•r—

-t-> s_ o-

Q. O 4->

t_3

•i- <4- ni

s_ O ro O O O O LT) ro

O CL)0) O i-l CM CM .-I CM LO LO> x: o o O O O O O O

co i

—

< I— 1 1 1 1 1 1 1 1 1 1 i • • •

CL> -QQ fO1—

1

<4-

CO Oa-

0)CL <c <c s: s: _1 o o LU _l —1 O O < <t

•r- E •4-> >1 N Isl CO 03 _j _j h- > > Q_ o t— I— > > 03 03

OO CL)O h- o O LU LU «t <: <c <c t_3 LU Q- O. =3 z O O D. Cl. <c <:

+J 4->

CO fOCD1— CO

CD +-> i_ 4J+-> 4-> X o cS- O LU »—

1

LU •—

'

LU i—

•

LU — LU I—

I

LU *—

1

LU i—i LU 1—1 LU 1—

1

LU 1—1 LU i—i

a) 2:>i— CL)

3 CL)O 00

CD

rd

-72-

T3a;13

C•r-4-3

Coo

co A

•1

—

to4-> c:ro ou •i

—

o +j_l 03

*i

—

o >e O)<rs S-

^Dc Xlo <c•r™+-> S-Q. O•r- M-S-

o CUto ^~0) -OQ ro

1—

<+-

CO Oa>+> -a•r~ cCO <D

4-> +->

00 reaj

I— toO)

+> 4->

S- O<u ^>

I

—

CD13 CDO CO

it- s_ O oqj

er > 4- ci> 4- a>r— o *-> o +j

+J 3</> o l/l to0) LjJ LU~q X 4-> 4->

O Irt (/>

oo

M- 2o <u

4- CO -i- OJ

. to - (Or— tO O J=O ID I

s* tJo •—i s-4- 01

- o •* 4->o o ci_> to CJ •—

1

UJ LU 4- o re o re «4-> *J <D o c to LU +J zc 3 to

• 1/1 • W > =S •»— a> OJ o-^ OJ •r- ai <t >> o . X oC en a>E 1— E 1— +J CD •r— •r- OJ c L. ai

QJ E E <c u. CD10 ai to 0) S_ <_> OJ ai +-> c

• en • Cn ro to en LO C to 1/1 s_ <£.

c c ^ * C • "i— o o» re •• re re X) re CL j= _l -Q <s>O je o _c i

—

S- ID lo -C *• f— +-> s- Or-» u r~- u eu o r-

4

i—

i

i_> CO «t CO * re _J•—i s_ •—

• s- o 4- s_ CT\ zc0) O) c 0) •« T3 « Q.

« +J •> +-> re o +J > • Eo c O C o +J < CO =t c =c -o <: r— < +> reCJ >-> (_> 1—

1

o CO o '—

1

tJ 1—

1

o cc o CO t_> to or

0) -^> JZ<C I—

en Q.

O I—1

ro OCD csiO O

O O CO o oo in "3- to o«-( o o o o

O. O. a: <t «c =£ < <a: <:CO CO< <

CU

-73-

oCD13C

•i

—

+->

eoo

co r>

•1

—

CO+J cCO oU '1

—

o 4->

_l ro"f—

73 >c O)rO s_

.ac .qo «=c•r—4-> s-Q. o

•1

—

<^-

s-

u O)CO 1

—

QJ -QQ n3f—

tf-

CO oCD-M "O•i

—

c:CO Ol

+> +J00 f0(D

h- COCD

4-> +->

S- OCU 2=>

i

—

<D=3 CDo oo

s-o

in J=t—

i

0)

<4-' o>r— 00t— n- o(0 o •

CJ CO. CM

<4- zo

0) 00 o1/0 C71 ro Q.

c T—1 QJc 10o lt> -C • *r—

u ^ s:4-> —

i

s- Ofw <u #*

1.) 4->

o =t c- <: -o_i <_> •—

<

o cc

i! .2 ^ iS i! JS

CM QJ

<C CJ

CM 01 ro qj CM 0) CM 4) CM QJ4-> !-> +> +J +->

+J 1/0 +J CO +-> to +J to +J OOOf Q£ en cc OT

+-> +-> M +-> t->* to M t/1 #. (/) * I/) •> to

<C Ol «=c o «=c a> <: a; •a: ai

CM QJ CM Ol

1/5 i OO

CM 01-t->

1 COa: or ce

+j +j +j* to * to * to

«t <jj < aj <: aj

ai T-> x:>=E I—

O to o oo -a- «*LO LD «3- to coO O o o o o

too .—i •—i oo oi «3-cm to *3" r^ir>CM «—

<

LOtT) i—If-H O *—t O •—I O «—

*

oo oo oo oo oo oo

o t—1 efx <c «t

_l <C =£: _J(— CO CO h-O ct et C_> «=c <: <c <:

CJ oQ- Q.

LU •—i LU •—

i LU •—i CU i

CD

-OfC

-74-

(1)

13e

•i

—

4->

EOO

co «N

• 1

—

00+J crC oU •1

—

O 4->

_J fT3

'1

—

a >E CDro S-

-QE -QO <•r*"

+-> S-Cl o

•i

—

<+-

S-o aito i

—

a; joQ ra

h-

4-00 Oa;•»-> -o

s-

a>-D.

OTOIE

csj a>+->

-4-> OO

o

z cj*->

en uc _•-<

n- 4- u- >o O o 1- QJ

-J Q. tl- (Jo s-z <u Z° <1) z aj It- Q. «s

+J +J *j O z <uOl oo en oo CTi 00 • o Q.o o O z en

to to uo r**.

• 0) • a> . ai ro . o »+J (— +j i— 4-> »— tD OO -H Ccc Q£ ce o

•< * »> • => • +Jro to —TO—</> to f +J -*-> C7>—I J* —1 -^ —i .* ce « cc c

i. S_ s- c•» TO M (^ *• TO « o • >

«£ +-> < +J •a: -u <c +J <E O

QJ -r-

> .c=1 (—

?|

LT> oo to ro "3" lt) to ro to o oO i-H O <-i r-. r-» O CvJ to too o o o o O O o o o or~~ r^ "=t ~h to roo-j oo o —< O —

I

o o o o o o

CO CO (—

r-~ o oo•—

<

LT) LDO O O

oo co

OO CL)

4-> 4->

00 <ocu

h- 00

O)4-> +Js- oCD S>

i

—

o>13 cuO oo

TO C•r- -i- COCD-

CD

JD "OCD

S- 1

—

TO to01 4->

>- l/->

c1—1

-75-

-ocu=3C•i—

-MEOO

co rs

•r- 00+-> c:re ou •i

—

o 4-J

_l ra•r—a >

c CUrC S-

-Qc -Oo<a:•r™+-> S-a. c•p— <+-

s-cj CDoo 1

—

ai .aQ fa

I—

<i-10 oa>+> -a•r— dt/> a>

+-> -M<S> roO)1— to

cu-l-> +jS- o<u 2:>r— O)13 (1)o (/">

r— >,C0 >> >>IO LD re re> 2 2 2 ». 1—

1

t-H »—

1

r—

1

<*- C j*: -t-> .*: .*; •» f-H "CM •CM ••CM • CMO Ln s_ to S- S- CO CM CO «3- CO «s- CO tj- CO ^flo cr> re O re re ld «d- vo LD LO CO•p- 1 0. Q. a. 0. Lf) IT) >, ir> >, in >, in >,z "O r- 0) >> :*? s ^ 2IO CTl ^ ,

— J^ >> >>^ >> >> >> >>z: >d ii -t- ^ ^ l-> 4-> 4- +-> 4- +-> 4- +J <4-

00 s: O 4- O O O O O Or- c c c OO «3- 10 i- •> i- s_ C C • c • c • C •

.—

i

c 4-> 0J ai ai a> re • JC re 2 -C re z .c re z x: ozrO • 00 +J Ln +-> CM +j 00 1— z. u <J (J u 1— u4-> 1/5 1 LO Ci (/1 .—

t

u) L. s- S- •!- s- L. -i- s- s_ •<- &. V- ••- s--u en • 10 0) »3- ai Ln a; ro re ••- «C re E < re E«£ re E <C re E <:0: c :d w 3 3 3: E 3: 31 31 X„ >

ID* re „ re „ re #>LO s.

Ln•TO s.

LOa.

Ln••LD 8.

Ln cj•-LD Q.

«C >- >> >- 4-> >- +J >- +j >- CO >- . >- . >- . -r-_J O ^ CO :»£ i/o ^ LO i^ to :*: • 0. i^ ^H Q- ;*: cm 0. :»£ cm 0. i<i ro D-

> -t:< I—

COCOo

O CO LnO 1-H LTlr-l T-l O

?! <c <cu cuc co o co co

<D

CO

-76-

<urjc

•i

—

+->

c:oO

co CV

•1

—

(/)

4-1 cre oU •1

—

O +->

_1 roi

—

•o >E <D03 S-

-QC -Qo<•r"4-> i_Q. o

•i

—

4-S-

o O)10 i

—

ru .aQ ro

h-1

<4-

O0 OQJ+-> "O•r- cOO <d

4-> 4->

00 oCIJ

H- 00(1)

+-> -MS- OO) 2:>

1

—

OJ=5 cuO CO

4-> p* 4-> X • o «*•

1/1 •^ O Ol O0 O i a:oo oi > i ro 01 » 4-1 4-1- i— c 4-1 «oo r- C +J 00 Xc o ro S r^ •i- S- oo f^ O 01o—

-

oo OJ Ol O • E 3 O Ol cz:4-> 01 •i- 4-1 •-H Z • T3 .0 _c Q.«* Ol Ol Ol 01S_ i— IT T" CO •o- o: IT) _C U Ol CMO I— ro oo -H K_ • V) S. 1— CM »- >> t— *r— •f—

e: t- s: O 1- >> lti ro «t s: 4-> -h s: s. s:> +j >> £ 4-> 2 z: oi <_) ro» c • oo 4J • (_> * 01

co <+- 2.CO * ro

<-t o -H 01 t_) O0 CO •> 3 n iflOl «3" *3" «a-^- t/1 «3" 1— CM Ol CM O f- ro o «3- oo co on in LTl

p- >> " ro c ro O- oi .c i-H -r- UO IS) uo• T3 • "O i- r^ Ol Ol

i/0 ro OO Ol S- CM • -* • z « • o . z: oo •

Z: • VI Ol CM c 4J (J 4-> «-~ 4-> 4J 4-1 3 4J 4-1 4->=>-^ =1 o a. t— o c£ ro ce "CO Of cc » © ce co oc t»s cc CO

01 5- 4J X 4-1 o 4J ro .c i-H =3- <n. D-4-> • o en • wo X »<a-

>- ro r— >- s. X TJ C a: i*- lOh X 01 X • o X r-~ X CO X COS^Olfl ii o. O cr: •r- o o O O---' o z O VO 4-> O —i o •—

1

O i-H

0) •!-

> ^<: i—

cr> co i-H•—1 o CMo o o

o. a. i

UJ >-<I

<C < —ICO CD >—<C «C UJU

OlC LUOl—

i—

I

UJ •—

I

a)

JQro

-oOl

s- r^10 ro01 4J>• 00

ct—

«

-77-

<D-a^

i

—

uEOc_>

eo r»

•r

—

(-0

+-> Ert3 OO • i

—

O +->

_J ro•t—

-o >e CD(0 s_

JDE -QO <C•r-+-> S-Q. C

•i

—

M-S_u CDto i

—

a> .aQ ra

h-1

M-l/l OCD4-> -a•r- ECO CD

4-> 4->

c/> ft3

aj

t— 00

<D+J -t->

S- Ocd ^>

1

—

cu13 CDc_> O0

+J * > 0)O O l/> »a-• >> >> CD vo a> coz +j 4-> d) S_ c Ol 4-o C_3 — 3 "O (D • o

•r- -O -Q ce t-J »s-

E ai ai O 0) •Z i— >> " a) ujl£> x> -O TT Po p—VO o o -o -o <_}** F" *

• E z z C •>- e: xi0) id s: E l/> r—

<C =3 * Q)LT> (0 ID *3" «3- 4- Ol cr> -i- c->«3" I/) «3- >n VD O C <4- lOlt-Mt—

l

I-H ud LT> CO LO •i- O lO-DI-c s_ vr> CM .1* (Uo • 0) • • Z m • • q: •

!-> X *J CO +-> r-» *J 3 i/i +-> T3-t-> ce o 0£ o£ Q£ CM E o: m- ec»o -J a> a> •^ o<J » E >!o n: *t- i 3Z +J a: 31 • -P_i o o © s: o y o a: O JX> O 3 O

•X c

• oQ> 1-> JC«S r-

C\JO O

—I «S <t

LU <_> < «SI— co co

CD

-Of0 ai

s- i—

>-I— 4->

3 fOcS

cIM

3

-78-

-o C <D<d 3+j CD c<o (J C Es- c r-vj <o

CO o QJ ut oCO n- COo i. 3 o o

l/> O) -O 4->

-i: +J V X(J • 0) 3 s_ CDo o> •o 1— s_ OlOT +J o c +J

<_) I/) o <_) o o(1) +J -uo • c cy -Cs_ c •r- 0) E CL c<C t- o 3 o (/I

_j s: ^> D 1/5 (O Si

<_> +J (-> o>i.

i/) <V QJ CLC •*-> c b

o c IDi- c i/o 3 1/1

o

i

ID ID

CM CM

!-> JD1/) (O ii <—0) 1— _J *»•.

(— >=t crl+- E E

-o o0) TD

E QJ3 CDO0 to

CD

IX) O O o o o o o O OIX) «3- O o o o *3- o ID CO

C E IX) CTi in "3- ID ID «3- ro CM r»i- <_>

E I«3- O i—

l

CM -H O) i—i COa a ro CM CM ro " CM ro

o o o o OE o o o o o

*Y r^ U3 O en "3-

cr> IX) r-~ p~- CMcs CM 1-1 *-«

o o o o «3- O ID Oo o o o O O CO oO ID O "3" Oi CO co «d-

O) CO CO CO O r^ ro «3-

—i en CM I-H «—1 F—

1

-I

ooQ CM Q QOOQ CM

i— en cm eno El «-< «-•

CM CD

I*- CM

O r- O CO

r~- cm

«3- CM ID «3- «3- ID O) <X)

«3- IX) IX) VO ID IX) ID IX)

_ LU

-79-

!-> IO in• 4-> c S_ l*- +->

s_ c1

o o ct_ tn 11- +->

o m in -u lt- 4J k. • X oo E m c c o M- C CD S. CD •1-5

m 10 O o o oO en ^- •^ O. S_>> +J 0) r— -O • +-> lO O O c cn+->

= IO IO l/l in in i— S +J u u x: TO' ' o +-> IOro xi CO in +J •i- CD trt x CD c oo -i- CM = TD J* m in o e «* CD +-• o i_ +-> IO oB cO i— o in 4) u 01 CD c o IO s CD IO .£3 en oB o

• o f» in o c s_ » 10 • c X 3 m +-> •r—

a: o o —Ol o

o CT1.Ce u

o,_

o CD en CD u CD +JO (Om Cn • en X oO en en -i-j c •r- .O s_ c s-

t-> +-> c 4-> o CD .c s- s- S. +-> C 3 c c CD 4-> -Q CD Ol Oc o o (_J +-> in +J s_ CD O <_> -r- CD <*- n; o (J 3 oflj •r— (— +J o +-> i— ^. r— £ 3 CD X> c 3 S-

• in • in IO 10 OJ u in it- • o o. M- +J -O c 3 JZ) CD£ c o C 4-> +> CD E •r- C CD E i^ o it- S- c <n S- +JO •1- i- Ol i- o +J i— c r- .C IO ro CD CD 3 O IO C 3 CDO £ LU E Cl- s: u I/) IO CD -i- s: o in :r in r~ 1— O CJ3 M t— -o

s_ in in in in inJ2 CD CD CD o O CD CD O O o<c >- >- >- z z >- >- Z z

ID >3-

"OO)3e•r-4->

Co

f0 C_+->

fO *>

Q CDr~~

4-> .c

to ro<u h-h-

4-T3 Ocu -a•i- c:

n

^ r—_i

X EI4- +JO fO ro o CSJ o CSI o O

CO o 00 o -H C\J Os- cu

c Ei- <J

PO LO «3" O <y> o OE I i—i t^ «3- «1" CM <—

1

CMro i^ a a ro t-H i—l i—

l

EE CU3 CDCO CO E

oo OCO Lf)

oQ. ° Ol in LT) CO

1 ai C\J LO roc;

oCD

IT)

O

o o o o o o o o«3- O O tn o o o oCO O 00 00 O 00 O C7\

00

ai

-OfO

«^

l~~ LT)

CM CMO «3-

o O

i— CT "3" Q 00 O<_> E z z

^J

f~ CM CO r-l CM t-l

l£) p-

•3"

CO rt"3"

CO »3- CO i-HCM CO^-i r~~

ro to O Ol CD CM «3- «H

r-^ r-. r- IO r^ r~ r^ r-

4-> L. 4->

X o c

-80-

E 23 M o s-

c r— a> -o QJo _, A3,_ 4-> on reo D T3 o ai 3 3 S

<v O CD +J > c O O+J O X: <D o = 210 en o «3- -o o i- •^ +j on o o oS- 4-> lo -o c CM cn on 3 re CM CM —o C 0) o o> o o O o E o O CD

• .4-> XI o L. *-> re ' X!S- ^-» c 0) l<- U <D0) • c T3 re re i-

on +J s_ O) o en on Cn en c a> t. +j +-> Cn Cn+J+J 01 o 4-> T- 4-> 4-> +j o 4-> 4-> a> c on +J +> 0)c "o •^ o +J <-> E <_> <_j JD ^ CD t-> •f— O <_> T3

s_ o i. re a>£ CD • -o • +J S- s_ on u i • S-

cn+-> +J c C 4-> o c o c cn <d c c a>o +-> X X O •i- o o o o >> u x: •r- -Uu <_> (1) LU (_> s: co Q. SI a. LO (_> xi <-J I e: s <_>

s_ onXI o o o o o o O o o o 01< z *- 2 z z Z z z z >-

LO LO

-a

c

oro o4Jro »Q CD

o ID LO «3-

CO

-*u CD

C >>L. o s_I— z a *j-

CO COd) I—

<+-

-a o

_i

•r- cU_ CD

<t- -MO ro

>> >>S- CDro ^iEE CD3 QJO0 00

cl

C E•i- u

o oon o CO o—i lo

LO ro

00 O LO OHlfl LO Oion m lo

coLO

00LO

ro ooCSJ

o oLO oCM LO

oooo oo oIX> LO

cn loi

oro

LO CM

oro

oCO

oo

00LOLO

CM LO

CM CM

OLOi—

I

a-.

<L

ro-I .

LOoo

ooLD

ooLD

oO o o o O LOO Q O Q ro lo Q O LO LO P0 1^ LO 00 CM00 Z 00 Z LO V Z 1 z v 1 V r^ lo 1 V «3- -i

O ElO LOZ -H

lo a\cm >a-

lo encm «r

o oV V

uro eft ro en ro o^ O ro

LO LO 00 t-- oo r-» <o vo

o o<—

1 f—

1

V Vo1—

t

1 V

O<-< CM

CM

r^ r^ o CM «3-

r^ lo 1 r-~ f-. r-»

00 oen lor-H on

CM LO

X o c

-81-

5•o o

1/1 oo o 1/1

4-> en ail/> 3

J= <u o "o.+J Cl-Q cn Es- t/> 1/1 +J n- E <oo> <C ro o O l/l

> i/i S- s.^~ i- E * CU 14- .

3 <o s O s +J -t->

(_> QJ o i- 0> o ID i/> x.>> o H- CM 2 o cu

I/O 1—4 o. O c<o i/i E • a) o c

C3 0) <o s- -O oC -u </) (U 10

1/1 en Ol (O our •4- I/)

*-> 3 +J c +-> 5 o oc o -o <_) +J O •4->

g•r- o> E X • 1/1

> ^~ f0 oj o j= a>E 0) •r" c r— c c O-jQo S- ID o> c o i/i t/io CL S- s T3 o £ o «* ID

i. VI-O CU O o O*< >- z z Z

<_J

s-

oo 1 UD «3- l£>

x a>O -4-1

D. idUJ S

s- s_

o os_ s.cu cu+> "O -4-> T3CU QJ CU CUT3 -i- T3 -r-

S- S-CU 3 0) 3+J -Q +-> XICU CUi— CU >— cua. s_ CL !_

E cu E cuO JZ O -Ct-3 S o s

us oo

"OCUZ5C•r—-l->

EO

(0 O+->

fT3 »*

O CUr-~

-M .a(/) COcu h-I—

4-o Or—cu -a

•i

—

Cu_ CU

4- +->

o (O

>> >.S- CDrn ^EE CU=3 cuoo C/)

c E oi- <J o oE I <: oo 00O.G z ro CM 1

s_ D>>S-o

o o o © oo r» o CO olh oo o 00 rom 4—4 1 oo i *—

1

lO o

cu cu cucu c c cu o o oz z z

ro O ro Omm ir> inoo ^»- ro «d-

oo Otn uooo *!

CU

-Io•3-

oooo

ooo

It- 1— o o oO o o o3 CT) <Z CM f—

1

o oc^ E Z ^H CM 1 oo oo

o00

i mo or-~ oco mcm —

I

o o o or- o I

s- oco in j» in

<-> E|

SI

oet CM VDZ 0~ CM I

in or^ oo' i

co mf-I CM

Or-4

CM 1

OCM 1

in o CTi o CM

r- 00 <£> 1 r-~ i r^

I rH IICSJ Csl CM CM CSJ C\J

oo ro -82-

-oCU

c X!o s- Ol S_ 010) o c c •i- XI

4-> c •r— cu o re cfO s. -* os. X> ^~ 4-> ai >>•-> u E O -Oo 01 XI re 4-> -C u J-> o - o +J l/l

x> 3 X +J 0) m = !_

s_ c E en <u QJ +J O M- X) XIOI o t/i c >>o c *j ro 014-> xi _o O +-> •^ +-> <— s_ o re o • i/> enHI Ol l/l +J «3 XI re O CL • E o cX) i/i C l/l Q. *> l/l CD re Clt-

s_ x> Oi C o E c E O (0 E cz X +->

t/1 01 3 jD o J3 03 o <0 D C s_ re en oi l/l oi reu +J J= +-> on •r— l/l a> •f— ai QJ o 01 S- +J X! o oc 0) re t/i a t/i 10 t/1 "O t/1 t/1 Ol t_> -w • o

0) (0 o X> o C S- s_ +-> t/i t/1 J->D. s- -C E s_ O) -i. s_ -ii o a> 01 ^£ l/l E • -£= Ol 14- .

E E Ol Q. o S- 2 _l u s_ <_> j= 4- »-> t_> •f— re C Q..O uo O J= (/) s. o t— o o o c M- Ol o Ol •1— l/l t/1 O c<_> o 2 «c 4- <_> oO O *~~ o •— =i o o m t/i e: >=t IO f—I 1—

s-

-O o o o o o o o O o o<=: z s z: z s z z z z z

t_>

t.oo ro VD Ln *3" •d- "

'

TT ID CM ID

5̂1

tj o O in CM T3- o «3" <s- ?—

1

ID

oro o+->

ro c*

a a>^—

+-> .ato (OCD h-I—

M--a on—cu -o

•1

—

Eu_ a>

<*- •Mo ro

>> >>S- curo i^EE cu13 cuoo co

o a a

oooro

3C E

c El ro o ro o ro o(J LO LTl U1 LTl lo in

= ro «- ro n- ro >o-

Q. csl

o o

a i

<D

J3ro -I

COoo

<*- .

—

o o o o O Of~» o r-« o r>. o

3 cr CO tn CO LTl CO Lfl

*/5 E CM ^H CM ^-1 CM ^-tt—l LTl I

OOQ CO

r^ OCO

oCO

oCO

i— CT CM CM CM CM CM CM o o oo E «a- « <" «3" «3- •« •"-"1 l-< 1 i-H

oj r^

CO CO

X o c

-83-

<o in

QJ

c +->O IC

Cn » 0)

<_> o s-

O (1)

= H- Q.T3r~ o ci£> c 3O Ol o

• C v- Cr UNCD.

QJ

O C Cn

3-0

•1- c x o +->

E jliJU U

«1

* aj qj= TT >O C 1-N 3tt

C_) TO• -o

. s_ •.-

C S- 1—•1- o oz <-J 3;

<£> l£>

<L)

-3c•r—+JEO

ro O4->

<TS *»

Q (U^™

+-> -Qco n3CU (—h-

4--O Or—cu -01— Cu_ <u

4- +->

O 03

>> >>S- a)(0 ^EE a>3 <uoo 00

i^> ID LO

111

>> >1 01 •^1 ins- s- (J s_

1— CM

c E•r- <JE 1

a. C3

r-~ o,-H O1—t CT>

roCM

00 id en 01r^ i-H >3-

1 «3"

COCTl

f0

CTl1—

I

o ooVDoo

CM

ID

oroen

Oo roro <£>

>— D- COCOCM IDLO CM

OO

HCM ID .-< ID

CO f-«.

ro vo

1^ r-~

*j s- (->

x o cLU •—

1

1—

1

LU _

1

-84-

ro01

C71 s_M T3 ICO 0)

tn ,

—

O t. .«.,-ro c CM 0) cnO r~J

oO *->

IT3

C 4J•r- <o «a-

t- s- £ o> J^o u. c (J "C

(/I en en to o o> o> D•!-> *-> c c 4-1 S. en .c -o cc o o o o Ol u c O1

t-1 >-, o o• t/i </> in X co -O

E c o o c •r- o E t/i _)o •1- i. s- Q. O !- (—<_J s: uj LU z: m LU CO Q o

s_ i/i (/i

XI 0) 0) O O<t >- >- z. z

c_>s-

o

en4-1

<_5

en en+-> +j

<J3 us

I+->

co

ro C_>+->

ro *S

O <Dr—

+-> Xto ro

<D h-h-

<+-

-a Otu -a•r- dli_ a)

q- 4->

o r*3

>1 >,s- (1)

ro ^eE <D3 CDCO oo

>1s-o

i£ i—_J "-«

a: E

o o oo o o o o o00 r—* CO —

<

CO *—

I

ro ro oo ro mmo Oo o O OCO i—i CO i—l

ro ro ro ro

o o o ooo oo oo ooCO*""1 00 »—

I

00 *—

i

CO »—

'

cor*") ro ro ro ro ro ro

!—

1

«3"

«* «*VO CMo !—

1

cu

-Qra

r-

</>| roaJ o

oo10

3 cr Q

<-> El

CM

oj oo'

O O OCO CM CO CM CO CMCM Ol CM Ol CM OlCM »-l CM ^H CM >—

I

ro i—" ro <—

i

r^ cc r~ oo

o o o o o oCOCM O0CM 00 CM C0CM CO CM C0CMCMCi CM C7l CM CTl CMCT1 CSI Ol CU OlCM i—

I

CM"—

I

CM •-< CM>—

'

CM i—< CM i—

I

ro i—i ro •—• ro —i ro •—i ro •—

i

r-. 00 r-» CO r-~co r^oo r-~ CO

u '- •*->

x o c

-85-

E E>D +-> oOJ 3 s-to o 4-o 4- !_ LO i-H

O ^ +-> O cnu (O c

-o ro o s_ •.- s_ D -o Chc 0) i. <u 73 0) C c Oo s_ <D +J c =o O o Oo 113 +-> <-> c O IC (J u -t->

(C to -O i— c(/)

en f— to f— c i- t/> cn cn £ cn+-> A3 i- -O o -o o *-> -U T3 4->

<_> E 01 +J <-> o c <_)

00 +-> s. to • 10 oI/) o OJ -c a> • -Q

+-> 0) IP- c Cn CO. 4-> *-> to CX c T3 to to X X f-LU o -O e: OJ «£ (0 LU LU O •^

LO l/) oE 0)

= ro = <_D CO 01 cn roc O to cm a>o o o a.u • 4-> • a. •

.(O ro -c

Clcn cn to cn to to4-> +-> =*> -t-> o roo (_> (O <o o +->

X! +> to L.. T- 0) • 01 01

+-> C I— > CUT3X •f- O f- •r- to CLU Z :c t. s: <: =

•-O to

-ocu3c•1—

+->

c:o

ra o+J<T3 *>

Q cu^—

4-> -Oto fO<1J 1—r—

4-TD O^~cu -o•1— cLL. cu

q- +->

o ra

>> >>s- CUro ^EE CU3 cuCO CO

Ll. +J3 4-

i^ f—-J -^=C cr

I—

f

c. d

o oo o Is* o r^ oo i-H in o LO O

co ro CO LO co locr> I-H CSl i—1 CM

o or-^ o r^ OLO O LO OCD LO CO LO—

I CSI —' CSJ

o o o or-. O ^ O r-^ o r^oLOO LOO LOO LOOcolo colo co to color-H CM l-HCNJ l-HCNJ I-HCVJ

o oo o to otD o to LO

cu

fC

-I ,

«s- ro«3- C\JO OO o

ooto

o o o o o o

u

oCO CMcm enCM i-H

COCO V

CO00 V

00CO V

00CO V

00CO V

COCO V

00CO V

CO00 V

00 LOI-H

CO o•-H

ro i-h CM CO CM CO CM 00 CM 00 CM 00 CM CO CM CO CM 00 cm ro O r~

r-. co LO LO in lo U") LO LO LO lo tn Lf> LO lo in LO LO r^ lo r- to

*J J- +->

x o c

o00 -86-

a<u^c•r™4->

EO

CO (_)+-»

ro *%

Q a>^—

+-> -O(/) fOO) I—1—

*4-

a Or—a> -a

•I'- cLL. <u

<+- +jO 03

>> >>5- CDta ^EE <L)

=3 CUCO CO

O u O . > cn cc s- a. c JC 1_ >> ro jr >> lo c "a o

ro CU o 4-> CU +-> ai -r-+-> o cn CU ZC cn CU re +->

TD i- A" c T3 C lO +J a. -o c M Q. c +j roai Q •r— en en CD CU S- a> CU CU 01 •f— ro s_TT o c c .*: r— CO I— ^ 1— oO +-> s_ cj CU LD Q.O cj 'a.o +J +j -i-s- 4- ai -o +J QJ +> a. E o CU 4-> e= O 1— c s-a> o cn O rt) .C 3 cxro o _^ jc: 3 o ^ •r- gj ai

i/> CU +-> s- o u o ID cj (J o o cnj-> •tJ c c ai cj c =C -C "O CUc 0) ai o r— M- cn cn cn Q. CU o cn cn CU D oO c -o

1c S- O S- i- c 3 c +-> CU e 3 +J CU oc c_> 1/1 •>-> D <U 0J •f- O •1— 1— •f— o •r— O •r— o ^: -C >,

E CO c o c (-> -o > s. > o c o > S- c O O 1/) r—O -C o s_ +-> CO CO c (0 -C CO O 3 s_ ro -C 3 S- O •i- cl_3 O u LU <D c/1 3 3 a. »— o. ^ C3 CU a. 4-> O cu a: o o

S_ c/l in to in in

£ £ o £ £ CU>-

CU>-

CU>-

CU>-

CU>-

o

o£2 Jl

- ro LD LP •& -1 f-H -1 «*- "* «3- "3-

5 1

<-> o ID U3 CO CM •—

'

o ~* o CNI w

«I cnE E

oC E o o o o o o o oi- (J O 00 o o o o o o o oE I cm ro o CM o LT) ro un cn IDa a C\J—( 1 CN1 00 CO CM >-( LD ro 1 CM

G

a»

-Oro

Joovc

co

co

o oo oI ro I «3" —

I

l r-

<

II I U3

O E| V ^h i ro

QJ «3- CO

CM U3^3- CM

Q 00 001 1 1

001 V

*J- ^3- cm r-^ CO o oco ro VD CO 1 CO 1 LT) 1 CO

oi ro

X o c

COco -87-

i— i

—

+j »D o c 0) <D E (/)

1/1 c oi I/) o -o 01 4-1 0) TS i-i

-i-> n> to u o c 4-1 s- ai4-> <o o l/l ^ • l/l

is- (J 4-1 c

l/l o j* 4-1 ID <J J= o E O) S- •!-

0) 4-1 c u S_ o Q.+J tO > s 4-1 o4-1 c o o o <_1 i. i/i to o C LO TO 0--0

g.o r— ID <a <D O 01 s_ r— i—

I

01 QJ

</) -o l/l T3 +-> T3 -O +-> l/l 0> o c (- +->

*J HI o •*— 4-1 O ID C •» i/> 4-1 c o o IOc to c •o IO i- IO 4-1 ro -o <o 4-1 i-

o o t/> O) -* S_ qj 5 1/1 s- o£ a. -O. E 4-1 z u +J 0) E >-> 3 4-1 • ID 0) "4-

E o 1/1 13 i/i CM _l •1— ^~ > o i

—

it- T3 i— en s_ > i.o s_ ai o 1 1— -C c s- C •r- +-> XI C 01CJ Q. a i/> E "~

'

CJ +-> </) •r- <*- l/l Jf" IC <s> o <z •—• o.

s- i/> l/l l/>

-O O oi o CD o O o o o 01 ocC z >- z >- z z z z z Ji- z

VD tO to «* O

"O<u3c*r—+->

Co

rc C_)+Jro nQ CD

^~"+J -Q</) fOQJ h-h-

M-T3 O^*"

cu -ai— Eu_ <D

M- +->

O ra

>> >>s- a>ta ^Efc a>3 cuC/3 00

n

i£ i—_J —

C E O•I- O O Lf)El O l-H

oor-» CO

ooo

to

o

oO r-Ho o

oCO

oo oIS> C\Jo ocvj ro

to to

ai aic co o

IO CM O

OO OO inin «g-

CM CM

CO to

CM CM

otDCO

O roO ro to<C i—i t—i r-*Z LTl -H CTi II

a)

-Q _l

.

1/1

1

CO+J 00^—

1 COos*!

'1

co

CO

CMtoO-i

— "^J to toD Cfl Mrsi/> E| i-l -H

CM O O O LT) o oOl o "» o O CM o o.—1 CM i—l CM in i—

i

ro cm

o ocr, CO

I II i-H CO II

CO COv «r

oo toi-H ro

O O•-H CM

O to—1 CM

O COi—i i—

i

CTio oi-i ro

in ino o-> o o ro r-. co in to CO i—i to

«3- CM to «3" to ro to *J- r-s ro in ro

II II

cm in

i r-» r-~ «3-

x o c

-88-

•O 0) 1—CD f—T3 C oSO *- t/>

Ol J- c +J4-> <U T3 o onc O) OJ

i• +J </> s_

-C ro «3 uE Clj-> S-o 00 O XI +Jo < S_ «=c to

s_ to 00jO CD CJ<: >- ;—

os-o

-adja3r—oco

fO CJ)+>03 r<

Q CD!—

+-> -Q00 03CD I—I—

H-T3 Oi

—

O) -ai

—

c:u_ CD

M- +Jo 03

>> >>S- CD03 ^EE 0J=3 CDCO CO

i«: f—_i ^-> o

•3"

c E «3" O cr> oo *—t LT> ro r-»

b i«3- CM d «3"

a c; CM —i cr> r-n

Ea. <->

o oCO oo

oo->

cu

.a03

-I

o o3 crJ LO CM

ro in

O oro id

SiOl o-. oE] CM ,-H

»—* f—

(

ml • •

o.—1

1 V 1 1

O 00CM l~~

H f—

1

CM i—1 CM

D CO i r~~ r^ «3-

w '— -t->

x o c

-89-

Abbreviations

Al - aluminumA - asphalt dippedAM - asphalt masticABA - asbestos bonded, asphalt dippedC - concreteCTL - coal tar laminate - U.S. Steel NexonCTE - coal tar epoxyE - epoxyGS - galvanized steelPE - polyethylene - Inland Steel Black KladPVC - polyvinyl chloride - Wheeling steel plasticotePAE - polyamide epoxyVP - vinyl plastisol - Bethlehem SteelVC - vitrified clayGA - galvalume - Bethlehem SteelIZ - inorganic zincOZ - organic zincS - steelALM - aluminized

90

Key to Terms used in Table IV

1.

2.

3.

4.

5.

7.

8.

9.

10.

11.

12.

No.

6

5

Ext. or Int. - External (soil side) and Internal (water side)

pH - pH of soil or water if present. Soil side data not taken if no external coatingpresent or representative sample unobtainable.

CI - Chloride content in mg/1

Sulf - Sulfate content in mg/1

E - Potential of pipe exterior with respect to a saturated copper sulfate referenceelectrode in volts.

I c- Corrosion current on exterior pipe surface measured by polarization curve.

Tafel slope extrapolation or polarization resistance, amps

p - average soil resistivity taken by 4 pin Wenner method, reported at 5 ft. depthin ohm-cm.

P min. - minimum resistivity of saturated soil or resistivity of water, ohm-cm.

H - Hardness of water measured in mg/1 CaCO.

mALK - methyl orange alkalinity expressed in mg/1 CaCOj

WFR - Water flow rate at time of inspection, essentially stagnant water indicated by"None"

1

13.

CtgC - Coating condition

Asphalt

No Deterioration

Checking, Erosion of Invert

Chipping at Crests, incompletectg .

Disbondment at joints

Ctg. removed from one side ofCorrugations

50-90% removal from invert

Complete removal from invert

Polymer

No Deterioration

Roughening of Ctg. at Corr.Crests, Blisters at edges <l/4'

Disbondment at edges/seams to1/4" Blisters to 1/2" dia.

Tearing of coating at Corr.Crests

Disbondment at edges/seams to1", Blisters to 1" dia.

Epoxy

No Deterioration

Checking, some deteri-oration at ends

Disbondment at jointsto 1"

Disbondment >1" lessthan 6"

Disbondment >6", <2'

Disbondment sufficient to collect Disbondment >2', <3'debris - create blockage

Complete disbondment/delamination Complete removalor removal

14.

15.

16.

CorrC - Corrosion Condition

No.

6 No Corrosion

Zinc Corroding, Concrete eroding on invert or where exposed

Mild General Corrosion of Steel, Concrete erosion undermining ctg.

Steel pitting

Layer type corrosion of steel

Steel perforated at waterline or joints

Invert missing

Abr - Abrasion a factor

Min. Ctg. - minimum coating thickness measured

Metric Equivalents:

2.54 cm3 0.40 cm

1 inch1 foot

91

wJ

CO

UCU

c•H«4H

QJ

Pi

CUCO

4J £•Hg >%•H PQJ

X)C CU

o j-i

•H >-i

4J oCfl ftU QJ•H Pi<4-l

•H CO

U <0)

ft CO

OT 4->

rHH cd

03 Xio ft•H CO

ft <!>>H +j

V)CU

>H3c_>

UHo

omo•H4-)

CO CO

sn id

a) Ba, 3 mOh 4-1 CMO -H 1

W pq m

CM

r4CU O

Ph S

^ OCO CM

5 I

EC 00

5-1

OCO

C I

•H U!3 CM

ON O"

I I I I

or>8 -d-m CMCM 1

I mm l orH CM

CO HI I

CM CM

I I CO H

00CO vOCO I

cfl oft v£)

mCM

Im l l

• ti X OX -H cd <fcd B B cmB i

b-s e^s 6^2 mt>S cr\ cm o o

oo H as

mCO •

CO CT\

03 I

co cm p-i ft oo

ooCO MDCO I

cd oft vO

c•Hs

o CTn

b^s m m •

<r r>- m oo o CT\

CO rH oo <* • o-n

I I

CO co ^1- CM CO

r^ CO en m O CO

H 03 cd • • cd

CM P-I ft CO rH ft

S . C•H X •H

o s + 2 Sm o BI + B-? O r>«

o m m m r>8 ONco CM 00 <t rH o>

I I

oCMCM

IOoCD - CO

O

Io co

co co <3 o co <3cd cd cdNfuf^lzHMZ

o O O o CO O coCO o\ c^ CM CM CO CT\ o> <f on <r-J- iH rH -* rH r^ <T rH rH CM rH CMCM m a CM

0>s o

CM CMrHCM

CM g a a a ao a o Q o Q Q Q o o o o o o

• H H H H H H H H Ha EC S pa S EC S S s P3 a HI EC EC EC01 W H 00 H CO E—i t—

i

^i C/l w 00 w CO 00ftCO 1

co3 3

co CO CO ^^^^ 33

ECo

ft p=< rJ • PE( 4-1

o o 4-1 o CO

r-~ CM u 01 • CU

r^ n cu H 4-» n rH >. E-4-t ft 00 a 4J cd 4-t

U 4J <4H o C • 0) C 60 •H CD CO

cd cd <J H U 4J •H -^ > CO XIrv 4J c O • cd CU -H

a CS c 4-1 cd «• 4-1 O P-i Xi H ti rHo o o c 01 >, C u 4J rH o CD O•H •H •H •H m 4-1 CU 00 CO 4-1 3 C/3

>, 4-> 4-1 4-» o •rH 4-) 4J a <u CTj ft a O4_> cd cd cd ft (3 rH ti a •H H 0) E»> •H •H 4J

M r4 U u o •H o 0) C H 4J m > C01 4-1 4J 4-1 jd ^ o B CU 4»S •H •H M CU

ft CU a) CU CO CO 3 3 4-1 CJ S CO a cu oa c c PS cd CO .H J3 4-1 i+-i O o Pi 0) a, um cu a) cu H o O X •H O J3 rH cu ft S <u

Pm P-I Ph ft ft (J CO < cq W C/j fx. Q 00 H P-i

o

4-1

oc4J

3rQ

CO

cu

•HrHftftcd

CO<!CM

OHECI-O

CO

cd

B

x)cu

•HrHftft

rH Cd

cdCJ T3•H rHft cu

>> -HH pt)

/—\ ^~\rH CM

92

m

co

pHB-HHi

coHP(0

O•H

-HCJ

<U

ftCO

oEh

CO

3

+J•H&

o•H-PHCO

Oft

euprH

ftCO

<;

cHcoHP(0H

>

CO

co

<U Mp td

u o

-P TJ

co X

CO

cu

c0)

rH CNm *>

a,CO

CO

(D

Ccu

,T)

fd

u

CO

CH

fO

uaa,

CD ro

en o

tn cn

• coco n •

rH CN CMI I

r^ cnrHo •

CM UD

I

in

IoI I

H r~ 1 i

I

o

CO

+J • •

-H c c£ H •H•H o £ e1-3 in

I <X> o\»• o LO 01O ro CO Ocu

ftCO

cu

3 •

T3 -Pfa •H CUo CO Ehr» a>

r» KUP Mh

td

c fi

O >1•H •H PP -p HtO rd Mu M •rH

p +-> X!p cu cu 3co c c rHCD ai cu

En Cm a, CO

CO .—

^

a> CO

ft (1)

>i ftp >1pCDCM r-—' ^CO CO

a c•H •HCn tn•H •HH SH

O

p 4->

c CCU

n ucu CU

m <4Hm mrH •rH

t! T3

00 CN

» ^

<tf [^U3 CD

cu CU

u Ua c

CU

*H SHCD CU

<4H <4H

cu CU

« arH CM

93

TABLE VII - Criteria for Using Metallic Coated PipeWithout a Supplemental Organic Coating

State

Alabama

Arizona

Arkansas

California

F lorida

Hawaii

Idaho

Indiana

Kansas

Louisiana

Mississippi

Montana

Nebraska

Nevada

New Jersey

New Mexico

Oregon

Criteria

GS , AL: pH > 8.5 but not >> 8.5

GS, AL: pH 6-9, SR > 2000

SER: pH 4.5-7.5

S'ER: pH 4-9, SR 100-100,000Use graphical relationship between pH,SR and service life

SER: pH 5.2-8.4, SR 107-83,000GS, AL: pH > 5.9, SR > 4 500; use CAmethod where above criteria not met

SER: pH 4-9; use CA method and 40 yr designlife

AL: pH 5-9; SR > 1000; use CA method and 40yr design life

AL: pH 4-8, material depends on road type

SER: pH 5.7-8.5, SR 700-2300

SER: pH 4-9, SR 200-30,000AL: pH 5-9, SR > 1500, no iron ore present

GS : use CA method and 40 year design life

SER: pH 2.5-9; GS : pH 6-9, SR > 1000;AL: pH 5-8.5, SR > 800 and flow < 2.1 m/s

SER: pH 6.5-7.5, SR > 1000

GS : sulfates > 1500 mg/1

AL: pH 4.5-8.5, flow < 2.4 m/s

SER: pH 5.9-11.4, SR 100-55,000; use CAmethod and 50 year design life

SER: pH 4-10, SR 250 and greater; GS , AL:pH 4.5-10, SR > 1500

94

Pennsylvania

Texas

Utah

Washington

Wisconsin

Wyoming

Notes

:

1.

2.

SER: pH 3-9; GS , AL : pH 5-8

SEP.: pH 4-10, SR 150-4000; AL: pH 5-9,SR > 1000

Use chart relating pH, SR, soluble saltswith desired service life

SER: pH 4.4-8.9, SR 200-30,000, AL : pH5. 8-. 5, geographic criteria

AL: pH 5-9, SR > 1000

GS: sodium sulfate < .2%; AL: pH 5-8,alkali < .1%, sulfate in water < 150 mg/1

SEP = Service Exposure Range in stateSR = Soil Resistivity in ohm-cmGS = Galvanized SteelAL = AluminumStates not listed either do not allowuncoated pipe or have no formal criteria

95

TABLE VIII - Typical Material Costs for Flame SprayCoatings for Abrasion Resistance

Material Cost/Ft. 2 *

Type 316 Stainless Steel $1.91

Aluminum Bronze 2.6 6

Molybdenium 13.81

60 Nickel - 16 Chromium 3.27

Monel 3.7 5

Chrome Oxide 14 . 51

Aluminum Titania 6.03

Aluminum Oxide 6.22

Phenolic Sealer for coatings .04 9

*Based on coverage rate of 1/2 pound for .010 inch

Metric Conversion:

1 pound = 454 g1 foot = .305 meter

96

-PM-l

tJ1

CO\+J

CO

oU

OO CM in KD OO «*^o t-~ po en in *x>

rH rH CN CN OO 00

</>

^LnmoococNiH^xiuDCNcoiHoor^cr>ix)p-icNooLnLn^or-coooooLnLncNV£)(x>^i—ioo'3''3, '!3H '3i '3, '3, ^r'3'm*x>t^r--cNr^r~Ln

rH CN CN CN

CD

Urd

Cm

6

U to

SM CD

4-> -PW CO

>iU CO

tn CnC cH •H-P -Pcd rd

u u0) C7>

> C•H •H-p -Prd CO

c •HM XCD w4->

rH

<C -P

X

r4

4->

*4H

r> CN r-in "vf m

• • •

«3< r- <Ti

o r~- ^ ^r1

r- asCN V£> ro o O 00

oo -^ in 00 CO CTi

</>

r^oocorHcrir^r~i—icyiiHoooo'sj'CTiooocN(^(N ,*o\o]'>j| ^)crio^^^ir)HOOoinCTiiHi—IrHCNCNCNCNOO^OOOOOOLnLncn

HrHrHiHrHHrHiHiHtHCNCNOOr^r--VD

CP s

c•H m> =

rd •

CH in-P

+ rd •

= =

TJ T3 Oi4-> LDCD • a) •H rd m = -*a g a - Q • 00cm Cn (^ 00 U - • 00 •

•H C C "H 4-> rd "* pQ •H Q rH En rd 4-> • -P

s > = r • rd = rd rd

-P O E rd -P in £1 H • u = CN 4-> = 5rH rH O CM rH in 4-> r a rd CD^DrHTJ rd^mLOrd O m rd rd ^T CO 4-> g r-{ O CD U <D <H CN

x: • -P A • • ^-\ < U rd H O • -H Q) >, g O r-\

Cm • U Cm >lO u rH 4-> X • •

CO -p CD CO 4J -P X • + + U Cm 4J Cm CO 4->

<! rd 4J > <C rd rd CD -pfdCMCDa1 C04->-Ps fd c •• Cm4-> U U -Q + rd < >i W rd rd rd

O TJ -p H T3 CO rH >iB rd CD CD £) >l rH rHm CD rd CD cn CD X 6 6 3 O >lX >lO SHW4-14-) U

• o TJ •H + T3 C g T3 >i-H rH X rd Cm rd fi CD Q)

co • C O -P C •H rd a cd X 5h 5h 4-> Cm U Eh CD -rH CD U &tn O CD co a. -P C W tJ Cm Cl, rrj cn Cm W Cm CO C rd X CD ,£!

C -P PQ Sh rd h pq rd W c a CD rd W WCflTirdCMCn^^-H rd Cm a c CD w u 4-) a CD rd rH Xi ^3 K4-> 0) CO U 5h C CQ -H -H rd SHTJ SHrH-H4J CD CD 3rd -P 5h p •P rd •H 4J 4-> 4-> C 4-) rd-HrdOcJCDCC Cm" rH

•-i +J CD -H rH -P CD E-f 6CH CD CD -H rH Eh g Eh PQ U Q) CD rd

u rd CO g rd rd co > rd rd X3 X! 5h rd fd CD 3 iH Sh rH SH

A CD >i rC x3 a; •H rH >i-H x: 4J4J ox:h >,rH ^ xi >. a a >, 3tr> Cm-Q <-\ CI4 Qa& 4-> rd rH CO ftCCH Oi rd H rd rd (T>rH 4-> 4-1

G CO CO to CO CO rd O 3 CO >i >i£ CO OOOrH-HOCDCD 3 rd

-H <; <; p-i <: < <c C Udih<cncoU<ua<UfaKft2 2PQ!2•P UCO

•HX

0)

-PrH • oh (N ro^r m ko r>- 00

W rH on 00 "^ m <wD <: <-\ CN 00 ^r in <o r~ CO <7^r-ir-lr-i(-ir-ir-\r-\ r-\ f-\

coCD-MOC4-J

OO

o

0001

a)CD(T3

Q.

<ucuco

97

>1Xcd o

•H C aX rH rd CD

nfl n tn^J\d) !D C Ocn C CVH T3

1 -rl G Gcm rd en rd

-P r- CD XI^ • rHJ) o*o u c •

Cn -w- T33 01 O -rH CD

rd cd m o en tJtr> rj^ o • 3 3

G ^ftHHvo rd in -to- urH U -P -PC

U) CO CVH^ }-l CD CD

CD O ^ u wa ? x uH CO C rH CD CD

aojo o grjvH £ ^— -H

p rd P -H • 5H

M >h rd paCD CD u io m> > -rH rH 4HrH fO rH 1 •

3 CL.CN tji

CD Oi W pSH < • W

rH fCj 01 "*CD rH UCD W • 2 •

+J+JH r-01 01 rH g CM •

-H CD — CJi

-OUg4J ccd tn C'HN . >, +J

-H • CQ 01 -H 01

A to • -p rd

rd CD O CD rd rH

> T} . +) o rQrH -H Em rd -Hrd W C rH PCn •> U CVH

,C 01 CD O. Sh

jS +) u -P fd Cn

O CD rHC XI -H rd M M•r) rH "H g

1 ^ Ol >-( <+H

co oi ao G •

-^ C d 4H ro D1

oi r- -H en

rd -H — • -P+J • CM rd <7\

rji rd £ P </> U CN

C tn-P mh rd <T* g•H 3 TJ a O+J M p • C CD •

rd U 01 rj1 rd 5h «*

rd 01 Oi IIm

U O (U\ CJi •

rH 00 C ,C • CN

C £ <tf -H O PO P f>l W -rH m II

a rd • rd x:

T3 -H —' CD £ • •

CD g H tJ1 c01 cn • tn U 01 •Hrd \ >H 4-> CD

CQ rH HH <4H T3 <4H rH r-i

01

CjJ

p .—

.

<-+

rH CN

2 »-' "

—

98

EH

WHWEna

DUQWNH

<

oEh

SoHCOwQ

wsHPh

I

Xw

coHCO

x: uT3 o< rH

CNp(U -

-P -Hcd co

g a,ouHw

coH UCO oCD H£ CN"3<:

CD

ftrd

poCO

to

MCO

CD

u•H CO

Eh -HS

tnG•rH

-p

Ou

pCB

-HHaa,

CO

o O o O O O O o ID LD O Oo 00 o 00 CO CO CO ID CN CO H enCN H CN H rH rH rH CN CN H CN H

oin

oo

LD

OCO

oo

oo

oo

oLDin00

o o omomco r- <=?

H

mo

oCN

rH

O oco

m om oCN

m o or- o m^r CN o

CN «3<

o

xf

om^5

1 PU CD

> CO1

CNCN m

H ^D k£> ro CTi 4-> 1 i

in CN O H m H o O <J\ CD Cm h)rn [» r^- r*» <C r-~ o m r- l-D 1 i

H rHC C C £

"nT r- HCO -H

CO

Tl T3 p P P PU P -p +j 4-1 P CD 0) H Hrd cd CO CO CO CO CJ o P ft -H •HCn Cn >H >1 >1 >i 13 T3 P ft X! rQc C C C Qa w Ph (X « cc w H Ph « s 2

HcH

>i CO

C P c3 13 CD

-H XJ 0) p a> CO J3 * rH P CD *

cd-9

C CD CD g >! T3 C T3 >i». £ •H X! rQ -rH X \ •H rH >i X

CD PS CO X! rQ P H O CO ^4J tj d) 3 3 ft ft rd < CD H ftrd H T3 rH « « « w P p PC < wg O cd CU ii >i O CD

o cO P g U U u CO CD P •H g O P CD

£cd -rH •H H -H rd -0 3 P •H •rH CO c

H c rH +J P P 5 H PQ P P ^3 •rH

u cd P -rH P CM CD CD * CD He g £ ft CD Q grH CD rH CD rC rC c rC C H rd H 0. £, rd

o >l g o g rC P P •H P •H >i >i >i CO rC P U >ic 4-1 H H H CO C C 10 C CO a H fl o CO C C H

-rH 2 P rC Sh rd >i >1 CD >1 CD -H H rC rd >i -HN X! C^ u ft & W w PS CO cz > Ph > CU & CO N Ph

99

co•HWa)

42TS u< o

U CN01

p0>

IuHW

o OLD OCN CN CN

O00

OCO

o

go-HCD

<01

rd

UO00

uo

CNOo

oo

oon

om

m oomm

wto

Q)

G

u•H en

42 HEh -hg

enG•HPrd

ou

0)

•H.Hcmcm

to

01

CM>1Eh

c•HPrd

ou

m

CO

oCM

m

CTl

rd

01 ilin • }H 01

rH T3 4H 0) CM O1 rH > o CN O CO« ^ SH O •H -rH CN -H rH CTl

1 CO Cm U a en rH tn rd 1

ro r- Cm D tO * rd -r) <rH m >i g g r) Cm

+J rH >1 >i -prH rH •H rH X H ^ w •

•H H rH M rH O rH U 3 T)43 43 H 4-> •H rd H fO T3O P d) 0> rH D rH G Ug g D Cm a; pq C£ pa H Cm

Ua> X * *e SH G G

>i H 0) •H -HX in g CO CD

a, •H D 0>

Cm u a «W 42 cm

en uM rd rl •H •H(0 S rd +J PEh &H 0) 0>

>. H 42 42X rH >i •-{ -P P

rd c rd G GCm •H >i >iW U > U CO CO

G

•H pP rH•H (d

T) 42C ft

CO

u (d

U -P0)

Cm 42

W 42tn P 6G -H u•H ST3 rfrd >i m0) P CNr) -H oH o

CN -rl •

434-1 -r) II

Prd rd 42

0) Cm CM Utn g X, Grd •HM O II

01 rH> en in Ord oi en oP CO •

oi rd •

u o yo II

rd -HT> X H

rd G •rH

P -H H erd COQ * CM rH

0)

01

po

100

Eh

WWECCO

g>DUQWNHs

8

OHCOWECQ

EC

PQ<En

GO•Hen

<D

,G

< HCD &(-P0)

gourH

co•HCO

0)

T3 g

CD &>

oCO

to

en

<D

GMo•H,G en

Eh rHH

tn gG-H+JfO

oo

ooCN

ooo

ocn

ocn

m o inCN o cnCN CN ro

ocn

oo

oino

ftft

oCN

ooCM

OOrorH

OCOrH

OO00<-\

Oin

ooo

oooo

ocnrH

oo

in o or- O oCN CN CN

o o Oo o or> in CN00 00 rH

oox>

om o o mm o r»tf> en r^

c

HP

r^ CN in r> o in o O •rH

T3o Cn rH 00 "ST >£> r- cn cCN 00

ocH

CD

U•H>

rH

ucH

rH CN 00

.G

CNO

Mcu

ft

cn

tn

•rH

rd

CD

Sh

CN

MH

fd

gCJ

mCNoo

II

•Hen

ft

mCN+1

G

-rH

en

CU

Xi

cd ». ft CU ft u <CO <D en Cn ^ G

U • <C rd H in^-~. !H •H G U II CU

rH > cn ft rH a> rH PH en Sh G en •H > LD O CU

o TS CD O <C o rd cn O gG CO •H • CO • G

>1 •H -p rH >1 ,G CU • o-^ !S M rd •H M ft u <£>

II H rHen — G o en en rd en W3 cn •H G <C X rH -HEC en T3 X! >1 EC rd -H o— +J c g M rH P H g 0)

u -H en + •H rd CO M• 3 s U 3 o Q ft rH ft

ti 4-> EC +J

U rH 4-> CD >1!H ^ rd rH + CO M

rH ft c ,G rd en • • •

CD tfl ft Xi 00 p 3 rH CN ooCD rH Eh cn ft rH Q) ECP rH id < en r^ r3

CO •H P a> < +O <D CO cu \ c en

CD g G rH M !h o ••

M >i u Q CU 4-> QJ 00 en

•rH M cd rd £ g en ft cn CU

ft in G n 3 •-H >i ft i P3 rd >1 >H M Q <

w EC tJ CO Eh ft ft-

*4 Ch S

101

TABLE TTT - SUMMARY OP STATE SPECIFICATIONS

? z i3 s

COMMENTS

AI.ABAMA

ALASKAAH I ZONA

ARKANSASCALIFORNIACOI,oKAI)0

ConnecticutdelawareKl.uk IDA

(;i:ulo;iA

1 1awA I I

IDAHO

ILLINOIS

INDIANAIOWA

KANSASKENTUCKYI/XIISIANA

MAINEMARYLANDMASSACHUSETTSMICHIGANMINNESOTAMISSISSIPPIMISSOURIMONTANA

NEBRASKANEVADANEW HAMPSHIRENEW JERSEY

NEW MEXICOHEW YORK

N. CAROLINA

N. DAKOTAOHIOoklahomaohe«;on

pennsylvaniarhode islands. carolinaS. DAKOTATENNESSEETEXASUTAH

VEHMONTVIRGINIAWASH 1 NOTON

WEST VIRGINIAWisconsinWYOMING

X

XX

X

X

X

X

X

X

X

X

X

X

X

X

XX

X

X

X

X

X

X

X

X

X

X

X

XX

X

X

X

XX

X

X

X

X

X

XX

X

X

X

X

X

X

X

X

X

XX

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

Specify Mln. Asphalt Application Temp, of199°(390°P).

X

XX'

X

X XXXX X

X X

Specifies tar based pitch for structural plate.

X X Asbestos Bonded smooth lined spec.

No. Spec, on asphalt mastic.

X

Asphalt application spec.

X X

X X X Require 1.5 oz./sq. ft. zinc on asb. bonded.

X X M190 only used on aluminum.Require 1.5 oz./sq. ft. zinc on asb. bonded.Water vapor permeability test on asphalt mastic.Coated pipe no longer used.

X FSS TT-C-494 used on structural plate.

Detailed asphalt physical property tests.

XX X

Aluminized pipe permitted.

X XX Asphalt application spec.

Previously used M190.

X X

No specifications received.

Application Spec.

Application Spec, Require 1.35 oz. per sq. ft.

zinc on asb. bonded.

102

APPENDIX A

CORROSION AND EROSION EFFECTS ON CULVERTS

Corrosion is defined as the deterioration of amaterial by chemical interaction with the environment inwhich it is exposed. While this definition allows anymaterial, metal or nonmetal to corrode, corrosion is gen-erally thought of as occurring to metals. The corrosion ofmetal under drainage pipe exposure conditions is caused byelectrochemical reactions. That is, both electrical andchemical phenomena occur on the metal surface.

The classical corrosion cell has four characteris-tics, all of which must be present for corrosion to occur.First, there must be an anode where oxidation or the loss ofelectrons takes place. Secondly, a cathode must be present.The reaction at the cathode involves the gain of electronsand is called reduction. The anode corrodes, the cathodedoes not. Differences in metal composition, structure ordifferences in the environment surrounding the metal cangive rise to anodic and cathodic areas on the sane piece ofmetal. A single piece of steel may rust because of micro-scopic anodes and cathodes on its surface created by com-positional differences.

The third item required in the classical corrosioncell is an electrolyte. An electrolyte is defined as anychemical substance or mixture containing electricallycharged atoms or groups of atoms which can move in an elec-trical field. Natural water and soil containing solublesalts are electrolytes.

Finally, in the classical corrosion cell, theremust be an electrical connection between the anode andcathode. If a copper bar (cathode) and an iron bar (anode)are suspended in water, without a metallic path betweenthem, no corrosion current can flow between the two metals.But, if they touch or are otherwise joined through a metall-ic path, current will flow causing corrosion of the iron.Note that the iron and copper will both corrode individuallyeven if not connected due to microscopic anodic and cathodicareas on their surfaces.

Chemical reactions occur simultaneously at theanode and cathode. For the reactions to proceed, the gainin electrons at the cathode must equal the loss in electronsat the anode. Examples of reactions at the anode are:

103

E , voltso

Fe Fe+2

+ 2e~ +.440

Al Al+

+ 3e~ +1.662

Zn Zn+

+ 2e~ +.763

Where E = standard oxidation potential at 25°C using ahydrogen electrode

The metal is transformed into an ion or charged atom andelectrons. The metal ion passes into the electrolyte. Theelectrons move via the metallic path to the cathode. Possi-ble reactions at the cathode are:

M + e > M

+ +2M + e -+ M

2H+

+ 2e" +- H

O + 4H + 4e -* 2H

+ 2H O + 4e -+ 40H~

where M is the metal corroded such as iron, aluminum orzinc. The chemical reactions proceed because the anode andcathode are electrically different. If one could separatethe anode and cathode and place a voltage measuring devicebetween them, a voltage would be observed. Voltage is ameasure of the driving force of the reaction.

All corrosion affecting drainage pipe is caused byvoltage differences from one surface to another. Corrosionoccurs from several causes, such as: local compositiondifferences on the metal surface, bimetallic couples, oxygendifferential cells, pH differential cells and interferencefrom nearby cathodic protection systems. Anaerobic sulfatereducing bacteria (sporovibrio desulfuricans ) usually foundin wet clay soils of neutral pH, cause corrosion attack byacting as depolarizing agents. Aerobic sulfur-oxidizingbacteria, such as thiobaccillus thiooxidans cause corrosionin low pH environments by the production of sulfuric acid.Bacteria have been reported as causing corrosion of steelculverts and are known to attack aluminum but corrosion ofaluminum culverts by this mechanism has not been reported.

104

Corrosion can occur in a uniform manner over alarge surface area or in localized areas. There are dif-ferent forms of localized corrosion that can affect drainagepipe. Pitting corrosion causes the formation of cavitiesextending into the thickness of the metal. Other forms oflocalized corrosion in addition to pitting are known toaffect aluminum alloys under certain circumstances. Exfoli-ation is the preferential corrosion of aluminum grain boun-daries which causes a scaling off of the surface in flakesor layers. Intergranular corrosion of aluminum is prefer-ential grain boundary corrosion extending into the metalthickness

.

Erosion can accelerate the effects of corrosion.Aluminum and iron (under certain conditions) form protectiveoxide films which can inhibit further corrosion. Stones andother debris can abrade this film allowing further corrosionto occur. This results in a more rapid deterioration thanwould normally occur with corrosion or abrasion alone.

105

APPENDIX B

LITERATURE REVIEW SUMMARY

The results of interviews and summaries of techni-cal reports are included in this Appendix. Appendix Cpresents the published references. Other sources of infor-mation are listed after each state's summary.

Alabama

Alabama has acid soil areas, coastal areas andswampy areas. Galvanized steel, uncoated concrete and claypipe are used for culvert pipe. Aluminum pipe is used butnot extensively. Galvanized steel is always installed withan asphalt coating and paving applied according to AASHTOSpecifications. Some polymeric coatings have also been usedbut in-service performance is not known. Metal pipe can beused with a soil pH of 4.5 or greater. Concrete or claypipe is used in acid areas, coastal and swampy areas.

A study conducted in 1969 2 3 concluded that asphaltpaving deteriorated in areas of high erosion and extremeacidity or alkalinity and that water eventually penetratesasphalt.

Service life prediction charts using pH, soil resistivityand dissolved oxygen content are presented. Criteria formaterials used are presented as follows:

pH < 4.5 - use concrete or vitrified clay. Inareas of extreme acidity, use coatingon concrete.

4.5<pH<8.5 - use concrete, bituminous coated gal-vanized steel or bituminous coatedaluminum

pH > 8.5 - use concrete galvanized steel oraluminum

pH >> 8.5 - use concrete", coated galvanized steelor coated aluminum

Additional References:

Highway Specification Section 530Special provision 1147, 6-1-77

106

Technical Memorandum No. 3-70(2), 6-1-7 rev.3-23-76, "Criteria"Mr. Edward Eiland, Engineering Dept., HighwayDept . , Montgomery, AL

Alaska

Both galvanized steel and aluminum culvert pipeare used at the contractor's option. Concrete, clay, as-bestos cement and bituminous fiber pipe are also used fordrainage pipe. Asphalt coating and asbestos bonded asphaltcoated galvanized steel are used in areas of high acidity oralkalinity. Most culverts are installed uncoated . Coatingsare reported to be performing satisfactorily. Abrasion onthe invert is reported to be the worst problem.

Additional References:

Highway Sepcifications , Section 602, 603, 706, 707Mr. Ray Shumway, Highway Dept. Juneau, Alaska

Arizona

Galvanized steel, aluminum, aluminized steel (noinstallations yet), bituminous fiber, asbestos cement andcast iron are permitted under state specifications. Gal-vanized steel and aluminum are normally used uncoated.Galvanized steel is coated when the soil resistivity isbelow 2,000 ohm-centimeters. Asphalt coating and paving areused on galvanized steel and applied according to ASSHTOSpecifications. Culvert durability problems are causedmostly by external corrosion.

A study completed in 1973 10 reports good corre-lation between electrochemical measurements and soil corro-sion. Cathodic protection studies using sacrifical anodeson galvanized steel culvert pipe yielded unsatisfactoryresults. Cathodic protection tests used magnesium anodes toprotect uncoated galvanized culvert. The cost of providingprotection in this manner was judged unreasonable. Bitumi-nous coatings reportedly increase service life by at leastfifteen years.

Developments since the 1973 report indicate thataluminum culvert pipes are giving satisfactory service insoil with resistivities greater than 500 ohm-centimeters.Field coatings on galvanized steel are not as durable asfactory-applied asphalt. Aluminized steel pipe is beingevaluated in the laboratory and in field tests.

107

Additional References

Highway Specifications, Section 707Mr. Steve Dana, Research Engineer, Phoenix, AZ

Arkansas

Concrete, galvanized steel and aluminum areallowed as contractor's options. Asphalt coating of gal-vanized steel was used in the past but state experienceindicates that the coatings are not needed. No durabilityproblems for culvert coatings were reported.

Additional References:

Highway Sepcifications, Section 606Mr. A. E. Johnson, Asst. Chief Engineer, LittleRock, AR.

California (Selected for Field Study)

Galvanized steel, aluminum, asbestos cement andconcrete are permitted for use as culvert pipe. Coatingsused include asphalt and polymeric coatings applied toAASHTO Specifications. Polymeric coatings are currentlylimited to areas where interior abrasion is not severe. Aprocedure was developed in California 3 ' 37 for estimatingservice life utilizing resistivity and pH. Fifty years isused as the basic design life criteria. A reported 8 to 20additional years service life are obtained for asbestosbonded bituminous coated and paved invert, 5 to 15 years forbituminous coated and paved and zero to 8 years for bitumi-nous coating (Highway Design Manual, Table 7-851. 3A). Thelower life expectancy is for abrasive bed loads at flowvelocities greater than 7 feet per second (.213 meters persecond). California has new criteria for using aluminumusing a 50 year design life. Aluminum is permitted wherethe pH is greater than 5.5 and less than 8.5 and the resis-tivity is greater than 1500 ohm-cm. Aluminum culverts arenot permitted where experience shows that deterioration cantake place. Additional provisions are made for backfill andmarginal pH and resistivity situations.

A study completed in 1973 1

8

evaluated 4 coatings,including asphalt, United States Steel "Nexon" , DaubertChemical Co. "Pioneer Culvert Mastic No. 100 8" and PacificCorrugated Culvert Co. "Copoly X Terrashield" and "Raceway".Laboratory abrasion tests indicate that the "Nexon" coatinghas a better abrasion resistance than asphalt or Pioneermastic but would wear through faster than the asphalt due to

108

the thinner film. Salt spray tests revealed a loss of bondbetween the asphalt and test plate with the others retainingtheir bond.

A study completed in 1964 2 recommends using coatedaluminum where the pH is less than 5 and the pH is greaterthan 8. The researchers found no relationship betweenresistivity and corrosion and also that cathodic protectioncan control corrosion when soil resistivity is below 1,500ohm-centimeters. Other findings were that aluminum shouldnot be used under abrasive conditions or when the resis-tivity is below 2,000 ohm-centimeters.

In a 1977 study on abrasion 77, hot dipped asphalt,

coal tar laminate (Nexon) and PVC (Plasticote) were comparedusing a rotating drum abrasion test. Test results indicatedthat the two polymeric coatings have abrasion resistanceequal or better than asphalt but do not have equal resis-tance to abrasive flow. Polymeric coatings are approved foruse on the soil side of culverts and on the interior whereabrasive flow is not expected.

Additional References:

Highway Specifications, Jan. 1979 Ed, Section 66California Test Method 643, 1978Standard Special Provision 66.01, 1-3-78Highway Design Manual, 5-1-7 2, Sections7-841, 7-851Mr. J. Robert Stoker, Cal Trans, Sacramento, CAMr. Roy Chalmers, Cal Trans, Sacramento, CA

Colorado (Selected for Field Study)

Galvanized steel, aluminum and concrete are usedfor culvert pipe. Specific restrictions are based on acriteria utilizing pH, chloride, sulfate and resistivityinformation. Coatings used include asphalt, asbestos bondedasphalt and polymeric coatings.

A culvert performance study completed in1968 found that asbestos bonded asphalt coated galvanizedsteel and reinforced concrete were useful in acidic oralkaline waters. Uncoated galvanized steel and aluminumwere usable in all but highly alkaline environments. Stain-less steel culverts exhibited corrosion in alkaline environ-ments .

Another study published in 1977 39 states that cladaluminum performs better than galvanized steel and that

109

galvanized steel corrodes in high alkali, acid or saltenvironments. Stainless steel culverts were found to cor-rode rapidly in high salt and alkaline soil. Asbestosbonded asphalt coated pipe was found to offer only temporaryprotection due to abrasion and thermal cracking. An un-bonded plastic wrap over an epoxy coating was tried butfound to lead to worse deterioration than if the plasticwrap was not used. Several polymeric coatings, includingU.S. Steel Corporation "Nexon", Wheeling Corrugating Company"Plasticote" and Inland Steel Company "Blacklad" are in testat various test sites throughout the state. Some loss inbond was noted at the seams of the Nexon coated sample butno other deterioration was reported.

Additional References:

Highway Specs., 1976 Ed and 8-1-76 Supplement -

Section 624Mr. Herb Swanson, Research Engineer, Denver, CO

Connecticut

Galvanized steel, aluminum, concrete, asbestoscement and bituminized fiber pipe are used in Connecticut.Galvanized steel is always installed with asphalt coatingand paving. There are no formal material selection cri-teria. Abrasion at the invert is the worst problem. Fieldapplied tar based pitch is used on structural plate cul-verts.

Additional References:

Highway Specs. & 11-77 Supplement, Section M.08Mr. George Upton, Highway Engineer, Hartford, CT

Delaware

Galvanized steel and aluminum are used in Dela-ware. Concrete is no longer used. Galvanized steel isinstalled with asphalt coating and paving. There are noformal material selection criteria. A granular backfill isused around culverts under major highways.

Additional References:

Highway Specs., Sections 617 & 618Mr. Watson Baker, Jr. -Chief, Materials & Research,Dover DE

110

Florida

Galvanized steel and concrete are used as alter-nate materials. Aluminum is used but only at selectedinstallations. Bare galvanized steel is used where the pHis greater than 5.9 and resistivity is greater than 4,500ohm-centimeters. Where the pH is less than 5.9 and resis-tivity is less than 4,500 ohm-centimeters, the Californiatest method is used. Specifications are currently beingwritten for aluminum. Galvanized steel is asphalt coatedand paved. A polymeric coated culvert was installed fouryears ago. The basic durability problem with asphalt is acoating breakdown caused by exposure to water and light.

A 197 5 study 5 found that bituminous coatingsextend culvert life ten years and paving adds another thirtyyears to culvert life. Principle metal loss in Florida isdue to interior corrosion.

Additional References:

Highway Specs., 1977 Ed., Section 943-6Mr. R. P. Brown, State Materials Engineer, Talla-hassee, FL

Georgia

Georgia utilizes galvanized steel and concrete forculverts. Asphalt and polymeric coatings are used where pHand resistivity criteria show it to be needed. Paving isused where an abrasive load is expected. Only limitedservice history is available for polymeric coatings. Alumi-num pipe is not used. An uncoated galvanized steel culvertfailed in a stream with a pH of 2.8 in less than six months.Failures originate on interior surfaces.

Additional References:

Highway Specs., Section 844Mr. Hugh Tyner, Chief, Research & DevelopmentBureau, Forest Park, GA

Hawaii

Concrete, galvanized steel and asbestos cementpipe are used. Concrete is used in the more corrosiveenvironments. Asphalt coating was used but has been dis-continued because of rapid failure due to erosion by siltand sand. Hawaii uses California Method 643 to determineculvert life and a 40 year design life criteria. Pipe

111

thickness is increased in acidic soils to give the requireddesign life.

Additional References:

Highway Specs., Sections 603.03, 707.03Mr. Walter Kuroiwa, Materials Testing & ResearchBranch, Honolulu, Hawaii

Idaho

Galvanized steel, aluminum, concrete, asbestoscement and acrylonitrile-butadiene-styrene (ABS) are usedfor drainage pipe. Coatings are applied to both steel andaluminum and include asphalt and asbestos bonded asphaltdipped. Paved inverts are used. California Method 643 isused to estimate culvert life. Deterioration of both in-terior and exterior surfaces occur with coatings judged tobe successful.

Several studies were performed in Idaho on culvertdurability 7 ' 1 5 ' 3 5/ 3

6

. Uncoated galvanized pipe is recom-mended unless the California test method indicates a life ofless than 4 years. Uncoated aluminum pipe is acceptable ifthe resistivity is greater than 1,000 ohm-centimeters andthe pH is between 5 and 9.

Additional References:

Highway Specs., Sections 601, 706Mr. J. W. Hill, Research Supervisor, Boise, ID

Illinois

Drainage pipe materials include galvanized steel,aluminum, concrete, asbestor cement and clay. Concrete isused on major projects. Asphalt and polymeric coatingsapplied to AASHTO Specifications can be used. Paved invertsare sometimes used. Deterioration of the asphalt and pavingat the water line was reported. In general, asphalt coat-ings have not performed well. Asphalt deterioration at thewater line, and resulting corrosion, are considered majorfactors. Corrosion mainly occurs on the inside surface. Analuminum culvert installed 10 years ago in a pH 4 environ-ment is reported to be performing well where both a steeland concrete culvert previously failed. Fertilizer and minerunoff are major problems in Illinois.

112

Additional References:

Highway Specs., 7-1-76 Ed, Sec. 602, 710Supplemental Spec, for Sec. 603, 7-1-77Supplemental Spec, for Sec. 602, 2-1-77Culvert Design Manual (6-200), 12-75Subsurface Drainage (6-400), 3-78Mr. Don Fowler, Design Engineer, Springfield, IL

Indiana

Indiana utilizes galvanized steel, aluminum, con-crete, vitrified clay, asbestos cement, ABS and cast iron.Asphalt coating and paving are used as well as asbestosbonded asphalt coated pipe. Both asphalt paving and con-crete paving are used. Highway specifications requireconcrete paving to be 2-inches ( 5 cm) thick over corrugationcrests and also that wire fabric, welded to the invert, beused as reinforcing. Aluminum is used where the pH isbetween 4 and 8 and traffic flow is not greater than 200vehicles per hour. Asbestos bonded asphalt coated pipe isused if the pH is less than 4 . The type of roadway to beconstructed generally determines the culvert material used.No known coating problems were reported. Internal deteri-oration is considered to be the worst. A 2 year study hasstarted to evaluate culvert coatings. The study includeslaboratory evaluations of coatings in a low pH environment.

Additional References:

Highway Specs., Sections 715, 907Drainage Design Manual, Section 7-400, 2-77Mr. S. R. Yoder, Chief, Div . of Design, Indian-apolis, IN

Iowa

Galvanized steel, concrete and clay are used forculvert pipe and aluminum pipe is under study. Bituminouscoating without paving is used. The designer chooses thematerial to be used. Corrosion from both acid and alkalinewaters is reported to cause corrosion on the pipe bottomfrom the soil side. Cathodic protection was considered atone time but never used because of cost considerations.

A 1970 study of 52 culverts which were 2-6 yearsold indicated that the asphalt coating on all of the cul-verts had deteriorated. - Temperature fluctuations arethought to be the cause. The worst deterioration was at theculvert ends and at the silt line.

113

Additional References:

Highway Specs., 1977 Ed, Sec. 4141Mr. R. H. Given, Chief Engineer, Ames, IA

Kansas

Concrete, galvanized steel and aluminum are usedas culvert materials in Kansas. Bituminous coatings areused but do not perform well. Studies indicate that bitumi-nous coatings crack and disbond after about threeyears'* 3

th h

. Most corrosion occurs on the interior of theinvert. A 1968 laboratory study 44 indicates that etchingthe zinc coating .improves adhesion of asphalt by 21 percent.The study reports that alternate wetting and drying loosensthe bond of asphalt to zinc. Both studies recommend thatthe use of coatings in Kansas be discontinued.

Additional References:

Highway Specs., 1973 Ed, Section 1009Mr. Carl Crumpton, Assistant Engineer, Planningand Development, Research Section, Topeka, KS

Kentucky (Selected for Field Study)

Coated galvanized steel, coated aluminum, asbestoscement, bituminized fiber and concrete are used in Kentucky.Coatings include asphalt, asbestos bonded asphalt and pav-ing. A study is now in progress to develop useful materialselection criteria. The most severe problem in the state isacid mine runoff. A study at the acid runoff test site atMorton's Gap, near Madisonville, indicates deterioration ofasphalt coated pipe 22

. Studies are in progress in the acidstream, pH between 2-2.5, to evaluate aluminum and stainlesssteel

.

Concrete pipe is sometimes paved with vitrifiedclay or epoxy. The epoxy paving consists of a primer andepoxy mortar top layer. The primer consists of an epoxyresin with polysulfide polymer and amine catalyst. Themortar is made with the same epoxy resin as the primer plussilica mortar sand and mineral fiber as fillers.

Additional References:

Highway Specs., 1976 Ed, Sections 809.01, 810Special Provisions Nos . 7(76) & 14(76)Special Specifications for Protective Coating of

114

Reinforced Concrete Culvert Pipe, Ohio County,WK5-2Mr. W. B. Drake, Assistant State Highway Engineer,Research, Frankfort, KY

Louisiana (Selected for Field Study)

Uncoated aluminum, asphalt coated steel and con-crete pipe are used in Louisiana. Asbestos bonded asphaltis also used where the pH is less than 5 and the resistivityless than 1,500 ohm-centimeters. Adhesion problems withasphalt coatings are being experienced. There is no seriousabrasion problem but there is a sulfide problem.

A study in 1971 1 indicated that the Californiatest method was applicable to Louisiana and that asphaltcoatings added 8 years to culvert life. Asbestos bondedasphalt was considered superior to asphalt.

Studies are ongoing in the state to evaluatevarious coatings, including asphalt, asbestos bonded as-phalt, polymeric coatings (coal-tar laminate and polyethy-lene) and uncoated steel and aluminum28 /

29. Results re-

ported to date indicate that asbestos bonded asphalt isout-performing other coatings. Coating disbondment isoccurring on asphalt coated aluminum, asphalt coated steel,polyethylene and coal tar laminate in some environments.Complete pipe perforation occurred on the galvanized steel,aluminum plate arch, asphalt coated steel, coal tar laminatecoated steel, and polyethylene coated steel at the mostsevere test locations. The test samples were installed in1973.

Additional References:

Highway Specs., 1977 Ed, Section 1007Engineering Directives & Standards Manual, No. II,2.1.6, 2-24-77 and 2.1.1, 1-3-77Mr. E. J. Breckwoldt, Research Div., BatonRouge, LA

Maine

Galvanized steel, aluminum and concrete are usedfor culverts in Maine. Bituminous coated and paved gal-vanized steel culverts with granular backfill are used.Asbestos bonded asphalt coating is used in coastal areas.Corrosion problems are limited to the pipe interior. Onesmooth bore culvert was installed with U.S. Steel Nexon

115

coating in 1972 in Palermo, Maine. The environment isnon-abrasive and the culvert is reported to be performedwell. A five year old aluminum zinc coated test culvert isout-performing a galvanized test culvert. Personnel reportproblems occurring in the last two years with disbonding ofbitumenous coatings. Studies are currently being conductedon aluminum pipe, asbestos bonded pipe, aluminum coatedsteel, polymeric coatings (some freeze thaw separation isoccurring), epoxy coatings (bubbles and disbonds at pH 4)and aluminum zinc coated steel.

Studies in 1974 2 5 and 1976 2 " indicate that tenextra years of life can be obtained with bituminous coat-ings, six if there is high stream flow. Aluminum culvertsperform better than galvanized steel if the pH is between5.8-7.9 and resistivity is greater than 8,2 50 ohm-centi-meters. A study in 1975 determined the average metal lossprobability of galvanized steel culverts.

Additional References:

Highway Specs., Section 7 07Mr. K. M. Jacobs, Materials and Research Div.,Bangor, ME

Maryland

Galvanized steel, clay, concrete, asbestos cement,bituminized fiber and aluminum pipe are used. Aluminum isbeing used at test locations only. Bituminous coated gal-vanized steel was used in areas of high sulfate mine waters.Aluminum coated steel is being considered.

A study completed in 1971 2 7 concluded that bitu-minous coatings add four years life and paving adds eightyears life to culverts and was, therefore, not warranted.Bituminous coatings are no longer used.

Additional References:

Highway Specs., 1968 Ed., 197 5 Revision and Draftfor 1979 Ed.Mr. Nathan L. Smith, Jr., Assistant Chief Engi-neer,Materials and Research, Brooklandville, MD

Massachusetts

Concrete, coated galvanized steel and coatedaluminum are optional culvert materials. Asbestos cement,

116

clay and cast iron are also used for drainage pipe. Bothgalvanized steel and aluminum are coated with asphalt andpaved according to AASHTO-M190. Polymeric coatings are alsoapproved for use. An aluminum-zinc coated steel culvert wasinstalled at one location but no performance data is avail-able. There are no formal material selection criteria.

Additional References

:

Highway Specs., Sections 230, M5.03Mr. Gene Bastansa, Research & Materials Engineer,Wellesley Hills, MA

Michigan

Concrete, galvanized steel and aluminum, clay,bituminized fiber, cast iron polyvinylchloride (PVC), poly-ethylene (PE) and ABS pipe are used for culverts and otherdrainage pipe. Bituminous coatings are used for some pipes.There is no written selection criteria. There is a sectionalong Interstate Highway 7 5 where deteriorated galvanizedsteel culverts are being lined with polyethylene.

A study completed in 1975 33 used polarizationcurves and potential measurements to evaluate culvert corro-sion. A good correlation was found between the corrosioncurrent determined by polarization and the visual per-formance rating. Resistivity, chloride, sulfates and pHwere found to be good indicators of corrosion but no rela-tionship was reported. Bacterial corrosion is considered animportant factor.

Another report presented results of laboratorytests on three polymeric coatings 17

. Results of immersiontests in several chemicals indicate disbondment of coatingdue to dissolution of steel or zinc.

The most recent study, completed in 1979, includedonly two coated culverts, a nine year old asphalt coatedculvert and a nine year old asbestos bonded asphalt coatedculvert 49

. Both coatings had deteriorated in the invertportions and the metal corroding. It was felt that thecoatings did not provide a significant increase in lifeexpectancy. Service conditions for these two culverts arenot provided in the referenced report.

Additional References:

Highway Specs., 1976 Ed., Section 8.08Mr. Glen Caldren, Design Engineer, Lansing, MI

117

Mississippi

Concrete, PVC, aluminum and galvanized steel areused for culverts and other drainage pipe. Bituminouscoatings are used on both aluminum and steel. Invert pavingis sometimes used. Mississippi uses the California TestMethod 643 method to determine whether or not to use metaland if coating is needed. Corrosion on interior and ex-terior surfaces is a major problem with culvert durability.

A 1964 study indicated that bituminous coatingsadded 2-7 years to the life of galvanized culverts'* 8

.

Additional References:

Highway Specs., Section 709Mr. Buford Stroud, Design Engineer, Jackson, MS

Missouri

Clay, concrete, galvanized steel and aluminum arethe culvert and drainage materials used. A bituminouscoating for galvanized steel was used but was dropped fromthe specifications because of adhesion problems. The causeis thought to be water absorption and freezing. Polymericcoatings are used but no experience data is available yet.Material selection is based on an evaluation of existinginstallations. Corrugated metal pipe is not used undermajor highways.

Additional References:

Highway Specs., 1977 Ed., Sections 725, 1020,1021, 1024Mr. Gerald Manchester, Design Engineer, JeffersonCity, MO.

Montana

Galvanized steel, concrete and aluminum pipe areused in Montana. Asphalt, asbestos bonded asphalt andpolymeric coatings are used. Material selection criteria isbased on a test of soil conditions. Non-metallic or bitumi-nous coated aluminum pipe are used if the soil resistivityis less than 1,000 ohm-centimeters and the pH is greaterthan 5. Bituminous coated galvanized steel is used if thesoil pH is less than 6 or greater than 9. Charts are givenfor corrosivity of steel versus alkalinity and pH. Uncoatedgalvanized steel is considered usable in the pH range 6-9and uncoated aluminum is approved if the pH range is 5-8.5

118

and resistivity is greater than 800 ohm-centimeters andnon-abrasive flow velocity is less than 7 fps (2.13 m/s).Concrete is approved if the pH is greater than 6 and thesoluable sulfate content is less than .25 percent. Bitumi-nized coatings and vitreous clay are used in other areas.Concrete paving is used on large culverts under abrasiveflow conditions.

Additional References:

Highway Specs., Section M-170Special Provisions 190-1(77)22, Unit 7

Montana Dept. of Highways CriteriaSurvey & Plans Manual, Part 15, Section 15-333Mr. Carl Peil, Design Engineer, Helena, MT

Minnesota

Aluminum, vitrified clay, asbestos cement, gal-vanized steel and concrete are used for culverts and otherdrainage pipe. Asbestos bonded asphalt and polymeric coat-ings are used. Asphalt coatings tend to check and oxidize.Asbestos bonded asphalt is performing well. Paving isspecified for both bituminous coated and polymeric coatedculverts in abrasive conditions. Corrosion deterioration onboth interior and exterior surfaces occur. The state isdivided into four zones which are used as criteria for theuse of metal culvert pipe. The four zones are based on theresults of a study published in 1969 (Investigation No. 116,"Serviceability of Corrugated Metal Culverts") which relatecthe life expectancy of metal pipes to soil acidity andwetness. Increased metal thickness and coatings are usedwhere flow velocities are expected to exceed 5 feet persecond (1.52 m/s).

Additional References:

Highway Specs., Sections 2501, 2502, 2503, 3226Drainage Design Manual 5-294.300, 12-12-77;5-294.343, 4-23-74Mr. F. W. Thorstenson, Director, Materials, Re-search & Standards, St. Paul, MN

Nebraska

Concrete, galvanized steel, aluminum, clay andcast iron are used for culverts and other drainage pipe.Bituminous coating and paving were used in some areas de-pending on local conditions but coatings are no longerconsidered needed and are no longer covered in the specifi-cations.

119

Additional References:

Highway Specs., 1965 Ed. and 2-19-70 SupplementalSpecs., Section 99Mr. Donald Swing, Materials & Tests Engineer,Lincoln, NB

Nevada

Aluminum, asbestos cement, bituminized fiber,clay, concrete and galvanized steel are used for culverts.Bituminous and coal tar laminate coatings are approved foruse. Material selection is based on a chemical analysis forsulfate salts. A coating is used if the sulfate contentexceeds 1,500 ppm. Asphalt performance is not consideredsatisfactory since salts are trapped under the coating andaccelerate corrosion. Most corrosion occurs on the bottomexterior culvert surface.

Additional References:

Highway Specs., 1976 Ed., Sections 601, 709Mr. J. M. Desmond, Asst. Chief Materials Engr.,Carson City, NV

New Hampshire

Concrete, galvanized steel, aluminum, asbestoscement and plastic pipe are used for culverts and otherdrainage pipe. Coatings include bituminous, asbestos bondedasphalt, polymeries and aluminum coated steel. Polymericcoatings are limited to non-abrasive conditions. An alumi-num coated steel culvert is currently being tested. Thereare no written materials selection criteria. The designengineer evaluates the bed load, slope and type of runoff todetermine the type of pipe and coating. Interior abrasionis the worst problem although future problems from roadsalts are anticipated. Differences in asphalt performancewith manufacturer are reported. Problems are reported inmaintaining the structural integrity of aluminum duringconstruction.

Additional References:

Highway Specs., 1974 Ed., with 11-77 Addendum #3,Section 591, 603Mr. Joseph Grady, Design Engr., Concord, NH

120

New Jersey

Galvanized steel, concrete and aluminum are usedfor culverts in New Jersey. All metallic pipe is coatedwith asphalt and sometimes paved. Material and coatingcriteria are based on a statistical method developed by NewYork. The New York method uses probability curves forcorrosion rates to determine service life. New Jersey'slargest durability problem is on the inside of the pipe.

A 1974 study34 presents criteria for using alumi-num pipe, steel and concrete. Concrete is used where the pHis less than 4.5 and aluminum can be used if the pH isbetween 4.5 and 8.5 and the flow velocity is less than 2.4meters per second ( 8 feet per second )

.

Additional References:

Highway Specs., Section 8.1.3Mr. Kenneth Afferton, Design Engr., Trenton, NJ

New Mexico

New Mexico generally utilizes uncoated galvanizedsteel, aluminum and concrete. Only one coated installationwas reported. This involved a galvanized steel culvert in ahighly alkaline area (pH 12) of northwestern New Mexicowhere the zinc deteriorated. In this case, an asphaltmastic coating (Fed. Spec. WW-P-40 5B, Coating F) was used.Erosion of this coating has been observed after 1 year,California Method 64 3 is used as well as test coupons buriedat the construction sites for larger culvert structures.Mechanical problems during construction of aluminum culvertsare reported.

Additional References:

Highway Specs., Section 501Mr. R. S. Busch, Bridge Construction Engineer,Santa Fe, NM

New York (Selected for Field Study)

Bituminous coated and paved galvanized steel anduncoated aluminum pipe are used. New York experiencesdurability problems with plain asphalt coated pipe. Thestate is divided into corrosion rate zones according to a

new study which is not yet published. One zone corrodessteel at the rate of .00 50 8 cm per year (.00 2 inch per year)and the other at .01016 cm per year (.004 inch per year).

121

Corrosion occurs primarily on the inside surface. Polymericcoatings are used in three installations. One installationon the Genessee Expressway is experiencing abrasion anddisbondment problems after one year. Concrete pipe is usedexclusively on Long Island. Coated and paved pipe isthought to extend culvert life by twenty-five years, butthis is to be verified by a test program. Reinforced con-crete paving (Class D Portland Cement Concrete) can be usedon uncoated steel or aluminum. Highway specifications allowthe use of smooth lined steel or aluminum pipe (corrugatedouter surface and smooth interlocking interior).

The study on uncoated steel and aluminum is to bepublished in 1978- or 1979. A previous study had establisheda procedure using corrosion rate probability curves and acorrosion rate zone map for estimating service life. Thepresent study indicates that this method might not be realis-tic because of the test methods and conditions at the time 1 3

of the original study.

Additional References:

Highway Specs., 1978 Ed. and Addendum No. 1,1-2-79,Sections 600, 501, 707Mr. Pete Bellair, Senior Civil Engineer, Albany,NY

North Carolina

Coated galvanized steel and concrete pipe are usedfor culverts. Asphalt coating, sometimes with paving, isused in the central and western portions of the state whileconcrete is used on the eastern coastal plain. Interiorcorrosion is a major problem and paving is sometimes lostduring periods of high runoff. Specifications allow the useof aluminized pipe in place of galvanized steel and the useof polymeric coatings in place of asphalt.

An early study in 1944 32 recommends that asphaltcoatings be used in areas of continuous flow and persistentorganic water. Asphalt coated and paved galvanized steelperformed well in this investigation.

Additional References:

Highway Specs., Section 824, 932, 1978 Ed.Project Special Provisions, Corrugated SteelCulvert Pipe, Section 932, 8-3-79Mr. M. P. Strong, Research Engineer, Raleigh, NC

122

North Dakota

Galvanized steel, aluminum, cast iron and concretepipe are acceptable for drainage pipe in North Dakota.Asphalt coating was used at one time but is now consideredunnecessary. There is no criteria for material selection.Severe corrosion on a structural plate pipe is now beinginvestigated to determine the cause.

Additional References:

Highway Specs., Section 830Mr. Robert T. Peterson, Research Engineer, Bis-mark, ND

Ohio (Selected for Field Study)

Concrete, vitrified clay, coated galvanized steeland aluminum are used for culverts and underdrains. Asphaltlined and paved pipe and asbestos bonded asphalt pipe areused to coat galvanized steel. A vitrified clay lining issometimes used. New design criteria call for asphalt coatedand paved steel pipe and field paving of structural platestructures where the pH is between 4.0 and 5.5. Vitrifiedclay or lined reinforced concrete are to be used where thepH is below 4.0 or above 9.5. Concrete pipe is sometimeslined with a polyamide cured coal tar epoxy (MIL-P-23236 ) ata dry film thickness of .0762 cm (.030 inch). Asbestosbonded asphalt with paving, structural plate steel pipe withstainless steel bottom plates and structural plate steelwith concrete paved bottom and vitrified clay invert arealso to be used where industrial wastes are expected. Linedconcrete, asbestos bonded asphalt or vitrified clay are tobe used in highly acid or alkaline soils. Coal tar laminate(Nexon) coated culverts have been used in acid mine waterareas.

An as yet unpublished study on culvert durability6indicates that adherence problems of asphalt coatings arelargely due to improperly cleaned pipe and improper pipetemperature during coating application in addition to lowresistance to thermal cycling. Bituminous coatings aloneare thought to be of little value. Field applied concretepaving erodes rapidly under acidic conditions. Vitrifiedclay lines and asbestos bonded asphalt are reported resis-tant to acidic conditions. The coal tar laminate tested hadpoor resistance to abrasive flow and bed loads over 2.54 cm(1 inch) and also exhibited delamination at lock seams.Corrosion of rivets on laminate coatings in acid waters isreported. Recommendations for applying asphalt include an

123

acid wash of the zinc, and asphalt bath temperature of 204 °C± 15° (400°F ± 5°) where the pipe is not preheated and193. 3 °C ± 15° (380°F ± 5°) where the pipe is preheated to149°C (300°F).

Another unpublished study of culverts in Ohio bythe University of Akron indicates that, of the culvertsexamined in the previous study 6

, the average service life ofbituminous coatings was 3.16 years and bituminous coatingwith paving was 18.71 years.

Additional References:

Highway Specs., 1977 Ed, Section 1104.25,1104.26, 707Draft of ODOT Location & Design Manual, 2-78Mr. John Herl, Hydraulic Engr . , Columbus, OH

Oklahoma

Galvanized steel is the culvert material used inOklahoma. Bituminous or coal tar laminate coating is usedin the eastern third of the state and paving is used wherethe velocity is expected to exceed 2.44 m/s (8 fps ) . Theeastern part of the state has low pH and high resistivitysoil. Coal tar laminate coatings are used in western Okla-homa where salt areas exist. Coal tar laminate is thoughtto have 2-4 times more abrasion resistance than bituminouscoated pipes but paving is considered necessary in areas ofhigh flow velocities and abrasive loads. Aluminum is notused for culverts because of problems. Differences inasphalt performance and lack of durability of unpaved pipeare reported. A test installation yielded three months lifewith an asphalt coating and six years life with a coal tarlaminate.

A study published in 1971 2 l evaluated culvertperformance and California Method 64 3 with respect to itsapplicability in Oklahoma. The accuracy of the Californiamethod was found to vary with location. The same studyrecommended that pavement cover at least 25 percent of theperiphery and be at least .318 cm (.125 inch) thick abovethe top corrugation.

Additional References:

Highway Specs., Section 726Mr. Curtis J. Hayes, Project Engr., Oklahoma City,OK

124

Oregon (Selected for Field Study)

Aluminum, galvanized steel, concrete, asbestoscement and bituminized fiber pipe are used for drainage inOregon. Steel and aluminum are coated with asphalt west ofthe Cascades. Concrete or asbestos bonded asphalt culvertsare used in coastal areas and aluminum seems to perform wellin swampy areas. Uncoated steel and aluminum are generallyused east of the Cascades.

Studies on culvert durability 11'41 recommend that

uncoated galvanized steel be used where the pH range is4.5-10 and the resistivity is greater than 1,500 ohm-centi-meters and east of the Cascades. Asbestos bonded or bitumi-nous coated and paved pipe are recommended in western Ore-gon. Uncoated aluminum can be used where the pH is between4.5 and 10 and the resistivity is greater than 1,500 ohm-centimeters.

Additional References:

Highway Specs., Sections 603, 707Mr. Stephen Macnab, Hydrologic Engineer, Salem, OR

Pennsylvania (Selected for Field Study)

Galvanized steel, aluminum, stainless steel,concrete, vitrified clay and PVC pipe can be used for drain-age. Design criteria are:

a. pH 3.5 or less, use stainless steel, vitrifiedclay, clay lined concrete, coal tar epoxy linedconcrete, coated galvanized steel or aluminum

b. pH 3.5-5.0, use coated galvanized steel or coatedaluminum, concrete

c. pH 5.0-8, use galvanized steel, aluminum or con-crete

d. pH 8 and above, use coated galvanized steel,coated aluminum, concrete

e. High sulfur content - same as pH 3.5 or less

f. Highly abrasive conditions, use a .0254 cm (.010inch) polymeric coating on the outside surface

g. Water pH less than 5 or greater than 8 and soil pH5-8 and soil resistivity 6,000 ohm-centimeters or

125

greater, use a .0254 cm (.010 inch) interiorpolymeric coating

h. Water and soil pH as in (g) but soil resistivity2,000-6,000 ohm-centimeters, use .0254 cm interiorand .00762 cm (.003 inch) exterior coating

i. Water pH less than 5 or greater than 8 and soil pHless than 3.5 or greater than 8 and soil resis-tivity less than 2,000 ohm-centimeters, use .0254cm (.010 inch) interior and exterior coating.

Pennsylvania used asphalt coatings with paving but onlyfound a three to five year coating life. Polymeric coatingsare now used exclusively.

Laboratory studies in 197 2 indicated that poly-meric coatings are superior to asphalt coatings (plain orasbestos bonded) in abrasion resistance, chemical resistanceand weathering. A study in 1976 16 reports that bituminouscoatings are lost in the majority of culverts within threeyears.

Additional References:

Highway Specs., Supplement Section 707, 1-17-78Design Manual, Change 24 to Publ. 13, Sections2.12.15.02-2.12.15.05, Part 2, 4-12-77Mr. Eugene Eckert, Engineering Coordinator,Harrisburg, PA

Rhode Island

Galvanized steel, aluminum and cast iron are usedfor drainage. All galvanized steel is asphalt coated andsome is paved. There are no formal material selectioncriteria.

Additional References:

Highway Specs., Section M.04.02Mr. Nicholas A. Giardino, Assistant MaterialsEngineer, Providence, RI

South Carolina

Concrete, galvanized steel, aluminum, vitrifiedclay and cast iron are used for drainage. Asphalt coatingis used in some cases and paving is used depending on streamflow conditions. Concrete is generally used because the

126

other materials are more costly. Coated pipe is used whereerosion is expected.

Additional References:

Highway Specs., Sections 714, 715, 716Mr. Richard Stewart, Research & Materials Engi-neer, Columbia, SC

South Dakota

Concrete and galvanized steel are used for cul-verts. Coatings are not used. Soil maps, showing areascorrosive to steel and concrete, are used for design.Deterioration is basically corrosion from the outside of thepipe. Severe corrosion is reported in shale soils.

Additional References:

Highway Specs., 1977 Ed., Section 1000Mr. Merle Buhler, Design Engineer, Pierre, SD

Tennessee

Concrete, galvanized steel and aluminum are usedfor culverts. Bituminous coating of the invert was useduntil six years ago. Coatings are no longer used since thecoating tended to crack and erode. Resistivity and pH areused as material selection criteria.

Additional References:

Mr. J. B. Wilee, Materials & Test Engr., Nash-ville, TN

Texas

Galvanized steel, aluminum, concrete, clay, asbes-tos cement, bituminized fiber, ABS and PVC are used fordrainage. There is no aluminum pipe in service yet. As-phalt coated pipe is occasionally used and asphalt bondedasphalt coated pipe is permitted but has not been used.Interest in polymeric coatings was expressed. Internal andexternal corrosion problems exist. There are no formalmaterial selection criteria.

Additional References:

Highway Specs., 1972 Ed., Items 460, 464

127

Special Provision to Item 460, 8-78Special Provision to Item 464Special Specification, Item 4233, 8-78Mr. Gilbert Barr, Supervising Field Engr., Aus-tin, TX

Utah (Selected for Field Study)

Galvanized steel, aluminum, concrete and clay pipeare used for drainage. Asbestos bonded asphalt, coal tarlaminate and asphalt coatings are used. Design criteria aredetailed in a 1974 study . Reinforced Portland bementpaving is specified in some installations.

The 1974 study reports that the total soluble saltcontent is the most important factor in pipe corrosion. Theeffects of pH and resistivity are greatest at soluble saltcontents of less than 1.5 percent and both lose dominance athigher salt concentrations. The corrosive effects of saltspeak at 5 percent. Additional service life is reported tobe 2 6 years for asbestos bonded asphalt coated steel and 16years for asphalt coated steel. Sulfate contents exceeding.5 percent might be the principle deteriorating agent forconcrete according to the study. An expression was derivedrelating the pipe condition to the soluble salt content, pH,soil resistivity and age.

Additional References:

Highway Specs., 1970 Ed., Section 715, 514Mr. Dale Peterson, Research & Development Engi-neer, Salt Lake City,, UT

Vermont

Galvanized steel, aluminum, concrete and asbestoscement are used. Asphalt coating is used on major highwayconstruction and polymeric coatings are approved but havenot been used. There are no material selection criteria fordurability. No problems are reported with coating dura-bility.

Additional References:

Highway Specs., Section 601, 711Mr. Robert F. Shattuck, Hydraulics Engineer,Montpelier, VT

128

Virginia

Galvanized steel, aluminum, concrete, vitreousclay, PVC and bituminized fiber pipe are used for drainage.Aluminum pipe is used uncoated. Bituminous coating, pavingand increasing metal thickness are used to prolong pipelife. Concrete pipe is used where the pH is less than 4 orgreater than 9 or abrasive conditions are present. Bitumi-nous fiber pipe can also be used in acidic areas. Fiber-glass pipe was installed in an area of acid runoff on Inter-state 95 near Coiantico. Criteria were developed for pipeselection based on road classification and project location.A study is in progress to evaluate coating performance.

Additional References:

Highway Specs., Section 240Instructional & Information MemorandumLD-76(R)11.7Mr. R. V. Fielding, State Materials Engineer andMr. C. F. Boles, III, Design Engineer, Richmond,VA

Washington

Galvanized steel, aluminum, concrete, asbestoscement, bituminized fiber and vitrified clay are used fordrainage. Bituminous coating and paving is used west of theCascade Mountains, asbestos bonded asphalt coatings are usedin areas of salt water exposure and concrete or coated steelare used in alkaline conditions. Uncoated galvanized steelcan be used east of the Cascades. Coal tar laminate hasbeen used in several locations for about ten years but itsperformance has been inadequate in abrasive situations.Interior pipe surfaces offer the worst durability problem.Aluminum is used in a pH range of 5-8.5 but not if tidalaction or a heavy bed load are present.

A 196 5 study 1

* found no correlation between servicelife, pH and resistivity. Cathodic protection is not con-sidered practical as interior deterioration is the primaryproblem in Washington. Comparison of results with theCalifornia method indicate that the California method givesconservative information as to service life. Bituminouscoating on aluminum was found to be poor and unpaved bitumi-nous coatings were found to be ineffective against erosion.

Additional References:

Highway Specs., 1977 Ed., Section 9.05

129

Mr. Roger LeClerc, Materials Engineer, Olympia, WA

West Virginia

Galvanized steel, aluminum, stainless steel,concrete, clay, asbestos cement, bituminized fiber andplastic are used for drainage. Asphalt coated, asbestosbonded asphalt and sometimes paving are used for steel.Coal tar epoxy coating or a clay plate liner are used onconcrete in high sulfate environments. Some adhesion prob-lems are encountered with asphalt coatings if they arestored above ground for an extended period. Specificationhas a provision for asphalt sealed hot applied asphalt-sandmix reinforced with wire mesh and applied to primed gal-vanized steel. Alternate provisions for Portland cement orShotcrete are provided.

Additional References:

Highway Specs., 1978 Ed., Section 604, 713Mr. John Judy, Director, Materials Control, Soil &Testing Division, Charleston, WV

Wisconsin

Galvanized steel, aluminum and concrete are usedfor culvert pipe. Coatings are generally not used becausedeterioration from ice occurs. Bituminous coatings do notadhere to pipe. The California method is used to determineservice life. Sulfides are measured to detect anaerobicbacteria. Aluminum is used where the pH is between 5 and 9

and resistivity is more than 1,000 ohm-centimeters. Alumi-num is installed uncoated. Aluminum is successfully used inacidic (pH5) and agricultural runoff areas.

Materials currently in test are aluminum cladsteel, stainless steel and a polymeric (polyethylene) coat-ing. Bacterial corrosion, boggy acidic soils and agricul-tural runoff are major problems. An evaluation in 1972reports that aluminum and stainless steel are outperforminguncoated and asphalt coated galvanized steel . Test condi-tions are generally acidic (pH 4.7-7) with relatively highresistivity soil (15K to 30K ohm-centimeters). A laterreport12 indicates that both aluminum and stainless steelare performing better than galvanized steel after 9 yearsexposure. The bituminous coatings on both aluminum andgalvanized steel were removed from the metal surfaces inimmersion areas.

130

Additional References:

Highway Specs., Section 520, 521, 523Mr. George Zuehlke, Wisconsin DOT, Madison, WI

Wyoming

Concrete, galvanized steel and aluminum are usedfor culvert pipe. Present selection criteria call for usingconcrete pipe or coated metal pipe when the percent ofsodium sulfate in the soil is greater than .2 percent.Aluminum pipe is installed uncoated. Coatings includeasphalt, asbestos bonded asphalt and precoated polymeriesall applied according to AASHTA Specifications. They reportthat asphalt coatings and aluminum are performing well.

New material selection criteria are proposedutilizing analysis of soil and water pH and sulfate content.Under the new criteria, guidelines for using uncoated steel,coated steel, aluminum, the types of coating and Types II orV cement are given in a table.

Additional References:

Highway Specs., 1974 Ed.Special Provision for Corrosion Resistant Pipe toSections 603, 607Pipe Selection Chart Design Criteria - ProposedMr. Robert G. Warburton, State Materials Engineer,Cheyenne, WY

Other Studies

1. In a project for U.S. Steel Corporation, theSouthwest Research Institute evaluated U.S. Steel's"Nexon" coal tar laminate in comparison with asphaltcoated and asbestos bonded asphalt coated galvanizedsteel 9

. Among the findings of these laboratory testsare: both Nexon and bituminous coatings disbonded fromsubstrate, corrosion deterioration did not occur on anyof the coated samples, coating disbondment did notoccur during freeze-thaw cycling and lock seam jointsare superior to riveted joints. Asphalt spalling wasobserved during the freeze-thaw cycle. Abrasion testsindicated higher wear rates for asbestos bonded asphaltand roughly similar wear rates for asphalt and Nexon.However, depending on the bed load, the thinner Nexoncoating abraded away at about the same time as bitu-minous coatings. Extensive delamination of all coat-ings occurred during chemical exposure tests in calcium

131

chloride, sodium chloride, sulfuric acid, sodium sul-fate, ammonium carbonate, fertilizer, seawater anddeionized water.

The bond between Nexon and asphalt paving wasnoted as poor. Chemical attack on asbestos fibers insodium chloride, sulfuric acid and sodium sulfatecaused delamination in some cases. Chemical attack onzinc also occurred leading to delamination.

The investigators noted differences betweenmanufacturers regarding bituminous coating adhesion tothe metal on the as-received samples.

2. A study by U.S. Steel Corporation38 comparedthe performance of pot galvanized steel with continuousgalvanized strip. Zinc coating by the pot dip processwas heavier than by the continuous process, .079 g/sq.cm. (2.59 oz/sq. ft.) versus .065 g/sq. cm. (2.12oz/sq. ft.). Although pot dipped galvanized steelshowed an advantage at one test site in Washington, nosignificant differences in overall performance werereported. The study also indicated that there is noadvantage to galvannealed steel over galvanized steelin either corrosion or erosion performance.

Tests utilizing 66°C (150°F) 2.5 percentsodium hydroxide wash on zinc coated steel indicated animprovement in asphalt adhesion where abrasion issevere.

Type 409 stainless steel performed well inthe U.S. Steel exposure tests when exposed to aggres-sive conditions, including acid coal mine drainage andabrasion.

3 . Bethlehem Steel Corporation is testing Galva-lume (aluminum-zinc) coated steel culverts at 16 loca-tions along the east coast. Test sites range in resis-tivity (soil & water) between 315-168,000 ohm-cm and pHbetween 5.1-8.5. Galvalume is performing slightlybetter than galvanized steel on the average 4 5

. Thisinvestigator inspected the test site in New York asdescribed in the Field Survey section.

4

.

Armco Steel Company has conducted exposuretests of aluminized and zinc coated steel at varioustest sites throughout the country 46

. The results ofthese tests seem to indicate that aluminized type 2

steel is superior to zinc coated steel from both a

132

corrosion and abrasion standpoint. The test sitesincluded fresh water marsh, alkali soils and "normal"soil. Range of pH was relatively narrow, pH 6.3-8,1with one value to pH 9.5, so that information on acidicand extreme alkaline conditions is not known.

5. A study by the U.S. Department of Commerce in1964 involved an inspection of culverts along the BlueRidge Parkway in North Carolina 1* 7

. Their findingsindicate a reduction of 3-11 points in asphalt pene-tration and an increase in the softening point up to10.8°F after 19 years. Most changes occurred on thesidewalls

.

6. A study of the Soil Conservation Service 50

indicates that:

a. Corrosivity is not closely related to soil resis-tivity

b. Asbestos bonded pipes perform better thanother pipes and that the asphalt is retainedbetter when asbestos bonding is used.

c. Asphalt coatings do not stay on the invert ofpipes subject to wet conditions.

d. Paving is of little use in pipes subject to highvelocity flows. Turbulence at pipe joints isinstrumental in removing paving from the pipes.

e. Use of non-corrosive backfill is desirable.

The study encompassed 72 corrugated steel pipes inIowa, Kansas, Missouri and Nebraska. It is SoilConservation Service practice to apply cathodicprotection to pipes where the soil resistivity isless than 4,00 ohm-cm or the pH is less than 5.

Details of the method of cathodic protection isnot given in this report, however, another SoilConservation Service report provides a method 51

using sacrificial magnesium or zinc anodes. Thisreport recommends cathodic protection for asbestosbonded asphalt coated pipe and presumably allcoated pipe.

7. A study at California State PolytechnicUniversity 52 attempted to develop a method of evalu-ating coated and uncoated culvert pipe performanceusing polarization methods. Corrosion rates were

133

calculated for uncoated galvanized and asphalt dippedgalvanized steel culverts, using the E-log I polariza-tion method. No attempt was made to compensate for theeffect of coating defects when calculating corrosioncurrent densities.

Asphalt properties were recently investigatedby the Asphalt Institute 53

. The study indicated thatalthough asphalt properties have not changed over theyears (adhesion to metals was not tested), asphaltproperties do vary substantially depending on grade,method of manufacture and source. No single factorcould be isolated to explain this. The properties ofasphalt are reportedly changed on heating.

In a study conducted for Armco Steel Corpora-tion, Battel le Columbus Laboratories 55 inspected asbes-tos bonded asphalt coated corrugated sewer pipe over 20years old. The investigators found that asbestosbonded pipe was generally satisfactory except in areasof: salt water tidal action, low resistivity soil withdecaying vegetation, acidic waters, highly abrasiveenvironments and organic solvent containing environ-ments .

134

APPENDIX C

References

1. Azar, D. G., Drainage Pipe Study, Research ReportNo. 57, Louisiana Department of Highways

2. Beaton, J. L., Comparative Corrosion Resistance ofAluminum and Galvanized Steel Culverts, CaliforniaDivision of Highways, August, 1964

3. Beaton, J. L. & Stratfull, R. F., Field Test forEstimating Service Life of Corrugated Metal PipeCulverts, HRB Proceedings, Vol. 41, 1962, p. 255

4. Berg, V. E., A Culvert Material Performance Evalu-ation In The State of Washington, Washington StateHighway Commission, April 1, 1965

5. Brown, R. P. & Kessler, R. J., Performance Evalua-tion of Corrugated Metal Culverts in Florida,Florida Department of Transportation, Nov., 1975

6. Culvert Durability Study, Ohio Department ofTransportation, unpublished paper

7. Culvert Inspection of 1957, Idaho TransportationDepartment

8. Culvert Performance at Test Sites in ColoradoResearch Report 68-8, Colorado Division of High-ways, August, 196 8

9. Curtice, D. K. & Funnel 1, J. E., Comparative Studyof Coatings on Corrugated Metal Culvert Pipe,Southwest Research Institute, SWRI Project No.07-2733, Final Report, March 15, 1971

10. Dana, J. S., Corrosion of Highway Metal Structuresand Cathodic Protection, Final Report, ArizonaHighway Department, November, 197 3

11. Deocampo, S. et al, Condition Survey Report,Asbestos Bonded Steel Culverts in Western Oregon,Federal Highway Administration, 1976

12. Dunn, K, H. & Martinelli, T. J., Evaluation ofCorrosion Resistant Type Metal Culvert Pipes,

135

C.T.H. GG, Ashland County, Project ID 0624-50-52(7604) Wisconsin Department of Transportation,July, 1976

13. Durability of Corrugated Metal Culverts ResearchProject 29-1, New York State Department of Trans-portation, Status Report, May, 1977

14. Durability of Drainage Pipe, National CooperativeResearch Program, Synthesis of Highway Practice50, 1978

15. Durability of Metal Pipe Culverts, Research Pro-ject No. 16, Idaho Department of Highways, April,1965

16. Eckert, Eugene B., Performance of Asphalt Coatingson Corrugated Metal Pipe, Pennsylvania Departmentof Transportation, March, 1976

17. Evaluation of Polymeric Coatings for GalvanizedSteel, Michigan Department of State Highways,Office Memorandum, April 13, 197 7

18. Evaluation of Protective Coatings for CorrugatedMetal Pipe, Research Report CA-HY-MR-6655-1-73-05,State of California, Division of Highways, Febru-ary, 1973

19. Experimental Project - Stainless Steel CulvertPipe, Status Report No. 4, Project FH 32-2, U.S.Department of Transportation Federal HighwayAdministration, January 25, 1972

20. Fontana, M. G. & Greene, N. D., Corrosion Engi-neering, McGraw-Hill Book Company, NY, 1967

21. Hayes, C. J., A Study of the Durability of Corru-gated Steel Culverts in Oklahoma, Oklahoma Depart-ment of Highways, 1971

22. Havens, J. H. , Durability of Culvert Pipe, Ken-tucky Department of Highways, August, 1968

23. Hyde, L. W. et al , Detrimental Effects of NaturalSoil and Water Elements on Drainage Pipe Struc-tures in Alabama, Alabama Highway Department,August, 1969

136

24. Jacobs, K. M., Aluminum Culvert Corrosion, Techni-cal Paper 7 6-5, Maine Department of Transporta-tion, October, 197 6

25. Jacobs, K. M., Culvert Life Study, Technical Paper7 4-1, Maine Department of Transportation, January,1974

26. Jacobs, K. M., Durability of Galvanized CorrugatedStructural Steel Plate Culverts, Technical PaperNo. 75-9, Main Department of Transportation,September, 197 5

27. Kalb, M., Statewide Survey of Bituminous CoatedOnly and Bituminous Coated and Paved CorrugatedMetal Pipe, Maryland State Roads Commission,April, 1971

28. Kinchen, R. W. et al , Evaluation of Drainage Pipeby Field Experimentation and Supplemental Labora-tory Experimentation, Interim Report No. 1, Re-search Report No. 101, Louisiana Department ofTransportation, March, 1977

29. Kinchen, R. W. et al, Evaluation of Drainage Pipeby Field Experimentation and Supplemental Labora-tory Experimentation, Interim Report No. 2, Re-search Report No. FHWA-LA-78-115, Louisiana De-partment of Transportation, March, 1978

30. Lindberg, R. I., Method of Estimating Corrosion ofHighway Culverts by Means of Polarization Curves,Highway Research Record No. 204, 1967, pp. 1-10

31. Lowe, T. A. & Koepf, A. H., Corrosion Performanceof Aluminum Culvert, Highway Research Record, No.56, 1964, p. 98

32. Morgan, C. D. , Pipe Culvert Investigation, NorthCarolina Highway Department, 194 4

33. Noyce, R. W. et al , Corrosion of Galvanized MetalCulverts, Transportation Research Board #539,National Research Council, 197 5, p. 38

34. Ray, D. & Croteau, J., Corrosion of CorrugatedMetal Pipe, Final Report, New Jersey Department ofTransportation, February 1, 1974, unpublishedstudy

137

35. Stanley, A. F., Aluminum Pipe Culvert, IdahoDepartment of Highways Condition Survey, 1977

36. Stanley, A. F., Corrosion Resistance of AluminumCMP Culvert Pipe, Idaho Department of Highways,Special Report No. 15-73, June, 1973

37. R. F. Stratfull, A New Test for Estimating SoilCorrosivity Based on Investigation of Metal High-way Culverts, Corrosion, Vol. 17, October, 1961,p. 493T

38. Swan, J. D., Steel Products for Culvert Applica-tions, U.S.- Steel Corporation Applied ResearchLab, Monroeville, PA, 1972

39. Swanson, H. N. & Donnelly, D. E., Performance ofCulvert Materials In Various Colorado Environ-ments, Final Report, Colorado Division of High-ways, September, 197 7

40. Uhlig, H. H. , Corrosion and Corrosion Control,John Wiley & Sons, NY, 1963

41. Wolfe, U. D. & Macnab, S. H. , Corrugated MetalPipe Comparison Study, ODOT Publication 7 6-3,Oregon Department of Transportation, July, 1976

42. Welch, B. H., Pipe Corrosion and Protective Coat-ings, Final Report, Utah State Department ofHighways, November 12, 197 4

43. Worley, H. E., Corrosion of Corrugated Metal Pipe,Final Report, State Highway Commission of Kansas,1971

44. Worley, H. E., Effectiveness of Bituminous Coat-ings on Corrugated Metal Pipe, Kansas State High-way Commission, 1970

45. Allegra, L. , Bethlehem Steel Corporation, HomerResearch Laboratories, Bethlehem, PA. Test Data

46. Armco Steel Co., Presentation to FHWA in Washing-ton on June 22, 1978

138

47. Welborn, J. Y. & Olsen, R. E. , Materials Division,Office of Research, Report on Inspection of Bitu-minous Coated and Uncoated Galvanized Metal Cul-vert Pipe, U.S. Department of Commerce, Bureau ofPublic Roads, Oct., 1964

48. Brown, R., Mississippi State Highway Dept., Missi-ssippi Pipe Evaluation Study - 1964, Nov. 10, 1964

49. Noyce, R. W. & Ritchie, J. M., The Michigan Gal-vanized Metal Culvert Corrosion Study, 58th AnnualTRB Meeting, 1979, Washington, D.C.

50. Hall, Howard W. , Study of Corrosion of CorrugatedSteel Pipe Spillways In Structures Designed by theSoil Conservation Service, Broomall, PA, Sept.,1978

51. Tokach, D. J., Corrosion Protection on CorrugatedMetal Pipe, USDA Soil Conservation Service, SouthDakota, Technical Notes, Engineering Design SD-12(Rev.), April 12, 197 8

52. Pye, E. L., Program to Establish Test Methods forCorrugated Steel Pipe Systems, California StatePolytechnic University, Pomona, for AISI, April30, 1977

53. Puzinauskas, V. P., Properties of Asphalt Cements,The Asphalt Institute, College Park, Maryland,1979

54. Sarikelle, S. & Simon, A. F., Field and LaboratoryEvaluation of Energy Dissipators for Culvert andStorm Drain Outlets - Evaluation of Field Per-formance of Corrugated Metal Culverts, FinalReport, Part III, ODOT Project 14263(0), Univer-sity of Akron, Dec, 1978 - Draft

55. Payer, J. H. et al , A Study of the Durability ofAsbestos - Bonded Metal Pipe in Sewer Service,Battel le Columbus Laboratories, Columbus, Ohio,July 23, 1976

56. Miller, I. and Freund, J. E., Probability andStatistics for Engineers, Prentice-Hall, Inc.,N.J., 1965

57. Kunz, R. G., Cooling Water Calculations, Chem.Engr . , Aug. 1, 1977, p. 61

139

58. Ryznar, J. W., A New Index for Determining Amountof Calcium Carbonate Scale Formed by a Water,Journ. AWWA, Apr., 1944, p. 472

59. Duckworth, W. E., Statistical Techniques in Techno-logical Research, Methuen & Co., London, 1968

60. Vrable, J. B. Unbonded Polyethylene Film forProtection of Underground Structures, MaterialsProt. & Performance, Vol. 11, No. 3, Mar. 72, p.26

61. National Corrugated Steel Pipe Association, Pro-duct Data Bulletin, Guidelines for AsphaltCoating Practice for CSP Products, NCSPA, Glen-view, 111.

62. Bestougeff, M. A., Chemical Composition and Struc-ture of Asphaltenes, Their Importance Among Na-tural Organic Compounds, Bulletin Soc. Chem. Fr.,1967 (12) pp. 4773-82

63. Abraham, H., Asphalts & Allied Substances, 6thEd., Vol. 3, Van Nostrand, N.Y. 196 2

64. Ibid, Vol. 2

6 5. Jongepier, R & Kuilman, Characteristics of theRheology of Bituminous, Proceedings of the Associ-ation of Asphalt Paving Technologists, Vol. 38,1969, P. 98

66. Marvillet, J., Influence of Asphalt Composition onits Rheological Behavior, Asphalt Paving Tech-nology, 1975, Vol. 44, p. 416

67. Brodnyan, J. G., The Use of Rheological and OtherData in Asphalt Engineering Problems, FranklinInstitute Research Laboratories, Phila., PA

68. Patent 2,313,596, 1943, Shell Development Co.Patent 2,295,974, 1943, Shell Development Co.

69. Patent 3,259,512, 1966, Petrolite Corp.

70. Brown, B. W., Improvement of Sands and Gravels foruse with Asphalt, University of Mississippi, 1965

71. Kartashevskii, A. I. et al, Production of AdhesionAdditives for Highway Bitumens, Neftepererab.

,

Neftekhim, 1971 (9), 6-7 (Rus.) Abstract only140

72. Scott, J. A., Adhesion & Disbonding Mechanisms ofAsphalt Used in Highway Construction and Main-tenance, Asphalt Paving Technology 1978 pro-ceedings, Assoc, of Asphalt Paving Tech., Vol. 47,p. 19

73. Hercules Powder Company, Wilmington, Del., ProductBulletin PC-132A, Hercules Rosin Amines

74. Japanese Patent 75,136,324, 1975

75. Japanese Patent 75,136,323, 1975

76. Water Absorption of Bituminous Pipe Coating Ma-terials, Technical Sales Bulleton TS-4-32, TheBarrett Co., Chicago, 111.

77. Crozier, W. F. and Stoker, J. R., AcceleratedAbrasion Tests of Polymeric Protective Coatingsfor Corrugated Metal Pipe, Calif. Dept. ofTransp., Report No. CA-DOT-TL-6738-1-77-01 , Jan.1977

78. Colgate, D., Hydraulic Model Studies of CorrugatedMetal Pipe Underdrain Energy Dissipators, Engi-neering and Research Center, Bureau of Reclama-tion, Jan. 1971, Report REC-ERC-71-10

79. Johns, H. , Erosion Studies of Pipe Lining Ma-terials Fourth Progress Report, Report No. ChE-97,June 1969, Dept. of the Interior, Bureau of Re-clamation, with update

80. Hayes, Harry, Service Life of Coal Tar EnamelProtective Coatings, Jour. AWWA, Vol. 32, No. 10,Oct. 1940, p. 1705

81. Stewart, J. C, Wax Bead Sealing of ConcreteDecks, Oklahoma Dept. of Trans., ReportFHWA-RD-OK-7 9-0 2-1, Dec. 197 8

82. Polymers in Concrete, American Concrete Institute,Detroit, Publ. SP-40, 1973

83. Abrasion Data on Coating Systems, Carboline Co.,St. Louis, MO, Sept. 1976

84. Leidheiser, H., Electrical and ElectrochemicalMeasurements as Predictors of Corrosion at the

141

Metal-Organic Coating Interface, CorrosionControl by Coatings i Science Press, Princeton,1978

8 5. O'Brien, H. C, Some Laboratory Evaluations ofBituminous Coatings, Industrial and EngineeringChemistry, Vol. 58, No. 6, June 1966, p. 45

86. Standard Specifications for Transportation Ma-terials and Methods of Sampling and Testing, PartsI & II, AASHTO, Washington, D.C., July 1978, 12thEd.

87. Gehring, G.'A., Jr., Simulated Testing and Evalu-ation of Protective Coatings to Control Corrosionon Aircraft Launching and Recovery Equipment,NAECENG-7839, Dec, 1973

88. Gehring, G. A., Jr., Protective Coatings forPostal Items, Final Report to United States PostalService Research and Development Laboratory,Rriflfvil 1p. MA. Ann. . 1Q7fibervice Researcn andRockville, MA, Aug., 1976

142

APPENDIX D

Corrugated Steel Pipe Fabricators Interviewed and Visited

Asphalt Refiners Interviewed

1. American United Products Co., Natchez,MS, Bob Swanzy, Plant Manager 601-442-5405

2. Armco Steel Co., Middletown, OH (Visited)Gil Morris, Senior Sales Engr. 513-425-2083

3. Bancroft & Martin, Portland MEHenry Darvey 207-799-8571

4. Brown Pipe Co., Montgomery, AlR. C. Brown 205-262-6444

5. Caldwell Culvert Co., Greenville, MSW. Loveless 601-332-2625

6. Clinton Culvert Co., Clinton, IAJeanette Refbord, President 319-242-6864

7. Culverts & Industrial Supply Co., Inc.,Casper, WY, George M. Royer, VicePresident 307-234-7121

8. Empire Steel Manuf. Co., Billings, MT (Visited)Thomas Brien, President 406-252-0101

9. Florida Steel Co., Tampa, FLJim Goddard 813-621-3511

10. Lane Metal Products, King of Prussia, (Visited)PA, M. Cathers 215-272-4531

11. Lane Metal Products, Bedford, PA (Visited)R. T. Way, Plant Manager 814-623-1191

12. Lane Metals, Pulaski, PAL. Lanthiter 412-652-7747

13. The Levine Co., Inc. Des Moines, IAH. R. Craig, Jr., Vice President 515-262-5613

14. Midwest Culvert Co., Sioux City, IAJ. Troel 712-255-3503

143

15. Northeastern Culvert Corp., Westmin-ster, VT, A. Rogers 802-722-3359

16. Penn Culvert Co., Billerica, MA 617-667-3837

17. Ramco Steel, Inc., Houston, TXJ. Morrison 713-443-3400

18. Republic Steel Co, Canton, OH (Visited)B. J. Andrews, General Manager 216-493-2680

19. Smith Culvert Co., Enid, OKR. Vise 405-233-5555

20. Southern Tank & Culvert Co., Cowpens,SC, D. R. McCaa 803-463-4311

21. Southwest Manuf. Co., Oklahoma City,OK, L. Hall 405-236-3056

22. Syracuse Tank & Culvert Co., Syracuse, (Visited)NY, S. Steel, Plant Manager 315-476-3181

23. Texas Steel Culvert Co., Arlington, TXG. Wetzel 817-265-2255

24. Tri State Culvert Co., Laurenceville,GA, D. Johnson 404-963-9256

Asphalt Refiners Contacted

1. Husky Oil Company, Cody, WY

2. Windsor Chemical Co., Reading PA

3. Trumble Asphalt Co., Summit, IL

4. Vulcan Refining Co., Cordova, AL(no longer in culvert asphalt business)

5. Warrior Asphalt Co., Tuscaloosa, AL

6. Standard Oil Co., Cleveland, OH(no longer in culvert asphalt business)

7. Lion Oil Co., El Dorado, AR

144

APPENDIX E

Suggested Culvert Coating Specification

1. Application

1.1 This specification applies to coatings defined in1.2 applied to corrugated galvanized steel (AASHTO M36),corrugated aluminum (AASHTO M196 & M219) or any other metal-lic coated corrugated drainage pipe.

1.2 Coatings included in this specification includedasphalt applied by immersion, cold applied asphalt mastic orcoal tar, asbestos bonded asphalt, precoated polymer typesand pavings.

2. Coatings

2.1 Asphalt

2.1.1 Asphalt used to coat corrugated drainagepipe must meet the requirements of AASHTO Ml 90 and, inaddition, the following physical properties:

1. Penetration at 0°C (32°F), ASTM D5 or AASHTOT49 - 25 minimum at 200g for 60 seconds

2. Penetration at 25°C (77°F) - 35 to 55 at lOOgfor 5 seconds

3. Flash point, ASTM D92 or AASHTO T48 - 232°C(450°F) minimum

4. Specific Gravity, ASTM D70 or AASHTO T229 -

.98 minimum

5. Softening Point, ASTM D36 or AASHTO T53 -

93°C (200°F) minimum/110°C (230°F) maximum

2.1.2 Tests shall be conducted with each day'scoating run to include:

1. Softening Point

2. Penetration at 25°C (77°F)

The test results shall not vary outside the limits given in2.1.1. Samples shall be removed directly from the vat.

145

2.2 Asphalt mastic and coal tar

2.2.1 These coatings shall meet the require-ments of AASHTO M243 and:

1. Penetration at 25°C (77°F) / ASTM D5 or AASHTOT49 - 5-25 at lOOg for 5 seconds

2. Softening Point, ASTM D36 or AASHTO T53,30-34°C (205-240°F)

3. Percent Solids, AASHTO M243 - 60 to 68

Asphalt mastic shall meet the additional requirements fromFed. Spec. WW-P-4 5B as follows:

Min. Max

Filler

Percent by weightInorganic filler containing long fiberasbestos and fine inert minerals,percentVehicles, percentAsphalts, percentPetroleum, percentNon-volatile bitumens, calculated bydifference, percentPigment, percent by volume of non-volatileFlash point of solvents, °F

.

20.

100604535

30.025.0

80.0

40.0

6555

Coal tarTT-C-494.

shall meet the requirements of Fed. Std FSS

2.3 Asbestos bonded asphalt

Steel sheets for culverts shall be coated with alayer of asbestos fibers, applied in sheet form by pressingthem into a molten zinc bonding medium. The galvanizing orspelter coating shall meet the requirements of the pro-visions of AASHTO M36. It shall be applied at such a rateper square foot, that, when sampled in accordance withspecified methods, the recoverable amount of spelter, afterthe asbestos-bond has been removed, shall be not less than1.5 ounce per square foot of double exposed surface. Asbes-tos-bonded metal pipe culverts shall be fabricated from as-bestos-bonded sheets, the base metal of which shall meet therequirements of AASHTO M218. Both sides of the metal sheets

146

shall be coated with a layer of asbestos fibers. Immediate-ly after the metallic bond has solidified, the asbestosfibers shall be thoroughly impregnated with a bituminoussaturant. The finished sheets shall be of first-classcommercial quality, free from blisters and uncoated spots.After the asbestos-bonded sheets have been fabricated intoculvert sections, an asphalt coating meeting the require-ments of section 2.1 shall be applied according to section3.

2.4 Polymeric Coatings

Polymeric precoated sheet shall meet the require-ments of AASHTO M24 6. Coatings shall be completely free ofdefects such as pinholes, blisters or cracks.

2.5 Pavings

Asphalt paving material shall meet the require-ments of section 2.1.

3. Application

3.1 Asphalt Dip (Plan galvanized and asbestos bonded)

3.1.1 Pipe must be clean of all dirt, greaseand water prior to coating. Minimum pipe temperature priorto application shall be 5°C (41°F). Aluminum pipe should belightly sandblasted or chemically etched prior to coating toprovide a good surface profile.

3.1.2 Pipe shall be immersed in asphalt at204°C ± 3° (400° ± 5°F) according to the following schedule:

Thickness Minimum Immersion Time

.132 cm ( .052 in.

)

2.0 minutes.163 (.064) 2.5.201 (.079) 3.0.277 (.109) 5.0.351 (.138) 6.5.427 (.168) 8.0

The pipe shall then be dipped a second time for a sufficienttime to obtain a minimum asphalt thickness of .127 cm (.050in. )

.

147

3.2 Paving

3.2.1 Asphalt shall be poured into the pipeinvert with dams constructed at the ends in order to coverall corrugations and provide a smooth invert. Asphalt depthabove the crests of the corrugations shall be .318 cm (.125in. ) . Minimum asphalt temperature during the casting opera-tion shall be 93 °C (200°F). The extent of paving shall bein accordance with AASHTO M190 unless otherwise specified.

3.2.2 Paving applied to polymer coatings shalldemonstrate satisfactory adhesion. Asphalt shall not peelor disbond from the polymer when a putty knife is forcedbetween the asphalt and polymer.

3.3 Asphalt Mastic

3.3.1 Pipe must be clean of all dirt, greaseand water prior to coating. Pipe temperature must be above4.4°C (40°F).

3.3.2 Apply the mastic by either troweling,brushing or spraying to a thickness such that the final dryfilm thickness will be no less than .127 cm (.050 in.). Thecoated pipe shall be free of missed spots, pinholes, blis-ters and other defects.

3.3.3 The coating shall be allowed to fullycure prior to allowing water flow through the pipe.

4. Transportation and Handling

4.1 Coated pipe shall be handled at all times withequipment such as wide canvas slings and wide padded skidsdesigned to prevent damage to the pipe coating. Barecables, chains, bars or narrow skids shall not be permittedto come into contact with the coating. All handling andhauling equipment shall be approved by the inspector beforeuse

.

4.2 In truck shipments, the pipe should be supportedon wide cradles of suitably padded timbers. All chains,cables or other equipment used for fastening the load mustbe padded.

4.3 Pipe shall be stored along the trench side sup-ported on wooden timbers placed under the uncoated ends ofthe pipe to hold the pipe off the ground.

148

4.4 Pipe shall be hoisted from the trench side to thetrench by means of a wide canvas or leather sling. Chains,cables, tongs or other equipment, likely to cause damage tothe coating, or dragging or skidding the pipe, will not bepermitted

.

4.5 Any damage shall be repaired before lowering thepipe into the trench. Repairs should be made using asphaltmastic or coating approved by the manufacturer.

4.6 At all times during construction of the culvert,the contractor should use every precaution to prevent damageto protective coating of the pipe. No metal tools or heavyobjects shall be unnecessarily permitted to come in contactwith the finished coating.

4.7 Any damage to the protective coating from anycause during the installation of the culvert and beforefinal acceptance of the purchase shall be repaired as di-rected by the inspector by and at the expense of the con-tractor.

5. Preparation of Ditch and Backfill

5.1 Bottom of ditch shall be free of rocks or otherdebris which would tend to damage the coating during orafter lowering in.

5.2 Ditch shall conform to the pipe so that no forcingof the pipe is required when lowering in.

5.3 In extremely rocky areas, ditch shall be paddedwith sand or soft fill to prevent excessive damage to theprotective coating. Measures required to be at discretionof the inspector

.

5.4 Backfilling shall at all times be conducted in amanner to prevent damage and abrasion to pipe coating.

5.5 Placing of backfill about pipe shall only be donein the presence of the inspector after his final inspectionand acceptance of the pipe coating.

5.6 After placing and aligning pipe in the trench,loose backfill shall be placed about the pipe to a depth ofone foot above the pipe. This backfill shall consist offine soil or sand.

5.7 Settlement of backfill in the trench shall be asdirected elsewhere in the construction specifications.

149

APPENDIX F

Recommended Test Program

We recommend a test program to evaluate alterna-tive coatings, surface preparations and asphalt modifica-tions. The test program should be divided into two phases,that is, laboratory tests to screen a large number of alter-natives and then field tests to evaluate the best methodsdetermined in the laboratory.

Laboratory Program

The test program should include:

1. Evaluation of methods to improve the bond betweenasphalt and galvanized steel and aluminum. Methodsinclude surface treatments and primers.

2. Evaluation of methods to improve the abrasion resis-tance of asphalt. The program should include the useof fillers and blends of asphalt.

3. Evaluate alternative organic coatings. The evaluationshould include those coatings listed in Table VIII ofthis report and Zincrometal.

4

.

Evaluate the effects of asphalt composition on itserosion resistance and adhesion to galvanized steel.

5. Evaluate alternative means of protecting culverts otherthan organic coatings such as flame sprayed coatingsand fiberglass. Reinforced Portland cement, concretesealants and polymer-concrete mixtures might best beevaluated in field tests.

Laboratory tests should be used to screen thevarious coatings and methods. Tests should address thevarious physical and mechanical properties of the coatingsystem under test, and should include:

1. Abrasion Resistance - see Discussion

2. Water Penetration - ASTM G9

3. Outdoor Weathering - ASTM Gil or G2 3

150

4. Impact Resistance - ASTM D2794 or G14

5. Ductility - 180° bend test

6. Freeze Thaw Resistance - AASHTO M246

7. Chemical Resistance - ASTM G20

8. Adhesion - ASTM D2197

There are several methods for evaluating abrasion, that is:air blast (ASTM D658), falling sand method (ASTM D968),pipeline coatings - rotating drum (ASTM G6 ) and Taber abra-sion test. Non-standard tests, such as the rotating drumtests used by Johns and Crozier and a rocker arm testused by Dr. Emery at McMaster University, have been appliedto culverts and drainage pipe. The rotating drum test usedby Crozier accommodates the most samples in the least spaceand would probably be the most practical.

Field Tests

Laboratory tests are not suitable to accuratelypredict field performance, but should, instead, be used toselect coatings for field evaluation. Several sites through-out the country should be selected to represent variousexposure conditions. Exposure conditions should include:abrasive stream flows (both mild and severe), salt water,brackish water, alkaline, acidic, freeze-thaw, ice, decayingvegetation, swamp and neutral non abrasive exposure. StateDOT research personnel should be consulted to determineexact locations. Sites should be selected so that all testculverts are exposed to the same stream flow conditions.Culverts should be installed and be large enough (at least3 6 inch diameter) that the interiors can be inspected with-out excavation. Exterior coating performance can be evalu-ated from coupons on smaller pipe sections buried nearby.

Detailed information should be obtained to charac-terize the soil and water at each test site both initiallyand at each inspection. Information documented shouldinclude: pH, resistivity, chloride, sulfate, nitrate,phosphate, hardness, alkalinity, stream level, stream flowvelocity and bedload type and size. Optimum data would alsoinclude maximum stream flow depths and velocity and theexposure time of those flows.

Inspections should be made initially, after sixmonths, then yearly until the rate and mode of deteriorationcan be established. Measurements to be obtained include:

151

film thickness (including abrasive loss), blistering, crack-ing, disbondment, film continuity and evidence of metalsubstrate corrosion. Detailed photographs are needed ateach inspection and a uniform rating system should be estab-lished and used at each test site.

American Society for Testing and Materials (ASTM)Subcommittee A05-17 is currently developing a field testprogram. ASTM is located at 1916 Race Street, Philadelphia,PA.

152

it U.S. GOVERNMENT PRINTING OFFICE: 1980 O— 628-296/2584 REGION 3-1

FEDERALLY COORDINATED PROGRAM (FCP) OF HIGHWAYRESEARCH AND DEVELOPMENT

The Offices of Research and Development (R&D) of

the Federal Highway Administration (FHWA) are

responsible for a broad program of staff and contract

research and development and a Federal-aid

program, conducted by or through the State highway

transportation agencies, that includes the Highway

Planning and Research (HP&R) program and the

National Cooperative Highway Research Program

(NCHRP) managed by the Transportation Research

Board. The FCP is a carefully selected group of proj-

ects that uses research and development resources to

obtain timely solutions to urgent national highway

engineering problems.*

The diagonal double stripe on the cover of this report

represents a highway and is color-coded to identify

the FCP category that the report falls under. A red

stripe is used for category 1, dark blue for category 2,

light blue for category 3, brown for category 4, gray

for category 5, green for categories 6 and 7, and an

orange stripe identifies category 0.

FCP Category Descriptions

1. Improved Highway Design and Operation

for Safety

Safety R&D addresses problems associated with

the responsibilities of the FHWA under the

Highway Safety Act and includes investigation of

appropriate design standards, roadside hardware,

signing, and physical and scientific data for the

formulation of improved safety regulations.

2. Reduction of Traffic Congestion, andImproved Operational Efficiency

Traffic R&D is concerned with increasing the

operational efficiency of existing highways by

advancing technology, by improving designs for

existing as well as new facilities, and by balancing

the demand-capacity relationship through traffic

management techniques such as bus and carpool

preferential treatment, motorist information, and

rerouting of traffic.

3. Environmental Considerations in HighwayDesign, Location, Construction, and Opera-

tion

Environmental R&D is directed toward identify-

ing and evaluating highway elements that affect

• The complete seven-volume official Btatement of the FCP is available from

the National Technical Information Service, Springfield, Va. 22161. Single

copies of the introductory volume are available without charge from Program

Analysis (HRD-3), Offices of Research and Development, Federal Highway

Administration, Washington, D.C. 20590.

the quality of the human environment. The goals

are reduction of adverse highway and traffic

impacts, and protection and enhancement of the

environment.

4. Improved Materials Utilization andDurability

Materials R&D is concerned with expanding the

knowledge and technology of materials properties,

using available natural materials, improving struc-

tural foundation materials, recycling highway

materials, converting industrial wastes into useful

highway products, developing extender or

substitute materials for those in short supply, and

developing more rapid and reliable testing

procedures. The goals are lower highway con-

struction costs and extended maintenance-free

operation.

5. Improved Design to Reduce Costs, Extend

Life Expectancy, and Insure Structural

Safety

Structural R&D is concerned with furthering the

latest technological advances in structural and

hydraulic designs, fabrication processes, and

construction techniques to provide safe, efficient

highways at reasonable costs.

6. Improved Technology for HighwayConstruction

This category is concerned with the research,

development, and implementation of highway

construction technology to increase productivity,

reduce energy consumption, conserve dwindling

resources, and reduce costs while improving the

quality and methods of construction.

7. Improved Technology for HighwayMaintenance

This category addresses problems in preserving

the Nation's highways and includes activities in

physical maintenance, traffic services, manage-

ment, and equipment. The goal is to maximize

operational efficiency and safety to the traveling-

public while conserving resources.

0. Other New Studies

This category, not included in the seven-volume

official statement of the FCP, is concerned with

HP&R and NCHRP studies not specifically related

to FCP projects. These studies involve R&Dsupport of other FHWA program office research.

DOT LIBRARY

0005 L.&3M