All Category




The qualrty of a nation’s Infrastructure IS essential to ItS continued competitiveness and economlc growth. A good economy as weil as healthy and liveable cities, towns, and villages depend on rehable and adequate transportation networks (i.e., airports, roadways, waterways), clean and efficient water distribution systems, and safe disposai of domestic and hazardous wastes. Their detenoration or failure to perform adequately will create an Intolerable hardship to everyday life and endanger the productivity of a natlon’s economy.

Many concrete structures are now approachlng or have reached or exceeded their design service life and are being used ln excess of thelr design capacity. In reviewing government studies and surveys conducted in Canada and the United States, there is convincing evidence that the quality of the Infrastructure in the western world is barely adequate to meet current requirements, however, it IS Inadequate to meet the projected demands primari~y due to increasing budgetary constralnts. For instance, a recent report by the Road Transportation Association of Canada (RTAC) 1 revealed that 38 percent of the national highway system was found to be substandard and 22 percent (790) of the bridges needed major strengthening and rehabilltation, requiring a total rehabllitation Investment of $13-14 billion (1993 dollars) in the next five years.2 A similar study conducted ln the United States on the state of the nation’s infrastructure revealed that, based on an academic scale, their performance (in 1987) would receive a ·C • barely adequate to support the required demands·. 3 The cost to upgrade public facilities were reported to be in the range of hundreds of billions of dollars. Similar results have been reported in Europe, the U.K., and other countries.

Part of the problem has been lack of public awareness, mainly due to the fact that public works facllities are often taken for granted, since many of them are out of sight. A major contributing factor to the infrastructure problem has been the decline in government assistance programs throughout Canada and other cou nt ries, mainly due to large existing deficits. The fear of increaslng their deficit further, governments are unwilling to accept responsibility for rebuilding and upgrading urban infrastructure facilities.4

However, studies by the Federation of Canadlan Municipalitles (FCM)!I and others have demonstrated th~t infrastructure renewal can dlrectly Improve the economlc vlablhty of commerce and industry, thereby generatlng Increased national revenues whlch can reduce deflclts Just as eHectivelyas reducing expenditures. For every bilhon dollars Invested ln Canada’s construction Industry, 20,000 jobs can be created.2 Simllarly ln the United States, for every public dollar spent annually to bUild and malntaln the Infrastructure, the prlvate sector spends $15 (US) to move people and goods. For every aViation dollar, prlvate flrms and Indlvlduals spend nlne dollars J Taking into account unemployment insurance and welfare savings, and Increased government revenues, these benefits are compounded. Although public awareness to the CrlSIS and government fundlng has increased, the problem still remalns a large one. The n ,atenal ln thls chapter has been adapted from dlHerent avallable references, especially 7 and 13.



An obvious solution ta the prob”’m may be to make avallable the funds necessary ‘or rehabllitatlon. However, a dependable, high quahty Infrastructure IS not attained by money alone: bUilding and maintalning public works requires the sklll, and commltment of time and energy of people throughout the pubhc and prlvate sector. For example, Curtis” reports that rehablhtatlon costs can be reduced ln Canada by:

  • ReSistance ta public pressure for elaborate and expensive faclhty treatments.
  • Elimination of unnecessary delays in correcting problems.
  • Improving technical eHiciency. Generally these are associated with standardized treatment and automated mass production techniques.
  • Greater rehance on payment by users.


A report on the state of public works ln the United States3 recommends that a strategy to upgrade the infrastructure must include other mechanlsms ln addition to increased Investment. These include:

  • Classification of the respective raies of allieveis of government (federal, state, and local) in the construction and management of infrastructure ta Increase accountability.
  • More flexible administration of federal and state mandates to allow cost-eHective methods of compliance.
  • Steps to upgrade the quality and quantity of basic pubhc works management information ln arder ta measure and Improve the performance and efflclency of existing faclllties.
  • Financing of a larger share of the cast of public works by those who benefit from service.
  • Additional support for research and development to improve technologies and far training of public works professionals.


All of the above leads to SIX main categories of needs to Improve response to the mfrastructure crisis:

  • Apply the best management technrques to infrastructure management;
  • Elimlnate corruption wherever It exists ln the infrastructure field;
  • Apply the best skills of the pnvate sector;
  • Reduce bUilt-in industry structure problems such as fragmentation and harmful duplication and competition;
  • Improve the process of education for the Infrast’ucture managers; and
  • Conslder trade-offs, and where a level of infrastructure service cannot be maintained, It should be reduced.


Clearly, the requlred task dictated by these recommendations is indeed a challenglng one. In fact, It wIll require a shared vIsIon and effort by bath the public and the pnvate sectors. Wlth cooperation, the ,nfrastructure problem can be resolved over a reasonable period of time without unreasonable budget Increases. The sooner the problem is addressed, the easier it Will be ta resolve. It should be kept ln mlnd that major public works prolects have a long lead time, especlally if they are located ln a metropolitan area or Involve controversial issues. Continued delay will only allow deterioration to advance to the point where costs will escalate.



Many Industnalized nations of the warld have come ta realize the importance of Investing in their public infrastructure. Due ta increased competition for the tax dollar, finding Innovative ways for financing current and future public works needs has become equally important. While several dlfferent options have been proposed, they ail depend on general tax revenues and user fees. A brief 2 submltted ta the federal gavernment of Canada ln February 1993 provided the following options as possible financing mechanlsms:

  • Build-operate-transfer systems
  • Lease-to-purchase options
  • Dedicated taxes
  • User fees and other pnvatlzatlOn schemes, and
  • The issuance of tax exempt bonds


Consequently, a national poli commlssloned by The Road Information Program (TRIP) of Canada ln April 1993 determ!ned that 58 percent of the respondents questlOned supported the Idea of a road user fee ta rehabliltate the National Hlghway System.

Simllar recommendatlOns were developed by the National Councii on Pubhc Works Improvement in the United States.3 These included:

  • Users and other beneficianes should pay a greater share of the cast of infrastructure service;
  • The federal government should be an equal partner ln flnanclng public works;
  • States should develop comprehensive finance strategies; and
  • Local governments should glve f!nanclal pnonty ta fundlng the maintenance of eXlsti,.,~ facilities.


ln planning for pubhc works investments, both the public and private sector should be glven clear stated performance objectives, conslder alternative ways of achievlng them, and have easy access ta Information about costs of operation and malntenance.3 Implementlng these steps will not be an easy task, but the Increase ln costs due ta deldy Will hlnder a nation’s ablhty ta cape wlth the Infrastructure crlsis.



Rehabihtating concrete structures IS a speclahzed field requiring skill and expertise. Due ta the complexity of the restoratlOn process, the design englneer must be able ta perform several raies: that of an investlgator, designer, matenals performance specialist, and construction Inspector.8 Ta simplify the rehablhtatlon process, englneers have anempted ta devise a systematlc (or decision-making) approach ta arnve at an appropnate solution. One possible approach IS that developed by Tracy and Fhng,6 whlch emphaslzes the Implementation of three distinct phases ta the rehablhtatlon process (Figure 1.1): the concrete condition survey whlch identifies the cause, rate, and extent of detenoration; the structural aspects investigation whlch places structural, functional, and operational constralnts on the rehabilitation process, and the repalr program which glves vanous solutions to the identifled problems.

A slmllar but more ngorous approach was developed by Chung7 for repairing concrete structures damaged by steel relntorcement corrOSion, although the approach can generally be adapted and apphed to structures in general. The inter-related factors involved in the repair strategy are shown ln Figure 1.2. The baSIC steps involved ln the procedure are summarized below.



The tlrst step in the restoration process IS to conduct a condition survey of the structure to coUect suffie:””nt data to determine the cause, extent and rate of deterioration. The survey should not be limlted to the damaged portions only, but should include the structure as a whole. Test results from sound areas of the structure are essential in providing necessary baseline data for comparison. It is good practice to obtain a sufficient number of test results ta perform the reqUired statistical analyses.


From the condition survey an assessment of the visible and hidden damage ln the Indlvidual structural members 15 made, wlth each member or portion of the structure belng classlflBd ln accordance wlth the intenslty of damage (I.e., depth of carbonation, chlorrde content, degree of spalling and cracking, and degree of rustlng)/ The classification shown ln Table 1.1 may be used as a gUide. Although classification of damage intenslty IS mostly arbltrary, It serves as a useful gUide for selecting the approprrate method of repalr.



The decision to execute repalrs and the complexlty of the repalr system ta be used depends on the extent of deterioratlon and the length of time the structure IS still required ta functlon after the repalrs are made. This tlme span IS referred ta as the pro/ected service IIfe (PSL). According ta Chung7 , the PSL of a structure can be categorrzed as follows

  • Short projected service life: < 5 years
  • Medium proJected service life: 5-15 years
  • Long proJected service hfe: > 15 years


If the structure has completed ItS full deSign life when deterioration reaches the unacceptable level, then repairs will not be necessary, unless the structure is required ta continue its intended functlon. Often, the structure will be ln active service wh en detenoration becomes unacceptable. Sometlmes detenoratlon has progressed to the point where the structure or member is elther technlcally or economlcally beyond repalr. In such a case, the only alternative is ta rebuild the structure 111 ItS entlrety, or ln part.

If the structure IS ta have a short PSL, extensive repalrs may not be economically Justifiable. In most cases, no repairs are needed, unless safety is a concern. With the PSL falling in the medium or long range, the degree of detenoration whlch requlres immediate repair may vary accordlng to opinion. Some owners or managers tend to postpone any action until the operation or safety of the structure IS Impalred, whlle others prefer to perform minor repairs or maintenance more often. The declding factors Include accesslbility of the repair areas, provIsion of facllities for repalr/malntenance and avallability of funds.

If the intenslty of damage is hght and the concrete and reinforcing steel are in good condition, Immediate repalrs are not necessary. However, Implementing preventive maintenance procedures, such as applying protective coatlngs, frequently helps to prevent hght damage from becoming critical, thereby extending the service IIfe of the structure.

Any repalr work or protective measure chosen for the structure must perform adequately for some period of time before another repalr Job becomes necessary. Chung7 defines thls penod as the trouble-free penod (TFP) of the structure, and IS charactenzed as follows:

  • Short trouble-free penod: < 5 years
  • Medium trouble-free penod: 5-15 years
  • Long trouble-free penod: > 15 years


The TFP of a repaired structure will vary wlth the complexlty and thoroughness of the repalr work performed. In general, a TFP as long as the PSL is preferred, however, availablhty of funds IS usually limited, thereby affordlng a less durable repair. The relatlonshlp between TFP and PSL IS represented schematically in Figure 1.3. A flow chart whlch may be used for selectlng an approprlate TFP is shawn ln Figure 1 .4.




Once the probable cause or causes of deterioratlon have been determined, an appropriate repalr option can be selected. In selectlng a repair option, the owner of a deteriorated structure must first make a decision, based on economlcal reasons, whether:8

  • To do nothing and allow the structure to deteriorate and eventually demohsh it;
  • To Implement short-term repalrs knowlng trlat further repair work will be needed before the end of the useful hfe of the structure; and
  • To undertake a major rehabllrtatlon or replacement project to provlde a structure whlch will provlde a trouble-free perrod equal to Its proJected service IIfe wlth only routine maintenance.





Depending on the rehabilitatlon scheme chosen the objective of the repair is ta restore the structure to a point where It can provlde the required trouble-free perrod. The possible options of repair include cosmetic, repeated, extended, preventive and replacement. 7 The decision tamake extended repairs or replacement IS usually dlfflcult to make and de pends on several interdependent elements including, the extent of damage, the temporary shonng requlred dunng the repair, and the extent or interference wlth the operation of the structure. The cholee ln the end will be an economical one.



There are several repalr systems available for performmg repalr work, mcludlng crack Injection, patchlng, shotcrete, protec!;ve coatlngs, cathodic protection, chlonde removal and replacement. Sorne systems, such as cathodic protection are deslgned ta make the concrete InconduClve 10 reinforcement corrosion, while others will prevent the Ingress of solutions that promote corrosion (e.g., protectlve coatlngs). However, systems that leplace the damaged concrete (I.e., patchmg, shotcrete) or flll the cracks wlthln the concrete (I.e., crack InJection) are more commonly used

Although it may seem deslrable to use the same repalr method throughout the entlre structure, a combinatlon of one or more of the above repalr melt.ods is often requlred and is usually more economlcal. In any case, If the cause of concrete detenoratlon IS not removed, the repair method chosen will only conceal, and otten, a~gravate the problem.



Chooslng an appropriale repalr option and repair system IS not a simple matter, although several options and systems may be equally effective in restonng the structure. In sorne cases, dlfferent options or systems may either be reqUired or may be more economlcal for dlfferent parts of the structure. In any case, each possible solution should be costed for the projected service IIfe of the structure. The solution whlch is chosen IS typlcally the one wlth the lowest cost.

The estimated cost of any repalr option or system should include the expenses mcurred for access, equipment, labor and matenals, and future maintenance. The total cast is the sum of the capital cost for repalr and the proJected cost for operatlon/mamtenance of the repalr system, including adJustment tor Interest rates and IOtlatlon. Any Indirect cast assoclated wlth loss of revenues due to interruption of dally operations dunng the repalr and maintenance penods sr.ould also be Included. Otten, It IS necessary to prepare a detalled estimate for companng the relatIVe beneflts of repeated repalrs and extended repalr. In many cases, due to budgetary constraints, a less desirable solution is chosen Instead of the optimal one.7



ln the past decade or so, there has been a worldwlde effort Into developing new construction matenals and techniques for rehablhtatlng deteriorated concrete structures. The main thrust of development, however, has been ln the field of synthetlc construction materials. Technologies for applylng these matenals are for the most part Improvements on eXlsting construction techniques, although sorne new methods have been put forth. Nevertheless, upgradlng the eXlsting Infrastructure and bUilding more durable structures in the future does not rest on developlng materials alone. The avallable research and development capablhties must be applied to the innovative questions assoclated with alternative Infrastructure technologies.

Numerous studies conducted ln the mld 1980s assessed research needs in various categories concermng Infrastructure management. The studles focussed on both management and technologlcallssues. For Instance, a common conclusion was that the process of public works management and the effectiveness of public works managers needed to be vastly improved. This would reqUire an analysis of Institutlonal problems of planning and management of facilities, including the process used for decision maklng.

The identification of the effects of new technology on urban infrastructure are important factors whlch need to be considered to Improve the performance and reduce the cast of existing systems and faclhtles. ThiS would Include adJusting Infrastructure management for future patterns of living. Equally Important is to develop standards and criteria for the design and performance of urban public facllltles, agalnst which national and local needs for investment can be measured. Also, constraints caused by the existing codes and standards should be minimized.

On the technological front, there is a need for technologies far beyond those in use today for quality assurance in construction. The construction industry IS becoming worldwide, demanding access to the world’s best products and services. A prerequiste to acceptance of products or services ln International trade IS a demonstration that they conform to international safety and performance standards. ThiS may be achieved through hlghly developed ·infratechno-Iogies •. 9 These are tools used by engineers through the entirety of the lite cycle of the structure or facility for developing and applying information, standards, codes and quality assurance. Infratechnologies requlred ta demonstrate such conformance will consist of:

  • Performance standards and codes
  • Procedures to assess the conformity of Innovatlve products and services
  • Computer-based knowledge systems
  • Automatic information exchange systems
  • Open systems for products and services


Predictlng the remalnlng selVlce hfe of eXlstlng deteriorated concrete structures has galned interest in recent years. Present methods and tools to predict future performance and selVlce are limited. There IS a great need for reflnement of avallable hfe prediction methods and selVlce hfe design cnteria must be further deflned. In a recent revlew of some case histones, Hookhamlll cited the following important research needs:

  • Further research IS needed to charactenze degradatlon processes ln terms of thelT rate of attack, threshold level, and treatment ln hfe prediction models.
  • Developing appropnate accelerated aglng techniques and tests 15 needed to Improve mathematlcal degradatlon models Instead of emplncal data
  • Defining periodlc nondestructive testlng and inspection methads and approprlate acceptance criteria are necessary ta provlde the data reqUired ta predlct the remalmng service life.
  • The comblned effects of several detenoratlon processes acting slmultaneou51y needs to be investigated and ,”cluded in hfe prediction modelling.


It shauld be realized that there are thousands of research needs in ail areas of Infrastructure management, and the only source for satisfylng thls demand IS dedicated research efforts by the engineering community. However, the role of the government and the public ln provldlng the necessary support for accomphshlng such an enormous task cannot be underemphaslled.



Marine structures have always been crltlcally Important to the operation and econamlc growth of ail nations. They have been ln existence for centuries and have been constructed of stone, limber, concrete, and steel to wlthstand the harshness of the manne enwonment. They are not only designed ta carry thelT selVlce laads, but loads fram shlp and wave Impact as weil. Although the average service IIfe of a marine structure IS approxlmately 25 years, marine structures 50, 75 and over 100 years old are still ln service provldlng the necessary means for movement of goods, vital to a nation’s growth. 11 Therefore, il becames Imperative to keep these faclhties aperatlng at servlceable level and to malntaln their capacities .

Marine structures include a variety of structures and are normally grouped either according to thelr functlon or thelr generallayout and overall geometrical conflguratlon.12 The term is normally apphed to berthmg and moonng facliities, container terminais and oil letMs, breakwaters, embankments, slope protection stllJctures, tldal bamers, dams, navigation locks, and outlet tunnels. It is not the Intent of thls thesls ta provlde an exhaustive revlew of ail the types of manne structures in use, but to summanze the main features of those most commonly encountered ln the manne environ ment. These Include: 13

  • Berthlng facilities for moonng and providlng support to ships and craft.
  • Drydocks used for construction of shlps and ta expose the underside of ships for inspection, maintenance, repalr, or modification.
  • Coastal protection structures deslgned ta protect shorelines or harbors.



The basic berthing faclhties that provide berthlng support for ships and craft are piers (normal to the shore) and wharves (parallel to the shore). Piers and wharves provide a transfer point for cargo and/or passengers between water carriers and land transport. The three major structural types for piers and wharves are open, sOhd, and floating.

Open-type piers and wharves are pile supported platform structures which permit water to flow beneath. Sol id-type piers conslsts of a retainlng structure such as an anchored sheet pile wall or quaywall, behind which earth fllI is placed to create a working surface. A floating-type pier is a pontoon structure that is anchored ta the shore by trestles or ramps. The top of the pontoon can be utllized as the working deck. These structures are not affected by tidal fluctuations but obstruct water flow to sorne extent. Open and solid type structures can be combined to provide a more advantageous layout. These structures are discussed ln more detail in the following sections.



Piers are structures which extend outward from the shore into the waterbody. Piers may be used for berthing on one or both sides of their length. The length of a pier is usually equal to or greater than the length of the longest ship to be accommodated. The wldth of a pler is usually established by functional, geotechnical, and structural conslderations.’

Open piers are plle-supported platform structures whlch allow water to flow underneath. Open piers are usually single-deck structures, although recently a double-deck pier was constructed by the U.S. Navy.’3 A schematlc of a single and double-deck open pler 15 shown ln Figures 1.5 and 1.6, respectlvely.




Closed piers, or solid fill piers, are constructed so that water is prevented from flowing underneath. The solid fill pier is surrounded along Its penmeter by a bulkhead or wall which retains the fill. A schematic of a typical solld flll pier 15 shown ln Figure 1.7. A special type of solid fill pier IS a mole piero Mole piers are earth-filled structures that extend outward from the shore. The sides and offshore end of the pier are retalned and protected by masonry or concrete sheet pile walls. ‘

Floating piers are connected ta the shore with access ramps. Ta prevent lateral movement and allow vertical movement of the pier with the tidal fluctuations, guide piles ln the center of the pier, or a chain anchorage system is utllized .


The floating pler may also be a single or double-deck structure. A ‘Ioating pier concept whlch was developed by the U.S. Navy is shown in Figure 1.8.13 A more detailed discussion of the design and configuration of piers is provlded in the U.S. Navy Design Manual NAVFAC DM 25.01.15




Wharves are structures which are constructed approximately parallel ta the shore. A marginal wharf is attached to the shore along Its fuillength and a retaining structure IS used ta retaln earth or stone placed behind the wharf. The retarning structure IS usually referred ta as a bulkhead or quaywall. With this structure, the ships can benh along the outshore face only. The typical wharf types are simllar to the basic pier types and include open and closed (or soUd fiiO layouts Wharf length and wldth is based on the same considerations as those for piers. Examples of open and solid fill wharves are shawn in Figures 1.9 and 1.10, respectlvely.


Wh en the depth of water adJacent to the shore IS tao shallow for deep draft ships, the wharf, which consists of a platform on piles, IS located some distance away from the shore ln deeper water and IS attached ta the shore by pile-supported trestles. The trestles are usually oriented perpendicular ta the wharf (Figure 1.11). If the trestle is located at the center of the wharf, the structure IS referred ta as a T-type wharf; If the trestle 15 located at an end, It IS called an L-type wharf; and if trestles are located at both ends, It IS referred ta as a U-type wharf.



Drydocking faclhties are used for construction of ships and ta expose the underslde of shlps for Inspection, maintenance, repalr, or modification. There are va nous types of drydocks that eXlst, Jncluding graving drydocks, floatlng drydocks, marine railways, and vertical syncrohfts. 13 These are briefly descnbed below. A more detalled diSCUSSion of drydocklng facllities can be found ln References 16 through 18.



Gravlng drydocks are flxed baSins adjacent to the shore and are constructed of stone, masonry, concrete, or sheet plie cells. They can be closed off trom the outshore si de by a movable watertlght barner (entrance caisson or flap gate). After the barrier is closed, the water is pumped out of the baSin to allow the shlp to settle on blocklng set on the dock tloor. A schematic of a typical gravlng dock IS shown ln Figure 1.12.



Floating drydocks are U-shaped structures that are used to raise ships or vessels out of the waterway. The structure IS flooded, permltting the vessel to enter, and then it IS pumped dry.



Manne railways provide an aceess for a vessel to enter the waterway from land and vice versa Manne rallways consist of él ramp whlch extends Into the water, a mobile shlp cradle on wheels or rollers, groundway shlp cradle tracks, hoistmg machmery, and chams or cables for hauhng the ship eradle. A typlcal marine rallway IS shown ln Figure 1.13.



Vertical syncrohfts conslst of platforms which are lowered Into the water to recelve shlps. The shlp is then lifted out of the water on the platform by electncally powered hOlstlng equlpment. Figure 1.14 shows a typlcal vertical syncrohft drydocklng system.



The primary funetlOn of these structures !S to proteet harbors from the erosive effects of wave action. Structures whlch commonly fall ln thls category Include seawalls, bulkheads, groins, jettles, and breakwaters. A bnef descnptlon of each follows; more detalled informatIOn on the design and configuratIOn of these structures IS avallable ln References 19, 20, and 21.



Seawalls are massive coastal structures bUiIt along the shoreline to protect coastal areas from seour caused by severe wave action and flooding during storms.19 Seawalls are constructed of variety of materials including reinforced concrete, rubble mounds, or granite masonry. Figure 1.15 shows three basic types of seawall configurations.

A curved-face seawall (Figure 1.15a) uses a sheet pile eut-off wall to prevent loss of foundatlon materlal by wave scour and leachmg from overtopplng water or storm drainage beneath the wall . The toe of the curved-face seawall conslsts of large stones la prevent or reduce scour

The stepped-face seawall (Figure 1 15b) IS designed for stablhty agalnst moderate waves. This seawall type uses reinforced concrete sheet plies wlth tongue-and-groove JOints The space that is created between the piles may be filled wlth grout ta form a sandtlght cut-off wall. Alternatlvely. a geotextile fabric can be placed behind the sheetlng ta provlde a sandtlght barner, wh Ile permltting the water ta seep through the cloth ta prevent the buildup of hydrostatlc pressure.

Rubble-mound seawalls can wlthstand severe wave action (Figure 1.15c). Although scour may occur, the quarrystone comprising the seawall can readjust and senle wlthout causlng structural failure.



Bulkheads are flexible sOli retalnlng structures whlch analn their stabllity from the structural members and the shear strength of the sOII.22 The pnmary function of bulkheads IS to retaln flll, and although not usually exposed to severe wave action, they are still reqUired to resist eroslon Bulkheads are generally either anchored vertical sheet pile walls or gravlty structures. ID The two baSIC structural types are shown ln Figures 1 .16 and 1.17. Cellular steel sheet pile bulkheads are sometlmes constructed where rock IS close ta the surface and sufflcient penetration cannot be provided for the anchored bulkhead type.



Groins are structures deslgned ta reduce the effects of eroslon to the shoreline by altenng offshore current and wave panerns. Groins are normally constructed perpendicular ta the shoreline and can be made permeable or Impermeable. ID Matenals used for constructlng groins include stone, concrete, timber, and steel. 13 Figure 1.18 shows an example of a concrete sheet pile groin .



Jettles are structures which extend from the shore Into deeper water to prevent the formation of sandbars and control water currents. These structures are ordlnarilv located at the entrance to a harbor or a river estuary. Jetties are usually constructed of rubble mounds to a height above high Me.13 A typical rubble-mound Jetty configuration is shown ln Figure 1.19.



Breakwaters are large r’,bble-mound structures constructed outside of a harbor or coastline to protect inner shorelines from severe wave action. These barri ers help to create a safe envlronment for moonng, operating, loadlng, or unloadlng of ships within the harbor. There are three general types of breakwaters, and may be elther connected to or detached from the shore. 13 Figure 1.20 shows a cross-section of a typical rubble-mound breakwater .



A vast number of concrete structures are ln direct contact wlth seawater or are exposed la seawater spray carned by wlnds. 1 The manne environ ment IS one of the harshest enwonmenls known to man. Concrete exposed ta the manne environ ment may detenorate due ta the comblned effects of va nous physlcal and chemlcal phenomena. Sorne of the most comman forms of deterioratlon include, chlonde-induced corrosion of the relnforclng steel. freeze-thaw attack. alkah-aggregate reaetion. sulfate attack, and physical eraslon due ta wave action and floatmg debns.

According to a study of case histones of concrete fallures ln seawater (Appendlx A), Investlgators have determined that the degree of df:!tenoration (physlcal or chemical) IS dependent on where the structural member 15 located wlth respect ta Mal activlty. Therefore, Mehta2 grouped manne concrete Into three exposure zones: submerged, splash, and atmosphenc (Figure 2.1). The atmosphenc zone, which IS always exposed to the atmosphere, IS susceptible ta cracking by several causes Includlng, fr~eze-thaw action, wetting and drylng, thermal cycles. and corrosion of ambedded steel remforcement. Also, concrete ln the tidal or splash zone, whlch IS locatad between hlgh and low Me, may experience cracking by impact of floatlOg debns and by deletenous chemical reactlons between the seawater and cement paste constituants. The submerged zone, whlch is cantinuously covered wlth seawater, IS susceptible ta chemlcal attack only.

Clearly, the most severe deterioratlon will occur in the tldal zone because the structure IS exposed to nearly ail the physlcal and chemlcal attacks. Concrete deterioration caused by any one of these phenomena Will increase the permeablhty of concrete whlch Will cause further detenoratlon by other types of attack. This chapter outhnes the vanaus physlcal and chemlcal phenomena whlch cause deterioration in the manne environment, and measures ta control such deterioratlon are al&o presented. A bnef revlew of cancrete detenoratton by bactenologlcal attack and hard impact are also provided. The materialm this chapter has been adapted from differant available references, especlally ~, 4, 6 and 45, and IS presented here for completeness .



According ta Mehta and Gerwick,3 there are two classifications of physical causes of concrete deterioratlon: surface wear and cracking (Figure 2.2). The various phenomena in each classification are dlscussed in the following sections.



The maJor causes of concrete detenoration are attnbuted ta cracking and subsequent corrosion of embedded relnforcing steel (Section 2.3).3 The causes and types of cracking are many and can occur ln bath plastic and hardened concrete. “n excellent review of the causes, mechanlsms, and control of ail types of cracking in concr’Jte is provided by the ACI Committee 224.4 ,5 The baSIC mechanlsms by which cracking stralns may be generated in concrete are:8

  • Internai movements caused by drying shrinkage, plastiC settlement or shrinkage, and expansion or contraction .
  • Expansion of embedded metals, such as relnforcement corrosion.
  • Externalloading conditions, such as deformatlons caused by dlfferentlal foundation settle· ment.


A summary of the vanous possible causes of cracking is provided in Figure 2.3. A general guide of the age at which these cracks occur in concrete is glVen in Figure 2.4. A summary of common defects occumng dunng construction IS provided ln Appendix B. The various types of cracks which occur most often ln practlce are summanzed below, and are adapted trom a review of References 4, 5, 6, and 7.



Plastic shrinkage cracking usually occurs on the surface of freshly poured concrete when It IS subjected ta rapid loss of mOisture. Cracking usually occurs wlthln the first two ta four hours alter placement If the loss of mOisture exceeds the supply by bleed water

The subsequent shnnkage at the surface and the restralnt provlded by concrete below the drylng surface layer Induce tensile stralns which cause cracking. These cracks are usually short and run in several directions (Figure 2.5). The wldth of a typlcal crack IS about 2 ta 3 mm (0.08 ta 0 12 in.) at the surface and decreases as the depth from the surface lncreases. The length of the cracks can vary trom a tew centlmeters ta over 1 m (3 ft.) in length and are spaced Irom a lew centimeters to as much as 3 m (10ft.) apart. In sorne cases, plastic shnnkage cracks can extend the full depth of the member.

Measures to prevent plastic shnnkage cracking Include the use of ·’og nozzles’ to humidlfy the concrete surface and covering the concrete surface with plastic sheetlng. The concrete can also be protected by erecting wlnd breakers ta diminlsh the wind velocity, and sunscreens to lower the surface temperature. ACI Commlttees 224R,5 302.1 RB and 30SR9 provlde other recommendations to prevent rapld mOlsture loss.



After the concrete is placed and compacted, It will continue to consolidate due ta the movement of mlxlng water toward the surface. If settlement of concrete IS restralned by reinforcing steel or formwork, cracking will occur near the element where It is belng restrained. In the case of relnforcing steel, longitudinal settlement cracks will form along the top of the rebar (Figure 2.6). Increaslng the rebar slze and slump, and decreaslng the concrete caver will increase settlement crackmg. If the relnforclng bars are closely spaced, hOrizontal settlement cracking may occur (Figure 2.7). These cracks over the top layer of the relnforcement will cause the concrete cover to spall. Steps to prevent settlement cracking include proper form design (ACI 347R) 10, adequate compactlon, the use of the lowest possible slump, and increasing the concrete cover.



Restrained drylng shrrnkage is caused by loss of mOlsture from the cement paste constituents.

The degree of drylng shrinkage orimanly depends on the amount and type of aggregate and the water-cement ratio of the mlx. As the amount of aggregate IS Increased the amount of shnnkage will decrease. Surface crazlng on walls and slabs IS an example of drylng shnnkage. This orten occurs wh en the concrete near the surface contalns a hlgher water content than the Intenar concrete, resultlng in a senes of shallow, closely spaced, fine cracks.

Drying shrinkage can be mlnimized by Incorporating the maximum practlcal amount of aggregate and the lowest possible water content in the mix, or uSlng shnnkage-compensating cement. However, thls requlres careful control and proper consolidation. The use of properly spaced contraction jOints IS an effective means of controlling shnnkage cracking. ACI Commlnee 224R5 provldes more dptails and other construction practices whlch help to control drylng shnnkage ln concrete.



Hydration of cement paste and changes in ambient conditions may cause thermal gradients wlthin a concrete structure, which Will in turn create differentlal volume changes. These drfferential volume changes Will create tenslle stralns that may exceed the tensile strain capaclty of concrete, causlng It ta crack. Mass concrete structures, such as piers, wharfs, and dams are prone ta thermal cracking. The larger the structure, the greater the nsk for thermal gradients. The cracks are usually found on the surface of the concrete, mostly ln the form of map cracking, and are normally a few millimeters or centlmeters deep.

Reduclng the temperature of the Internai concrete, delaylng the start of cooling, controlllng the rate at whlch the concrete cools, and Increaslng the tenslle straln capacrty of the concrete, ail help ta reduce thermal cracking. These and other methods ta reduce cracking ln mass concrete are dlscussed ln ACI 207.1 R, 11, ACI 207.2R, 12, and ACI 224R.



Structural cracks can occur as a result of externally applied loading elther during construction or during the service life of the structure. Construction loads are often significantly more than service loads. Since these conditions usually occur when the concrete is most vulnerable to damage, the cracks which develop are usually permanent. Overstresslng the concrete locally may also cause the concrete to crack. For Instance, concentrated wheelloads may cause cracking along the direction of the relnforcing bar as a result of hlgh bond stresses. Concentrated loads at anchorages of prestressing tendons can cause cracking along the direction in which the load is applied. Figure 2.8 summanzes the various forms of load-induced cracking.



There are several deletenous chemical reactlons whlch may cause cracking in concrete. These reactlons may be caused by reactive aggregates in the concrete or substances that come into contact with the hardened cement paste. Certain aggregates containing active sUica react wlth the alkalies found in the hydrated cement paste ta form an expansIVe silica gel. The resulting local stresses which occur as a result of thls expansion causes the concrete to crack and may often lead to complete detenoratlon. Water whlch contains sulfates also reacts with the portland cement paste constltuents to form an expansive product resulting in high local stresses which cause the concrete ta crack and deteriorate. Deterioration may also occur from the repeafüd application of deicing salts ta the concrete surface. The effects of these and other chemlcal reactioRS felatlng to the durabihty of concrete are dlscussed ln greater detall ln SectIOn 2.4.



Relnforclng steel ln concrete 15 usually protected by a passive oXlde coatlng whlch forms ln the highly alkahne pore solution ln hydrated concrete. However, relnforclng steel may corrode l’ the alkalimty (pH) of the concrete IS reduced by carbonatlon or by destruction of the passive film by aggresslve ions such as chlondes. The resulting steel corrosion produces expansive products which occupy a much greater volume than the onginal steel. This increase in volume causes hlgh radial bursting stresses around reinforclng bars whlch lead to cracking of the surroundlng concrete. The princlples of relnforcement corrosion are dlscussed ln more detailln Section 2.3



CrystalhzatlOn of sulfate salts ln concrete pores can lead to slgniflcant damage. This occurs when one slde of a concrete member is exposed to a salt solution and the other sldes are exposed to the atmosphere. Many porous matenals are susceptible to cracking from crystallizatlon pressures. 1



Errors in design and detaihng may lead to unacceptable cracking of concrete. The effects of cracking range from “poor appearance ta lack of servlceabllity ta catastrophlc failure”.” Common errors in design and detalhng that may lead ta cracking Include poorly detalled corners, sudden changes in cross-sectlOnal area, Improper selection and/or detaihng of relnforcement, member restraints, insufficient number of contraction JOInts, and Improper design of foundations, leadlng to differential settlement. The degree ta whlch Improper design and detailing will cause cracking depends on the partlcular structure and loading conditions involved.



The vanous weathenng processes which cause concrete ta crack Include freezing and thawing, wetting and drying, and heatlng and cooling. Ali of these processes create volume changes ln the concrete which lead ta excessive cracking. The best defence against deterioration due ta natural weathering is ta provide a concrete wlth the lowest practical water-cement ratio, durable aggregate; adequate air entralnment, and proper cunng. A more detailed discussion of these forms of detenoration IS provlded ln the following sections.



Detenoratlon of concrete due ta freeze-thaw action is one of the major durability problems with structures in cold chmates. The cause, rate of detenoration, and extent of damage depend on the charactenstics of the concrete pore matrix and speclfic enwonmental conditions. The frost damage to concrete usually manlfests ln cracking and spalling, and scaling. Cracking and spalhng are the most cam mon forms of damage and are caused by continuous expansion of the concrete from repeated freeze-thaw cycles. Scahng usually occurs as a result of freeze-thaw action ln the presence of delclng salts.1 The vanous mechanlsms by which frost damage occurs ln the cement paste are descnbed below. Since hardened cement paste and aggregate particles in the concrete behave differently when subjected ta freeze-thaw action, these are described separately.



Powers1314 theonzed that frost damage in cement paste IS caused by hydraulic pressures generated dunng the freezing of water ln the capillanes or pores of the concrete. When water begins ta freeze there IS a corresponding increase ln volume of nine percent.1 • 6 The resulting hydraulic pressure depends on the distance ta an “escape boundary”, the permeabllity of the concrete. and ·”e rate of ice formation. In the case of completely filled water pores, the concrete will crack.

Powers also suggested that osmotic pressure, caused by differences in salt concentrations in the pore fluid, can be another source of destruction ln cement paste.1 Since solutions treeze at lower temperatures than water, the hlgher the salt concentration in the pore fluid, the lower the freezing pOint (Figure 2.9). When the temperature of the concrete drops below the freezing point, ice crystals Will form ln the large capillaries, resulting in “an increase in the alkali content in the unfrozen portion of the solution ln these capillaries”.15 An osmotic potential is created which causes water in the adjacent unfrozen pores ta move towards the solution in the frozen pores (Figure 2.10). ThiS decreases the alkali content of the solution adjacent to the ice and causes the ice crystals to grow (ice-accretion). When the pore IS completely filled wlth Ice and solution, any further crystal growth produces expansive pressures whlch eventually lead to destruction of the cemenrpàste. According ta the theory proposed by Lltvan,16 the process explalned by Powers causes a portion of the paste in the unfrozen reglons to dry up, and the frozen reglons to expand. ln addition, damage occurs when the mOlsture wlthln the pores 15 not adequately redlstnbuted ta accommodate the conditIOns, e.ther due ta a hlgh degree of water saturation, rapld coohng, or low permeablhty.l ln such cases, when the pore water freezes It forms a seml-amorphous sohd which produces high Internai pressures. 1~



Powers17 found that aggregates bleed Internai water during freezlng, and the hydraulic pressure theory developed to explaln the damage to cement paste by frost action is also believed ta be applicable ta porous aggregates, such as sand stones, shales, and certain cherts. The behavior of an aggregate particle when exposed ta freeze-thaw action depends on the pore size distribution and permeabllity.l Wlth regard to reslstance ta frost action, Verbeck and Landgren18 proposed three classes of aggregate: low permeablhty, intermediate permeability, and high permeability. The first eategory Ineludes aggregates of low permeability and high strength. In these aggregates, when water in the pores freezes, the elastie strain ln the aggregate partiele is absorbed without eausing any damage.

Aggregates of Intermedlate permeabllity have a signifieant amount of small pores (500 nm or smaller). Verbeek and Landgren lB showed that for any natural rock, there IS a eritical parti cie slze below whlch internai water ean be frozen wlthout damage. There is no umque entieal slze for aggregates beeause thls depends on several factors Ineluding the degree of saturation, freezlng rate and permeablhty of the aggregate. However, some aggregates, sueh as granite, basait, dlabase, quartzite, and marble do not produee stress when freezing oceurs regardless of the partlele size. 15 If aggregates larger than the eritlcal size are used in conerete, the primary failure mode is accompanied by pop-outs, as shawn in Figure 2.11.

Aggregates having a significant number of large pores are eonsidered to be highly permeable. Although the hlgh permeability of the aggregate allows water ta penetrate and exit easily, trost damage ean still occur. When pressurized pore water IS forced out from an aggregate partiele, the interface between the aggregate surface and cement paste may be damaged. In this case, frost action does not cause damage to the aggregate partiel es. 1



The eharacteristies of the cement paste and aggregate bath have an effect on the frost resistance of concrete. In each case, the resulting behavlor IS dependent on the interaction of several factors, such as the location of escape boundaries, pore structure, the degree of saturation, the rate of eoohng, and the tensile strength of the concrete. Providing air entrain ment in concrete, using frost-resistant aggregate, and the use of proper mix proportloOing and cunng Increase the resistance of concrete ta frost damage.’ These are summanzed below. These and other measures to protect concrete agalnst frost damage are descnbed ln more detailln ACI 201.2R.’~

(a) AIr Entrainment. The frost resistance of concrete can be substantlally Improved by provldlng an adequate air-void system wJthln the concrete ta reduce its permeabillty (Figure 2.12). The addition of sm ail amounts of air-entraining agents to the fresh concrete mixture (e.g., 0.05 percent by welght of cement), small bubbles ranglng from 0.05 ta 1.0 mm (0.002 ta 0.004 in.) in dlameter are created for protection of concrete agalnst frost damage. 1

Depending on the aggregate slze and exposure conditions, the dosage can be varied ta produce the desired air content. The recommended air content vanes wlth aggregate slze because concrete mixes that contaln large aggregates require less cement paste than nch concretes with smaller aggregates. Therefore, the latter would need m\lre air entrainment ta provlde the same protection agalnst frost damage. 19 The recommended air contents for frost-resistant concrete, according to ACI318-92,20 are shawn ln Table 2.1. TheACI318 permrts a one percent decrease ln total air content for concretes having a speclfied 28 day compressive strength ln excess of 34 MPa (5000 pSI).

Ounng the placement of concrete, the air content of the concrete should be measured frequently. According ta ACI Committee 201,’5 the followlng te~t methods may be used: volumetrie method (ASTM C 173), pressure method (ASTM C 231), or the umt weight test (ASTM C 138). An air meter may also be used ta estlmate the air content. For hghtweight concrete, the volumetrie method is recommended. The air content in hardened concrete can be determined using microscope techmques ln the laboratory (ASTM C 457).

(b) Law Water-Cement RatiO. Verbeck and Khegef’ confirmed the hypothesis that at a given freezlng temperature the amount of available water whlch can be frozen will be more with higher water-cement ratios (Figure 2.13). When the water-cement ratio is decreased and the cement content IS increased, the frost resistance of the concrete Will increase.

Accordlngly, ACI Committees 201.2R’5 and ACI 318-9~o have set guidelines for producing frostreslstant concrete for a vanety of conditions: for frost-resistant normal weight concrete, the watercement ratio should not exceed 0.45 for thln sections (bridge decks, railings, curbs, sllls, ledges, and ornamental works) and sections with less than 25 mm (1 ln.) of concrete cover over the reinforcement, and any concrete exposed to deicing salts; and 0.50 for ail other structures. Also, for lightweight concrete, a mimmum 28 day compres~l”e strength of 28 MPa (4000 psi) is recommended .

(c) Frost-Resistant Aggregates. Sound coarse aggregates will produce frost-resistant concrete. Aggregates whlch are crushed to a nominal size of 12 ta 19 mm (‘/2 ta 3/. In.) will usually produce satisfactory results, since the crushing process tends ta break the aggregate along its weaker planes. Aggregates whlch are highly porous, such as some cherts, sandstones, hmestones, and shales are more susceptible ta trost damage than aggregates like granite, basait, quartzite, or marble. 19

Natural aggregates should meet ASTM C 33 requrrements and lightweight aggregates should meet the requlrements of ASTM C 330. The best way to evaluate aggregate performance IS by field experience, but if this is not feasible, the engineer must rely on laboratory testlng, such as petrographiç examination, rapid freezing and thawing tests (ASTM C 666), and dilation tests (ASTM C 671).

(d) Adequate Curing. Proper consolidation and curing are ais a important factors influencing the frost resistance of concrete. The ACI Committee 201.2Rl~ report recommends that alr-entrained concrete should reslst the affects of freezlng and thawing (one or two cycles) as soon as a compressive strength of 3.45 MPa (500 psi) is attained. Concrete should have a compressive strength of about 28 MPa (4000 psi) before it is exposed ta freezlng temperatures. For moderate exposure conditIons, a specified compressIve strength of 21 MPa (3000 pSI) should be attalned . For water-cement ratios of 0.50 or less, at least seven days of mOlst curing at normal temperature is recommended before exposlng the concrete ta freezing condItions. 1



The comblned effects of frost action and the presence of deiclng salts on concrete produce a more severe attack than frost alone. Applylng deiclng agents, such as ammonium chlonde, calcium chlonde, and sodIum chloride ta a concrete surface covered with ice wIll cause the concrete surface ta experience temperature shock when the ice melts. The temperature gradient which 15 created between the surface and the Interior of the concrete will generate Internai stresses that may cause the outer layer of the concrete ta crack.6

As prevlously described ln Section, the change ln the freezlng behavlor of the pore water is due ta the Ingress of salts (deicing agents) from the outside of the concrete. The content of deicing solutIon will decrease wlth Increasing dIstance from the surface of the concrete, creating a freezing temperature gradient wlthln the concrete.22 As a result of bath the change in temperature and salt concentrati~.n gradIents, sorne layers of concrete will freeze at different times, causing scaling (Figure 2.14). Researchers have noted that the most damage ta the concrete surface by scahng occurs when salt concentrations reach about four ta five percent. 1 The use of chlorides as delcing agents also Increases the risk of reinforcement corrosion (Section 2.3).



The reslstance ot concrete ta surface wear IS detlned as the “abllity of a surface ta resist being worn away by rubbing or friction”. 24 Concrete does not have a high resistance to repeated abrasion cycles, especially if the concrete IS very porous or has a low strength, and contains an aggregate of low wear reslstance. 1 Surface wear can occur due ta abrasion, erosion and caVItation. From a review of References 1, 6, 15, and 25, the three phenomena are summarized below.



The term ‘abrasion’ is usually used ta describe wear on pavements and industrial floors by vehlcular traffic. 15 Although this is not a significant prob!em in the marine environment, such wear may occur on surfaces of concrete piers or wharves. Accordmg to Prior,26 abrasion of pavements can be classlfied into two types:

  • “Wear on concrete floors due ta foot trafflc and hght trucklng, sklddlng, scrapmg or shdlng of abjects on the surface (attrition)”, and
  • ·Wear on concrete road surfaces due ta heavy trucks and automobiles wlth studded tires or chalns (attrition, scraplng and percussion)’.


Tire chains and studded snow tires can also cause slgnlfieant wear damage to good quality concrete surfaces. Tire chains cause wear by ·flaihng and scufflng action” as the metal studs contact the concrete surface. The damage caused by studded snow tires IS due to dynamlc impact of the small tungsten carblde tip of the studs. A study eonducted by Smith and Sehonfeld27 in Ontano, Canada, determined that ruts from 6 ta 12 mm (V4 ta V2 ln.) deep may form ln a single season where traffie is heavy. In general, wear due ta abrasion does not affect the concrete structurally, but in some cases It may cause servlceablhty or dusting problems .



The abFeslon reslstance of concrete IS dependent on compressive strength, aggregate propertles, use of topplngs, and finlshlng and cunng methods. Test and field experience have shown that compressive strength IS the most Important factor Influenclng the abrasIon resistance of concrete. Accordlngly, ACI Commlttee 201 15 recommends that the compressive strength of concrete should be more than 30 MPa (4000 pSI). This can be achieved by using a low water-cement ratio, proper grading of fine and coarse aggregate (hmlt the maximum nominal size to 25 mm), and minimum air content as dlctated by exposure conditions. USlng hard, tough, coarse aggregates will provide additional abrasion reslstance.



Erosion damage occurs as a result of abrasion caused by silt, sand, gravel and rocks which are carned by flowlng water over a concrete surface (attrition and scraping).29 Erosion is recognized by the smooth, worn appearance of the concrete surface. This damage IS common to hydrauhc structures at bndge piers, and structures protecting embankments or coasts. Due ta the presence of high water veloclties, spillway aprons, stilling basins, sluiceways, and outlet tunnel hnlngs are especlally susceptible to eroslon damage.

The rate at which eroslon occurs depends on several factors including the porosity or strength of concrete, and on the amount, size, shape, denslty, hardness, and velocity ofthe particles being transported. Dependlng on flow conditions, erosion damage can range between a few centimeters ta several meters. The relatlonshlp between fluid-bottom veloclty and the size of particles whlch a specific velocity can transport IS shown in Figure 2.15. If the quantlty and size of the partlcles are small, erosion will not be slgnlflcant at water velocities of up to 2 mis (6 ft/s).



Numerous matenals and techniques have been utilized for constructlng and repalring structures damaged by severe eroslon. Investigations by LIU30 have shown that “abrasion-resistant concrete should include the maximum amount of the hardest available coarse aggregate and the lowest practlcal water-cement ratio· (Figure 2.16). For instance, concrete containing chert aggregate will provlde approxlmately twlce the abrasion-eroslon resistance of concrete containing limestone .

When a structure will be exposed to severe eroslOf’ or abrasion conditions, ACI201.2R1~ recommends that, in addition to using hard aggregates, the concrete should have a minimum 28 day compressive strength of 42 MPa (6000 pSI) and moist-cured for at least seven days before exposing the concrete to the aggresslve envlronment If hard aggregate IS not avallable, the use of sillca fume and hlgh-range water-reduclng admlxtures will produce a very strong concrete . .’o

Vacuum-treatedconcrete, polym€’. concrete, polymer-Impregnated concrete, and polymer-portland cement concrete will provlde a higher resistance to abraslon-eroslon damage than conventlOnal concrete. Accordlng to ACI Commlttee 223,31 concrete produced wlth shnnkage-compensatlng cement, Will provlde an abraSion resistance from 30 to 40 percent hlgher than portland cement concrete with simllar mixture proportions. The vanous materials and techniques used for repalnng erosion damage to hydrauhc concrete structures are dlscussed ln Chapter 7 .



Erosion damage to concrete hydraulic structures can also be caused by cavitation, resulting from the sudden collapse of vapor bubbles ln water that is flowing at velocities in excess of 12 mis (40 ft/s) , or 7.6 mis (25 ft/s) in closed condurts.29 ln flowlng water, vapor bubbles form when the local absolute pressure in water drops to the ambient vapor pressure of water corresponding to the amblent temperature. Figure 2.17 shows examples of concrete surface irregularities which can cause vapor bubbles to form.

Cavitation damage is produced when the vapor bubbles collapse or implode. The collapses that occur ;.lt’cir the concrete surface produce very high instantaneous pressures that impact on the COUI , -t,~ surt,.:Aces. Repeated impact of these high-energy pressures will cause severe local Plttlh9 The Jamage caused by cavitation IS different from that caused by erosion because cavitation plt~ eut around the coarse aggregate panlcles. ThiS continuous action eventually undermines the aggregates causing them to come loose .



According to ACI Committee 210R29 , the cavitation resistance of concrete can be Increased by using high-strength concrete wlth a low water-ceme,’t ratiO, provlded It IS not subJected to abraslon-erosion damage. The use of hard, dense aggregate wlth a nominal mruumum size of 38 mm (1-Y2 in.) is recommended for produclng a good bond between the aggregate and the cement paste. This IS essentlal for achlevlng Increased reslstance to cavitation damage.

Cavitation damage has been successfully repaired using steel flber- reinforced concrete.32 ThiS material provldes good impact reslstance to cavitation damage and appears to reduce cracking and disintegration of the concrete. The use of polymers has also shown to Improve the cavitation reslstance of both conventlonal and fiber-reinforced concrete. 33 ,34 Va nous matenals and coating systems have been tested at the U.S. Army Detroit Dam (Oregon) High Head Erosion test flume.35 Figure 2.18 shows the performance of several of these materials subjected to water flows with velocities of 37 mIs (120 ft./s).

Although uSlng the proper matenals will increase the cavitation resistance of concrete, the best defence IS to mlnimize or eliminate the factors causlng cavitation, such as misalignments or abrupt changes of slope. In sorne cases, thls may be unavoldable and the designer can minimize the effect of cavitation by the use of aeration devices designed to supply air to the flowing water.29 Research has shawn that Irregularities on the concrete surface will not cause cavitation damage if the air/water ratio in the water adjacent ta the concrete surface is about eight percent. Such devlces are shawn ln Figures 2.19 and 2.20 .



Another approach to controlling surface wear of concrete is by testing and evaluatlng the materials prior to using them in hydraulic structures. A variety of standardized test methods are available for determining abrasion-erosion reslstance of concrete surfaces ln terms of weight loss after a specified time.’ ASTM C 779 descnbes three methods for testing the relative abraSion resistance of hOrizontal concrete surfaces ln terms of welght 1055 after a speCitled lime. These tests Include an abraSive type apparatus, such as steel dresslng wheels and rolling steel balls under pressure. ASTM C 418 describes the sandblast test, which determlnes the abraSion reslstance of con crete by subjecting it to the abraSive action of alr-blown sihca sand. The moditled Los Angeles rattler tests (ASTM C 131 and C 535) have also been used to determine abrasion-erosion resistance of concrete surfaces .


These tests are designed to slmulate erosion caused by heavy foot or wheeled traHie on concrete surfaces and are not appropriate for modelling eroslon ln hydraulic structures. Accordingly, the U.S. Army Corps of Engineers has developed a test method for simulating and measuring abraslon-erosion in hydraulic structures. This test (CRD-C 63-80), “Test Method for AbrasionErosion ReSistance of Concrete (Underwater Method),” subJects concrete specimens to abraslonerosion under the action of steel grindlng balls and circulating water with an approximate velocity of 2 mIs (6 ft./s). The damage IS measured by determming the amount of lost material as a percentage of the original mass. The development of the test procedure and data from a vanety of tests of vanous concrete mixtures have been reported i:ly Liu.3O



A racent raview by Mehta36 of proceedings of the cement Chemistry Congresses and other symposia held by ACI, ASTM, and RILEM in the last 50 years observed that ‘corroslon of relnforcing steel is considered to be the most serious problem responslble for lack of durability· ln modern (post 1960) structures. He states that bndge-deck and parking garage detenoratlon due to reinforcement corrosion has become a major concern in the United States and Canada. A recent report by Gerwlck37 discloses that many worldwide concrete-lined tunnels are leaklng as a result of corrosion of reinforcing steel. Accordlng to Mehta1 , a survey of collapsed bUildings dunng the 1974-78 penod ln England showed that relnforcement corrosion of prestresslng steel was the cause of failure in at least eight structures (12-40 years old). In manne structures, the most slgnificant damage from corrosion of steel occurs wlthln the splash zone, where the structure IS exposed ta alternatlng cycles of wetting and drylng38 .

The damage caused by reinforcement corrosion consists of expansion, cracking and eventual spalling of concrete cover. Corrosion IS often readily identlfled by rust stains which are bled through cracks on the concrete surface. Unfortunately, such signs usually indicate that the damage is already weil advanced. In sorne cases, reinforcement corrosion can result ln loss of bond between concrete and the steel, resulting in structural failure. 1 . 38



Reinforcement corrosion of steel in concrete is an electrochernical process whlch involves the transformation of metallic iron (Fe) ta rust [Fe(OH)3J. The phenomenon can be represented by an anode process and a cathode process3 as shawn by the reaction below. This is iIIustrdted also schematlcally in (Figure 2.21):

When metallic iron is converted ta rust, it produces an increase in volume which may be as much as six ta seven times that of the original steel, causing expansion and cracking of the concrete .

Embedded steel in concrete is usually protected from corrosion bya passive film of iron oXlde formed on the steel surface due to the high alkalinity (pH of 13.5) of the pore solution ln the hydrated cement paste. ThiS passive film must be dlsrupted before the anodic reaction can begin. Similarly, for the cathodic reaetlon ta occur, oxygen and water must be continuously available. Mehta36 and Hertlein39 state that for the transformation of Iron to rust to occur, ail of the followlng essentlal conditions must be satisfled:

  • For the anode process to occur, metallic (Fe) iron must be available at the surface of the relnforcing steel,
  • There must be voltage patent lai differences along the steel surface or the surrounding concrete,
  • Oxygen and mOlsture must be available for the concrete ta have eleetrical contact with the steel, and
  • The electrical resistivity of concrete must be low enough to allow electrons ta flow in the steel from anodic to cathodic areas.


To explain the deterioratlon of reinforced concrete structures in the marine environment, Mehta and Gerwick3 developed a cracking-corrosion interaction model (Figure 2.22), according to which an Increase in the permeabihty of concrete caused by enlargement and interconnection of microcracks is necessary for reinforcement corrosion ta occur. As the rate of corrosion increases, the formation of rust will increase microcracking further, thus increasing the risk of reinforcement corrosion. Ultimately, this process leads to severe detenoration of bath the steel and the concrete.



Depasslvation of steel in concrete is caused by the removal of calcium hydroxide around the rebar, and the breakdown of the Iron oxide film present on the surface of the rebar.36 Once the passive film is destroyed, the corrosion activity will depend on the electrical reslstlvity of the concrete and the amount of oxygen available at the cathode.3 The followlng sections pravide a brief summary of sorne important factors contributing ta steel depassivation. The principles of depasslvation are also valid for prestressing steel.a



Carbon dloxide (COJ present in the air or in some waters penetrates the concrete and reacts with the pore flUld ta form carbonic acid (Section This reacts with the al kali ne calcium hydroxide [Ca(OH)2J ln the hydrated cement paste ta form calcium carbonate (CaCOJ, which reduces the pH of the concrete ta around 9.4.39 This may be represented by the following reaction:

Research has shawn that reducing the pH of the pore solution in the cement paste ta below 11.5 destroys the passIVe film on the steel, and initiates the corrosion process. 1 ,3 The rate of carbonation (increase of carbonation depth with tlme) depends on the rate of CO2 penetration into concrete, and appears ta follow a square-root tlme law (Figure 2.23).8 Penetration of CO2 can only occur in air-filled pores. Concrete whlch is completely saturated with water will not carbonate, unless it IS subjected ta repeated wetting and drylOg cycles. ThiS is why in marine structures deterioration of concrete from corrosion of steel is more of a problem in the splash zone .



The presence of free chloride (CI) ions ln concrete destroys the protective iron oxide film on steel. 3 When free chloride Ions are present, the electrical conductivity of the concrete is Increased, and by chemical reaction, depassivate the steel.39 The Ions may be introduced in the concrete in several ways: as a secondary effect of carbonation (breakdown of chloro-aluminates, releasing cr Ions), as an accelerating admlxture, chloride-contaminated aggregates, or from delcing salts and seawater spray.

Although chio ride Ions are an essential catalyst of the corrosion process, the mechanism through which the protective film is destroyed IS not fully understood. Three theories have been postulated to explain the electrochemical effects of chlonde ions on steel corrosion:7

(a) OXlde FIlm Theory. Chloride ions penetrate the protective film through pores or defects in the film. Also, the chloride ions may “COlloidally disperse” the film, thereby facilitating penetration of ions .

(b) Adsorption Theory. As chlonde ions are adsorbed on the surface of the relnforclng steel, metal ions are freed more easily. As a result, the chlonde Ions react more aggressively with the dlssolved oxygen or hydroxyl Ions.

(c) Transitory Comp/ex Theory ·Chlonde Ions compete wlth hydroxyllons for the ferrous Ions produced by the corrosion pro cess· ta form Iron chlonde whlch diffuses away from the anode. The protectlve fernc oXlde layer IS destroyed when Iron chlonde breaks down to form iron hydroxlde and releases the chlonde ion whlch removes more ferrous Ions from the anode.



For the corrosion process to occur at any appreclable rate, It is clear that a speclflc chlorrde ion concentration must be present. This hmlt IS often termed the threshold chlonde Ion concentration. Exceeding this hmit will Increase the rate of corrosion. 7 The concept of threshold chlonde concentration IS shown schematlcally ln Figure 2.24. From the graph it is clear that an Increase in chloride concentration will not have an adverse effect on the concrete as long as the pH value IS also Increased.

Empirical data shows that, when the chloride to hydroxyllon molar ratio is higher than 0.6, steel is no longer protected against corrosion even at pH values greater than 11.5.1 The relationshlp between chlonde concentration and pH at the iron-liqUid interface IS shown in Figure 2.25. A chlonde ion concentration of 0.2 percent by weight of cement IS normally consldered as the threshold hmlt.40 For con crete mlx proportions typically employed ln practice, the threshold chlonde content required to start the corrosion process van es between 0.6 to 0.9 kg of CI per cublc meter of concrete (0.2 to 0.3 Ib/cy).’ However, an exact value has not been firmly estabhshed.



There are many techntques avallable to help predlct the start of corrosion, assess corrosion damage, and identlfy the cause. However, if they are to be used effecllvely, Il IS Important to understand the vanous processes that can Inltlate corrOSion, and how th,s corrosion Olay affect the structure.

To integrate the various factors influenclng relnforclng steel corrosion ln concrete structures, Tuutti41 proposed a model (Figure 2.26) whlch suggesls Ihat corrosion damage can be predlcted by assumlng two separate rate determlntng penods: the corrosion InitiatIOn penod and the corrosion propagation period. Each stage IS a~ected by dlfferent paramelers. The Initiation period IS influenced by the rate of CO2 and chlonde Ion diffUSion, white the propagation penod is Influenced by the rate of oxygen and water diffUSion These parameters control the depassivatlon of steel and the cathodlc reactron, respectlvely. Accordlng to lin and Jou4J the second stage IS more dlfficult to pradlct because Il not only depends on the rate of oxygen diffusion through the concrete coyer, but also the degree of corrosion Ihe structure can endure. The propagation or detenoratlon penod is Influenced by other factors, such as mOlsture content of the concrete, its quahty, strength, and mechanrcal requrrements. A hlgher temperature and a more rapld loss of mOlsture of the concrete can produce short propagation perrods (from SIX months to flve years).43 ln thls case, the effective service IIfe of the structure can be consldered as the initiation penod and also the design IIfe of the structure.42



Corrosion of metals can occur ln many ways. The type and rate of corrosion depends on the matenals present ln the concrete, and on the mOisture and gases avallable because they will affect ItS permeablhty, denslty, and reactlvlty.39 Corrosion may elther manltest Itselt ln the form of general rustlng, or locahzed attack known as plttlng corroSion. Other forms of corroSion Include galvaOlc corrOSion, stress corrosion, hydrogen embnttlement, and bactenal corrosion. These are dlscussed below.



General rustlng 15 the most common form of steel corrosion. As prevlously descnbed, this occurs through a complex electrochemlcal process ln which the metal is oXldlzed when exposed to air and water SmalJ electncal currents that flow between areas of dlfferent voltage potentlal on the steel transport metallons from anodlc to cathodlc areas, thereby reducing the steel cross-section at the anode, and deposltlng metal at the cathode. As a result, rust occurs untformly over the entlre steel surface. Since rust typlcally occuples approximately six to seven tlmes the volume of steel, thls Increase ln volume creates stresses that cause cracking and spalling.39



Pltting corroSIon IS a locahzed form of attack whlch Inttlates when the protectlve film breaks down over smalJ surface areas. This often occurs when there is a large concentration of chi onde ions ln the concrete ln small depasslvated areas. ThiS decr’3ases the electrical reslstivity of the concrete locally and Increases the rate of dissolution of Iron at the small anodic pit (Figure 2.27).39 This ohen results ln a substantlal local reductlon ln steel cross-sectlOnal area at the pit wlthout any external sign of damage to the concrete, because the corrosion products are soluble and are absorbed wlthln the concrete.44 ThiS form of rusttng otten leads to sudden fallure of prestressed or post-tensloned structures.

Anodlc and cathodlc areas can create concentratIOn cells whlch may elther be mlcroscoplcally separated (microcell corrosion) or locally separated (macrocell corrosion).8.39 These concentration cells are shawn schematlcally ln Figure 2.28. When the amounts of chloride ion concentratIon ln concrete vanes, electncal potentlal dlfferences along the steel surface are created which permit corrosion to Inltlate. Corrosion may also occur as a result of different amounts of oxygen that are avallable to vanous areas of the relnforcement, creattng what IS known as a dlfferentml-oxygen cell. 7 For instance, Funahashl et al.38 suggest that ln manne structures contamlnated wlth chlondes, rebar ln concrete located under the seawater is anodlc ta the rebar ln concrete exposed ta the atmosphere. Research by Okada et al.3 showed that the ratIO of the cathodlc area (Ac) ta the anodlc area (Aa) affects the rate of corrosion ln macrocells of relnforclng steel ln concrete. Their observations also showed that wettlng and drylng cycles, as opposed ta contlnuous ImmerSion, Increases the corrosion rate by Increaslng the Ac/Aa ratio (I.e., oxygen supply ta the cathode).



Galvanie cells are created when the steel is ln direct or Indirect contact wlth another type of metal whlch occuples a different position ln the ‘Galvamc Senes’ 39 The electncal potentlal that IS developed, and the relative sizes of the two dlsslmllar metals, will determlne the rate and extent of steel corroslon.7 The further apart the metals are ln the galvanlc senes. the more aggresslve is the reactlon.39



This form of corrosion may lead to bnttle fallure of reJnforclng or prestressJng steel. Localized anodic processes produce hlgh permanent stresses that can lead ta cracking. The anodlc process occurs at the root of the crack dunng the crack propagation stage as shown ln Figure 2.29.6



Another type of bnttle fallure whlch can occur in steel is the result of a cathodic process known as hydrogen embnttlement. Under certain conditions dunng the cathodlc process, atomlc hydrogen is produced as an Intermedlate product and can penetrate Into the steel. The hydrogen recombines to form molecular hydrogen wlthln the steel and produces a high Internai pressure whlch usually leads ta cracking (Figure 2.30).6 Both types of failures can be prevented If the steel is encased by sound hardened concrete or cement grout.



ln anaeroblc (oxygen-free) conditions, bactena can form on the surface of the concrete and penetrate to the level of the steel. Although oxygen IS not present, the bacteria produces iron sulfide which initlates the corrosion reaction. This reaction IS often severe and can lead to significant structural damage.44



Using a good quality concrete of low permeabllrty IS essentlal to control the various mechanisms involved in the corrosion process. Although no conventional concrete is completely impermeable, proper and careful attention to concrete mixture parameters, workm.lnship, and cur:ng will ensure a good qualrty concrete wlth a low permeabllity. The vanous parameters are summarized in the followlng sections.



Low water-cement r::ltios produce less permeable concrete which in turn provides greater protection agalnst reenforcement corrosion. Figure 2.31 shows the influence of water-cement ratio and the degree of hydratlon on permeability. Accordingly, the ACI Building Code 318-9~o specifies a maximum water-cement ratio of 0.40 and a concrete compressive strength of at least 35 MPa (5000 pSI) for normal welght concrete exposed to delcing salts, brackish water, or seawater (Table 2.2). However, if the concrete cover IS increased by 13 mm (% in), the code allows a ma>umum water·cement ratio of 0.45. The ACI Committee 357R-8445 report for the design and construction of offshore concrete structures recommends similar water·cement ratios for various exposure zones (Table 2.3). Wh en severe detenoration of concrete is anticipated, a 28-day compressive strength of 42 MPa (6000 pSI) is recommended.

The ACI Committee 201.2R’5 report recommends that for structures located above the seawater and seawater spray zone for a height of 8 m (25 ft.), or wlthin a horizontal distance of 30 m (100 ft.), the water-cement ratio should be less than 0.50 by weight .



The capaclty of concrete ta bind CO2 and cr willlnCrea&B as the cement content also increases (Figure 2.32). The rate of carbonatlon and chloride penetration in concrete are influenced much less by the cement content than by the water-cement ratio, quality of compactlon, and curing. However, the cement content Will influence the workabllity of concrete, and ta a lesser degree, the cunng senSltlvlty.6

ACI 318-9~o does not provlde any cement content requlrement for a manne envlronment. Normally, a cement content of about 300 kg/m3 (500 Ib/cy) IS enough to produce a low permeablhty concrete with adequate durabllrty, provlded the water-cement ratio IS below 0510 0.6.8 The ACI 357R-8445 report recommends a mlnrmum cement content of 355 kg/mJ (600 Ib/cy) of concrete. If more than 415 kg/m3 (700 Ib/cy) of portland cement IS used, special steps must be taken to reduce the hkelrhood of cracking ln thln members due to thermal stresses Thermal cracking can be reduced by replaclng part of the cement wlth a pozzolan 48



The durability of concrete IS greatly affected by cement composition. The tncalcium aluminate (C3A) content in portland cement concrete has a slgnrflcant ef’ect on the corrosion process. Increasmg the C3A content Increases the reslstance to corrOSion, slnce the chlonde Ions react wlth the hydrated tncalclum sulfoalumlnate ln the hardened cement paste to form an Insoluble Frredel salt. Recent research by Rasheeduzzafar47 showed °that corrosion Initiation tlme, IIme-tocracking of cover concrete, and chlonde threshold values Increased, whereas loss of metal from relnforcement corrosion d~creased as the C3A content of cement Increased”. Figure 2.33 shows the effect of C3A content of cement on tlme-to-Inrtlatlon of corrosion of relnforclng steel.

Simllarly, the loss of metal data for relnforcmg bars taken from ASTM Type 1 (CSA Type 10) and ASTM Type V (CSA Type 50) cement concrete test specimens Indicate that the corrosion performance of Type 1 (C3A: 9.5%) cement IS better than the performance of Type V (C3A: 2.8%) cement (Figure 2 34).47 Nevertheless, as the amount of chlondes Increases, the beneflt of addlng more C3A becomes less notlceable slnce C3A ln cement combines wlth only a IImlted quantlty of chlonde Furthermore, Increasmg the C jA content reduces the resistance of concrete ta sulfate attack. 7 ln such Situations, uSlng Type V cement would provlde adequate protection agamst sulfate attack but It would not rem ove free chlondes ta protect the steel from corrosion. ACI 357R-8445 permlts the use of ASTM Type l, Il, and III (CSA Types 10, 20 and 30) portland cements, but recommends that the C3A content should be between 4 and 10 percent.



ln marine enwonments, the use of pozzolans, such as sllica fume, fly ash, and blast-fumace slag, are commonly used to produce a concrete which Will slmultaneously resist sulfate attack and chlonde-Induced corrosion. Pozzolans combine wlth the calCium hydroxide and water ln the fresh mix ta form a hardened cement paste \Allth a hlgher strcngth and a reduced permeablhty (Figure 2.35). Pozzolans also combine chemlcally wlth the lime and reducc the effects of lime leaching. Typical mlx proportions include (by welght .’)f cement): 15 ta 20 percent fly ash, 50 ta 70 percent of granulated blast-furnaee slag, or 5 to 10 percent of ecndensed slliea fume. 46

ResearGh by Rasheeduzzafar47 on blended cements made by replacmg a portion of ordlnary Type 1 (high C3A) portland cement wlth 10 percent slilca fume, 20 percent fly aSh, or 70 percent blast furnaee slag, produced concrete wlth a hlgher resistance to corrosion and to sulfate attack Results of thls research are shown ln Figure 236. ACI 318-9220 reqUires a Type” cement or a Type 1 cement plus a pozzolan to reslst moderate sulfate attack ln seawater Il should be noled that when pozzolans or other cementltlous admlxtures are used ln addition 10 portland cement, It is more useful to conRlder the water to cementitious matenals ratio rather than simply the walercement ratio.46



Numerous organic and Inorgamc chemleal admixtures have been used to prevent or reduce steel conosion ln concrete. Water-reducing ad mixtures and superplastlclzers are commonly used to provide workable mixes at low water-cement ratios. To protect relnforeing and prestresslOg steel from corrOSion, calcium chlonde (CaCIJ or admlx1ures contalnlng chlorides should not be used.46 Chemical admll(fures used in portland cement concrete must meet t’le requirements of ASTM C 494.



Siflce 70 percent of the concrete mlx volume IS occupled by aggregates, thelr presence has a significant effect on concrete f.i~rmeabihty. For Instance, concrete permeablhty Will increase with r,reasing mruomum coarse aggregate sile. This is because most minerai aggregates have a permeablhty 10 ta 1000 tlmes gre=i sr than that of the cement paste.7 Therefore, it is essential that the moisture content of aggregate.:i useu ;;1 maklng the concrete is included ln water-cement ratio computations.

Aggregates whlch conta,” a sufficlent amount of chlorides may have a deleterious effect on reinforcement corrosion. Aggregates that conform to ASTM C 33 requirements can be used as weil as marine dredged aggregates, provlded they are washed wlth fresh water to reduce the chlonde Ion content.46 However, international expenence has shawn that reduclng the chlonde ion content ln marine aggregates ta an acceptable level IS very dlfflcult, even after double washlng.49



To provlde adequate corroslcn protection, AC1318-9220 hmlts the maximum water-soluble chlonde Ion concentration ln hardened cement at 28 days to the values shawn ln Table 2 4



Concrete caver depth over relnforcing steel 15 thought by many ta be the most Important factor influenclng reinforcement corrosion. Additlonal concrete caver delays the ingress of moisture and chloride Ions, which ln turn increases the tlme-to-corrOSlon period.48 The effect of the concrete coyer thickness on relnforcement corrosion IS influenced by several parameters, as shown by the expression below

The ACI 357-R8445 report on offshore structures provldes recommendations for concrete cover for varia us exposure zones (Table 2.6). Current practice recommends providing a minimum 65 mm (2.5 in.) of concrete caver for conventlonal concrete and 90 mm (3.5 in.) on t>restressing steel for structures in the splash and atmospheric zone exposed ta seawater spray. AASHTO recommends 100 mm (4 ln.) of concrete caver for such exposure except for precast plles.15

Aesearch by Lin and Jou42 confirmed that chloride ion penetration in concrete marine struc’ures can be predlcted by diffusion theory and Fick’s second law. Based on test results, the required con crete covers for effective service lives of 10,30, and 50 years are 51 mm (2 ln.), 88 mm (3.5 ln.). and 114 mm (4.5 ln), respectlvely, with a survival probability of 0.95 .



The quality of concrete compactlon dlrectly affects relnforcement corrosion. Inadequate compactian dunng placement faclhtates the Ingress of elements conduclve to relnforcement corrosion. For Instance, reducmg tl1″3 degree of compact Ion by 10 percent can reduce the concretE” compressive strength by 50 percent. reduce bond by 75 percent, and Increase chlonde permeablhty by 100 percent. 46

A study conducted by the Federal Hlghway Administration (FHWA)50 ln the USA., demonstrated the Importance of proper consolidation The study showed th;::t poorly compacted concrete wlth a water-cement ratio of 0.32, was less reslstant ta chlonde penetration than well-compacted concrete wlth a water-cement ratio of 0.60. Good compactlon can usually be achleved by using internai or Immersion-type \~oker) vlbrators.



Careful cunng, with control of both temperature and moisture, is essentlal for reduclng concrete permeablhty. If the concrete IS Inadequately cured, the permeabihty of the surface layer of concrete may be Increased by flve ta ten times. 6 If the cunng period IS too short, the protectlve passive film Will not develop before chlonde Ions penetrate the concrete.7 Accordingly, ACI Commlttee 201 15 recommends at least seven days of umnterrupted mOlst cun!1g, or membrane cunng. ACI Commlttee 30a51 also provldes current recommendatlons for cunng concrete. The effect of cunng tlme on concrete permeabll:ty IS demonstrated ln Table 2.7.

If concrete members are cured wlth low pressure steam. addltlonal mOlst cunng at normal temperatures IS usually beneflclal. ‘5 However, unless appropnate precautlon~ are taken. steamcured IltaSS concrete structures Will expenence Internai mlcrocracklng as a result of dlHerenlia1 thermal stralns. J The relatlonshlp between steam and mOist cunng and the corrosion Initiation lime of embedded relnforclng steel ln concrete can be represented by the followlng ernpmcal expression: 7



As prevlously stated, the thlckness of the concrete caver plays an Important role wlth regard tn the mfluence of cracks on relnforcement corroslon.6 AC1224-905 hmlts the maximum permlsslble crack width ta 0.15 mm (0.G06 ln.) at the tension side of relnforced concrete structures whlch are exposed to wettlng and drylng cycles or seawater spray (Table 2.8).

The CEB Model Code states that the width of a crack adjacent to the reinforcing steel should be less than 0.1 mm (0.004 in) If the concrete member is exposed to frequent flexuralloads, and 0.2 mm (0.008 in) for other structures. The International Prestressing Federation (FIP) recommends a maximum crack w/dth of 0.004 t/mes the nominal concrete caver.

A direct relat/onship between crack width and corrosIon has not been f/rmly established, however, exposure tests and site Inspections seem ta /nd/cate that the influence of crack width on the carros/on rate /s relat/vely sma” for crack widths up to 0.4 mm (0.016 in.).o Mehta and Gerwick3 suggest that by increasing the permeability of concrete and exposing it ta numerous processes of deteriorat/on, existing microcracks will eventually cause severe deterioration.

Cracks that propagate transversely to the reinforcement are not as harmful as longitudinal cracks (along the relnforcement). This IS because in the case of transverse cracks, corrosion is limited to a small area, 50 that the nsk of spalling of the concrete cover is low. In cases where the horizontal surfaces are in contact with chloride-contam/nated water, transverse cracks may cause senous detenoration. Under these circumstances, limiting the crack width will nct reduce the risk of re/nforcement corros/on.6 As a result, special protective measures must be implemented.



P{ûviding good quality concrete wlth adequate caver is just one of many ways to protect concrete from reinforeement carros/on. Many proteetive systems have been used with varying degrees of suceess. 1~ Reinforcing bar coatings and eathodic protection are other forms of corrosion protectIon and are usually more expansive than producing and placing low-permeability concrete.

The two basic types of protectlVe coatings are: anodie coatings (e.g., zinc-coated steel) and barrier coatings (e.g., epoxy-coated steel). Cathodic protection techniques on the other hand, render the concrete environment inconduc/ve to reinforcement corrosion either by forcing the ionic flow in the opposite direction or by using sacrificial anodes. 1 Both systems have been used with mixed results. Other systems include the use of concrete surface sealers or coatings. Reference 52 describes advantages, disadvantafj6S, and cost impact of various corrosion-protection systems .



Another form of concrete detenoration IS caused by “chemical interactions between aggressive agents present in the external environment and the constltuents of the cement paste”. t The rate at which these reactlons occur Will determine the durabllity of a concrete structure. The ease wlth which the aggresslve substance penetrates the concrete determlnes the rate at whlch deterioration progresses. The accessibllrty of these substances will be determlned by the permeablhty of the onglnally sound concrete, temperature, or by the passlvatlng layer of the products that are produced as a result of the reaction.8

The rate of chemical attack on concrete Will also depend on the pH of the aggressive fluid. t A well-hydrated portland cement paste, will contain high concentrations of Na+, K+, and OH’ ions which produce a high value of pH of about 12.5 to 13.5. Therefore, when portland cement concrete is exposed to an aCldic solution (Iow pH), the alkalinity of the pore fluid will decreasa which leads to destabilization of the cement paste constituents. For instance, free CO2 found in soft and stagnant waters, acidlc ions such as SO./ and cr in groundwater and seawater, and W ions in sorne industrial waters will usually lower the pH of concrete to below 6. Therefore, most industrial and natural waters can be consldered to be aggressive to portland cement concrete. t Acid rain, which has a pH of 4 to 4.5, can etch concrete surfaces.53

The chemical reactions that may lead to a decrease in concrete quality can be divided Into three subgroups as shawn ln Figure 2.38.7 The most important are:8

  • The reaction of acids. and salis of ammonium or magnesium, and soft water with hardened cement
  • The reaction between sulfates and aluminates ln the concrete
  • Alk~li-silica reactivity
  • Corrosion of embedded reinforcing steel


The Portland Cement Association (PCA) has published a report on the affects of the various substances on concrete along with a guide to protectlVe ireatmems.53 The affects of sorne of the more common chemlcals on the deterioration of concrete are summarized in Table 2.9 .



Water which cenntains chlondes, sulfates, and bicarbonates of calcium ~nd magnesium is generally not aggressive to concrete. On the other hand, water from condensation of fog or water vaper, and water from rain or melting snow and ice, does not usually contaln calcium ions and tends to dissolve the calcium-containing proclucts in portland cement paste. When the water is stagnant, the solution in contact wrth the concrete achieves chemical equilibrium and ceases to dissolve the cement paste. However, when the contact solution is continuously diluted by flowing water, the hydrolysis of the cement paste will continue. The reaction will continue to occur untll ail or most of the calcium hydroxlde has been leached away, leaving behind weak sllica and alumina gels. The leachlng of calcium hydroxlde from concrete Interacts wlth carbon dloxlde present in the air and produces efflorescence (white crusts of calcium carbonate). 1



Portland cement IS generally not resistant to aCld attack. Concrete deterloration by aCld attack IS caused by the reactlon between the aCld solution and the calcium hydroxide ln the portland cement paste. The chemlcal reaction produces water-soluble calcium salts whlch are removed by the eroslve action of flowlng water (Figure 2.39). In sorne cases, the resulting calcium salts are Insoluble and are not easlly removed from the concrete. 15

The rate at which a reactlon occurs wlth the concrete depends more on the solubility of the resulting calcium salt than on the aggressiveness of the aCld. The rate of deterioration will also be much hlgher ln flowlng water conditions. It should also be realized that, with acid anack, the hardened cement paste is completely converted to a soluble salt, thereby destroying the entlre pore structure. Therefore, unhke other types of attack, the permeabllity of the originally sound concrete 15 of minor importance.6 The three types of deletenous cation-exchange reactions that can occur between aClds and concrete are summarlzed below.



Solutions containlng hydrochlorlc, sulfunc, or nitnc acid are often found in industrial practlce. The reactlons that occur between these acids and the portland cement paste produce soluble calcium salts, such as calcium chlonde, calcium acetate, and calcium bicarbonate, which are subsequently removed by leaching. Also, reactions of solutions of ammonium chloride and ammonium sulfate wlth concrete produce highly soluble products as shown by the reaction below: 1

Carbonic acid is also very aggresslve to concrete. The aggressiveness of the reaction between carbonic aCld and calcium hydroxide [Ca(OH)21 present in hydrated portland cement paste is dependent on the amount of free dissolved CO2 in the a,!~cking solution. The reactlon can be shown as follows: 1,36

If there is a sufficient amount of free CO2 present in the solution, the Insoluble CaCOJ IS transformed into soluble calcium bicarbonate as shown by the second reaction Any amount of free CO2 over and above that whlch is needed for equihbnum would cause the second reactlon to move to the right, thus acceleratlng the hydrolysis of calcium hydroxlde. When the pH of groundwater or seawater IS above B, the amount of CO2 present IS neghglble. However, If the pH is below 7, it may contain slgnificant amounts of CO2•



Certain waters wlilch contaln agressive anions may react wlth cement paste to form Insoluble calcium salts, whlch may N may not cause damage to the concrPte dependlng on whether the reactlon product 15 8Ither expansive or IS removed by eroslon. Chemicals belonglng to thls category include oxahc, tartanc, tannlc, humlc, hydrofluoric, and phosphoric acid.



Seawater or groundwater often contain solutions of magnesium bicarbonate and wh en these waters come ln contact with concrete, they react with the calcilJm hydroxide ln portland cement paste to form soluble calcium salts. Magnesium ion attack is consldered to be the most aggresslve because It eventually extends to the calcium silicate hydrate (C-S-H) in the cement. Prolonged contact wlth magneslum ions leads to the formation of a weak magnesium silicate hydrate.



Concrete wlth a low water-cement ratio and a well-graded aggregate may provide adequate reslstance to mild aCid solutions. In situations where the acidic solution is stagnant, a ·sacnficlal” calcareous aggregate may be beneflclal. The acid may sometlmes be neutralized by replacing the siliceou5 aggregate wlth limestone or dolomite havlng a minimum calcium oxide concentration of 50 percent. In thls case, the aCld attack will be more unlformly distributed, reducing the rate of attack on the cement paste and minimizing loss of aggregate partlcles.53

No concrete, regardless of ItS quality, will resist long exposure to high acid concentration. In such cases, It may be possible to apply an adequate protective surface coating to the concrete.15 ACI Commmee 51554 and the Portland Cement Association53 provide recommendations for barner coatings to protect concrete from various chemicals.



Five types of reactions that involve the formation of expansive products have been identified as being deleterious ta portland cement concrete: sulfate attack, alkali-silica attack, alkali-carbonate attack, delayed hydration of free CaO and MgO, and corrosion of steel in concrete. 1 These reactions can cause closure of expansion JOints, deformatlon and displacements ln vanous parts of the structure, cracking, spalling and pop-outs. A discussion of each reaction follows.



This form of atté!rl< usually occurs when concrete IS exposed ta solutions contaming sulfates of sodium, potassium, calcium, or magneslum. Ammonium sulfate, whlch 15 often found ln agncultural sail and waters, is al 50 aggresslve to concrete. Seawater, whlch has a high sulfate concentration, can be aggresslve to marine structures. Deterioration of concrete as a result of sulfate attack is known to manifest itself ln two distinct forms: expansion, and progres~lve loss of strength and mass (Figure 2.40).’


It 15 believed that there are two chemlcal reactlons that occur ln sulfate attack on concrete.’!! The first involves the combination of sulfates of sodium, calcium, or magneslum wlth free calcium hydroxide to form calcium sulfate (or gypsum), as shown by the following reactlons:


As seen by the second reaction, magnesium sulfe,te creates the most severe attack on concrete. ln addition to the formation of gypsum, the reaction produces a poorly al kali ne magnesium hydro)Clde, which creates an unstable envlronment around the calcium-silicate-hydrate (C-S-H) binder. In such cases. calcium silicates release calcium hydroxide, converting part of the C-S-H binder into a coheslonless granular mass, ln addition to the expansive cracking.47

The second reaction involves the combination of gypsum and hydrated calcium aluminate to form calcium sulfoaluminate (ettringlte). This IS represented by the following equations:1 • 15

It is belie\led that the formation of ettringlte is responsible for the expansion. However, the mechanism by whlch expansion occurs is not fully understood. Two principal theories which have coexisted for a long time are: exertion of pressure by growth of ettringite crystals, and swelling of ·colloidal” ettringlte due to adsorption of water. A new theory C’oncerning sulfate expansion in concrete developed by Ping and Beaudoin!!!! attempts to explain the process based on the principles of chemical-thermodynamlcs. Il suggests that sulfate expansion is caused by the conversion of chemical energy into mechanical work to overcome the cohesion of the system. The expansive force comes from crystallization pressures which occur as a result of the interaction between the ettringite and the cement paste .



  • The main factors influencing expansion are:6
  • Amount of aggressive substance present
  • Permeabilrty of concrete
  • Cement type (C3A content)
  • Amount of moisture avarlable


reasonable degree of protection against sulfate anack can be provided by uSlng a dense, low permeability concrete with a low water-cement ratio and a high cement content. Proper consolidation and curing of fresh concrete with adequate cover thicknes3 produce a hlgh quality concrete with low permeability.’ Additional safety against sulfate attack can be provided by uSing sulfate-resistlng cements. The ACI Commlttee 201.2R report 15 and the ACI BUIlding Code 318-92!141 provide recommendations for the type of cement and water-cement ratio for normal welght concrete for vanous degrees of sulfate exposure. These are classdied into four categories and are shown in Table 2.10.


ln general, ASTM Type V (CSA Type 50) portland cement which contains less than 5 percent C:sA provides adeqûate protection agamst mild sulfate attack. European standards limlt the C:sA content of cement to 3 percent for high sulfate resistance.8 However, ln severe sulfate exposure conditions concrete containing hlgh alumina cements, portland blast-furnace slag cements (with more than 70 percent slag) , and portland pozzolan cements wlth at least 25 percent pozzolan (natural pazzolan, calclned clay, or low-calcium fly ash) have provlded a higher resistance to sulfate attack.1 Pozzolans combine wrth the free lime resulting from the hydration of the cement, thereby reducing the amount of gypsum formed. 57 The best results have been obtalned when the pozzolan is a Class F fly ash meeting the reqUirements of ASTM C 618.58



Alkali-silica reactlon (ASR) IS a chemlcal reaction that can occur between aggregates containing certain forms of sUica and sufflclent d:kalies (sodium and potassium) in the cement paste. The phenomenon, whlch was flrst descnbed by Stanton59 in 1940, is reported by the Strategic Highw~y Research Progrsm (SHRP) as being one of the major causes of concrete deterioratlon in the United States. Published hterature Indicates that ASR IS also widespread in other parts of the world such as eastern Canada, Austraha, New Zealand, South Africa, Denmark, Germany, England, Iceland,l India and Turkey.15 ASR often occurs in marine structures, such as dams, bndge piers, and sea walls.1

ASR Involves the breakdown of silica structure of the aggregate by hydroxyl ions to form an al kalisUica gel which can swell by absorbing a large amount of watef through osmosis. The hydraulic pressure which develops can cause expansion and cracking of the concrete, leading to a loss of strength, elasticity, and durabllity.l ASR will typically produce an irregular crack pattern commonly referred to as map or pattern cracking (Figure 2.41). Such cracking may also be accompamed by displacements or misalignments of structural members.80 An excellent handbook for Identifylng ASR in the field has been produced by Stark (SHRP).81

Most of the alkalies in concrete come from portland cement, although they are also found in seawater, ground-water, and deicing salts. Reactive silica is found in a variety of minerai forms in aggregates. Their rate of reactivity depends on their morphology, whether amorphous or crystalline.80 A comprehensive hst of deletenously reactive rocks, minerais, and synthetic substances responsible for concrete deterioration by ASR has been developed by ACI Committee 201 and is shown in Table 2.11. Several of these rocks, including granite gneisses, metamorphosed subgraywacks, and some quartz and quartzite gravels may react slowly.15 The quantity and reactivlty of the reagents, avallable mOisture, and temperature are ail contributing factors to the rate of ASR deterioration. Severe structural damage can occur in as little as three years, although in sorne cases, ASR can take more than 30 years to create any visible damage,39



Test proced~r9s for identitying potentially reactive aggregates and specification procedures for preventing or minlmizing its effects have been in place for many years, ‘rhe presence and amounts of reactlve constituents in the aggregate can be determlned by petrograpl-tic examlnation. Table 2.12, which is adapted from Reference 60, shows the maximum amounts of reactive constituents that can be tolerated in aggregates, and can be compaied wJth petrographic test resu!ts. Ta ensure long-term serviceability of concrete exposed ta conditions which promote ASR, Ozol and Dusenberry83 suggest that quartzose aggregates are ta be consldered as potentlally reactive. In addition, there are certain test methods which can be used ta measure the reactivity or the potential ta cause expansion. The most commonly used tests include ASTM C 289 and ASTM C 227.60 The results of these tests are used ta verity the findings of the petrographie examinatlon with regard ta the reactivtty of an aggregate .

However, Lane60 reports that these test procedures may not be effective in detecting ail potentially reactive aggregates. To compensate for this shortfall, several new test methods have been developed which are suitable for detecting the reactivity of aggregates which contain microcrystalline or stralned quartz. ThiS method IS being evaluated by ASTM as P 214 ·Proposed Test Method for Accelerated Detection of Potentially Deleterious Expansion of Mortar Bars Due to Alkali-Silica Reaction” .

Since alkahes ln concrete are mostly found ln the portland cement, the tradltlOnal approach to mlnimlzing ASR has been ta use low-alkall cements. The ASTM C 150 specifications recommend using cement with an alkall content not exceedlng 0.60 percent Na,p equlValent (N~O + 0.658 ~O). However, research by Ozal and Dusenberryb3 suggests that a lower value may be more advisable. Tuthl1l64 suggests that a ma><lmum limlt of 0.40 percent should be used to prevent ASA. Mehta 1 reports that investigatIOns ln Germany and England have shown that a total alkall content below 3 kg/m3 , is not hkely to cause any damage.

However, slnce it IS difflcult for producers to control the amount of alkall content ln cement, an alternative measure is to use blended cements made by adding pozzolans or ground granulated blast-furnace slag.60 Glass C and Class F fly ashes and sllica fume are the most wldely used pozzolans. Sihca fume IS reported to be the pozzolan whlch provldes the best protectIOn agalnst expansion resulting from ASA. Kosmatka and Fiorat062 report that Class F fly ashes (Iow hme) are more effective when provided ln amounts of 15 to 20 percent of the total cementltlous matenal. If Class C tly ashes (high lime) are used, a replacement of 35 to 40 percent would provlde the same degree of protection. Hogan and Meusel50 suggest that cements contalnlng 40 percent or more slag can conslderably reduce expansion. The standard test method for determlning the mix proportions ta achieve the deslred performance 15 ASTM C 441 .



Cenaln IImestone aggregates, have been reported to be reactive in concrete structures in Canada and the Unrted States. HI The mechanlsm of attack, Involves “dedolomitizatlon” of the carbonate rocks which leads to the formation of bruclte and the regeneration of alkali. This effect is opposrte to what occurs with ASR ln which the alkalies are consumed as the reaction occurs. However, concrete deterioratlon by ACR is also characterized by expansion and map cracks and IS more severe ln areas where there is a continuous supply of moisture. Another difference between the two reactions IS that ACR does not exude silica gel.

ln addition to expansion, a phenomenon assoclated with ACR is the formation of rims around the aggregate partlcles and extensive carbonation of the surrounding paste. This phenomenon IS not ‘ully understood, however, ACI Committee 201 15 reports that It occurs as a result of “a change in disposition of sllIca and carbonate between the aggregate partlcle and the surrounding cement paste”. The ri ms seem to propagate concentrically toward the center of the aggregate with time. Damage due to ACR usually occurs ln less than three years.39



Hydration of free crystalline MgO or CaO, can also cause expansion and cracking in concrete if they are present in sufficlent amounts. Current ASTM C 150 restrictions require that the MgO content in cement should not exceed 6 percent. However, laboratory tests showed that concrete ma,1e wlth a 10w-MgO portland cement contalning sufficient amounts of CaO can also produce expansion. ‘ Current manufactunng practices ensure that the content of free crystalline CaO in ponland cement does not exceed one percent. 7




Bacteriological growth such as lichen, moss, algae, and roots of plants and trees penetrating into the concrete at cracks, can cause mechanical deterioration to concrete structures. Microgrowth may also cause chemical anack by producing humic acid which is aggressive to the cement paste.

However, the most common type of biologieal attaek on concrete is found ln the sewer systems . Sulfur-oxidizing bacteria, using hydrogen sulfide (H2S) derived from the sewage, produces high concentrations of sulfuric acid, thus resulting ln acid and sulfate attack in the concrete (Figure 2.42). Rigdon and Beardsley68 noted that this problem commonly occurs in warm chmates such as Califorma (U.S.A.), Austraha, and South Africa. pomeroy67 also observed that thls phenomenon occurs at the end of long pumped sewage force mains in the northern (colder) climates.


The mechanism by which sulfuric acid is formed Involves two distinct processes whlch will nct occur unless certain conditions are met. Generally, a free water surface is required, and a low dissolved oxygen content in the sewage that permit the bulldup of anaerobic (oxygen·free) sulfurreducing bacteria. Some of these bacteria reduce the sulfates or proteins that are present in the sewage to form H2S. Sulfur-oxidizing bacteria (thiobacillus concratlVorus) reduce the H2S to form sulfunc aCid, which lowers the pH of concrate to 2 or less. The destructive affect of the sulfate ions on the calcium aluminates in the cement paste account for the daterioratlon of concrate, which usually occurs in the crown of the sewer.

Information which may enable the engineer to design, construct and operate a sewer sa that the formation of sulfuric acid is reduced is provided in References 67,68 and 69.



Marine organisms are commonly found on the surface of underwater concrete structures. This marine growth (or fouling) can have significant adverse effects on the integrity of the structure. Firstly, It increases the surface area of the profile that is exposed to current flow, thereby increasing wave and current forces on the structure. Secondly, marine growth can also cause deterioration of some concrete structures due to galvanic amion between the organism and the concrete.70 ln tropical or semi-tropical waters, several types of marine rock borers can penetrate into the concrete, although this damage usually occurs in low-quality concrete.



Marine growth can be categorized into two basic types: soft fouling and hard fouling.72 Soft fouling is caused by organisms which have the same density as seawater. This type of marine growth creates bulk, but is usually easy ta remove. Hard fouling is caused by marine organisms that are denser th an water and are more firmly attached to the concrete surface. These organisms are usually more difficult to remove. The following sections have been adapted from a revlew of Reference 72, and are provided for reader information.



Those organisms which cause soft marine growth are as follows:

(a) Algse This is usually the first type of marine growth to appear on the surface of an underwater structure and is usually referred to as slime. Being sensitive to light, algae is not usually found below 20 m (66 ft.) of water depth.

(b) Bacre”s. This will also be one of the first organisms to develop on éI structure and will grow in water depths in excess of 1000 m (3280 ft.).

(c) Sponges. These are often found on the surfaces of deep offshore structures at depths greater th an 1000 m (3280 ft.). See Figure 2.43 .

(d) Ses squins. These organisms have a soft body and are sometimes found in large colom es. See Figure 2.44.

(e) Hydroids. These are related ta the sea anemone and also grow ln colonies. They resemble seaweed, and can produce dense colomes to depths of 1000 m (3~80 ft.). See Figure 2.45.

(f) Seaweeds. There are several varieties of seaweed that grow on underwater concrete structures. The longest of the se IS kelp, which can produce fronds up ta 6 m (20 ft.) in length, if the conditions are suitable. See Figure 2.46 .

(g) Bryozoa. This marine growth looks like moss and grows very taU. Bryozoa is actually an animal with tentacles, as shown in Figure 2.47 •

(h) Anemones. The anemone, sometlmes referred to as anthozoan, has a cyllndrical shaped body which is surrounded by tentacles. It attaches itself firmly to the concrete surface by a discshaped base and is difficult to remove wrthout tearing its body. The speCies is found in many shapes and colors.

(1) Dead men’s fmgers (Alcyomum digltalum). These colonies which can grow up to 150 mm (6 in.) in length can be found on pier piles and rocks on waterfront and offshore structures. When theyare below water, several small polyps grow out from the finger-shaped body, with each polyp having eight feathery tentacles (Figure 2.48). When submerged, its color is white to yellow or pink to orange. When it IS out of the water, it is flesh çolored and resembles the human hand.



This group of marine growth is comprised of calcareous or shelled organisms and includes the following:

(a) Barnac/es This l’Orm of manne growth is the one most commonly found attached to waterfront structures. The common species is called Ba/anus ba/anoides (Figure 2.49). These organisms grow ln dense colonies to a depth of 15 ta 20 m (49 to 66 ft.), but can also grow at depths of 120 m (394 ft.).

(b) Musse/s. A mussel is a hard-shelled mollusc which attaches itself firmly to the structure by very strong threads located at the hinge of the shell. The main species, Mytilus edulis, is usually found in dense colonies to depths of between 20 m (66 ft.) and 50 m (164 ft.)



The type of orgamsm which develops, and its growth rate, will depend on several factors including water depth, temperature, current, salinity, and food supply. In general, the formation of slime or algae on unprotected concrete surfaces occurs in two to three weeks. In some cases, marine growth has been know to develop within 24 ta 36 hours after a surface has been cleaned. On the other hand, bamacles and soft fouling can develop ln three to six months. Mussel colonies generally take two seasons to develop, and can also grow on top of existing dead fouling. The various factors affecting growth rate are summarized below.

(a) Depth. In general, marine growth density decreases wlth Increasing water depth, since an increase in depth reduces light Intenslty. This reduction in hght intensity reduces the ability of certain organisms, such as algae, to photosynthesize. A generally accepted schematic representation of manne growth denslty wlth varying water depth (in British waters) IS shown in Figure 2.50. From the diagram it is evident that the highest density of marine growth occurs near the water surface, which is where wave loads are the hlghest.

(b) Temperature. An increase in water temperature will typically increase the growth rate of marine organisms. For instance, an increase in temperature of 10″C (1S°F) will approximately double the growth rate. However, most organisms stop growing at 30 to 3SoC (86 to 9S0F). Since the greatest variation of temperature occurs near the water surface, marine fouling colonies near the surface will undergo seasonal growth.

(c) Water current. Water flow velocity is an Important factor influencing the type of colony that develops. It appears that many Jarvae cannot attach themselves to con crete surfaces if the water \lelocity exceeds 0.5 m/sec (1.6 ft/sec). However, once attached, it can withstand water velocities (If more than 3m/sec (10ft/sec). At higher velocities, fouling which is not firmly attached is easily removed. During slack flow periods, larvae can attach themselves to structures in cracks, where water flow is either slow or stagnant. Once the organism develops, strong water currents will provide more nutrrents which will accelerate growth.

(d) Salimty. In ‘resh water, the only organism whch can grow is marine algae slime. The amount and type of marine growth increases wlth increasing salinity. For example, the size of mussels will be five tlmes greater as the salinity of water increases from 0.6 percent to the normal salinity of seawater (3 to 3.5 percent). As the salinity increases, hydroids will grow first followed by mussels.

(e) Food supply. Marine growth depends on the amount of nutrient available. Growth rates in coastal areas (shallow water) are higher than those offshore (deep water). Research has shown that marine organisms found in water which circulates around offshore structures have a higher growth rate.



A marine structure may be subjected to several forms of collisions or impact. Examples of these Include ship collisions, wave action, and dropped objects. Depending upon the type and behavior of the Impactor, the response of the concrete structure may either be global such that little or no local damage will occur, or if the impactor hardness and/or velocity of impact is high enough, local damage will occur.73 This section focuses on the latter, and is adapted from a review of Reference 73.

As a consequence of several serious accidents involving concrete offshore structures in the North 98 Sea (bath tram dropped abjects and ship collisions), the Department of Energy ln the UK sponsored several studies ta investigate the problem further. A report by Wimpey74 identified the various types of abjects that are hkely ta be dropped on concrete offshore structures. A subsequent report by Brown and perry75 Identlfied the vanous forms (Jf local damage which can occur from fallen abJects and developed Simple design formulae for assesslng the damage. These are discussed below.



The report by Wimpey74 concluded that the abjects most hkely ta cause severe damage are the end-on impact of slender abJects and impact of bulkV abjects. Brown and perry75 reco9mze five forms of damage (Figure 2.51) and have adopted a standardized descnption for each as deflned below:

  • Penetration (the depth ta whlch an abject penetrates the concrete).
  • Spalling (the cratenng damage on the impacted surfé’ce).
  • Scabbing (the fracturing and expulSion of concrete from the opposite face of the Impact).
  • Perforation (the abject passes completely through the concrete).
  • Shear plug (formed by inclined cracking through the thlckness of the concrete).


The type of damage which will most likely occur depends on the ratio of concrete thickness to the object diameter (t/d). For high t/d values (of the arder of 6), penetration and spalling are hkely ta occur. For t/d values approaching 1, scabblng and shear plug formation are more hkely to occur. For very low values of t/d, a global response may cause the major damage, although scabbing can also occur. Factors Influencing the response of concrete slabs to impact are discussed in Reference 76.



The report by Brown and perry7S concludes that present impact damage assessment methods would provide reliable results, but that the basis for established scabbing formulae was inadequate. As a result, new empirical formulae for scabbing were developed. For soUd abjects, a non-dimensional number Np is given by the following formula:


M = mass of missile (or impactor)

V = velocity of missile

do = diameter of missile

t = slab thickness

E = modulus of elasticity

fT = concrete shear strength

If NI is greater than some critical value, the nominal shear stress will be greater than the nominal shear strength of the concrete and inclined cracking, shear plug formation, and scabbing are likely to accur. The same procedl.lre is used for pipe-shaped objects .



The service life of any structure, such as a dam, navigation lock, bridge, wharf, or other marine structure, depends on the preservation of the physlcal condition of both the portion of the structure above and below the waterline. Therefore, it is Important to develop and implement an adequate inspection, maintenance, and repalr program for the ent”e structure. In many cases, however, underwater inspections are seldom performed because the evaluation of the condition of a concrete structure under water IS usually more difficult and expensive than evaluating a structure located above the waterline. 1 However, aging structures reaching or exceeding their design life, and the growing concern with concrete durability both contribute to making underwater inspection an essential part of today’s infrastructure evaluation and preservation technology.

Various groups, such as Transportation agencies, and Port Authorities in Canada, the United States, and other parts of the world require penodic underwater inspections as part of a preventive maintenance program. They are also undertaken as a requirement prior to the purchase of a facility by a new owner, ta evaluate the strength of the structure for new loading conditions, to gather information needed for planning the expansion or modification of a facility or as an initial construction inspection to confirm that a structure has been constructed in accordance wlth contract documents. Catastrophic events, such as ship collisions, earthquakes, hurricanes, and floods also require underwater inspections for damage assessment. 2 ln addition, deteriorated structures that might be dangerous to public safety, or that can cause substantial property damage, need to be continually inspected to determine ItS capaclty to operate safely. 3

Underwater inspections can be performed by divers, remotely operated vehlcles (AOVs), or manned submersibles. The most common method employed is the use of commercial/engineer divers. They are readily available at almost ail waterfront locations and can be mobilized relatively quickly. ROVs are becoming used more frequently for visual inspections and have also been used for making repairs. The basic ROVs are usually equipped with cameras, videos, and lights, and are remotely controlled from the surface. ROVs can be very economical in deep-water or long penetration dIVes. Manned submersibles are used primarily for performing very deep dive inspections of structures such as offshore oil platforms and pipelines.4 These types of vehicles are rarely used for waterfront inspections. A more detailed discussion of each of the above diving methods IS provlded later in this chapter.

Evaluation of a structure must take into account several factors, including design considerations, existing operating, inspection, and maintenance records, condition surveys, in-situ testing, instrumentation, and determination of the phenomena causing the deterioration.3 This section provides a summary of techniques and equipment currently used by divers ta visually inspect concrete ln eXlsting underwater structures or underwater portions of concrete structures. The planning and preparation required for an underwater inspection program, and methods of documenting and presenting the results are included along with a description of the recommended underwater Inspection procedurrs. The material in this chapter has been adapted from a review of different avallable references, especially 3, 5 and 18.



The objective of an inspection is to obtain the necessary information to assess the structural condition of the struc,ture to determine whether it meets current design and future performance criteria. However, the primary reason for conducting an inspection is the structural safety of the structure. Although the nature of the inspection will determine the extent of information to be provided, the general objective of a condition survey should involve the following:3 • 5

  • Identifying and describing ail major damage and deterioration
  • Identifying the phenomana or materials causing the detenoration
  • Determining the extent and rate of deterioration
  • Determlning the structure’s performance characteristics under future service conditions
  • Documenting the types and extent of marine growth, water depths. water visibility, tidal range. and water currents which will help plan future inspections
  • Determining conformance with contract documents and verifying as-built conditions
  • Identifying any potential problems which may occur wlth mobilization of equipment, personnel, and materials needed to make repairs
  • Making recommendations for suitable methods of repair and maintenance
  • Obtaining and developing data needed for making cost estimates of these repairs and maintenance
  • Recommending the types and frequencies of future inspections



Underwater inspection of marine structures can be grouped into three basic types or levels. They are differentiated by the amount of preparation work required and the means by which the work is to be performed as descnbed below. The level of Inspection to be used for a particular inspection must be chosen early in the planning Table 3.1, whlch was developed by the U.S. Navy, summarizes the general purpose of each Inspection and the type of damage that each level will deteet.



This type of inspection IS the most rapid of ail three because Il does not requlre cleaning of the element being inspected. The various purposes of a Level 1 inspection include: to confirm asbuilt conditions; provide initial information for developing an inspection program; and detect obvious damage or detenoration caused by overstress, impact, corrosion, or biological attack.



This type of Inspection requires cleaning of the concrete surlace either before or du ring the inspection. This level is needed for detecting and identifying surlace damage which may be hidden by manne growth. A limited amount of information can be obtained to enable a preliminary assessment of the load carrying capacity of the structure or element of the structure. Since cleamng is time consuming, it is usually done to critical areas of the structure. The amount and thoroughness of cleaning is dictated by the amount of information needed to make a general assessment of the structure.



A highly detailed inspection IS primanly conducted to detect damage which is hidden or damage whlch is about to occur, and to determine material homogeneity. This type of inspection will usually requlre prior cleaning. This level often involves the use of Nondestructive Testing (NOT) and sometimes Destructive Testing (DT) techniques. The NOT techniques are usually perlormed at cntlcal structural areas which may be suspect or are represemative of the underwater portion of the structure. DestructIVe or partially destructive testing, such as coring or physical material sampling, IS usually perlormed to obtain specimens for laboratory testing. Generally, the equipment and test procedures will be more sophisticated, and should be perlormed by qualified engineer divers or testing personnel.



lhe time, equipment, and effort required to perlorm the three different levels of inspection are considerably different. The time required for each level depends on environmemal factors such as visibility, water depth and water currents, water temperature, wave action, amoum of marine growth, and the skill and expenence of the inspector/diver.’ A guide was developed by the U.S . Navy for estimating the time required to conduct Levell and Levelll inspections and is provided in Table 3.2.

The information in the Table is based on a water depth of 9 ta 12 m (30 ta 40 ft.); visibllity of 1.2 to 1.8 m (4 to 6 ft.); warm, calm water; moderate manne growth about 50 mm (2 in.) thick; and an expenenced engineer diver supervised by an engineer at the surface. Inspection times for Level Il assume that approximately 1 m (3 ft.) of the element in the splash zone, 0.3 m (1 ft.) at mid-depth, and 0.3 m at the bottom, will be cleaned of marine growth using the most efficient method.5 Since Level III inspections depend on the extent of existing damage and can vary significantly for structure to structure, estimates of time are not provlded in the Table.



Various methods are presently ln use or currently under development for the underwater inspection of concrete structures. 1 n the last 15 years, significant technologieal advances have been made by the offshore ail industry. Various individual skills, equipment, and techniques are used in underwater inspections and r,an be grouped into three basic categones: manned diving missions; remotely operated vehicles (ROVs); and manned submersibles2 • These are described below .



Manned diving is the most common method used for performing underwater inspections. The breathing gas is provided either by self-contained underwater breathing apparatus (SCUBA) or by an umbilical hose which extends from the surface (surface-supplled/tended air or mixed-gas)’ . These are the two methods of divlng operations most suited for underwater inspection of relatively shallow marine substructures. The main advantage of using divers is that it is a versatile system whlch in most cases relatively inexpensive. On the other hand, this diving mode is limited by time and dive depth. In addition, slnce the diver’s sense of perception is quite different under water than ln air, his observations will be more susceptible to error’. Appendix C provides some general characteristics of these types of diving modes.



ln SCUBA diving operatioils, the dlver’s breathing tanks are typically mounted on his back. The SCUBA diver has the highest degree of movement than in ail the other types of diving methods because he is not connected to the surface or an umbilical cable. This diving method is highly mobile and is especially suited for performing short duration dives’.

The disadvantages of SCUBA divlng are: depth limitation, Iimited air supply, and difficulty in communication with topside personnel. In SCUBA diving the maximum sustained depth at which a diver can work is about 18 m (60 ft.). An experienced diver can dive to depths of 37 m (120 ft.) for short periods of time without experiencing any difficulty. Air requirements will be different for each diver. Normally, a 2 m3 (72 ft~ tank is sufficient to allow a SCUBA diver to work at a depth of 9 m (30 ft.) for approximately one hour. As a rule, the amount of time which a diver can remain submerged will decrease wrth increasing water depth or level of exertion’.



ln this method of diving, the diver breaths the air or mlxed-gas through an umbilical hose supplied from the ~urface. The breathing medium is forced through the hose by a surface mounted compressor. The diver is also attached to a communication cable, a lifeline, and a pneumofathometer. The diver can use either a hard hat with a dry suit or a face seating mask with a wet, dry, or hot-water suit’ .

The main disadvantage of surface-supphed/tended air diving is the signlfieant deerease ln diver productivity. The diver is considerably less mobile than the SCUBA dlver due to the extra welght and diving gear he must carry. Another dlsadvantage is thls dlving method needs a considerable amount of additional divlng equipment. However, the main advantage wlth this divlng mode IS that the diver IS ln continuous contact wlth the surface personnel. 1



Lamberton et al.1 have made a companson between SCUBA diving and surface-supphed/tended air diving and is summarized below. A typlcal SCUBA diving mission will require the following equipment:

  • Van or truck to transport the gear
  • Boat, motor and trailer
  • Anchors, mooring line, out board ladders, and life jackets
  • SCUBA tanks, wet SUit, fins, weights, masks, regulator, etc.
  • Dive flag
  • Chlpping hammers, picks, pry bars, problng rads, and scrapping tools;
  • Underwater lights
  • Writing boards, drafting eqUipment, and underwater slates for recording data;
  • Underwater still or video camera
  • Aulers, tapes, cali pers, or other measuring devices”


For a simllar inspection performed by surface-supplied/tended air diving, the following additional eqUipment is reqUired:

  • “Larger vehicle to transport the gear
  • Larger boat to support diving operation
  • Diving compressor and recelver tank
  • Diver umbilical with air hase, communication cable, lifeline, and pneumofathometer;
  • Surface-supply head gear
  • Diver’s radio’



ln both diving methods, the diver is subjected ta the hazards and conditions of the underwater environmem which directly affect his performance and safety. In arder ta avoid serious accidents or ln jury, the diver must have a clear understanding of these factors and must be able to recognlze and handle them2 •

The various hazards and accidents to whlch a diver may be subjected to are listed below’. A detailed explanation of the causes, effects, and treatments of these hazards and accidents IS provided in Reference 7.

  • Decompression sickness or nitrogen narcosis (the bends)
  • Oxygen poisonlng
  • Bleeding
  • Overexertion and exhaustion
  • Hypothermia
  • Squeeze
  • Gas expansion
  • Blowup
  • Loss of communication
  • Fouling
  • Polluted water
  • Noxious air
  • Tides and currents
  • Marine traffic
  • Marine life
  • Floating debris



A number of organizations have developed dive tables to help divers coordinate dive depth with duration so they can minlmize the possibility of their developing decompression sickness. The tables contain time limits for a dive to a given depth. For instance, the U.S. Navy Standard Diving Tables indicate that a healthy 22 year old Navy diver can stay at: a 9 m (30 ft.) depth of seawater for an unlimited time; approximately one hour at 18 m (60 ft.); and 30 minutes at 27 m (90 ft.).2

ln deeper dives, these limits are sigmficantly shorter. In these cases, inspections are usually performed by more th an one diver in succession, or by uSlng an on-site recompreSSlon cham ber (saturation and nonsaturatlon diving). The recompression chamber IS an air chamber whlch gradually adjusts the diver’s body to the pressure at which he will be worklng. Dependlng on the depth of the dive, the adJustment penod can vary from a few hours ta several days. For dives exceeding 40 m (130 ft.), a specially formulated mixed breathlng gas 15 usually used ta aVOId the potentially hazardous effects related to mtragen absorption by the bOdy.2

The dlvlng industry has recently developed a ngld-shell “one atmosphere SUlt” flHed wlth a set of pincers for hand actuators. The suit permlts the diver to work ln an enwonment of one atmosphere (ambient air pressure), which eliminates the posslbllity of the diver develaplng decompression sickness. Hawever, the SUit is expensive, dlfficult ta work wlth, and reqUires special surface support.2



Remotely operated vehlcles (ROVs) are being used more often for underwater inspectIOns. They are slmilar to robots and are used extenslvely for inspecting deep structures such as offshore pipelines, deep bridge foundations, and hydraulic structures. They are usually connected to a support vessel or to the surface by a flexible communication cable. The vehlcle IS maneuvered by ballastlng and propulsion equlpment, and 15 equlpped wlth video and still cameras mounted on the frame. Sorne ROVs are equlpped wlth mechamcal arms (mampulators) whlch can operate equipment or retneve physical samples. The vehlcle is remotely controlled from the surface wlthout the use of divers.2

Sorne ROVs, such as MANTIS owned by International Underwater Contractors (IUC), can be operated by a pilot under a one atmosphere condition. There are three modes of operation: as a surface controlled ROV system; with a pilot as an observer in partial control assisted by a surface operator; or with a pilot in full control.



ROVs can range from small, relatively inexpensive systems to highly capable but expensive systems. The particular ROV system used for a project depends on the nature and the depth of the underwater Inspection being conducted •

According to a report by the U.S. Army Corps of Engineers,3 there are towed vehicles, bottom crawlers, self-propelled vehicles and vehicles remotely controlled from the surface. There are six basic types of ROVs and these are briefly summarized below. The requirements of a speclfic diving operation will dictate the most efficient, cost effective, and safest system to be employed. Examples of some commercially avallable ROVs are provided in Appendix D.



This vehlcle operates in mldwater, IS equipped wrth closed-circuit television (CCTV) cameras, and can maneuver in three dimensions. Most of these vehicles recelVe their power from a suppon plattorm, but many are self-powered by batteries carried on board. These vehicles can operate inwater depths ranging from 30 m (100ft.) to 3050m (10,000 ft.). Vehicle dimenSions rangefrom “basketball size to that of a small automobile” and welghs approximately from 32 kg (70 lb) to 5455 kg (12,000 lb) in alr.3



These vehlcles are propelled and powered by a cable connected to a surface ship. Real-time or slow-scan CCTV and photographlc cameras are typically carried on board. Two types of towed vehicles are described below:3

(a) Midwater These types of vehicles operate in mldwater, but can also make contact with the bottom periodically. Maneuverability in the horizontal direction is controlled by the ship’s heading, and the vertical direction is controlled by a reeling cable. These vehicles are designed for long range, long duratiol’ dives and can operate in water depths of 6100 m (20,000 ft.).

ThE’ Remote Underwater Manipulator (RUM III) is an example of this type of vehicle which usually operates in a towed mode, but can also operate in a bonom-crawling mode to retrieve samples and perform detailed work. Another example is the Towed Unmanned Submersible (TUMS) which operates in a t~ed mode and can be used to perform detailed investigations as a tethered, fraesWlmmlng vehicle by using on board thrusters.

(b) Bonom – and structura/ly – reltant. These vehicles are towed in contact with the bottom of the sea (bottom-rehant) or are structurally supported by a pipeline (structurally reliant). They are unique in that they are designed for a specific purpose. They are very large structures and are usually used for cable or pipeline burial.



This type of ROV is powered and controlled by the surface suppon shlp and IS equipped with CCTV cameras to monitor the work ln progress. The vehicle is propelled by wheels or tracks whlch are ln contact wlth the seabed. These vehicles are speclflcally built to perform tasks such as pipeline/cable trenching, bottom excavation and backfilhng, maintenance, inSpection, soli investigation, or nodule collection.3

A good example of su ch a vehlcle is that developed by the Dutch for underwater Inspection work at the Eastern Scheldt Storm Surge Barrier. The ROV (PORTUNUS) is a 6 m (20 ft.) long by 4 m (13 ft.) wide by 2 m (6-112 ft.) high, tracked, bottom-crawling vehicle which can perform inspections ln turbid water. It can travel at velocrtles of up to 2.5 mis (8.2 ft/s) at depths from 15 to 45 m (49 to 148 ft.). The vehlcle carnes SIX television cameras, four high frequency transducers, and sonar.8 This vehlcle can be used for inspection of sttlling basins.



These vehicles are similarly powered and controlled from the surface shlp and are propelled by wheels, tracks, or push-pull rams which are connected to a structure. Sorne of these vehicles are capable of operating in midwater and can travel to and from the structure. Most vehicles are equlpped wlth CCTV, and are specifically designed for pipeline trenching, oil tank sounding, and ship’s hull cleaning and inspection.3



These are self-powered vehlcles whlch are not physically connected to the surface vessel. They can maneuver in three dimensions and operate wrthin a preprogrammed schedule. In sorne cases, their direction and altitude can be modlfled by commands given from the surface by an acoustic link. These vehicles can operate ln depths ranging from 30 m (100 ft.) to 6,100 m (20,000 ft.) and can dive for four to SIX hour durations.3 An example of this type of vehlcle has been developed by the Dutch to inspect the deep foundations of the precast pIers for the Eastern Scheldt Surge Bamer project. The body of the vehlcle (TRIGLA) is tubular in shape and measures 900 mm (36 in.) long by 42 mm (1..1fa in.) in dlamet3r. Il is self·propelled, free-floating, and is equipped with lights, cameras, and pressure sensors.8



Hybrid vehlcles are a combinatlon of ROVs that are remotely controlled from the surface or support vessel, and directly controlled by the diver or pilot. These vehicles are a recent addition to underwater work and can perform a variety of tasks, such as pipehne trenchlng, diver assis· tance, structure inspection and maintenance, pipeline anchoring, and cable burial. 3

Further detalls regarding the structural aspects of the vehicles, tools, sensors, personnel, supporting systems, applications, and operational and navigational considerations, etc., are provided in Reference 9.



The advantage of uSlng ROV systems IS that they can be used in environments which are considered to be unacceptable for diver safaty. They can also be used for very deep and long duratinn diving operations. In addition, these systems can continuously and repeatedly conduct inspections wlthout performance degradation or concern with diver decompression.3



According to Popovics and MacDonald,3 the disadvantages of using ROY systems are:

  • They are very expenslve to use
  • They are less flexible and less reliable than using divers
  • ROV systems usually require greater maintenance than diver systems
  • Video cameras on ROYs may provide dist()rted vlews of the extent of deterioration or damage



Manned submersibles or minisubmarines are used to perform deep underwater inspections of offshore oil platforms and salvage operations and have limited use with nearshore marine structures. They are simllar to ROYs except that they are larger, more cumbersome and require significant surface suppon equipment. 2 They are controlled by a pilot inslde the vehicle and can fallow a preprogrammed schedule via a highly sophlstlcated acoustlc transponder navigation system. A manned submersible IS typically equipped wlth vlewports for forward and downward viewing, CCTV cameras, video tape recorders, still cameras, underwater telephone for dlver communication, mechanical manlpulators, navigation sonar, and acoustlc tracklng systems. Examples of manned submersibles are provided ln Appendlx E.



Diving bell systems can be rapldly mobihzed to perform deep dlving services at low cast. The skid mounted “bounce dive” system consists of a double-Iock deck decompresslon chamber, transfer-under-pressure locks, and a closed-bottom dlver work bell. A typlcal closed-bottom dlvlng bell system IS iIIustrated ln Figures 3.1 and 3.2. A closed-bottom bell dlving system has several advantages over an open bell system, such as: It can bnng an englneer down as an observer ta get a first hand look at the work area, it provides more tlme for Inspection repairs or malerial retrievals, Il can surface rapldly without the need for making decompresslon stops, and is safer far launch and recavery maneuvers in rough waters. The work bell mates with the main deck decompression chamber from above using a standard transfer-under-pressure hydraulic matlng clamp.

The divlng bell can accommodate two divers and is equipped wlth an enclosed control console cabin, standard life support equipment, Internal/external depth gauges, temperature monitoring equipment, internal/external lightlng and several communication systems. The bell IS also equlpped wlth several auxihary emergency life support and communication systems. The deck decompresslon chamber consists of sleeping quarters, an envlronmental control unit, and medlcal lock. Typical examples of deep dlvlng or saturation systems are Included in Appendix F.



Support vessels are an important part of the inspection equipment and can greatly influence the quality of the inspection and repair work. Highly sophisticated projects often require sorne type of support vessaI. Typically used support vessels are described in Reference 8 and are summarized below.



This IS the basIc support vessel which provides a working platform. These vessels must be stablhzed during rough waters, especially during placement of tremie or pumped concrete (Chapter 5). The vessel is stabilized by several anchor lines, a good metacentric height, a wide area at the waterline, or buoyancy tanks .



These barges are used for shallow water work and can provlde several advantages for underwater repalr and inspection work: they provide stabihty dunng surface preparation, repalr, and inspection operations; they can be used for gUlding and controlling surface eleanmg eqUipment, and monitoring equipment; and are sUltable for placlng prefabncated conerete panels.



Large ships have been designed to support several diving systems. The ship carnes communication and navigation equipment and other gear which are needed to support diver activltles. A typical ship has liVing quarters which can accommodate a large crew for as long as 30 days at sea. The shlp is equlpped with a machine shop for repair and bUilding of eqUipment. The shlp can accomplish several types of missions Including Inspection and survey, search and recovery, diving support, oceanographic, geophysical, and bathymetnc surveys, medium and long-range research programs, and installation and/or retrievéa of sensor packages and systems. More details on a typical support vessel are provided ln Appendlx G.



The evaluation of a concrete structure under water is more complex than that of a structure above water and must be carned out by a team of experienced divers and engmeers. The reason for this IS because although engineers have the necessary technlcal background for making a reliable evaluatlon, they are not trained divers or ROV operators and vice versa. J

The diver performlng the underwater Inspection must be weil informed of the proper use and care of dlvlng equlpment, safety requirements, communication techniques, and dlvlng operations. There are agencies, such as the Amencan Association of State Highway and Transportation OffiCiais (AASHTO) and the Federal Hlghway Administration (FHWA) ln the United States, that require divers to be certified and well-trained in structural Inspection.’ Organizatlons that certify divers include:

  • Professional Association of Diving Instructors (PAOI)
  • National Association of Underwater Instructors (NAUI)
  • American Diving Contractors Association (ADCA)


Engineer divers are belng used more often to conduct underwater Inspection of marine structures. Many agencles in the United States, such as the U.S. Navy, require that inspections be conducted by professlonal engineering divers. Engineer divers are capable of evaluating the extent and effects of deterioratlon on the structure.4 A report by Buehring10 states that:

“The englneer tralned ta dive can become more familiar with the underwater environment and can also obtain pertinent information first hand … ‘ and goes on ta state, ‘”. The professional engineer who dives IS hkely to make more judgmental decisions than the commercial diver. Also because the engineer’s authonty exceeds that of the commercial diver, his observations and conclusions are more likely to be accepted at face value … ‘.

The majority of agencles use commercial divers supervised by professional engineers. The englneer possesses the necessary technical expertise and can provide guidance ta the diver or divers dUring the Inspection, permitting a faster, efficient, and more accurate inspection.3 The Inspection team should have a good background in civil or mechanical engineering, as weil as ln design, construction, maintenance, and operation of underwater concrete structures. 11



There are no tirm guidelines on the frequency of inspection of concrete structures. Public works are usually inspected regularly at short intervals.12 Some agencies in the United States conduct underwater inspections of bridge substructures every IWo years, while others schedule inspections every five years. Others inspect thelr bridges Infrequently or only after indications of underwater problems. In sorne cases, underwater inspections are performed immediately after each major storm where scour problems are anticipated.

The frequency of inspection Will depend on the expected rate of deterioration and damage to the facllity or structure. The U.S. Navy recommends that plies above the waterline, including the splash and tidal zones, should be inspected annually. The underwater portions should be inspected every six years starting trom the splash/tidal zone and proceeding downward. The level and frequency of Inspection should be adjusted according ta the eXient of the observed deterioration.5 The ACI Commlttee 35713 recommends that concrete offshore structures should be surveyed annually for damage or deterioratlon and the Inspection findlngs should be carefully revlewed every five years.

Frequent and weil organized inspections are an effective method of keeplng maintenance costs to a minimum. They are also useful for obtainlng base·llne data on a speciflc type of structure. The rate of detenoration can be momtored and decislons can be made when repalr becomes economical. 12 The International Prestresslng Federatlofl (FIP) GUide ta Good Practlce14 Includes a table whlch provldes gUidance on the Intervals that might be appropnale for ·routlne· and ·extended” inspections. A routine Inspection is visual and does not require special equipmenl or access. Extended inspection, on the other hand, IS a more detailed investigation whlch reqUires special access and remote vlewlng techniques. Intervals for three ·classes· of structure and for three environmental and loading conditions are Illustrated ln Table 3.3.

The intervals shown ln the table should be regarded as absolute maxima and ln most cases inspections should be performed more frequently. Inspection frequency is based on engineenng and econor~lIc judgement, and should be established ta suit the ~tructure with regards to use, siting, construction, and design.12



A report by Dr. G. Watson 15 on underwater inspection of offshore structures recognizes threEJ phases to any inspection process: definlng the requlrement implementing the inspection, and assessing the inspection results. He further states that:

‘There is a close interaction between the requirement phase and the implementation phase, in that the implementation cannot generally be deflned until the requirement detail is available. The a~sessment phase interacts with both the requirement and Implementation, by providing feedback via the results themselves, and the degree of attainment of the deflned reqUirement by the prescnbed implementation. •

There 18 no set procedure for conductlng an underwater inspection since each will be different and will requlre a process of data collection. However, it is Important to define how a structure will be assessed as thls can have a signlficant effect on visual inspection in the implementation phase. Although inspections vary widely in type, scope, objective, and complexity, most require the basic activlty shown in Table 3.4. The arder in which these activities are listed is not necessarily the sequence ln which these activities will take place. The activities of the inspection process outlined in the table are descnbed in more detail in the following sections.



The initial phase of any inspection program should be the collection and review of ail available information on the structure.4 The planning ana implememation of an inspection cannot be properly and efficiently done without consideration of information relatad to the design, construction, operation, and maintenance of the structure or facility.3 Review of the available information often provldes an indication to what caused or might be causing the defect. This information can help save considerable time in the field and result in a more accu rate inspection. Belf!5 has developed a list of project specifie documents, and their sources, often used in an inspection process. These are shown in Tables 3.5 and 3.6, respectlvely.

Geotechnlcal investigations can also provlde important information in the inspection and evaluatlon of marine structures, especlally if the structure has undergone any settlement or movement. If there is eVldence of settlement or movement, a geotechnlcal investigation should be performed to determine If subsurface conditions Will affect the structure in the future.4 A geotechnlcal investigation can also provlde an indication whether further settlement will occur. This may have a direct impact on the repair solution.



Once ail of the available Information on the structure has been obtained and reviewed, an inspection plan should be developed. Since surveys are bath expensive and time consuming, the Inspection must be carefully planned and implemented to obtain the greatest amount of Information ln the shortest time possible. The purpose of the inspection and the desired technical results Will determlne the specific Information that IS needed, the level of detail required, and the format of the final report. Site logistics will often decide the form of dlvlng mode to be used. For Instance, shallow structures are typically inspected by traditional diving methods while deeper structures may be inspected using ROVs or mlnlsubmarines.2



Environmental conditions that may hamper the efficiency of the inspection should be considered. These may include atmosphenc temperature range, water temperature range, tidal range, water depths, water vlsibility, and currents. Any condition which will directly affect the time required to perform an Inspection, such as the extent of marine growth, ice, or seasonal flooding should also be considered.5



The proper and adequate selection of areas to be inspected IS crucial to the effectiveness of the condition survey. A sufficient number of Inspection areas must be selected to provlde representative Information on the underwater portion of the structure. Thus, It IS Important to know the areas of the structure whlch will be subjected to maximum stress. It IS also very helpful to know the processes Involved wlth concrete deterioratlon. A useful flow ch art for developlng an effective Inspection plan for evaluatlng underwater con crete structures was developed by Brackett et al.28 and is shown in Figure 3.3. Another report by Brackett,16 sponsored by the U.S. Naval Civil Engineering Laboratory (NCEL), provides a good guide on sampling cntena for inspection of plie supported wharf structures.



Avallability of manpower and equipment are key factors that should be considered when planning an underwater Inspection. The sklll of the diver or ROV operator as an inspector is also an important factor that should be consldered. Most divers (or pilots of ROVs and minlsubmannes) are weil trained m the use of diving technlque& but do not have the techmcal background to conduct structural InvestigatIOns. Sorne inspections can be performed by usmg ralatlvely unskilled divers. However, If the inspection calls for “mterpretlve skllls· an engineer diver Will usually provlde the deslred results.2 The Inspection plan should also indicate the type of inspection eqUipment to be used for each Inspection task.3



As previously stated, periodic Inspections are essentlal to the implementation of an effective maintenance program. Underwater inspections should also be conducted during and at the end of the construction phase of the structure to provide baseline data for future inspections. Underwater inspections should be scheduled dunng favorable conditions such as periods during low water, low pollution levels, minimum Ice, no flooding, or good underwater ViSlbllity.3



The documentation and form of the report reqUired is an Important consideration when planning an underwater Inspection. For example, the documentation that is nended for a research project Will not be the same as that required for repair or damage assessment. In addition, the level of detall needed for a facliity purchase baseline survey IS quite different from that required by a damage repair document. If the dlvlng report Will be issued as a repalr construction document, the damage noted dunng the inspection should be properly °quantified and qualified. 02 Photographs or Video cassette recordings should be used whenever pOSSible as they can provlde a more detalled and accu rate description of inspection results. A more detailed discussion on methods of recording and documentation IS provided later in this chapt’9r.



Inspection of marine structures should generally be conducted in two parts: the first part is an above-water survey, and the second part is an Inspection of the underwater portions of the structure.4 It is Important to coordlnate the two surveys 50 that no area of the structure within the tidal zone is left uninspected due to tide elevation changes. The underwater inspection is usually performed along the same guidelines as for the above-water inspection. Therefore, this section discusses the Inspection of the underwater portion of concrete structures only .

The underwater survey should also be conducted in two phases whenever posslble. 4 The tlrst phase Involves a qUick visual (Level 1) inspection to detect vIsible damage or malor detenoratlon whlch could be used to make a prehminary assessment of the condition of the underwater portion of the structure. Information obtamed dunng thls phase should provlde gUidance for developlng and performlng a final (Level Il or Levellll) Inspection. The second phase, or final Inspection, IS a complete Inspection of the structure from the splash zone down to the mudhne as shown III Figure 3.4.

Since the tidal zone is the area where most of the mechanlcal damage or deterioration usually occurs, it should be Inspected at several locatIOns, wlth most of the inspection points concentrated Immediately below the low water zone. Inspection and documentation of the structure in the submerged zone should be spaced unlformly to provide efficient data gathering .


At or near the mudllne It is recommended to Increase inspection and documentation to evaluate the patentlal for scour and abrasion type damage.4 A typlcal underwater Inspection procedure IS outlined below:3 ,5

(a) Inspect the structure stanlng at the splash/tidal zone..

(b) Remove ail manne growth from a section approxlmately 450 to 600 mm (18 to 24 in.) ln length.

(c) Visually Inspect thls area for cracks with rust stains, spalling, and exposed reinforclng steel. The location, length and wldth of the defects should be recorded.

(d) Sound the cleaned area wlth a hammer to detect loose concrete or hollow areas ln the structure.

(e) White proceedlng downward, visually inspect the structure where marine growth is minimal, and sound wrth a hammer.

(f) At the bottom, record (on a Plexiglass slate) the water depth along wrth any observations of damage. Carefully Inspect the base of mass con crete structures such as retalning walls and footings to detect any scour damage.

(g) After returnlng to the surface, record ail observations and measurements immediately into the inspection log. If voice communication between the diver and topside personnel is avallable, the data can be dlrectly relayed to the surface as the inspection progresses.

More detalled procedures for inspecting underwater bridge substructures are provided in Appendlx H.



To facliltate inSpection, so that a thorough and accu rate visual examination of the structure can be conducted, sorne form of surface cleaning will almost always be required. The eXlent of the cleamng is depandent on me amount of marine growth present, and the type of inspection being made. For Instance, routine inspections generally require only minor clcaning, whereas detaited Inspections require thorough cleaning of ce nain structural elements. According to Lambenon et al.,1 “Indiscrlmlnate cleaning should be avolded” because it is not only time consuming and expenslve, It can also cause funher damage to weakened areas.

There are four baSIC categories of underwater surface cleaning tools and each category offers a selection of techmques giving several methods. The performance of each type of equipment, such as cleaning rates, depends upon many factors. s,1e These include:

  • The physlcal and operatlOnal characteristlcs of the cleanlng tool
  • The operator expenence
  • The extent and type of manne growth
  • Water vislbility
  • Accesslblhty of the surface to be cleaned
  • Underwater worklng conditions


Assuming good worklng conditions, an average expenenced diver can achieve the cleaning rates given in Table 3.7. The following sections descnbe sorne of the tools listed in the table that are appropriate for cleanlng underwater concrete structures.



The effectlVeness of these types of tools depends on dlver effort. These tools are not powered, are usually used for hght cleanlng, and are unhkely to cause any damage to the structure ItseH. Examples of these tools Include scrapers, dlver’s knife, chipplng hammers, probes, wlre brushes, chalns, and wlre ropes. Chains and wlre ropes are wrapped around the member (such as a pile) and are then pulled back and forth to remove any fouling.5

Hand tools are small, lightweight, hlghly portable, and are also the least hazardous for underwater use. The main disadvantage is that they are slow and time consuming. The hlghest cleaning rate that can be achieved is approximately 0.1 m2 (1 Jt2) per minute. When heavy marine growth is present, cleaning rates are as low as 0.02 ta 0.03 m2/min (0.2 ta 0.3 ft2/min). For thls reason, hand tools are not usually used for cleantng large areas or for removlng heavy marine growth from concrete structures. ‘8



Pneumatic and hydraulic tools are used for cleaning thick manne growth from large areas and are more efficient than hand tools, but can cause damage ta the surface of the structure. Hydraulic tools are usually preferred since they are safer and easier ta use, and they are not limited by water depth. They also cause less diver fatIgue. Examples of these tools include chippers, grinders, needle guns, and rotary brushes. Brush systems are available in various sizes. Brush sizes up to 400 mm (16 ln.) in diameter are suitable for cleaning bridge substructures. 1.17

To effectively remove heavy, calcareous marine growth from concrete surfaces, a rotary cleaning tool, such as the ·Whirl Away or Barnacle Buster” is recommended. This tool can be attached to and operated by most standard hydraulic grinders, disc sanders, and polishers. The attachment consists of several hardened steel cutters that rotate in a direction opposite ta that of the tool shaft. The Barnacle Buster is the most effective and safest tool for removing marine growth from concrete surfaces because it IS easy ta use and does not require high-pressure water.5

There are several models available rang,”g from 80 to 170 mm (3-1/4 in. ta 7 in.) in diameter. The largest diameter tool can clean more th an 75 mm (3 in.) of hard shell growth and 150 mm (6 in.) of soft growth at rates of 0.28 ta 0.56 m2 (3 ta 6 ft~ per minute. The smaller models cannat remove heavy fouling effectively and should be used only ta remove marine growth less than 50 ta 75 mm (2 ”v 3 in.) thick.18



High-pressure water jets are used widely and are very effective for tough cleaning jobs and produce sorne of the highest cleanlng rates. Water jets can be used ta displace loose sediment and de bris, rem ove low quality concrete, provide a rough surface for better bonding, and eliminate feathered edges around the perimeter of an eroded area. Water jets can be either used by divers or remotely controlled from the surface. a

There are many types of underwater water jet tools available on the market. They ail ganerally consist of a surface pump, a high pressure hose and a gun (Figure 3.5). Watar Jat cleanlng tools generally fall into one of two categories: hlgh-flow devices or low-flow devlces. la Hlgh-flow cleaning systems operate between 27 and 80 MPa (4,000 and 12,000 pSI) and between 45 to 97 m3/min. (12 to 26 gpm). Most hlgh-flow tools are fitted with a retrojet to counter the reaction force generated by the water jet. Low-flow cleamng systems opel ate ~i apprQ)omately 68 MPa (10,000 psi) and 8 to 20 m3/min. (2 to 5 gpm). This tool does not develop anough backthrust to requlre a retrojet.

Most water jet tools h:,”e interchangeable fan and straight-jet nozzles (Figure 3.6). III Fan-jet nozzles will clean a wider surface area, while straight-jet nozzles provide a higher cleaning intensity over a very smalt area. According to research by Parker et al., 19 the most effective nozzle size ln removing marine growth is 0.80 mm (0.031 in.) ln diameter. The fan-jet nozzle can clean an area up to ten tlmes faster than typical straight jets. However, the intensity of the fan-jet reduces sharply as the distance from the work surface is Increased. III The straight-jet nozzle is very effective for cleaning the Interior of cracks .

The main disadvantage of the water Jet cleaning tool is that Il is potentially hazardous and if not properly controlled, an unskdled operator can damage the concrete surface.17 The high pressure and veloclty of the water jet will easily remove human flesh and bone. Ali personnel should be aware of the hazards Involved and should be properly trained before using the device. Highpressure water jets can also produce excessive noise under water which, over time, can be harmful ta the diver.’B



Self-propelled cleaning vehicles are used for removing marine growth and corrosion from large underwater surfacE.; which are readily accessible. Although they have been designed for use on steel surfaces (hu”s of ships and ail tankers), they can also be used for cleaning concrete. For instance, the sides of lock walls, faces of dams and stilling basins are areas where these vehicles can be used effectively.1a

A typlcal underwater self-propelled vehicle consists of three large rotating brushes and travels on traction wheels. Dependlng upon conditions, it can clean a path about 1.2 to 1.5 m (4 ta 5 ft.) wide and can travel up to 27 m (90 ft.) per minute, giving a cleaning rate of 42 m2 (450 tr) par minute. The vehicle is connected ta a surface control console with an umbilical cable and can be elther remotely controlled or steered directly by a diver. The control console shows where the vehicle is located with respect ta orientation, depth, and the distance travelled. ,a

The disadvantage of using a self-propelled cleaning vehicle is that it can only clean fiat surfaces which are unobstructed. Also, due to Ils size and weight, deployment and recovery of the vehicle requires the use of large handling equipment. These vehlcles are also more expensive to operate than most ùther types of cleanlng equipment.’B



Visual inspection IS the most common method used to inspect underwater concrete structures.’ It is quick, usually uncomphcated, and nondestructive. It is usually performed to detect severe damage (Levell inspection), and to detect surface damage (Levelll inspection).3 There are three basic methods for gathering visuai information underwater and are listed in decreaslng order of resolutlon: dlver’s eyes, photography and video.’5 ln some cases, underwater observations can be made from the surface by an underwater scope. For example, underwater inspection du ring erosion repair at Chief Joseph Dam was performed through a 10.6 m (35 ft.) long scope equipped with a 150 mm (6 in.) diameter bottom glass and a high-power telescope.20



There are several hand tools available for performing an underwater visuallnspection. Hammers, picks, pry bars, and probing bars are used for performing soundings of the concrete (acoustic ringing). This method can detect voids in the concrete and delamlnatlon of the concrete coyer. Chipping tools are effertive for prodding the surface of the concrete to determine the depth of deterioration.’,3,5 These methods are economical but qualitative in nature, and should be used only as a gUide ln evaluating the condition of underwater concrete.

Flashlights for improving vlsibility are almost always necessary. The use of quartz iodide and thallium lodide lamps have been successfully used by divers to improve vislbility underwater.3 – Visibllity can also be improved by attachlng a clear-water mask to the face plate of the diving helmet.’



The measurement of physical dimensions provldes a partial quantitative measure of the member’s strength or degree of deterioration. This method is nondestructive, economlcal, and requires a minimal amount of time and equipment. However, this method does nct provide Information on the condition of the remaining concrete.3 More sophistlcated (Level III) techniques for taking internai defect measurements and assessing the condition of the remaining concrete are discussed in Chapter 4. Based on a review of Reference 17, below IS a summary of techniques most commonly used by divers to measure defects ln underwater concrete structures.



(a) Ruler. A ruler IS used for measunng crack length, spaillength and width. Measurements with a ruler can be taken wlth an accuracy of plus or minus 0.5 mm (0.02 in.).

(b) Tape Measure. For measurements up to 100 m (328 ft.), tape measures are usually employed. This IS not as accurate as the ru 1er, due to problems with positioning the end and tape sag. Accuracy is plus or minus 5.0 mm (0.2 in.).

(c) Magnetlc Tape. This can be used for measurements up to 3 m (9.8 ft.) and is often used to take circular measurements. Measuring accuracy is plus or minus 1.0 mm (0.04 ln.).

(d) Sea/es. These are often made from a special vinyl embossing tape (OYMO) and can be used ln conjunctlon wlth photography. Accuracy can be plus or minus 5.0 mm (0.2 in.).

(e) Comparator. A comparator IS used to measure crack widths and consists of a small handheld microscope with a scale on the lens closest ta the surface being inspected. Crack widths can be measured with an accuracy of about 0.025 mm (0.002 in.) when used in dry conditions.21 However, the use of this device for measuring cr~cl( widths under water may be severely limited if water visibility is poor.



(a) Cali pers. These instruments can be used for taking measurements up to 2 m (6.6 ft.) in diameter. If carefully uSed, measurements can be made with an accuracy of plus or minus 0.5 mm (0.02 in.).

(b) Special Jlgs. Various jigs are available for measuring the ovality of members. Accuracy can be plus or minus 5 mm (0.2 in.) .



(a) Profile gauge. The profile gauge can record a mlrror image of the defect with a possible accuracy of plus or minus 0.5 mm (0.2 ln.).

(b) Taut Wtre. This IS used for large spalled or deformed areas. It can also be used to measure the out-of-plumbness of a member. Accuracy depends on the conditions, but IS approxlmately plus or minus 5 mm (0.2 in.).

(c) Casts. These materials are also used to obtain a mirror Image of a defect. The most recently developed product for underwater use is produced by BP Chemicals ltd., and is marketed as “AQUAPRINT”. The moulding agent (supplied ln a cartndge) is applied to the surface by a pressure dispenser and cures in about 15 minutes. The flexible, but non-stretchable cast is peeled off and taken to the surface for VISU al analysls.



Although visuallnspectlons are quick and relatively inexpensive, there are severallimitations which must be considered. Environmental hmltatlons Include manne growth, which obscures surface defects unless cleaned, poor vlsiblhty in turbld water and strong water currents which ninders the divers capacity to work. Limitations which are related to the diver himself vary from inexperience as an inspector to poor observation and performance resulting from underwater environmental conditions. 3



Underwater inspections must often be conducted in turbid water which severely reduces vislbllity. ln many cases, vislbllity is zero. In these cases, a tactile inspection is required in which the diver must use his sensory perception capabllities, such as touch and feel, to detect flaws, damage and deterioration. The task is usually difficult to perform and requires more preparation than when working in clear water. The diver must have a good understanding of the structure by performing an In-depth study of the existing drawings of the structure. During the Inspection, good communication between the dlver and surface personnells essential. 1 • 3 This type of inspection most often requires prior cleamng if the structure is covered with marine growth.1



For an inspection to be useful, a clear, conCise, and complete record must be established. The inspection information should be documented uSlng standard forms and report formats. ~ecording devlces are necessary to provide a complete and permanent record of the condition of the structure. A Plexiglass slate and a grease pencii is useful for maklng notes and sketches under water. These notes can be transcnbed Into the inspection log when the dlVer returns to the surface.3.~ Current methods of recordlng Inspection flndings include drawlngs, inspection forms, closed-clrcuit televislon (video), flash photography, and photogrammetry.17 Ali of these methods are used frequently and are an essential part of the final report.



Inspection findings are often recorded on existing drawings of ti le structure after the diver has completed the inspection. When drawings of the structure are not available, they are drawn from memory by the diver at the end of the inspection. More often, these drawings are developed by using surface data recorders at the time of inspection and are verified by the divers. 11



Recordlng inspection data can be slmphfied by using appropriate inspection forms. They are also useful for comparing past, present, and future Inspection results. Inspection information should be recorded during the inspection, or as soon as the Inspection is completed, and in accordance with generally understood termlnology.3.5 A standard form which may be used for reporting the condition of concrete piles is shown in Figure 3.7. An explanation of the condition ratings used on the form is provided ln Figure 3.8.



This method of recordlng IS used extensively for underwater inspec~ions. Video can expedite major underwater inspections and has the advantage of “real-time” display to the surface and “real-time” quality control of the video image, and can be used to monitor diver performance.3 Video systems can be used as diver-deployed (hand-held or head-mounted), or as remotely operated, mounted on ROVs, or used in piloted minisubmarines. Presently, these systems are available in monochrome (black and white), color, and low Iight.17



There are two methods of recording video inspections: video home systems (VHS) and U-Matic. The VHS equipment IS smaller, more portable, and is more readily available. VHS eqUipment made from different manufacturers have good compatlblhty. The main disadvantage is that the qualrty of the reproduced picture can be poor, and copies of the tapes are always of poorer quality. The U-Matic system uses a wider tape whlch can provlde hlghly detailed photographs and better quahty Video tape reproduction. 17

The “VIDICON” is probably the most widely used image sensor for underwater inspections. It is a tube-type device whlch provldes high quality images. It is sensitive to bright light and Is insensitive to low light levels. However, the standard VIDICON has been modified to provide better performance in these areas.3 ln some cases, “down-hole” cameras can be used to perform visual inspections from the surface. For ir.stance, ln the case of repairs at Chief Joseph Dam in Washington D.C., (U.S.A.), a down-hole camera was used to inspect the vertical monolith joints of the dam.22 Similarly, a down-hole camera was used for inspecting vertical grout hales dunng repairs at Big Eddy Dam in Ontario, Canada.23

The important aspect of using video recordings IS that comments can be recorded on the tape as the Inspection progresses, elther by the dlVer performlng the Inspection or by surface personnel vlewing the monitor. When the dlver IS breathlng a helium gas mixture, the vOlee changes pitch and a special vOlce unscrambler must be used 17 Video tapes should be provlded wlth a tltle, a brlef deSCriptIOn of what is on the tape, and the date the InspeCtIOn was performed. The description should Include the name, location, type, and slze of the structure belng Inspected and i3ny other pertinent Informatlon.24



Image enhancement is a common technique used wlth video recording. Video recordlngs are electronically enhanced to produce a better quality Image. Enhancement IS avallable in color, elec:tronic video enhancement systems, and stereo video systems. The latter IS very useful where an .assessment of depth IS deslrable.15 However, slnce currently avallable camera systems can take extremely accurate and highly detalled photographs, the use of image enhancement systems ma’V not be necessary. 17



Endoprobes can improve visu al observation and are useful for Inspecting the inaccessible areas of a structure. The system, which consists of fiber-light gUide cables, a light source, and an optlcal system, provides a 1350 field of vision. The lenses are color adjusted so that an accurate photograph can be obtained. The Inspection observations made with an endoprobe can be recorded by the use of “Reflex’ cameras, “Polarold” cameras, and television monitors. Size and depth perception can be aehleved through the use of graticules. 17



Flash photography IS used extenslvely for recordmg inspection flndlngs. The photographs can be in color or monochrome, and can vary from general stand-off eoverage to 100 percent eloseup mosalcs. 15 Water turbldlty and monochromatlc marine growth Will most often make photographie color distinction dlfficult. ThiS problem ean be reduced by prior cleaning of the structure, fitting a clear-water box to the lens of the camera, or by uSlng proper Iightmg.3

There are three baSIC ways ta use a flash: as a single unit on a stalk as two units on either side of the camera, each on its own stalk and as a ring flash built around the camera lens. The latter usually requires the use of specialized close-up cameras. Back scatter from suspended particles in the water and reflected light from the obJect are the problems usually encountered when uSlng the single or double Unit flashes. 17 This problem can usually be mlnlmlzed by placlng the flash at an angle as shown in Figure 3.9.

Ali photographs should be numbered and labelled with a slate. If color photographs are used, a col or chart Indicatlng color distortions should be attached to the slate.3



There are several types of speclalized cameras avaUable for performing underwater inspections. Spec.ahst companies have created a variety of camera types and can be grouped into the following categories: 17

(a) The Nikonos System. This system IS a totally waterproofed camera with interchangeable lenses. It can take close-up photographs, using either adapters or close-up lenses and can also use wide-angle or telephoto lenses. The camera uses standard 35 mm film cassettes and either dedicated or nondedicated flash units.

(b) Hydroscan. This is currently the most widely used 35 mm underwater camera. This camera, which has been specifically designed to take close-up photographs, consists of a close-up lens, fixed prods to provlde the correct stand-off distance, a nng flash, a 250-frame film cassette, an automatic film advance, and a special waterproof underwater houslng. It IS also avallable wlth digitalized electronics for pnnting Information on each frame.

(c) Hasse/blad. These cameras use medium format 6 cm x 6 cm (2.4 ln x 2.4 ln.), single lens reflex eqUipment. Several of these systems are housed ln waterproofing Unlts. This camera system provldes a very large negatlve whlch provldes a good quality pnnt.



The selection of a particular type of film IS based on three factors: film speed (how quickly It reacts to lia”t) whether it is monochrome or color and whether it is small or medium format. These are summanzed below.

(a) FIlm Speed Several numencal rating systems eXlst for categorizlng film speed and are loosely grouped Into slow, medium, and fast films (Table 3.8). The American Standards AssoCIation (ASA), the Deutsch Industnes Norm (DIN), and the International Standards OrganlzatlOn (ISO) systems rate 25 and 50 ASA films as slow films, 100 and 200 ASA as medium, and 400 and 800 ASA as fast fllms.’7

(b) Color. Monochrome films are rarely used for underwater Inspections. Therefore, selectlng a film type usually Involves a chOice between color pnnts or color positIVes (slldes). The advantages of color slides are that they can be vlewed wlthout havlng to process prlnts. are easy to develop on site, and are readily avallable. One dlsadvantage, however, is slldes cannot be presented ln a report. Also, slides are very sensitive to exposure errors. On the other hand, color pnnts are less affected by exposure errors. Celor repnnts are easy to obtaln and are easily presented ln reports. 17 To speed up the processing of color pnnts. a 35 mm Polarold film wlth a portable processor IS used. This eqUipment IS not expensive but does not provlde the same degree of photograph resolution that can be obtained by using standard 35 mm color film.3

(c) Format. There is a choice between two basIc formats: small (35 mm) or medium (6 cm x 6 cm). The small format can be enlarged to a greater degree than the medium format, resulMg ln less loss of qualrty ln the final pnnt. 17 Selectlng a particular film type for a panicular application IS summanzed ln Table 3.9.



Photogrammetry or stereo photography IS used when depth or size perception is desirable. Photogrammetry Involves analyzing “information contained in two (stereo) pictures of the sa me scene taken from different angles”.17 Photogrammetry was first developed for use ~n land but modlfied eqUipment for underwater application has been available for qulte some time.

The underwater method produces stereo plctures by uSlng two cameras which are usually based on the Hasselblad equipment. The syF.tem is usually equipped with calibration devices to allow for vanous camera angles and water conditions. The stereo photographs are analyzed by a computer which can provide pnntouts to suit specifie needs, and can be interfaced with other computer systems. Measurements in ail three dimensions can be obtained from the photographs wlth an accuracy ot “lUS or minus 0.33 mm (0.01 in.).’7



An effective repair and maintenance program cannot be selected until the basic cause of the detenoration is determined. Underwater inspection findings can be used to make an engineering assessment of the structure to determme the cause, extent, and rate of deterioration, as weil as Its structural capacity and safety. This information may also be used to predict the remaining usefullife of the structure and to develop a repair and maintenance program which will allow the safe operation of the structure. Similarly, the inspection results can be used to determlne the required manpower and financing needed to perform the repalr work. Once the findlngs of the engineenng assessment are available, a management decision must be made whether to allow detenoration to continue, to make repairs that will keep the structure ln Its present condition, to make repalrs that will strengthen the structure or to replace the structure. 12

The decision-maklng process can be slmplifled by developlng a condition ratlng Index of the structure whlch willlndicate the urgency of corrective action. The urgency Index IS developed to aS5ess the conditIOn of the structure and the posslblhty for further deterioratlon. The Index IS used ta keep the facllrty operating at a speclfic load and safety level. The rating Index decreases with increasing detenoration.4 An example of an urgency Index ratlng system IS shown ln Table 3.10. The rating can also be modlfied based on engineenng judgement of the deterioration. The modification is shown en Table 3.11.

Management decisions must be based on the consideration of several important factors, su ch as safety, the operational need for the structure, thp. adequacy of the structure to meet the needs of its intended purpose, economlc considerations, appearance, environmental conditions, and other factors having an Impact on the structure3 • A flow chart developed by Brackett et al. 211 summanzes the decision maklng process for a maintenance and safety program of waterfront facilities, and IS shawn ln Figure 3.10.



The final phase of any inspection program should be the preparation of the final report. In some cases, the reportlng procedure is dictated by the client. In other instances, the form of the report IS not speclfled and it is up to the engineer or inspector to develop a format. There are no set rules for presentmg reports as they vary from one company to another.

The basic requirements for good report writln9 are grammar skills, syntax, style, punctuation, usage, and organization.25 It IS essential for the writer to be clear, concise, and to the point. Technlcal aspects should be kept separate from qualitative discussion. Most clients do not have the technical background ta understand engineering princlples .



The report should be typed and bound. The report should have a title relevant to the project as weil as a table of contents, and executive summary. Ali pages should be numbered, with the latest issue number indicated, and dated. Sections within the report should also be numbered. Any drawlngs or photographs pertaimng to the Inspection must be included in the report and should be cross-referenced. Any relevant video recordings of the inspection should also be referenced. and should Indicate the place where they can be viewed. 17

If required by the client, the report should provide recommendations for repairs including drawings or sketches showing the work. Construction cost estimates for the proposed repair work should also be provided for decision making purposes. Lastly, the report must be carefully reviewed and signed by an authorized representative within the company. A general outline whlch can be used for most inspection reports is shown in Table 3.12. Sorne of these topics are summarized below.




This section should be limited to one page whenever possible. It should describe in concise terms what the report covers, what was done, and give a summary of the major conclusions and recommendations.


This section should outline the specifie purpose for the inspection and what work was performed. It should identlfy the structure that was Inspected. as weil as its location. The scope should also identify the parties who sponsored (or authonzed) the Investigation. and the personnel who performed the inspection and testlng.


A t)rief description of the structure or facility should be provided including vicinrty, locality, and hi~,(orical background if available. Whenever possible, plan-view maps, elevations, sections of the structure,and geological maps should be included along with operational information of the facllity. Ali information obtained from other sources which has not be venfied by the wnter should also be included.


Ali eXlsting documents and engineering data which was reviewed and referenced in the report should be listed, with their title, date and origin. The documents may include original construction drawings of the structure and previous inspection and repair reports.


This section of the report will provide a detailed description of the observed conditions of the structure including a brief explanation of the various ln-situ testing techniques used. It Is important ttïat this section only include facts and observations, and not opinions. Ali relevant photographs should be included and referenced. Each photograph should be numbered and should provide a brief description of ItS contents. The degree ~f deterioration should be diagrammatically iIIustrated whenever possible. Any Inspection forms used for documenting the inspection results should be referenced and ,”cluded as an appendix to the report. Any relevant video recording should also be referenced and made available for vlewing .



A description of the laboratory tests tt’lat were performed should be provided. As with Inspection findings, only facts should be reported in this section. Interpretation of test results is included in the discussion section of the report. Photographs or drawlngs of where cores or physical samples were removed should also be provided.



This section should descnbe any structural analyses or calculations that were used to determine the service load capacity of the deteriorated structure or member. The analyses should include the effects of deterioratlon as determined by field and laboratory investigations. Based on these, the reduced capacity of the structure or element is developed.



This section provides an interpretation of field investigations, laboratory tests, and engineering calculations. Ali discussion must be based on the facts already presented in previous sections of the report. If the discussion is extensive, dividing il into appropriate subsections may be useful.



This is the most important section of the report which provldes a general summary of the findings and characterizes the condition of the structure. If an unsafe condition exists, it should be identlfied and methods for temporary bracing should be suggested28 • It should specify the adequacy of the structure or facility based on current design, and operational criteria. The remaining useful life of the structure should also be estimated. Recommendations for possible repair techniques should be provided along with construction cost estimates. Recommended repair procedures should be schematically iIIustrated sllowing the major features of the repair work .



Organizations or owners concerned with the durability and continued operation of concrete structures must perform penodic visual inspections. During these inspections, the observed damage or deterioration may require a more critical examination. In other cases, concrete structures may experience internai deterioratlon (e.g., corrosion of reinforcing steel) which becomes visible only after the deterioratlon has progressed significantly. The qualitative data obtained from vlsual inspections is generally inadequate to accurately assess the condition of the strut.1ure. In such cases, these Inspections must be supplemented with more sophisticated methods for obtalmng quantitative data to determlne the cause, extent, and rate of deterioration.

A wide range of techniques and apparatus are available to the engineer to examine and evaluate concrete structures above water. Sorne of these methods have been adapted for inspecting concrete structures under water. These test methods can either be nondestructive or partially destructive ln nature. The latter is often used for retrieving samples for laboratory testing, thus reqUiring sorne form of repair ta the concrete after the testing is completed. Depending on the information which is sought, the inspector will often be required ta use a combination of these methods tO be able ta determine the pnmary cause or causes of deterioration.

Generally speaking, these tests can be used ta evaluate the following four areas of concern regarding the deterioration of reinforced concrete structures:

  • Concrete quality and composition
  • Concrete strength
  • Corrosion of embedded reinforcing steel
  • Structural integrity and performance.


Although not intended to be a complete guide, the following sections provide a summary of the most commonly used methods for testing and evaluating existing concrete structures. Though sorne of the methods described here were developed in Europe, they are ail available in the USA and Canada through specialist companies. An excellent review of in-situ/nondestructive testing

of concrete is provided in Reference 2. A list of in-situ inspection techniques and common laboratory tests for concrete are also provlded in Appendices 1 and J, respectively. The mate rial in this chapter has been adapted from dlfferent available references, especlally 1 and 5, and are presented here for completeness.




The rebound hammer, also known as the Schmidt or SWISS Hammer, is a surface hardness tester that determines the uniformity of in-situ concrete, sa that areas of inferier quality can be detected. It measures the rebound distance of a spring-driven mass after it impacts the concrete surface with a standard force. It can be used above or below the water and is usefulln new construction . to assist contractors in determining stripping tlmes for formwork.’ A cutaway vlew showing the various parts of a typical rebound hammer IS shown in Figure 4.1.

To carry out the test (ASTM C 80S), the impact plunger is pressed firmly against the concrete surface, thereby releasing the spring-Ioaded mass trom its locked position. The mass then strlkes the steel plunger which is ln contact with the concrete surface. The resulting ‘Rebound Number’ or rebound distance of the hammer is read on a linear scale attached ta the instrument. A recording type of Schmidt hammer IS available which records the rebound numbers automatically on a roll of paper.3 To obtain representatlve data, it is advisable to take a minimum of twelve readlngs at each test location and averaged (excluding the minimum and maximum values).4 Ti,is number can be used to check concrete unlformlty by compaflng test data from vanous parts of the structure. In general, the higher the rebound number the better the concrete quahty. If callbrated wlth laboratory tests on concrete cubes or cylinders, the hammer can glve an Indication of the In-situ compressive strength.5

Although uSlng the rebound hammer IS a qUlck and inexpensive way of determining concrete unlformlty, It has many limitations whlch must be recognlzed by the user. For example, the results of the rebound number are affected by smoothness and mOlsllJre condition of concrete surface, and type of coarse aggregate.2 Smoother surfaces usually gIlle hlgher rebound numbers with less scatter ln the data. 4 Thus, if the concrete surface IS rough, it should be smoothed with a medium-gralned SIlicon carblde stone.6 On the other hand, saturated concrete tends to give rebound numbers sllghtly lower than wh en tested dry.4 If repair patchwork has been done, test locations away from the patches should be chosen, since the characteristlcs of the repair material may differ trom those of the structure. Processes that harden the concrete surface such as cu ring membranes or carbonation, may also glve hlgher rebound numbers. Since the velocity of the Impact plunger IS affected by gravlty and fflctlon, rebound numbers obtained in the vertical and hOrizontal plane of the same test location will be different.5 Figure 4.2 shows the typical effect on the rebound number when the Impact plane is not horizontal.



This Instrument, although less useful than the rebound hammer, may also be used to determine unlformlty by measunng the penetration reslstance of concrete at different locations of the structure. The Windsor Probe whlch is standarized by ASTM C 803, cOl1sists of a gun loaded with a hardened alloy probe whlch IS driven into the concrete. The length of the probe which remains exposed provldes a measure of the penetration resistance of the concrete. To ensure that the test 15 performed with sorne degree of unlformlty, ASTM C 803 specifies a maximum probe veloclty variation of three percent, based on a minimum of ten tests. 1 As with the rebound hammer, a calibration chart must be made to correlate probe penetration with concrete compressive strength,6 because calibration curves supplied by the manufacturer are not always rellable. This method 15 partlally destructive and requires repairing the concrete surtace after the probe IS removed. The test resu~s are also affected by the hardness of the aggregate.2



The UPV method consists of measunng the tlme It takes a direct compression wave to pass through th~ concrete (Figure 4.3). The tlme of travel between the initiai propagation and reception of the pulse is measured by an electronlc tngger/timer device. The average wave veloclty is then computed by dlviding the measured path length of the wave by the tlme of travel. Ultrasonic test procedures are standardlzed by ASTM C 597.


Since UPV is a function of the modulus of elastlclty and denslty of the matenal through whlch It travels, this method provldes comparative data for assessmg concrete unlformrty, as weil as locatlng defects (i.e., cracks, voids, etc.). In sorne cases, it has been used for estimating ln-situ compressive strength. However, the relatlOnship between pulse veloclty and concrete strength are affected by several variables, Includlng age of concrete, mOisture conditions, aggregate ta cement ratio, type of aggregate, surface finish, and location of steel relnforcement. 2 Table 4.1 shows the relationshlp between pulse veloclty and concrete conditIOn, as suggested by Whltehurst,7 and the minimum acceptable velocltles for specifie structure types (In Great Bntaln) are presented ln Table 4.2, as suggested by Jones.a

The UPV method has tradltlonally been performed by passlng ultrasonic pulses through the concrete between fixed pOints as Illustrated ln Figure 4.4, The most common and most accurate method is direct transmission when the transducers are on opposite, parallel faces of the test location, Seml-dlrect transmission is less accurate and not normally used because it is difficult to dupllcate the transmiSSion path, The least accu rate IS indirect transmission and IS used when only one surface of the concrete IS accessible, such as a retaining wall,8

With the recently developed scannlng system, shown ln Figure 4.5, the source and recelver transducers are free to move whlle obtalnmg UPV measurements. The UPV scanners conslst of coated plezoelectric source/recelver ceramlcs ln a cyhndncal shape whlch allows them to roll whlle transmltting and receivlng wave pulses. The signais are flrst amphfled and flltered, and then transmltted IntO a data acquISItIOn system Typical scannlng speeds range from 0.15 to a 31 mis (0.5 to 1 ft/s) , and can collect data at rates of 5 to 10 pulses/second. This speed provldes test data every 2.5 ta 5 cm (1 ta 2 ln.) along the entlre path. Wlth current data acqUiSItIOn systems, it 15 possible ta make scans of over 9 m (30 ft.) ln length at a tllne. After scannlng, the data IS analyzed by a computer system whlch computes the concrete charactenstlcs at each test location.9



Sonic logging IS used to perform the UPV test ln areas which are inaccessible or under water Logglng conslsts of transmltting an ultrasonic pulse through the concrete between source and receiver probes whlch are placed ln water-fliled tubes (Figure 4.6). The probes are lowered Into preplaced access tubes (PVC sleeves) or coreholes by cables that are pu lied over a special winch, which measures and records the probe depth. Sonic logglng is performed as the probes are withdrawn simultaneously, th us provldlng a contlnuous profile of travel tlme The method 15 capable of taking measurements at 25 mm (1 ln.) 5paclngs.’o It can also be used ln corehole5 drilled through the base of the foundatlon to assess the Integrity of the interface between bedrock and the foundation concrete.5

The measured travel times between probes and the corresponding wave velocities are used to evaluate the concrete quality. For example, longer travel times Indlcate irregulanties ln the concrete, whlle complete loss of signal is indicative of a defect ln the concrete between the tubes. 10



Diagraphlc dnlling IS a destructive exploratory technique used to determine the mechanical charactenstics and quallty of large concrete structures. Ouring the drilling, several parameters are recorded, such as the instantaneous penetration rat~ of the drilling tool, the compressive force (thrust) on the drill rod, and the torque applied on the dnlling tool. These parameters are related ta the mechanical strength of the concrete and to ItS cohesion. Diagraphic dnlhng was used for the In-Situ Investigation of an old concrete quay wall in Zeebrugge, Belgium.ll A typical diagraphlc drilhng record is shown in Figure 4.7. Interpretation of the diagrams are typically supported by video inspection of the boreholes, and supplementallaboratory testing of concrete cores. The advantages of dlagraphic dnlling include ease of application, rapid execution, and the relatively low cost.



Radiometry IS primarily used to measure in-Situ denslty and thlckness of concrete members. The Smith and Whiffin method, and the Brocard method are the methods most commonly used for determining the in-Situ densrty of concrete. The Smith and Whlffln test consists of dnlling holes into the concrete surface and lowering a radioactive source, such as cobalt or radium, down Into the holes. Geiger counters, placed in heavy lead sheath, are then posltloned on thl~ outslde vertical face of the concrete member at the same depth as the radioactive source. Th.~ Geiger counters are calibrated by taklng readlngs on samples of known denslties. From these madlngs, charts are prepared which are then used for determlnlng the in-situ denslty of the concrete by companson with subsequent Geiger readlngs.’ The Brocard method IS vlrtually the same, dlffering only in the radioactive source used and the thlckness of the concrete member whlch can be tested (up ta 410 mm/16 In.).12



Concrete durabihty IS closely related ta ItS ease wlth whlch water (or other aggresslve fluids) can move thlough ItS pore structure The rate at whlch flulds penetrate the concrete determlnes ItS permeablhty, and hence, ItS rate of detenoratlon. The permeablhty of in-situ concrete can be measured by a senes of tests presently ln use whlch are covered by the Bntlsh Standards Institution (BSI), BS 1181, Part 5 The most commonly used Include the Figg Hypodermic Test and the Initiai Surface Absorption Test. 1 These are summanzed below.



This test simply measures the tlme It takes for a change in pressure ta occur in a fluid (air/water) sealed withln an evacuated vOid ln the concrete (Figure 4.8). The test procedure consists of dnlhng a 10 mm (311 ln.) dlameter by 40 mm (1-% in.) deep hale into the concrete, and plugglng it wlth a polyether foam and seahng It wlth a Silicone sealant. For a water permeability test, a hypodermlc needle 15 pushed through the plug and connected ta a water source (100 mm/4 in. head) and a manometer. The tlme It takes the water ta move a distance of 50 mm (2 ln.) down a capillary tube is noted.

For the air permeablhty test, the hypodermlc needle IS connected to a vacuum pump and the tlme taken for a pressure drop of about 55 to 59 kPa (8 tn 8.5 pSI) wlthln the sealed vOid IS recorded. The permeability of concrete to wateT can be Judged quahtatlvely accordlng ta the Infiltration tlmes shown ln Table 4.3.



The Initial surface absorption test measures the amount of water absorbed by the concrete per unit are a under a constant pressure head (Figure 4.9). r”e pressure head (200 mm) IS applied through a flexible mlet tube attached to a watertlght cap which IS clamped to the test area. An outlet tube is connected ta a cahbrated capillary scale which measures the water penetration Into the concrete after the water source 15 closed. Measurements are taken at lime Intervals of 10 min., 30 min .. 1 hour, and 2 hours from the start of the test. The expected durabllity of the concrete IS classlfied ln accordance wlth the tlme of InfiltratIOn rates shown ln Table 4.4.

The followlng is a hst of other permeability test methods currently belng developed:’

  • VTI test’3
  • ISE caplilary test
  • Water absorption test
  • Pressure differential water permeability test
  • lonic diffusion test
  • Gas diffusion test




Concrete core testing is still considered to be one of the most reliable methods to determine concrete compressive strength. Coring is also useful for Investigating a variety of other in-situ charactenstics, such as the depth of surface detenoration, and the presence and size of visible cracks. Concrete co ring IS o\~en used to verity the results of other in-situ tests and to provide a physlcal specimen for supplementallaboratory testing (Section 4.6). The cores should be taken in areas that are representatlve of the structure. Methods for achievlng random sampling are also descnbed ln ASTM C 823.

Mather’l suggests that if the concrete is in deteriorated condition or when drilling operations are questionable, better core recovery will be achieved with a 150 mm (6 in.) diameter diamond bit and barrel than with smaller ones. On the other hand, when the concrete is in fairly good condition, the dnller is nighly skilled, and the rig is operatlng efficiently, cores can be satisfactorily retrelved using 55 mm (2-118 in.) diameter (Nx) bits ..

Cores should be logged as they are removed trom the hole and core holes should be accurately located on approprlate construction drawlngs. Cores should be properly packaged so that they will not be damaged or mlxed up dunng shlpment to the laboratory. In sorne cases, It may be necessary to place the cores Into plastiC sleeves, or wrapplng them ln cheesecloth dlpped ln hquid wax ta preserve the field mOlsture content. Preservlng the neld mOlsture content IS usually very Important If some deletenous chemlcal reactlon IS suspected.6 ExamlnatlOn of the cores once they are received at the laboratory are summanzed in ASTM C 856 and preparation of a concrete cyhnder for compression testing IS described ln ASTM C 39 speCifications. It baslcally consists of capplng the ends to achleve a smooth beanng surface to minllT,’ze or ehmlnate eccentricity dunng load applicatIOn. The core II) usually dlmensloned so that the length IS at least tWlce tlle diameter. After the specimen IS prepared, It IS placed Into the testlng machine and the load IS slowly and contlnuously applied until idllure. At the end of the test, the type of tallure and gf!neral appearance of the concrete IS noted on the test log. The compressive strength IS calculated by divlding the maximum load at fallure by the average cross-sectional area of the core.’ Typica! concrete fallure modes are shown ln Figure 4.10. In cases where the samples do not meet the specified length to diameter ratios, applicable correction factors are apphed to the calculated compressive strengths (Table 4.5).’4



The pull-out test IS an In-situ method of determming concrete compressive strength by measuring the maximum force required to pull an embedded Insert from the concrete mass. The concept was initially suggested by SkramtaJew ln 1938 and Investlgated further by Kierkegaard-Hansen. Current ASTM C 900 standards are based on tests conducted by Malhotra, Richards, and Rutenheck ln the early 1970s.

ln general, a pullout test consists of pulling out a speclally shaped steel insert tram concrete (Figure 4.11) The required pullout force reqUired IS measured using a dynamometer. Due to its shape, a cone (frustrum) of concrete IS pulled out wlth the Insert, generatir.g failure planes at approxlmately 45° to the direction of the pull. The pullout strength IS approximately 20 percent of the concrete compressive strength.2

ln a recent $tudy, Collins and Roper15 used pullout tests to evaluate concrete spall repalrs. In the study, it was determined that pullout test methods can be used to simulate spalling concrete. Ali laboratory specimens were subsequently damaged by pullout testing, repalred with epoxy mmtar and subJt1cted ta & second pullout test. The test program showed that the major factor governing the success of a repair to con crete IS the soundness of the repair plane.

Due ta the nature of the test, pullout techniques cannot be used on hardened concrete. Ta overcome this shortfall, new techniques have been developed in which a set of standard anchors are pu lied out of standard dnlled holes. The anchors can elther be normal pullout Inserts or splltsleeve w~dge anchors. In the former case, a cone of concrete IS pL’lled out whlle ln the latter case, internai cracking of concrete IS prodlJced. Tests rnvolvlng rulling out bolts set by means of epoxy ln drilled holes have also been reported 2



This method provldes a means for determining concrete compressive strength by measunng the force requlred to pull free a steel probe which 15 bonded to the concrete surface wlth a highstrength adhesive (epoxy resln). An equivalent cube compressive strength IS obtarned uSlng a calibration graph.

Typically, the bond strength between the probe and the concrete surface is conslderably higher th an the tensile strength of the underlyrng concrete, thus elimlnating the posslblhty of fallure at the Interface. Fallure usually occurs wlthln the underlylng concrete mass. The approxlmate tenslle strength of the concrete IS computed by dlvldlng the load requlred to break off the concrete mass by the area of the probe. The compressive strength of the concrete IS obtalned by dlviding the cross-sectlonal area of a specimen by the load requlred to crush 1t. 1

ln cases where the concrete surface 15 too smooth to allow a good bond between the epoxy/concrete Interface, the concrete surface 15 partlally cored to expose a rough finish 50 that a better bond cou Id be ac.hleved for the probe.



This mettlod 15 used for determlnlng the flexural strength of concrete at sorne distance away from the surface 2 ThiS 15 done by breaklng a 55 mm (2-‘/& ln.) dlameter cantllevered core formed by cutting a clrcular slot ln the eXlsting concrete member. The core IS broken off at Its base by applylng a force at the top wlth a hydraullcally-operated Jack which IS cou pied ta a load cell. The force reqUired to break off the core 15 correlated Wlth compressive strength by means of calibration chans, developed by cyhnder compression testlng of retneved cores. 1



The Internai Fracture test (8SI 19868) consists of drilhng a 6 mm (% in.) diameter hole into the concrete test location to a depth of approxlmately 35 mm (1-% in.}, mstalling a 20 mm (3/4 in.) expandlng wedye anchor Into the hole, and pulling the anchor wlth a torque meter fltted on a 75 mm (3 ln.) reactlon tnpod. The maximum torque value, averaged over a minimum of SIX readmg5, 15 correlated wlth the compressive 5trength of the concrete by means of calibration curves. 1




The halt-cell potential test IS used for determining the probablhty of active corrosion of steel remforcement ln concrete.3 The test, whlch is standardlzed by ASTM C 876, measures electrical potential dlfferences between anodlc and cathodlc areas that eXlst ln an active corrosion process by means of standard half-celfs. Since the corrosion reactlon ln the concrete is dependent on the amblent temperatl’re, It IS reported that useful readlngs are usually obtained at temperatures ln ex cess of SoC (410F.) 1

The test is performed by connecting the negatlve terminal of a high-Impedance millivoitrneter to the embedded reinforcing steel and the positive terminai ta a halt-cell (Figure 4.12). The halt-cell conslsts of a copper electrode whlch is Immersed ln a copper sulfate electrolyte solution. The electrode 15 connected ta the voltmeter by a lead wlre. The other end of the cell consists of a permeable pad through whlch the copper sulfate solutIOn can make electrlcal contact wlth the concrete. Other types of half-cens such as sllver/sllver chlonde can be used, but copper/copper sulfate IS the most cam mon.

By taking readmgs at multiple locations, an evaluatlon of the corrosion activlty of embedded steel or other metals can be made. If sufflclent readmgs are taken in a predetermmed patturn, equipotentlal hnes can be drawn to create a dlagram whlch resembles a contour map (Figure 4.13) The Isopotentlal hnes are created by connectlng pOintS of equal electncal potential The general pattern of eqUipotentlal contours can readlly Identlfy areas of hlgh corrosion actlvlty and areas whlch are on the verge of developlng corrosion actlvlty.J,16 Recent developments ln survey techniques has made thls process much qUicker and le~s tedlous ta use. An example of thls 15 the potential wheel whlch glves a contlnuous prlnt out of electrochemlcal potenllals rather than spot readlngs on a flxed gnd (Figure 4.14). ”

According ta the current ASTM standards, areas that show potentlals more negative th an ·350 mV are sald to be actlvely corrodmg wlth a probablhty of more than 90 percent Corrosion IS neghgible (Iess than 10 percent probablhty) ln areas where the potentlal is less negative than -200 mV. At mtermediate potentials (between -200 rnV and -350 mV), the state of corrosion actlVlty 15 un certain (Table 4.6) .

Different investigators have assigned dlfferent values to these cnte ria. For example, recent work by Pfelfer et al.,e defined the threshold limlt of corrosion to be -230 mV. Also, work from ConcreteIn- The-Ocean projects in the U.K.’7 found that the corrosion nsk crtteria described in the curren\. ASTM C 876 standard is not applicable to ail corroding structures. On some of the structlJrflS surveyed during the proJect, very negatlve potentials were found with no sign of detenoration. Table 4.7 shows sorne of the results obtained and the associated risks found by breakout anL examining cracking.

It is important to realize that, while potentlal measurements glVe an indication of corrosion actlVity, it does not show the extent or rate of corrosion. Half-cell potential readings should be correlated with data from other test methods (descrtbed elsewhere ln this chapter) to determine the axte.,t and rate of corrosion activity .



As descnbed earlier, the presence of a conductive medium or electrolyte is one of the necessary reqUirements for the c.crrosion process to Inrtiate. Therefore, the rate of reinforcement corrosion in concrete de pends on, among other factors, the capacrty of the concrete ta resist the flow of electrical currents. Since flow of electrlcal current is Inversely proportional ta resistivity, a measure of the concrete reslstivrty is indicative of the hkelihood of corrosion in the reinforcing steel. Experience has shown that a high reslstivity is usually associated with a low corrosion risk and vice-versa.

Measuring the bulk resistivlty of concrete is usually done by Wenner’s method which is standardized by the British Standards Institution es 1881 Part 5. An array of four electrodes placed agalnst the concrete surlace pass a current through the outer two electrOdes using the concrete to complete the eleetncal circuit (Figure 4.15). The voltage drop whieh oecurs aeross the inner two electrodes IS reeorded. The resistivity is caleulated by using an empirical expression relating eurrent, voltage drop, and spaeing of the electrOdes. 1

It should be noted that resistivity measurements are affected by moisture conditions. In dry conditions, half-cell or resistlVlty tests may indicate no corrosion activity even though corrosion may be weil advanced. Tests should therefore be conducted during wet seasons or after the structure IS wetted thoroughly. Caution must be exercised when uSing these methods en posttensioned structures, as test results may not be penlnent ta the condition of the tendons themselves.5 A state-of-the-an revlew of electric potential and reslstlvity test methods is glven by Figg and Massden.



A pachometer survey is usually performed as pan of the detailed Inspection. The pachometer (or covermeter) is an instrument used to locate and map embedded rainforcing steel and ta measure the depth of concrete cover over the rebar. These instruments are commercially available for use in dry environments and are easily adapted for underwater use.

The pachometer typically consists of coils wrapped around U-shaped magnetic cores. A magnetic field is produced by sending an alternating current to one of the colis and measurlng the current which IS developed in the other coil. The magnitude of the measured current is affected by both the diameter of the rebar and the distance from the coils. The concrete surface is scanned with the probe until a maximum meter reading IS obtained, 9ivln9 the location and orientation of the embedded rebar. A maximum meter reading will be obtained when the axes of the probe pales are parallel to and dlrectly over the axiS of a reinforcing bar.4 A display dlal is graduated to indicate the depth of the steel. ‘8 ln general, pachometers are calibrated for rebars ranging from 10 M ta 45 M (ASTM N’I 3 ta NQ 16) in size, and can be used to measure depths of concrete caver ranging from 6 ta 200 mm (V4 ta 8 ln.) ln thickness.4

Other magnetic abjects in the vicinity of the rebar where the measurement is being taken will affect the pachometer survey. It may be un able to dlstinguish individual bars if the rebar is bundled or too close. The effects of parallel25 mm :1 ln.) diamater rebars, IOC2!dd 50 mm (2 in.) below the concrete surface, IS shawn schematically in Figure 4.16. It is reponed that, if the center ta center distance of two parallel rebars is at least three tlmes the thickness of the concrete caver, this effect will be negligible.·



The corrosion of embedded steel reinforcement in concrete is affected by the pH value of the surrounding hydrated cement paste. Concrete normally provides a high degree of corrosion protection to embedded reinforcing steel due to the stabilizing affect of the high alkaline (high pH) envlronment. However, the passlVity of the protective iron oxide film, which forms on the steel surface, can be disrupted by a reduction in the pH value of the pore fluid within the concrete. This usual~ occurs as a result of carbonation, or by the penetration of sufficient amounts of chloride ions.

A relatively simple means of determining the depth of carbonation, is by the use of a chemical indicator. The difference in alkalinity (or pH value) between carbonated and uncarbonated concrete IS indicated by a change in color. This requires spraying the indicator on a freshly broken concrete surface by using a chisel ta chip off the side of a drilled or cored hole.5 Also, it is essential ta ensure that the carbonated surface is not contaminated with dust from uncarbonated concrete. The indicator most commonly used is a solution of phenolphthalein ln diluted ethyl alcohol whlch remains colorless for carbonated concrete and changes to purple-pink wh en contacting uncarbonated concrete (pH> 10).3 Though It IS not as accurate as laboratory testlng, it provldes a good site indication of carbonatlon depth.



The effects of chloride ions on corrosion of relnforclng steel in concrete is weil documented ln the literature. Free chloride ions Increase the electncal conductivity of moisture in the carbonated concrete, and depasslvates the reinforclng steel, thus promotlng corrosion. Therefore, a measure of the chlonde ion content in concrete is indicative of the likelihood of corrosion activrty.

The presence of chlonde ions ln concrete can be detected and measured ln the laboratory by chemical analyses of powdered concrete samples, although sorne simple chemlcal tests have been developed ln the U.K. for site use.20 • 21 Powdered concrete samples are usually obtalned from several depths, extending from the concrete surface ta beyond the outer reinforcing steel. The samples are then dissolved in a chemlcal solution. Stnps of special indicator paper are dipped Into the solution and tt”e helght ta whlch a color change nses glves the chloride content in percemage by mass of concrete. In order to obtain the chloride ion content ln percentage by mass of cement, the cement content must be determlned.3

Values of 0.20 and 0.40 percent chlonde by mass of cement are generally taken as chloride threshold limits for prestressed and relllforced concrete structures, respectively.22 A survey23 by the Building Research Establishment in the U.K. has suggested that corrosion is not likely ta occur if the chloride ion content of reinforced concrete is consistently less than 0.40 percent by weight of cement and I1lghly probable if Il exceeds one percent. Labora~ory procedures avallable for determlning chloride ion content ln concrete include the VOLHARD method and the X-Ray Florescent Spectrometry Method.1 The former is a relatlvely simple chemical test whlch is standardized by British Sta.ldard BS 1881 Part 124. The x-ray method requires speclalized testing equipment.

The depth of carbonation affects the levels of chlonde content in the concrete. For example, the chloride ion profile obtained from a coastal structure in the Middle East (Concrete-In-The-Oceans Projects) indicates that the chio ride content peaked at the maximum depth of carbonatlon within the concrete caver rather than at the surface (Figure 4.17). According ta the report, it appears that the ability of the concrete to bind chlondes is severely reduced by carbonation. 17 Theophilus and Bailey24 dlscuss the importance of carbonation tests and chloride levels in durability analysis of concrete stra 1(‘!IJres.



This is a destructive test which requires removing the concrete coyer to expose and visually inspect the reinforclng steel. This allows the observer to make a visual assessment of corrosion damage. Samples of the relnforcing steel may be removed for laboratory testlng to determine various properties and charactenstlcs, such as steel type, tensile strength and corrosion resistance. 5




The tapping test IS a simple but labor intensive nondestructlve method for locatlng delamlnated concrete. It reqUlres the investlgator to stnke the concrete surface at predetermlned gnd locations. Oelaminated areas are easily detected by a dull sound. However, uSlng a hlghly resonant obJect to stnke the concrete surface may produce sounds whlch may make it dlfficult to distlnguish delamlnated areas from sound concrete. ‘



This method is also used for detecting delaminated concrete and requires the use of four 500 mm (20 in.) long chains, attached to a cross bar whlch ln turn is attached to a metal rod. To perform the test, the assembly of chains is dragged over the concrete surface in a swinging motion. As with the tapping test, a distinctly different (dull) sound is generated when the chains are dragged over delaminated concrete. Currently, thls method is used extensively because it has baen reported to give fairly accurate results and IS relatlvely inexpenslve.’



This method IS capable of detecting low density or honeycombed concrete, microcracking, and delamlnations. The test involves striking the concrete surface wlth a small hammer containlng a load-cell and monitoring the response (veloclty) of the impulse with a geophone. The transducer signais are fed ta a data acquisition system and processed by a PC computer. The valocity graph is divided by force ta glve the mechanical impedance response graph. thus providing information on dynamic stlffness, structural resonance, concrete quality and Integrity.!) The method has also been adapted to detect loss of support or voids beneath concrete pavements, floors, dam spillway linings and runways.’O A typical impulse response method for slabs and pavements is shawn ln Figure 4.18.



The lE methCJd is a sanie test used for evaluating member integrity and thickness. If Is a nondestructive test that only requires access to one sida of the concrete member. It can be used to evaluate the integnty of slabs, walls, bridge decks, dams, tunnel linings, and parking garage decks.’o The lE method was inltlally used in 1945 by Long et al.’ and developed by Carlno and Sansalone2~.26 in the early 1980s.

The lE method involves impactlng the concrete at a point wlth a transmitter. As the pulse travels through the concrete, It IS reflected by Internai defects, such as honeycombing, cracks, or material of different denSlty (Figure 4.19). A transducer “cou pied” ta the surface records these reflected pulses and Indlcates the presence of an Internai defect. The transducer signais are processed through a computer and are dlsplayed on a Sereen. The general shape and height of the pattern on the screen Indicatd~ the type and extent (surface erea) of the defect present (Figure 4.20).27

The lE scanning system is similar to the UPV scanner and uses basically the same hardware (Figure 4.21). In contrast to the UPV scanner, the lE scanner requlres only a single scanner unit which incorporates bath signal source and receiver. The lE scanner consists of an impulse hammer mounted on an electrically driven solenoid. The electrical impulses can be generated automatically or manually by an operator sWltch, which allows testing at various speeds, locations, and data densities.9



Infrared thermography uses remote sensing techniques ta record the heat emisslon from the surface of an obJect. It is a diagnostic tool used extensively for assessing the condition of concrete roadways and pavements, slnce heat emission IS affected by internai defects such as cracking and delamination. Since concrete is not a good conductor of heat, cracking and delamination will create different rates of heat transfer. 1 Delamlnatians are displayed as welldefined white colored areas on the infrared thermagram as opposed to a ·manled grey-white colar” that IS produced by sound concrete .

Heat radiation from the sun can help to produce a more noticeable contrast between sound and unsound concrete. For instance, a difference in temperature of about 1.5°C (2.7°F) will crea!e a clear contrast between the two. Thermographlc scanners currently in use can detect a temperature dlfference of up ta O.2°C (O.36°F). However, a greater accuracy 15 required to compensate for dlfferent heat emlSSlon rates due ta surface fimsh and debns



A quantitative evaluation of the structural integrity of a structure, such as a bridge pier or footir g, can be obtained by determining ilS eigenfrequency (dynamic response) when subjected to an impact vibration test. Since the state of a structure (condition of concrete, condition of bearing stratum which supports thEl concrete structure, etc.) affects!ts eigenfrequency, the integrity of the structure can be judged by companng the measured elgenfrequency with an established standard value. 28

The test, which was recently developed by the Railway Technical Research Institute (ATRI) in Japan,29 involves applying an impact load to the pier by means of a 30 kg (13 lb.) weight suspended from the girder, and its responses (displacement and acceleration) are measured (Figure 4.22). The weight can be separated into several sections to facllitate its transportation . A data acquisition system records and processes the responses ta produce the measured eigenfrequency of the structure.

A judgement of the structural Integnty IS made by a determmation of the ·,ndex of Integnty· whlch is obtalned by dlvlding the measured eigenfrequency by the standard value of elgenfrequency The standard value for the elgenfrequency of the pler ln the direction perpendlcular ta tho bndge axis can be obtalned by empmcal formulae. For example, the standard value for the eigenfrequency of a pler on a spread footing can be obtalned by the formulae shown ln Table 4.8. The Index obtained is then compared to the values ln Table 4.9 for judgement of Integnty.



Statlc load testing IS usually performed as a means for evaluating the load carrying capacity and performance of a structure, or component of the structures. Full scale load tests have been developed for a vanety of structures includlng, beams and girders (ASTM E 529), cladding components (ASTM E 997, E 998), roofs (ASTM E 196, E 695), truss assemblies (ASTM E 73, E 1080), and piles (ASTM D 3966, D 1143). The loadlng is usually applied by hydraulic jacks, mechanical Jacks, air pressure, or other heavy materials. Prior to conducting the test, the component or area of the structure belng tested IS Isolated from the structure to obtain an accurate response.3O

Dunng statlc load testlng, two types of responses are measured: deflection and straln. DeflectlOns are usually measured by defler.tion transducers, deflecting dial gauges, and high precIsion levels or laser equlpment. Strains are measured by the use of standard electromc straln gauges, electronic dis placement transducers, accelerometers, and pressure transducers. JO Stralns are erther recorded manually, wlth a portable straln mdicator, or automatlcally by a data acquisitIOn system.1 Followlng each load test, a detalled examlnatlOn (1 e., crack survey) of the structure or corn panent shauld be canducted 13 Althaugh very expenslve, these tests are very Informative.



There are numerous laboratary test methods far determlning the cause or causes of detenoratlan ln concrete, and many others far determlning Its compOSItion. When ordenng a specIfie laboratory test, It is important ta understand the Intent and purpose of the test whlch IS belng conducted and its signlflcance ta the InvestigatIOn before It IS performed. The lelevant parameters that may cause test results to vary from in-SItu conditions must be c\eally understood.30 It is not the Intent of thls section ta descnbe ail these tests, but rather to descnbe sorne of the types most commonly used when Investlgatlng concrete detenaratlon.



Petrographie analysls uses microscope techmques ta determlne the concrete compOSItIOn, concrete quality, and the cause or causes of dlstress or deterioration. ThiS test procedure, whlch is standardized by ASTM C 856, was anglnally developed ta descnbe and classlfy rocks and has been adapted to Include hardefled cancrete, mortar, grout, Il0rtland cement, and other construction materials.44 The analysis IS typlcally performed on 25 mm (1 in.) dlameter specimens obtalned from the site by core dnlling.

Petrographlc analysis can also be used for estlmating future durabllrty and structural safety af concrete elements. For example, sorne of the Items that can be evaluated bV a petrographlc analysis include cement paste, aggregate, minerai admlxture, and air content; trost and sulfate attack; alkali-aggregate reactlvlty; degree of hydratlOn and carbanatlon; water-cement ratio; bleedlng charactenstlcs; tire damage; scahng; pOpOlltS, and several other aspects.31



ThiS method IS used ta determlne the detalled nature of the ‘strength-conferring agents’ including the presence and distribution of “rellct” cement grains, and the extent and location of carbonatlOn in portland cement concrete. B The method involves x-ray examination of a paste concentrate made by breaking up some of the concrete. The mortar is removed from the aggregate and is sieved over a 150~ (NO 100) sieve and the material passlng the sleve is ground ta pass the 45~ (Nil 325) sieve, placed in a hOlder, and scanned on a diffractometer. Useful information for interpreting x-ray charts of hydrated portland cement IS found in References 32, 33 and 34.



Cement content tests are valuable tor determining the cause of strength loss or pore durability of concrete.3 ‘ Cement content can be determined by ASTM C 85 and C 1084 standard methods or by the malelc acid or other nonstandard procedures.3S ,38 The standard method is used to determine the amount of calcium oXlde and soluble silica content ln the concrete by performing an oXlde analysls. The cement content is then computed from each component through the use of a mathematlcal relationship, The cement content is taken as the average of the computed values from each component, provided they are within one percent (or 25 kg/m~ of each other. When the two computed values are not within these limits, the lower value is taken as the cement content.’



Sufficlent amounts of sulfates ln hardened cement paste can lead to sulfate attack, causing expansion and disruptlon of concrete. Sulfates can penetrate the concrete from exposure ta seawater or seawater spray, mix water, chloride-containing admixtures, or deicing salts. The sulfate content or sulfate attack in concrete can be Identified using chemical analyses of field specimens, The concrete sam pie is broken up, weighed, and dispersed in a solution of water and hydrochlorlc aCld. It is subsequently boiled and filtered, and methyl rad indicator is added. The solution IS then neutralized by addlng ammonium hydroxide, hydrochloric aCid, and baflum chloride. Next, the sam pie is bolled again and maintained for a period of 30 minutes, after which the sample is left ta stand for 12 to 24 hours. Desiccation of the sample will precipitate a mass of barium sulfate which IS th en weighed. The sulfate content (expressed as a percemage of cement content) is computed through the use of a mathematical expression relating sulfate content ta the mass of barium sulfate produced. If the sulfate content is in excess of three percent, chemlcal attack 15 hkely to occur.’



The amount of air voids ln hardened concrete signlficantly affects ItS permeability. hance its resistance to various deleterious attack mechanisms, such as freeze-thuw action, sulfate attack, and penetration of deicing salts. Accordingly, ACI 318-9237 has set forth minimum air content values for various environmental ~xposure conditions. The various types of air voids that eXlst in hardened cencrete Include: micropores or gel pores, capillary pores, and macropores (Figure 4.23). The capillary pores and macropores are those most relevant ta concrete durablllty.J8 The micropores are those formed within the interparticle spaces in the calcium silicate hydrate and its total volume is considered to be too small to have an adverse effect on the durabihty of concrete.’

The air content and air-void system characteristics of hardened concrete can be determined by ASTM C 457 test procedures. There are three standard test methods which are normally employed to determine air void content: Linear Traverse (Rosiwal) Method, Point Count Method and Modified Point Count Method.’ The tests typically involve microscoplc examlnation of concrete samples removed from the structure. The information obtained ‘rom these tests also include the volume of entrained air, its specifie surface, paste content, and spaclng factor.3 !



For any repair solution to be effective, the factor or factors causing distress or detenoration in the concrete must be clearly unders ood. This often Includes knowing If any cracks in the structure are still moving as load or temperature changes. This Information can most often be obtained by penodlc visual inspection or by employing passive monitoring methods. These Include devices such as crack momtoring devices and other movement measurement instruments. These are discussed below and are malnly adapted from a review of Reference 5.



This devlce IS the simplest of ail the crack mOnitors and consists of th in glass plates which are glued across the crack. Any subsequent crack movement will break the glass.5 Although, this method is easy to use, it does not provide an indication of the extent and direction of crack movement. A more sophistlcated Instrument IS the Avongard crack monitor which gives a direct reading of crack dis placement and rotation (Figure 4.24).39 The unit consists of two acrylic plates; one IS etched with a fine grid pattern and the other is marked with a cross-hair. The plates are glued on bath SI des of the crack, wlth the cross-hairs overlapping and the grid centered on the crack. Any movement of the crack can be measured from the position of the cross-hairs on the gnd.5



Crack movement can also be monitored and amplified (50 times) by the l’se of a mechanical crack indicator, shown in Figure 4.25. This device has the advantage of indicating the maximum range of movement occurnng dunng the monitoring period.39



Strain gages can be erther electrical or mechanical, and are used to monitor slow movements of cracks caused by load and temperature changes.5 To obtain a more detailed time history of the crack movement, a wide range of transducers (Unear and rotary potentiometers, and LVDTs) and data acquIsition systems (stnp chart recorders or computer based) are available.30 • 38



Acoustic emission is used to monitor and detect very small movements near cracks or voids within a concrete structure. These movements produce acoustic sound which can be monitored and recorded by sensors attached to the concrete sUiface. Once a change occurs ln the structure, such as an increase in crack length, a correspond mg change in stress will create a different acoustic sound.5



If it is suspected that settlement of the structure may have caused cracking, more sophlstlcated devlces are available for monitoring movement of earth near or below the foundatlon. Deviees, such as extensometers or inclinometers, are used to monitor the movement of earth over time to determine if settlement is still occurring. If this problem is not corrected, further settlement will continue to cause cracking in the structure, which in turn will make any repalr Ineffectlve .



The integnty of concrete structures ln seawater depends mainly on the presence of surface cracks, which in many cases, lead to corrosion of the reinforcement and serious deterioration to the concrete. Therefore, one of the concerns of underwater inspection is to detect cracks. Traditlonal Inspection methods can only detect damage when the corrosion process has developed extensive cracking which can be visually detected by divers.

Acoustlc inspection techniques do not require divers, can be used in low-visibility conditions, and can perform inspections through layers of sediment or 50ft manne growth. There are two modes of acoustic inspection techniques: sonar or echo sounding from the surface or by a towed underwater vehicle, or ultrasonlcs, “a local high-resolution underwater acoustlc system.·40 The side-scan sonar is a good method for mapplng the general conditions of large areas, such as stilling basin floors. 41 Ultrasonlcs is useful for determlntng concrete deterioratlon when cracking, spalhng and pitting has occurred. ConventlOnal ultrasonlc methods use bulk sound waves whlch are scattered by the aggregates in the con crete. Therefore, ultrasontC techmques that do not penetrate deeply into the concrete are more effective. Two such techniques are the leaky Rayleigh wave and acoustlc mlcroscopy methods.40 • 42 . 43 Ta apply these methods remotely, a hydrauhc mampulator has been developed which can be mounted on ROVs. The basIc pnncip’es governlng the operation of these two techniques are summanzed below.



This method employs Rayleigh waves for detectlng and measunng crack depth. By directing an ultrasonlc beam ta the concrete surface at a speclfic angle, a Rayleigh wave is generated and propagates along the surface of the concrete. The presence of a surface crack will affect the Rayleigh wave propagation and split it Into vanous components: a reflected Rayleigh wave, a longitudinal body wave, and a Rajlelgh wave which travels down along the crack wall (Figure 4.26). When the wave reaches the crack root, diffraction occurs and a Rayleigh wave travels upward along the far side of the crack and out towards the recelver. A wave is also reflected dlrectly to the recelving transducer. These waves do not penetrate deeply into the structure and will only approximate the location and depth of the crack .



Acoustic microscopy is used in conJunction with the leaky Rayleigh wave method to determine the width of the crack and Ils trajectory. This involves scanning a concrete surface wllh a transducer which emits a highly focussed ultrasound beam (Figure 4.27). The reflected image which is produced is simllar to a photograph, except it is less affected by turbid water, soft manne growth. and loose debris in the crack.

ln scannlng the surface, the beam focal point is located on the concrete surface and reflected back agaln to the same transducer. A short pulse signal is emitted immediately after transmitting the beam 50 that it can use the transducer as a receiver. By scanning the surface along closely spaced parallellines, an image of the concrete surface can be constructed. The width and length of the focal point depend on the transducer diameter, curvature (focal distance), and frequency of the ultrasound used .



The best quality repair of underwater concrete structures can be performed in dry conditions after dewatenng the structure. This can be usually accomplished by pumping the water from a steel sheet pile enclosure or cofferdam built around the structure, or portion of the structure being repalfed. However, in some cases, dewatering is impossible, expenslve and often pohtically sensitive. The U.S. Army Corps of Engineers reports that dewatering costs assoCiated with the repalr of the underwater portions of concrete hydraulic structures average approximately 40 percent of the total repair costs. ‘ ln these cases, It becomes necessary to repalr or place concrete in submerged conditIOns.

Many of the techniques avallable for above-water repairs can be easily adapted for use under water. However, materials used for dry repairs often cannat be used under water.2 As a result, there has been considerable effort in the past 15 years by government agencies and specialist contractors toward developing effective and affordable techniques for placing concrete under water.3

The major factorl.l whlch must be considered when developing an underwater repair scheme are summarizcd below:

  • The cast of carrying out underwater repairs is much greater than for similar repairs performed in dry conditions. The work carried out at the site should be as simple as possible.
  • Surface preparation of the damaged concrete requires special techniques ta ensure that the repalf surface IS not contaminated before placlng the repair concrete.
  • The repair material used must be able to cure under water.
  • Special formwork and placement techniques must be considered to prevent or minimize mixlng between the repair concrete and water.


The cause and extent of detenoratlon estabhshed dunng the condition evaluatlon of the structure, site logistics, and cast Will dlctate the method of repair. The repair technique selected must be designed ta suit the specific site conditions and meet the clrent’s needs and budget constralnts . ln sorne cases, it is necessary to perform laboratory and site tests on both repalr methods and materials to Identify potential problem areas. No one techmque Will be most efficient and cost effective for ail underwater repalr Jobs. The materlal ln thls chapter has been adapted trom a review ot dlfferent avallable reterences, especlally 2. 12 and 38.



Betore a repair operation is performed, the damaged area of the structure must be cleaned ta allow a detalled inspection by divers or ROVs. ThiS IS necessary so that an accurate assessment of the damage can be made and an effective repalr program can be prepared. The lirst step ln the repalr Will be to remove ailloose and unsound con crete, and severely detenorated or dlstorted reinforcing steel.

Rernovlng concrete and cutting reinforcing steel underwater IS more complex than performmg the same tasks above water. The nature of the underwater work Will often dlctate the selection 01 cutting equipment. For example, the thermic lance (Section 52.25) IS capable of cutting relnforcing steel and concrete at the same tlme, whlle high-pressure water Jets can be used ta rem ove only concrete.2 The followlng sections outhne the various techniques and equipment most commonly used tor prepanng underwater concrete surfaces for re~dlr and are adapted from References 2. 4 and 5.



Underwater cleamng 15 often necessary to remove marine growth to facllitate the inspection and ta be able ta define the eXlent of damage. It 15 also required to ensure a goad bond between the substrate and the repalr concrete. Typlcally, the repalr concrete, type and amount of manne growth, and accessibility of the concrete surface serve as a gUide for selecting the J}roper clean· ing equipment. 6 Hand-held or mechamcal wlre brushes, needle guns or scabblrng tools are good far cleaning small areas whlle for large areas, a high-pressure water Jet Will be more effective. If the marine growth present IS hard or the concrete surface IS contamlnated wlth ail, adding an abraSive slurry or detergent ta the water Jet will improve the cutting ablhty of the tool.



Once the area and extent of damage has been defined, the cracked and detenorated concrete can then be removed. The method selected must ensure that the remalning concrete and relnforclng steel IS not damaged. The following is a summary of techniques which can be used ta remove concrete and reinforclng steel.



ThiS method is used extenslvely for performing underwater work. The high-pressure water tool uses a thln Jet of water dnven at hlgh velocitles to remove the hardened cement paste mortar. The system operates ln the same way as that used for cleaning concrete surfaces, except that water IS delivered at much hlgh~r pressures (typlcally between 5000 and 30,000 pSi).2 The diameter of the nOllle onfice, and the water pressure at the nOlzle determine the flow rate of the jet. The depth of the eut mainly depends on the number of times the water jet is passed over the surface of the concrete.6 The water Jet can elther be frame-mo;Jnted and automated, or portable. If properly used, the water Jet can be used ta eut Irregular shapes with minimal damage ta the remalning concrete and steel relnforcement.” However, a high-pressure water jet is potentially dangerous as It will easlly remove bone and muscle.7 . 8 Therefore, it should be used only by expenenced divers or operators.



The concrete splitter IS a pneumatic or hydraulic expansive device that is used to break concrete Into sections. Expanding cylinders are Inserted into drilled boreholes along a predetermined plane and pressunzed untll splitting occurs.2 • 4 The device, shawn in Figure 5.1, consists of a hydraulic system and a splitter. It contains a plug, feathers, cylinders, a piston, a cam man ding valve, and a control lever. Several splltters can be used simultaneously with one hydraulic system.

The pattern. spaclng. and depth of the hales the orientation of the feathers and the number of splitters will determine the directIOn and eXlent of crack planes that develop. The spacing of the hales is determlned on the basls of the percentage of steel reinforcement present in the concrete. The dlameter of the hales range from 30 mm (1-3/16 in.) to 45 mm (1-0/. in.) and minimum hale depths range from 300 mm (12 in.) ta 660 mm (26 in.), depending on the type of splitter used.4

The advantage of using the concrete splitter is that it can be used to pre-split large sections of con crete for removal. The splitter IS safe to use and limited sklll is required by the operator or diver. The main disadvantage IS that the depth to which it can remove concrete from mass structures IS hmited. Also, secondary methods of removal are often reqUired to complete the work4



Recent developments have demonstrated that removlng concrete with expansive agents can be less costly and as effective as using the concrete splitter. The agent (or cement) is mlxed to a slurry form with water and poured into plastic bags. The bags are then placed mto pre-dnlled boreholes within the concrete by divers. As the slurry solidifies and expands, it prodJces tenslle stresses (as much as 30 MPa) that generally exceed the tenslle strength of concrete. Cracking of the concrete will begin to propagate out from the hole over the next 12 to 24 hour penod and may continue for a couple of days before stopplng. Secondary means of breaklng the concrete are also usually required wlth this method to complete the removal.

Expansive cements are relatively safe to use and reqUire hmited Installation skills. When the boreholes are located in areas exposed to ambient temperatures, the agent may freeze (during cold weather) or overheat (dunng summer months), causing it to loose its effectiveness. Its main disadvantage IS it could take a couple of days bAfore pre-splitting becomes optimum.4



The carbon dloxide blaster is a blastlng devlce whleh uses pressunzed carbon dloxlde gas to breakup large masses of matenal. It was flrst used in 1930 for breaking coal, and has since been reportedly used for breaklng concrete, rock, and stone. The blasting device conslsts of a reusable cartndge, sWlteh, electneal cable and power supply. The cartndge, shown ln Figure 5.2, is a hollow steel tube eontalOing pressunzed carbon dloxide and is fitted with a firing head screwed on at one end and a discharge head screwed on at the other end. The cartndges are placed and caulked firmly Into pre-drilled holes at predetermined spacings. The cartridges are then electrically detonated, produclng a mild explosion which breaks the concrete apart. The explosion causes mlnor damage to the remaining concrete.2 . 4

The dlameter of the boreholes range between 55 mm (2-v. in.) and 75 mm (3 in.), and are drilled approxlmately 3 mm (1/. In.) larger than the size of the cartridge being used. Only one cartridge per hole IS recommended. Since the technique is potentially hazardous, highly skilled personnel are reqUired for blast design and executlon of blast deslgn.4



The thermal lance is the slmplest and most commonly used method of cutting. It uses intense heat generated by the reaction between oxygen and mild steel rods to cut through concrete. The system consists of a long steel tube ~lIled with mild steel rods, a flexible pressure hose, an oxygen supply, and an acetylene or propane supply tank. Oxygen IS forced through the tube and Ignlted by the oxygen-acetylene or propane flame. The resultlng reactlon creates temperatures up to 3500°C, whlch allows the tlp of the lance to melt the concrete or relnforclng steel. The resultlng eut is approximately 40 to 50 mm (1-1/2 to 2 ln.) wlde and advances at an apprQ)umate rate of 215 mm2/sec (0.33 in2/slic). The higher the percentage of steel relnforcement ln the concrete, the faster the cutting rate.2 4 Underwater cutting rates decrease sharply wlth Increaslng water deplh and are usually limited to relatlvely shallow depths (60 m maximum) 2



Mechanical cutting tools for underwater work have been used extenslvely for many years. Of the many underwater mechanical cutting tools available, the ones most often used are the hydraulically powered dlamond tipped rotary abraSive saw and chlpplng hammer. The rotary abraSive saw is consldered the most useful because it can cut concrete and relnforclng steel slmultaneously. However, this type of cutting toolls relatlvely slow and can only be used to eut to a limited depth, depending on the saw blade diameter.2 The chipplng hammer IS very useful for removing detenorated concrete coyer to expose steel relnforcement that IS ta be reused in the repalr. Other underwater cutting tools include hydrauhcally-operated drills used for dnlling small diameter holes or taking concrete cores. Conventlonal pneumatic breakers and saws are limited to water depths of about 6 to 9 m (20 ta 30 ft.). However, recent developments uSlng a ·closed hydraulic system” have been successfully employed to greater depths. Diver-operated chlpplng hammers uSlng an oil compression system are not depth senSItive and can be used to a depth of about 6100 m (20,000 ft.).5



Reinforclng steel which IS badly corroded or damaged must be removed and replaced before applying the repair concrete. This Will require the use of underwater steel cutting techniques. Many underwater steel cutting methods have been used in the past and are slmllar to those used on land. Selection of the actual method used depends on how deep the work IS, and on the available equipment and fuel (or power).5 Ali of the known thermal cutting processes and their advantages are hsted in Tables 5.1 and 5.2, respectlvely.

The methods most commonly used for cutting steel reinforcement under water are oxy-fuel (acetylene or hydrogen), oxy-arc, or mechanical cutting. Oxy-arc is the most widely used underwater cutting technique with O)(y-fuel cutting being used only in special applications. New techniques have also been used, either on an experimental basls or for performing actual underwater cutting work. Underwater oxygen cutting requires prlor removal of heavy manne growth, scale and surface rust. 5



These underwater oxygen cutting techniques use intense heat to melt the steel being cut by a process known as “burning”.2 The steells preheated ta its melting temperature and then a highvelocity stream of oxygen is directed at the preheated metal ta produce the cut. The cut is produced as a result of a chemical reactlon between iron and oxygen. The molten metal is blown away by the oxygen stream.

The gases used for underwater cutting with oxy-fuel techniques are the same as those used in air (acetylene and hydrogen). However, due to the instability of acetylene at pressures over about 103 kPa (15 pS;), it is not used at depths greater than approxlmately 10 m (32.8 h.). Therefore • only hydrogen is generally used for underwater cutting. Stabilized methyl-acetylene propadiene (MAPP), has been used to a hmited extent. Propane and natural gas have also been used for underwater cutting, but are not as effective. Oxygen-fuel gas techniques are generally used when eleetric currents, produeed by the oxy-arc system, can cause electrolysis, spark formation, or electrocution.5 A typieal gas cutting torch tlp IS shown in Figure 5.3



Oxygen-arc cutting is simllar to oxyge’n-fuel gas cutting except that an electric arc is used to preheat the steel instead of oxy-fuel gas flames. A high velo city stream of oxygen is forced through an electrode to jet away the molten metal. Underwater oxy-arc cutting can cut steel thlcknesses ranglng from sheet gages to about 75 mm (3 in.).5

Steel tubular electrodes, ceramic tubular electrodes, and carbon-graphite electrodes have ail been used for underwater cutting. The steel tubular electrode was speclfically developed for underwater cutting and is the most commonly used electrode. The main dlsadvantages of uSlng steel electrodes are Its short IIfe and narrow “kert”. The narrow kert makes It difficult for the diver to look for incomplete cuts. These limitations can be overcome by uSlng ceramlc tubular electrode&, but these are brittle and expensive. Carbon-graphite electrodes are also bnttle.:I A typical thermlc cutting torch and steel tubular electrode are shown ln Figure 5.4.



Shielded metal-arc cutting is similar to oxy-arc cutting and can be done with virtualty any klnd of mild steel welding electrode provlded It is properly waterproofed. However, cutting rates wrth the shlelded metal-arc system are much lower than those attainable wlth oxy-arc cutting. The shielded metal-arc system IS especlally effective for cutting cast Iron and nonferrous materials.5



Mechanical cutting methods are usually employed wh en only a limited number of smalt diameter reinforcing bars need to be eut. The most commonly used machines are usually hydraulicallyoperated diamond-tlpped rotary saws. Hand-operated tools, such as boit croppers, have also been used.2



Before carrying out repairs, the causes of the damage or detenoration must be clearly Identified. ln the case of spalled con crete it is important to distinguish between damage caused by scouring or impact, and that caused by corrosion of embedded reinforcement. Each type of damage will require a dlfferent type of repalr procedure. Therefore, once the cause of damage has been determlned, the appropriate repair method can be chosen.

The repalr can be achieved by the use of either portland cement or resin-based materials. Thelr selection depends on the intended purpose of the repair, since they protect concrete in different ways. For Instance, cement-based materials provide an alkahne envlronment for the reinforcing steel which prevents or delays corrOSion, white resin-based matenals prevent the ingress of oxygen and moisture. The selected materlal should closely match the mechanical propertles of the substrate. Although this implies that, using cement-based materials may be more appropriate, resins are more suitable for underwater repairs.2 The following sections describe the surface preparation requirements and the general charactenstics of these two types of repairs.



Preparing thé substrate surface is probably the single most important factor for a successful repair. Applying a sound patch to an unsound surface willlead to failure of the repair, because the patch will spall away by removlng some of the unsound matenal. Therefore, the first stop must be to thorought)I rem ove the unsound or contaminated concrete. The penmeter of the deteriorated concrete area should be saw-cut to a depth ranglng from about 6 to 25 mm (‘/~ ta 1 in.) to provide a neat edge. The eut should be normal to the surface or slightly undercut, for a depth of a least 10 mm W. in.) as shawn ln Figure 5.5. Feathered edges are not desirable and should be avoided as much as possible. Depending upon the depth of detenoratlon, It is usually preferable to expose the full penmeter of the relnforclng steel because It provides a good mechanical anchorage for the patch repalr.

Once ail the unsound concrete is removed, the surface must be given a final treatment prlor to performing the repair. Any reinforcing which IS removed must be replaced with new pieces, either spliced wlth couplers or lapped with existing bars. If reinforcing bars are not available for anchorage, It is otten deslrable to install dowels dnlled and grouted into the surrounding concrete.2 Alternatively, metal fixings can be fired Into the concrete with a velocity-powered underwater stud driver (Figure 5.6). As a final step, the concrete surface and reinforcing steel should be flushed with clean water to remove any dirt, grease, rust or manne growth which may reduce the bond strength of the patch.



Conventional cementitious mortars are susceptible to washing out of fines when they are immersed in water. To prevent this from occurring, special admixtures have been developed which improve the coheslVeness of the mortar and reslst cement washout. Several proprietary grouts have also been developed, based on ‘special cementltious cements and sands with thlxotropic and adheslVe additives’,2 which resist washout when they are poured through water. The mixes are formulated to be self-Ievelling and can normally be used ln thicknesses of 19 mm (0/ .. in.) to 150 mm (6 ln.). For vertical surfaces, the mortar is poured through water or pumped to fill formwork as shown ln Figure 5.7.

Ouick-setting (hydraulic) cements are also sUltable for repairing sm ail vertical spalls or voids.7 A hydraulic cement is a singie-component material that can cure under water because of the interaction of water and the constltuents in the cement. Admixtures can be added ta the hydraulic cement for obtaining speciflc performance requirements (i.e., slow or qUlck-setting).9 Ouick-setting cement is prepared ln small quantities and elther hand placed or tool smeared by the diver. Quick-setting cement can attaln a compressive strength of up ta 41 MPa (6000 pSI) and bond reasonably weil to the eXlstlng concrete provlded the substrate has been properly prepared.7



Conventional epoxy or polyester resin mortars are unsuitable for underwater use. However, with the use of special currng agents, repair mortars have been developed that are insensitive to water and are capable of currng under water. 10 Epoxy compounds are 100 percent reactive, two or t …. ;ee component thermosetting polymers, generally formed by mlxing an epoxy resln and a hardenlng agent (sometlmes referred to as a catalyst). Sometlmes an oven-dried aggregate IS added to the mixture to alter the performance characterrstlcs.9 ln some cases, heavy aggregates, such as barytes are added to the resln 50 that water can be dlsplaced more effectlvely when the mortar IS poured into the formwork.

Thelr good adheslon to concrete, bond durabihty, and variable cure characterrstics over a wlde temperature range make them very versatile and ideal for patchlng sm ail spalls or voids ln concrete under water. Epoxles can be chemlcally formulated to suit the speclfic construction requirements rn terms of performance and environmental condltions,11 but must meet the requlrements of ASTM C 881.

The versatlhty of epoxy formulation is described by Mendis 10 and is shown by the wide range of pro pert les which can be attalned:

  • Physlcal propertles: Low to hlgh modulus
  • Rate of cure: Instantaneous to very long or moderately long cure times.
  • Temperature cure: Cure varies from very low to hlgh temperatures.
  • Water Insensltlvity: Ability to cure under moist conditions or under water.
  • Chemical resistance: Is resistant to solvents, alcohols, ketone, alkalies, bases, organic acids, and inorganic acids.
  • Handling versatility: Low to high viscosities or of gel consistency.


Epoxy mortar compounds must be mlxed immedlately prror to use. Correct proportioning and thorough mixing is essential for a good performance repair. Most epoxy mortar repair failures are due to incorrect proportloning or inadequate mlxlng.2 Thorough mixing can be usually accompli shed by the use of mechantcal mixing devlces, such as drrll motor paddle mixers. If the epoxy mortar constltuents are of drfferent colors, streaking ln the mixture will indicate that mixing is not complete, and should continue until a unlform color 15 achieved.

Epoxles cure by chemlcal reactlon whlch begins Immedlately after the constituents are mixed. The rate of currng or pot hfe of the mixture depends on temperature and time. Pot life is the amount of time after mlxlng for whlch the epoxy mortar can be used before It beglns to set. In general, the pot IIfe decreases wlth increaslng amblent temperature. The normal operatlng temperature range for most commerclally avallable compounds IS between 4°C (40°F) and 32°C (90°F). When the concrete or amblent temperature IS outslde thls range, It may be dlfflcult ta apply and cure the mixture. In these cases, the constltuents can be preheated or cooled ta a sUltable temperature to ensure effective and adequate cunng of the epoxy.1″ The water temperature ln which the repair will be made must also be consldered bec au se It IS usually much colder th an the amblent temperature and may affect the cunng process

Surface preparation and the patch work should be performed ln accordance wlth the applicable requirements of ACI Commlttee 50313 and the manufacturer’s recommendatlOns The proper safety procedures for the use of epoxles should be followed and should be ln accordance wlth the reqUirements of the Federation of Resln Formulators and Apphcators



LaboratoryI5,’6 and field studles’2 . 17 have demonstrated that pressure Injection is a Viable and cast effective method for restoring the structural integnty of cracked concrete, provlded that the crack is dormant (non-movlng) and properly cleaned. Pressure Injection has been slJccessfully used for repamng cracks ln bndge substructures, rlams, 1819 10cks,”O wharves,21 plles,22 and other types of concrete structures. InJection, whose use dates back several centunes,23 involves InJecting a sealant liquid that eventually hardens ln the crack. The matenals currently used for crack InJection are elther cement-based or epoxles, dependlng upon the wldth of the crack and thelr Intended functlon once hardened.24

A general range of crack widths that can be treated by epoxy InJection IS between 0.05 mm (0.002 in.) to 6 mm (1/4 in.). Narrow cracks (0.05 mm/O. 002 ln to 1 25 mm/O.OS in.) require a 10W-VISCOSlty epoxy with a rapld cure tlme. A higher VISCOSlty epoxy can be used ta repair wlder cracks but it should have a longer gel tlme ta avold excessive bUlldup of heat. Tao much heat can cause excessive expansion, resulting ln cracking when the epoxy cools. 12 Cementltlous grouts are suitable for cracks wlder th an 6 mm (114 ln.). However, due ta the nsk of washout of cement, epoxy IS usually preferred. In these cases, a fine aggregate IS added to the epoxy ta provlde a more substantlal filler material and ta reduce materlal cost. 7

Experimental work has shawn that penetration of epoxy adheslve into cracks is affected by crack geometry,24 temperature, viscosity, pot lite, and to a lesser degree, injection pressure.25 For instance, low viscositles, ln the range of 100 ta 1000 cp, and a pot hfe greater than four hours, are typical of high penetration epoxies. Viscoslty must be such that back pressure IS less than about 700 kPa (102 pSI) to prevent the concrete from cracking. Gel time must be long enough so that it does not affect the VISCOSIty drastically. A rapld increase ln VISCOSity may cause difficulty dunng injection. 11

Epoxles with a high modulus (high bonding strength and low elongation characteristics) are generally sultable for Injecting cracks which are stable. Low mOdulus, stress-relieving epoxles (Iower bond strength and high elongation characteristics) are used for injecting moving cracks. ln both cases, the material prevents water trom penetrating into the crack. A typical range of desirable epoxy resin properties are glven by Mendis10 and Bean12 and are included in Appendix K.



The inJection procedure generally consists of drilling holes at close intervals along the crack, installing entry ports, sealing the crack between the ports, and injecting the epoxy or grout under pressure. The injection usually begins by pumping the epoxy resin into the lowest port of vertical or inclined cracks, and the port at one of the ends of a horizontal crack. The pumping continues until a good flow of epoxy emerges from the next higher or adjacent port. The first port is then plugged, usually with wooden dowels, and Injection contmues into the adjacent port and sa on. ln sorne cases, where drainlng the water from the crack is not necessary, injection can be started at the port in the widest crack because Il is easier to fill a narrow crack from a wider portion of the crack rather than vice versa.2tI A typical crack Injection system is shown in Figure 5.8.



The most common method of installing entry ports involves inserting fittings into drilled hales. The fittings are usually surface mounted, however, in sorne cases can be socket mounted (recessed).2tI The hales are either drilled directly into the crack or at an angle to intercept the crack. The inJection ports can be bonded Into the hales with quick-setting epoxy resin ta prevent them from being ejected during pressure injection. The ports should be strong enough ta allow the epoxy to be Injected into the cracks, such as one-way pipe nipples, tire valve stems, and copper tubing 12 or patented packers.27 Ports fabricated from cutting nylon tubing have also been used extensively for underwater injection work. Other methods frequently used ta provide entry ports include bonding a fitting (with a hat-like cross section and a hale in the top) flush with the concrete face over the crack, or omitting the epoxy seal from a smaillength of the crack. The latter can be used with special gasket devices that caver the unsealed portion of the crack and allowepoxy Injection directly into the crack.28



Injection holes for most jobs are 13 mm (% in.) or 16 mm (% in.) in diameter. For massive structures, 22 mm (7~ in.) and 25 mm (1 ln.’ diameter hales are drilled ta intercept the crack al several locations.28 The depth of the hale into or at the intersection of the crack can vary from a minimum of 50 mm (2 in.) to 300 mm (12 in.) for thicker concrete sections. v

Drill hale spacing depends on crack width and ciepth.12 Injection holes are normally spaced from 100 mm (4 ln.) ta 300 mm (12 in.) apart.2 ln some cases, hales can be spaced as much as 1.5 m (5 ft.) apart. In general, if cracks are less than 0.125 mm (0.005 in.) wide, injection port spacing should nct be more than 150 mm (6 ln.). For cracks in members less than 610 mm (2 ft.) in thickness, ports should not be spaced more than the thickness of the member. For cracks that are greater than 610 mm (2 ft.) in depth, Intermediate ports should be installed ta monitor grout flow,12 and ta ensure full depth penetration as shawn in Figure 5.9



After ail the injection ports have been installed, the crack lengths between the entry polms should be sealed ta prevent the epoxy from running out of the crack during injection. The material often used for sealing cracks ln underwater concrete structures is a thixotropic epoxy paste. The epoxy should have adequate bond strength to withstand Injection pressures. A U.S. Engineer Army Waterways Expenment Station (WES) laboratory report12 recommends that a sealant be capable of containing the epoxy resin at an injection pressure of about 690 kPa (100 pSI) for up to 10 minutes.

Prior to Injecting the epoxy resin, thorough cleamng of the crack is essentlal. The method of cleanlng is dependent on the size of the crack and the nature of the contaminants.29 ln most cases, the crack is flushed with a high-pressure stralght-nozzle water let to remove internai contaminants (such as grease or marine growth) whlch can prevent epoxy penetration or Inhiblt the bonding of the faces of the crack. If necessary, water blasting can be comblned with wlre brushing, routlng, or the use of plcks or similar tools. 12 The bondlng characteristlcs of the substrate can also be improved by mixing a bio-degradable alkaline based detergent or speclally formulated chemlcals wlth the blast water.20



There are three types of equipment used for epoxy injection of cracks: a hand caulking gun, a pressure pot, and a dlspenslng machine.12 Wlth the hand caulking gun and the pressure pot the epoxy resin and hardener components must be mixed manually, whereas the dispensir,g machine mixes the components in the system immediately prior ta injection. Although the resin components can be mixed manually with graduated beakers, mlxing paddles, and power drills, the best method of mixing is do ne with dispensing machines. Ali of these injection systems are deslgned for low-pressure injection applications. A good revlew of these three methods Is provided ln Reference 12 with the salient points summanzed below.



Caulking guns are usually employed for small jobs involving low-pressure grouting operations. The standard caulking gun consists of a 325 ml (1/12-gal.) caulking tube wllh a 75 mm (3 in.) tapered plastic nozzle. The epoxy compound mixture is poured into the caulking tube, the cap is placed into the tube, and the cartridge IS then inserted into the gun. About 3 mm (~ in.) to 6 mm (V4 in.) of the tip of the plastic nozzle is eut off and the aluminum seal is pierced. The epoxy resin should be pushed to the tip of the nozzle ta force out the air at the top of the cartridge .

The grouting operation should be started as soon as possible to prevent the epoxy resin from gelling ln the cartndge. The tip of the plastic nozzle is insertec! into the entry port and the epoxy is inJected by squeezing the trigger. Flow of the epoxy can be monitored by watchlng for the movement of air bubbles ln the clear plastic nozzle. If the epoxy-resln mixture in the cartridge starts to generate heat, the pot IIfe IS about to be reached and grouting should stop untll a new cartndge 15 prepared. When the groutlng stops, the caulking gun should be cleaned wrth solvent, if necessary.

The Injection operation can be faclhtated and expedited by uSlng pneumatic-powered hand caulklng guns. The Injection procedure is identical to that with a hand-powered caulking gun except that hydraullcs are used ta dehver the epoxy into the cracks instead of the hand trigger.



The pressure pot apparatus is slmllar ta equipment used for spraying paint, and USf;;S an 8 liter (2·gal.) pressure pot as the reservoir for the freshly mixed epoxy-resin, which is poured into a 2 liter (%-gal.) plastic container, which is then placed in the pressure tank. A flexible rubber feed line is attached to the inslde of the outlet port on the lid of the tank extending to the bottom of the reservoir and the IId IS secured and pressurized. Once the tank is pressurized, the epoxy injection hase can be used to grout the cracks.

The pot uses either compressed air or an inert gas to provide the operating pressure (690 kPa/100 pSI minimum). Ta minimize pressure losses in the system, the injection hose is usually not very long (Iess than 3.3 m/10 ft.), and therefore it must be placed near the injection ports. For this reason, the pressure pot has seen limlted use for underwater applications. The pressure pot should be flushed at the end of each days’ work, or any tlme the injection work is stopped longer than the pot hfe of the mixture. Flushing could be done with methylethyl ketone, toluene, or any other recommended solvent.



Using epoxy dispenslng machines is the quickest and easiest method of injecting cracks. With this method, the epoxy compound is mixed as it is needed, th us, eliminating any concern about pot IIfe. Sevèral types of propnetary dlspensing machines are available which pump the proper proportions of epoxy resln and hardener ta a special intermixing nozzle near the injection port .

Pneumatically-operated, variable ratio dis pensers are most widely used for crack Injection operations. For this system, the epoxy resin and hardeOing agent are placed in separate canisters and the desired pumplng ratio is set. Each component IS then pumped by the proportioning pump ta the mixing nozzle by a remote control sWltch which IS attached ta the feed lines. This allows a diver ta operate the pump as needed while inJecting the epoxy IntO the ports. Epoxy injection continues until one or bath canlsters are empty. When thls occurs, fresh matenals are added ta the proper canisters and InjectIOn can proceed. If Injection IS stopped for any penod longer than the pot IIfe of the material, the complete system must be flushed wlth a solve nt. The system should be also flushed out with compressed air ta ensure that any remalning solvent is removed.

To ensure the dispensing machine is delivering the correct mlx volumes, two control devices are provided: the ratio check device and the pressure check devlce. The ratio check device is connected to the dlspensing machine and bath adhesives are pumped simultaneously through the device during the same time interval Into separate calibrated containers. The amounts pumped are compared ta determlne If the volume ratio is correct. Adjustments should be made if the amounts pumped vary more than two percent. The pressure check device ensures that the proportions are not changing due ta leakage or seepage. The devlce is connected ta the mixlng head and the pressure drop IS mOOitored once the pump is stopped. If the pressure drops more th an 140 kPa (20 psi) in three minutes, grouting should stop until the problem is corrected.



Occasionally, large volumes of concrete are required ta be placed under water, for example, ta repair eresion damage ta dam stllling basins, navigatIOn lock floors, spalled seawalls, or simply ta protect foundations against scour damage. Underwater concrete placement IS often carned out under conditIOns which adversely affect the characteristlcs of the fresh mlx. The quality and resulting durabllity (compressive strength. bond, permeability, etc.) of the concrete will depend on the composition of the mix and the method by whlch It IS deposrted. JO

The major concern in placing conventlonal concrete under water is the washing out of cement fines and sands as the fresh mlx moves through the surrounding water, resuhing in a higher water-cement ratio.31 Therefore it is, therefore, essential ta produce a mix which is coheslVe enough not ta segregate, but adequately workable sa that it can consolidate under its own weight without the need for compaction.32

The resultant demand for higher quality underwater repalrs, due mainly to the high cost and technlcal dlfficulty usually associated with dewatr -lng, prompted considerable research into developing concrete mixtures and techniques surtable for underwater repair work. The following sections provide a summary of the existlng and recently developed techniques for repairing concrete structures under water.



ln many respects, the mix design for undelWater repair concrete is normally designed using the same rules and recommendatlons as would be used for repairing concrete in dry conditions.2 • 33 However, depending on the nature of the repair work and the available resources, certain modifications may be required, as described below.



Different types of portland cement are manufactured ta meet various physical and chemical requirements of specific environments ta which it will be exposed.34 American Society of lesting and Materials (ASTM C150) lists eight types of portland cement, of which Type Il (Type 20: moderate sulfate resistance) is usuallv recommended for underwater concrete. In cases where sulfate exposure is more severe, Type V (Type 50) cement is usually more suitable.

To reduce the effects of washout and maintain a sufficiently low water-cement ratio, a relatively hlgh cement content is needed. A review of the literature7 ,32,35,38,37 indicates that a cement content of approximately 350 to 415 kg/m3 (590 ta 700 Ib/yd3 ) of concrete will be suitable for most undelWater concreting applications. For such rich mixtures, water-reducing admixtures are required to produce a highly flowable con crete wh Ile maintaining a low water-cement ratio. For large repairs, a portion of the cement content (up ta 15 percent) is sometimes replaced by pozzolans, such as fly ash or sUica fume, to reduce the heat of hydration usually associated with rich mixtures. Addition rates in excess of 15 percent can significantly reduce the workability of the concrete and decrease the strength gain.38 Lean mixes of less than 330 kg/m3 (556Ib/yd~ are hlghly susceptible ta cement washout and will probably not be suitable for underwater applications.2

The most sUltable cement content used ln a mix design should be determined by trial mixes performed at the site, and williargely depend upon the particular application (large volume or th in lift) and the method (tremie or pump) used for depositing the concrete.JO For large reinforced concrete repairs, where concrete is placed by bottom skip or toggle bag (Section 5.5.2), mixes are not usually designed. Instead, a corn mon approach IS to use existlng mlx proportions, known ta give the desired compressive strength at the workablhty normally used in the dry, and slightly oversanding the mlx and increasing the cement contem by approxlmately 25 percent. This results in a coheslVeness and workable mlx which d”es not require compaetion and resists loss of cement by washout.33



Ta produce a flowable, self-Ievelling concrete, while maintaining a low water-eement ratio, the use of well-graded rounded aggregate is generally recommended. Properly washed manne dredged aggregates and round river gravels will be most suitable.2 The maximum aggregate size is particularly dependent on the method of placlng the concrete. Pumped concrete, for instance, will require a finer partie le slze than tremie-poured concrete. JO For large unreinforced repairs a 40 mm (1-% in.) aggregate size is usually recommended, whlle for reinforced placements a maximum aggregate size of 19 mm (0/4 ln.) should be used.32

Research by Gerwick39 has shawn that a high sand/gravel ratio is beneficial ta the concrete mix with regard to segregation and washout. Accordingly, Gerwick recommends the use of 42 to 45 percent (by weight) of sand ta the total quantity of aggregate. To obtain a cohesive mix, a sand gradation without the finest particle slze should not be used. If such a sand is used, the addition of fine material, such as fly ash, should be used.30



Fresh, potable water is always an essential Ingredient for producing good quality concrete. This is especially important in reinforced concrete where certain contaminants in the water (i.e., chlorides, sulfates, etc.) not only can affect concrete strength, but also cause corrosion of the reinforcing steel. However, sorne waters that are not fit for dnnklng may also be suitable for concrete.34 Acceptable criteria for water to be used in concrete is provided ln ASTM C 94 specifications.

ln many parts of the world, however, thls restriction can presem major practlcal problems, and often a financial burden. For instance, many parts of the world, such as the United States (California and Florida), the United Kingdom, and France, de pend heavily on sea-dredged sands and gravels, whlch sven after double washing, contain harmfullevels of chlorides. In developing countries, construction is often plagued by shortages of fresh water, which may nead to be imported at a high cost. A recently developed ·seawater concrete process..o may prove to be a viable solution to both practical and financial difficulties imposed by water requirements. The new process makes it possible to mix unwashed beach sands and sea-dredged aggregates with seawater. The procass uses a chemlcally modified portland cement, mixed with complex minerai constituents in ratios that de pend on project specifications and the available materials at the site. The process allows hydration wlth water containing up to 100 grams par liter of salts, which is considerably greater than the 32 grams par liter content of normal seawater. Contractors must be licensed to use the process. The followlng advantages of the process have been reported:40

  • Provldes a protective Inorganic polymer coatlng on the reinforcement
  • Reduces settlng times
  • Improves compressive strength
  • Reduces permeability, shrinkage, and cracking
  • Increases modulus of elasticity



Recent research4′.~ has shown that certain concrete ad mixtures has made it possible to place higher quality concrete under water. Test results show that the incorporation of antiwashout admixtures (AWA) and water-reducing agents produces cohesive, flowable, and abrasion-resistant concrete which resists cement washout, and reduces segregation and bleeding. Weil proportioned concrete containing AWAs can decrease the mass loss of the fresh mixture when dropped through water by three times as compared ta conventional tremie mixes with an equivalent slump.43

AWAs are natural or synthetic water-soluble polymers which physically bind the mixing water in the concrete, th us increasing the viscosity of the mixture. A majority of AWAs consist of microbial polysaccharides, such as welan gum or polysaccharide derivatives, such as hydroxypropyl methylcellulose and hydroxyethyl cellulose.37 Optimum dosage rates of AWAs are small and decrease with a decreasing water-cement ratio. Tao much AWA can significantly reduce the workability of concrete. Studies show that mixes with water-cement ratios from 0.32 to 0.40 require only approximately one-tenth of the amount of AWA recommended by the manufacturer

because their dosage rates are based on water-cement ratios ranging from 0.45 to 0.65.Je The five categories of AWAs, as classified by Ramachandran,44 along with dosage ranges are summarized in Table 5.3.

Since AWAs increase the water demand of concrete mixtures, especlally those with a high cement content and a low water-cement ratio, high-range water reducers (HRWRs) are n8eded to maintain a flowable concrete “Ithout reducing its strength or durability.42 The type of HRWR used affects the washout characteristics of the mixture. For Instance, mixtures containing melamine – and lignosulfonate – based HRWRs have proven to be more resistant to washout th an mixtures containing naphthalene or synthetic polymers. However, mixtures containlng naphthalene improved the abrasion-erosion resistance of concrete more than other HRWRs.41 Some HRWRs and AWAs are incompatible. Cellulose-derivative AWAs are compatible only with melamine-based HRWRs. Many proprietary products are sold with the HRWR and AWA in the admixtures.3I

It is reported that combining AWAs and HRWRs may entrap up ta 15 percent air by volume, resultlng in reduced concrete strength.3I Natural gum AWAs, su ch as welan gum, do not entrap air and can be used wrth either naphthalene or melamine-based HRWRs.37 Alternatlvely, alrdetralning admixtures, such as trlbutyl phosphate or octy alcohol have been used to reduce the air content. 31

The use of pozzolans, such as sllica fume and fly aSh, are frequently used in concrete to enhance durablhty, strength, and adheslon, and are sometimes added ta improve washout resistance of underwater concrete. 37,3I,~ A portion of the cement is sometimes replaced with pozzolans of high fineness ta minimlze expansion due to alkali-sihca reaction and sulfate attack.45 admixtures, are sometir·-:es used to produce flowable mixtures, however, they do not adequately prevent washlng out of fines and cement. 38 The proper type and dosage of admixtures should be determined by trial batches at the site prior to the beginning of any concrete placement. A list of admixtures for use in concrete along with the applicable ASTM specification under which they are standardized is provided in Appendix L.



As underwater concrete must be able to compact under its own weight, the fresh mix must have a high workability (slump) and possess good flow characteristics. Ta achieve this, a slump of 150 to 200 mm (6 to 9 in.) is commonly used. For heavily reinforced repairs or when concrete must flow over long hOrizontal distances, a slightly higher slump may be required.32 The degree of workabihty of the concrete also depends on the method chosen for placing and finishing.3I

Research conducted by Heaton48 concluded that for concrete with slumps from 150 ta 200 mm (6 to 9 in.), there .s virtually no difference between the compressive strength of compacted and uncompacted concrete. The research revealed that uncompacted concrete placed under water can produce compressive strengths of about 35 to 50 MPa (5000 to 7200 psi). The results of the compressive strength of concrete as a function of slump and degree of compaction are shown in Figure 5.10 .



A review of the literature2 ,30.33,41 reveals that there are several techniques used for placing concrete under water, some of which have eXlsted since the turn of the century. In the earliest applications, massive volumes of concrete, where high compressive strengths were not requlred, were successfully placed under water uSIOg the weil known tremie method. Variations of the tremie method, such as the pumplng and hydrovalve methods, were later developed and used extensively ln Europe. Although these methods were designed to prevent cement washout, they did not reach thelr full potential unt!! the relatively recent development of suitable admixtures (AWAs) to minimize this problem. For many applications ln Europe and Japan, pumped con crete has become the preferred method over the traditional tremie pipe

Other underwater placement methods, such as the skip box and the recently developed tllting pallet, allow the concrete ta free fall through the water. These methods rely on the use of AWAs ta prevent cement washout. The tremie method and pumped concrete “are designed ta protect the concrete from exposure ta the water”.41 The method chosen for placing concrete underwater must not create tublulence sa that the contact between the concrete and the water IS mlnimized.2 The followlng sections descrlbe the possible methods for placlng concrete under water and are summarized ln Table 5.4.



For many years concrete has been successfully placed under water using the tremie method. It is best suited for placing large volumes of highly flowable concrete. This method allows concrete to be placed tram the surface ta the exact underwater location by the use of a tremie pipe. The pipe IS connected to a hopper into which the concrete is deposited by skips, belt conveyor, or by pumping. The lower end of the tremle pipe is kept immersed in the freshly placed concrete ta prevent the concrete whlch flows out of the pipe from intermixing with the water.30 • 32 However, with the use of AWAs, thls requirement may not be as cntical for ensuring a successful underwater repair. The following Important factors must be considered when placing concrete by the tremie method:

(a) Tremie Equipment. There are three possible configuratIOns for the tremle plpe.9 constant length pipe which IS ralsed as concreting proceeds; pipe whlch IS made up of a number of sections (with flanged and gasketed JOints) whlch are dlsmantled dunng concretlng, and telescoplc pipe (anached to the hopper IS a pipe of sm aller diameter than the actual concrete placing pipe).3o The pipe and pipe JOints must be strong and watertight. Typlcally, a steel tremle pipe is used, but a rigld rubber hose9 or a flexible Plpe2 could be used Instead An alumlnum alloy tremie pipe should not be used because It can produce an adverse chemlcal reactlon wlth the concrete.9 The tremie pipe should have a smooth Inside surface and be of adequate crosssection for the size of aggregate to be used.

Tremie pipe diameter usually ranges from 200 to 300 mm (8 to 12 ln.), though diameters as small as 150 mm (6 ln.) and up to 450 mm (18 ln.) have been occaslonally used. A tremle pipe diameter of 150 mm (6 in.) IS commonly consldered as the minimum for 19 mm (0/4 ln.) aggregates and 200 mm (8 ln.) as the lower hmlt for 40 mm (1-% in.) aggregates,33 and should be at least elght times the maximum coarse aggregate slze.9 Smaller diameters may cause pipe blockages, however, 100 mm (4 ln.) dlameter tremle pipes have been used for small repalrs.32

The hopper is used ta provlde a steady flow of concrete down Into the pipe, and should be large enough to enable the level of the concrete ln the hopper ta remaln constant. 33 The pipe and happer assembly is usually supported by a crane whlch can control vertical and hOrizontal movement of the tremle pipe. A typical tremle pipe arrangement is shown ln Figure 5.11.

(b) TremIe Seal. The tremle pipe is positloned over the area to be repalred Witt! the lower end of the pipe resting on the bottom. Various methods have been used to prevent Intermlxing of conerete with water in the pipe. Steel plates or wooden plugs are frtted to the end of the pipe, when the “dry plpe” method IS used for startlng the tremle pour. As the pipe is lowered to the bottom, the hydrostatic water pressure seals the gasket and keeps the Interier of the pipe dry. Once the tremie pipe IS filled wlth concrete, It IS raised shghtly (usually no more than 150 mm/6 ln.), allowlng the end seal ta break. Concrete flows out and aeeumulates up around the mouth of the pipe, creating a seal.32

For deep water dpplicatlons, the buoyancy of the empty pipe may be a problem during posltionlng. For this reason, the “\Vet pipe” techntque is more commonly employed.32 ln this method, a travelling plug is inserted at the top to act as a barrier between the concrete and the water. The water in the pipe is then pushed out as the weight of the concrete forces the plug ta the bonom. Once the plug reaches the bonom of the pipe, it usually floats back to the surface once the tremie is lifted. Foam plastic (or rubber) and inflated rubber balls have baen frequently used as a travelling plug.33 However, an inflated rubber bail may collapse at depths greater than 7.6 m (25 ft.) and may not be effective as a seal.32 To resolve this problem, wooden spheres, made trom low denstty wood, such as pine,·7 or a wad of burlap have been used.8

(c) Placmg the Concrete. Once concrete placement has started, the mouth of the pipe should remain buried about 1 to 1.5 m (3 to 5 ft.) deep in the fresh concrete. Concrete placement should be as continuous as possible making sure that the level of the concrete in the hopper is kept at a constant height to ensure a smooth continuous flow.32 The concrete flow rate in the pipe 15 controlled by raising and lowering the tremie.2 Ali vertical movements of the tremle pipe must be slowly and carefully do ne to prevent loss of seal. The volume of concrete being placed during the tremie operation should be contlnuously monitored to detect a loss ln seal Underruns are indicative of a 1055 in seal because washed and segregated aggregates occupy a larger volume. A noticeable Increase ln flow rate of concrete ln the pipe Will al 50 Indicate 1055 of seal.32

During concrete placement, the tremie pipe must remain flxed horizontally to avold damaglng the concrete surface in place, which cou Id lead to addltionallMance (weak mortar) and IOS5 of seal. Tremie pipes should be closely spaced so that concrete does not have to flow over long distances. Otherwlse, too much concrete surface area will be exposed to water, causlng segregation and formation of IMance. A pipe spacing of two or three times the depth of concrete being poured has been suggested.32 Distributing the concrete horizontally is either accomplished by flow of the con crete Itself or by stopping and repeating the process, Including reestablishing the tremie seal.

When depositlng large volumes of concrete, two methods are used to spread the concrete horizontally: the layer method and the advanclng slope method. In the layer method, the entire area is concreted at the same time using several tremie pipes, keeping a level surface as the concrete rises. With the advancing slope method, the area is concreted one section at a tlme by movlng the tremle pipe to adjacent areas.32 A single tremie pipe can usually ~oncrete an area of about 30 m2 (300 ft.~.33

(d) Flow Pattern. It was traditionally believed that dunng a tremle pour, concrete flows under and is protected by previously placed concrete. However, a recent study has shown that tremie concrete may produce different flow patterns which may expose more concret~ to water than was originally perceived. The study also concluded that the flow pattern IS affected by the shear characteristics of the fresh concrete. Two different flow patterns were observed: layered and bulging flow.-

A layered flow pattern was associated with concrete having a hlgh internai shear resistance. The new concrete flowed up and around the pipe and then outward over the previously placed concrete, producing very steep slopes. This created a significant amount of laitance at the far end of the pour. The bulging flow pattern produced a more uniform displacement of the concrete, resultlng in much flatter slopes and less formation of laitance. This preferred flow pattern was reportedly made possible by reducing the internai shear resistance of the fresh concrete.48

Since the flow pattern seems to be related to the shear properties of the fresh mix, It is important to produce a highly flowable concrete. For instance, conventional tremie concrete mixes did not perform the best in the studies. The studles found that replacing up to 50 percent of tne cement with fly ash improved the performance of the tremie concrete. Aiso air-entraining agents and some water-reducers decreased the shear resistance of fresh concrete, producing the preferred type of flow pattern.48 A sample of a good mix proportion and a typical aggregate gradation are shown in Tables 5.5 and 5.6, respectively.



Pumping concrete is an extension of the tremie method. Recent improvements in the design of concrete pumps33 and development of AWAs has made pumping the preferrecl method of placing con crete under water.37 It provides the most expeditious means of placing concrete under water in areas of IImited or difficult access, such as beneath piers .

Pumping concrete offers several advantages over the tremie method:

  • Concrete can be deposlted directly from the mixer into the formwork.
  • Concrete can be pumped to the bottom of the formwork to displace the water through a vent at the top2, or by inserllng the end of the pipe or hose Into the form from the top,T avoiding free-fall of concrete through the water.
  • Concrete is delivered under pressure rather th an fed by gravity, sa blockages in the pump line can be easily corrected.37
  • Use of a crane boom affords more precise positioning of the concrete during dlscharge. 37 .
  • Deposlting the concrete under pressure reduces the need to constantly lift and free the tremie pipe. This reduces the nsk of segregation within the concrete.


For small-volume pours, small-dlameter (50 to 100 mm/2 to 4 in.) pump lines can be easily controlled by divers.37 When using small diameter pump lines, the concrete must be flowable and cohesive enough to pass through the pump wlthout blockage. This usually requires a lower slump concrete than that USed for tremle mixtures. Siumps from 100 to 125 mm (4 to 5 in.) have been used successfully.30·33 However, mixtures which contain too much water tend to segregate and cause blockage ln the hose or pump line. Hlgher slump concrete (200 to 250 mm/8 to 10 in.) has also been used with the aid of AWAs to provide a hlgh degree of cohesion, needed ta prevent washout of fines. For instance, the underwater concrete that was pumped to repair the end sill at Red Rock Dam3e in south central Iowa had a slump of 230 mm (9 in.) and contained an AWA. The concrete was delivered using a 100 mm (4 in.) diameter pump and the mix proportions used are shown in Table 5.7.

Small, rounded coarse aggregates are preferred over crushed stone. If crushed rock is used, the coarse aggregate should have a maximum size of less than one-third the smallest inside diameter of the hose or pipe being used. Porous aggregates, such as expanded clay, foamed slag, pumlce, and many coralhne materlals, should not be used, slnce they tend to absorb water and stlffen the fresh mix.9 If these aggregates must be used, they should be presoaked as described in ACI Commlttee Report 304.2R.49 The properties of fine aggregates (sand) are more important than those of coarse aggregates. The sand should have a relatively high fraction of the finer sizes.9 A typlcal grada.,on of aggregates suJtable for use with a 50 mm (2 in.) diameter pump is shown ln Table 5.8.

Care!IJI planning of pump location and hose routing IS essential before starting an underwater repalr operation. The pump line should be placed horizontally or yertically to prevent the buildup of bleed water ln the pump hne. The hose or pump line should be lubricated with a lean cement slurry before pumplng commences to preyent segregation and blockage. e Segregation of the concrete can also be prevented by forcing a sponge plug into the top of the line to physically support the mix and preyent free-fall between pump strokes. For very deep water applications, this may need to be supplemented with bends in the pipe or hose to break the fall of the concrete .

At the end of the pour, the pump can be contlnued ta flush out weak concrete which may have developed as a result of segregation or washout. Also by closlng the Inlet and the outlet valves, the concrete wlthin the formwork can be pressurized ta mlmmlze bleeding of the mix.2 Concreting should proceed reasonably fast and should continue as long as possible without long delays. Care must be taken not ta pump fresh concrete under concrete which has already begun ta set. 33 A typical pump hne arrangement for underwater concretlng IS shawn in Figure 5.12.



The hydrovalve method, which was developed and first used by the Dutch in 1969,30 is a variation of the tremle method. This method uses a flexible (nylon) hase which is compressed by the hydrostatic water pressure ta dehver the concrete. As the concrete is placed in the upper pan of the hose its weight will eventually overcome the comblned hydrostatic pressure and fnction within the hase, allowmg it ta move slowly down the hase. The “slow and contained movement” of the concrete down the hase helps ta prevent segregatlon.Je The bottom pan of the placing hase is enclosed wllhin a rigld tubular section which IS placed at the desired level of concrete surface.30 The thickness of the concrete IS bU11t in successive layers and can be placed wlth a tolerance of ± 100 mm (± 4 in.).JB A typical hydrovalve apparatus IS shown in Figure 5.13.

An advantage of this method is that Il can place stlft mixtures (havlng a slump less than 140 mm/5-V2 in.) as weil as higher slump mixes usually employed with the tremle method. Il is also relatively simple and inexpensive. Further details on this method IS provided ln Reference 47 .

The Kajima’s Double Tube (KDl) tremie method, which was developed in Japan, is very similar to the hydrovalve method. The KDT method also uses a collapsible tube, but it is encased in a steel tube that has several vertical slits. The slits allow horizontal movement of the tube, often done when resetting the KOT. Field tests show that this method is reliable and inexpensive.3I Further information on the KDT tremie method is given in Reference 50, and the typical procedure used is shown ln Figure 5.14.



The Abetong-Sabema51 and the Shimizu52 pneumatic valves are attached to the end of a concrete pump line (Figure 5.15). The valves are used to control the flow rate and amount of concrete that is placed by, “permitting, restricting, or termlnating” the flow of concrete through the pump line. When the pumping boom is moved, the valve is closed to protect the concrete in the line.·’

The Shimizu pneumatic valve is similar to the Abetong-Sabema valve, but has a level detector attached ta the valve unit. When the level detector senses that the concrete has reached a speclfied thickness, the valve closes and allows the tube to be reposltioned. This method is currently considered to be one of the best methods for underwater repair.3I

There is another type of check valve which is available for use ln pumping underwater concrete. The valve, which fits a 125 mm (5 in.) diameter pump IIne, lB 450 mm (18 in.) long and has only one moving part. The valve is constructed on a “gum rubber reinforced with nylon fabrlc plles,JI and can operate in up to 52.7 m (173 ft.) of water with a maximum line pressure of 690 kPa (100 psi). With this valve, concrete can be placed without immersing the end of the hase in the freshly placed concrete.lI



Underwater concrete can be placed with the aid of bottom-opening sklps (buckets). This method involves filling a bucket wlth concrete above water and slowly lowenng it down through the water and discharging it at the repalr area. To minimize washout of cement, the skip must be equipped with two overlapping canvas flaps, which are pressed aga,”st the top surlace of the concrete by the water pressure. This prevents turbulence as the bucket is lowered through the water. After the bucket is lowered and is penetrated a smalt distance into the already placed concrets, the skip must be lifted slowly so that the discharged concrete does not intermix with the surrounding water as the bottom opens. The bucket should have bottom-opening double doors which can be operated automatically or manually.33 As additional protection against washout, skirts fitted at the bottom may be used to confine the concrete while it IS being placed (Figure 5.16).2

An advantage of this method is that very stlff, dense concrete can be placed If used ln combination with vlbratory or pressure compact Ion methods.38 A slump value between 100 mm (4 ln.) and 140 mm (5-% ln.) is commonly used.30 Since the nature of the skip work subjects the concrete to a greater nsk of washout, AWAs should be used. A recent laboratory study conducted ln Indla concluded that when concrete is placed under water using the skip method, the use of superplasticlzers results in segregation due to the air-entralning effect of the admixture.35 Therefore, the use of HRWRs wlthout the Incorporation of AWAs and air-detraining agents should be avoided when using the skip method for underwater concrete placement. Althougl’1 washout cannot be entlrely eliminated, it can be minimized by applying the following additlonal precautions:30

  • The sklps should be completely fi lied.
  • The sklps should be raised and lowered slowly.
  • The “advancing front” of the concrete should be built from the bottom upwards.
  • A continuous supply of concrete should be provided to prevent layering or washout while waiting for the next batch (delays should not be more than 10 minutes).J3


The main dlsadvantages wlth the sklp method are ItS slow rate of operation and the small volumes of concrete they carry. Also, il may be dlfficult to place the concrete in formwork with small openings. In thls case, divers are needed to control the placlng of the skips. However, the skip method is used best where small volumes of concrete are needed at different locations or where mass concrete IS required to stabllize the foundation of a structure.2



The tilting pallet barge was recently developed by the Sibo group in Osnabruck, Germany.38.41 This method 15 used to place thln layers of concrete in shallow water. The concrete is evenly spread on tlltlng pallets constructed along the deck of the barge and then dropped into the water in a free-fall. The method, which requires AWAs, can be adopted for use in deeper water by lowering a skip with tilting pallets to the repair area .



Preplaced aggregate (PA) concrete is an effective way of repainng concrete structures under water, especialty in areas where placement of convention al concrete would be either difflcult or impossible.2 ,7,9,311 ln this technique, coarse aggregate is placed in formwork and a cementitlous grout is slowly injected under low pressure from the bottom up, displacing water and flfting the voids between the aggregate. The resulting high aggregate-cement ratio and point contact between aggregate particles produces a signlficantly lower shnnkage straln, typicalty 50 to 70 percent that of conventlonal concrete. Bonding strengths of PA concrete to existing concrete surfaces are between 70 to 100 percent of that attainable ln conventlonal concrete.

The grout IS InJected at the bottom of the formwork to prevent the formation of air or water pockets. For this reason, the grout pipes are usually installed before the aggregate is placed and extend to the bottom of the formwork.30 During Injection, they are gradually withdrawn as the level of grout nses (Figure 5.17). Grout pipes may range from 19 mm (‘l, in.)2 to 35 mm (1-311 in.)30 in diameter and are usually spaced no more th an 1.5 m (5 ft.) apart.7 Sounding tubes are often placed alongside the grout pipes sa that the grout level can be monltored during placement. Alternatlvely, translucent panels can be provlded in the formwork so that grout flow can be monltored. :2

If grouting from the bottom reqUires tao great an injection pressure, injection tubes may be built Into the formwork at several levels. In this case, the grout would be Injected at the lowest inlet first and proceed upwards in a slmilar manner as with epoxy the injection method. For small repairs, injection can be done through an Inlet pipe at the bottom of the form.2 When grouting is completed, a pressure of about 70 kPa (10 psi) IS held for several minutes to allow any remaining air and water to escape through a vent at the top of the formwork.9

For thls method to be successful the formwork must be watertight and must be able to withstand the full hydrostatlc pressure of the grout.2 Vents must be provided at the top of the formwork to allow water ta escape as the grout fi Ils the form. If the forms are not sufficiently vented, back pressures will create voids ln the concrete fill.7 To prevent the loss of fines and cement at the top of the grout, the formwork usually completely encloses the aggregate. The top ventmg forms are usually made of a permeable fabric next ta the concrete face, and t’Iacked wlth a steel grillage or wire mesh. This backing is attached to a stronger backing whlch IS made of plywood and perforated steel to allow air ta escape. The formwork IS usually anchored wlth dowels to reslst the uplift pressure generated by the groutlng operation. J8

Selection and proportlomng of materials for PA concrete must be done very carefully. Aggregate must be graded so that grout can f10w easlly between the partlcle spaces. Coarse aggregate maximum size should not be larger than one-thlrd the minimum thlckness of the repalr concrete,~:1 but not smaller than 19 mm (314 in.).2 If aggregates smaller than 19 mm are used, the grout should not cantain any sand ta prevent “bndging” between valds, although thls may cause bleed lens to farm beneath the aggregate particles (Figure 5.18). The \lold content of the aggregate shauld be bp.tween 35 and 50 percent, which can be achleved by uSlng unlfarmly graded aggregate . ..:ach cubic yard of PA contalns about 27 cublc feet (bulk volume) of coarse aggregate, compared with 18 ta 20 cubic feet in convention al concrete.53 A typlcal PA gradation IS shawn ln Table 5.9.

Several proprietary grouts for PA concrete are avallable and normally consist of portland cement, a pozzolan such as fly aSh, fine aggregate, and a grout “fIUldifler”.53 To minimize bleeding, fine aggregate (well-graded zone M sand)2 is generally graded to a fineness modulus of 1.3 to 2.1, with most partlcles passlng a Nil 16 sleve. The fly ash makes the grout pumpable and retards settlng tnne. The grout fluidifier also retards sening time and keeps a low water-cement ratio, usually between 0.42 and 0.50. The flUidifier IS effective because it produces an expansive gas which prevents the buildup of bleed water under the coarse aggregate. The retarder found in most flUidifiers provldes about two percent entrained air, which improves the durability of the hardened concrete. Fluidifier IS normally added at a rate of one percent by weight of the total cementltlous materlal in the grout. 53 Also, adding AWAs to the grout avoids the construction of expenslve formwork.38

Preplaced aggregate con crete can also be made by using epoxy resin instead of cementitious graut. Although it IS signlficantly more expensive. epoxy resin is advantageous for several reasons: Il has a very small partlcle slze (typlcally less than 100pm), variable VISCOSity, and variable pot ,’f~ This results in a material which IS more versatile and can be used with much smaller coars~ :-g”j’egata slzes.2



This method conslsts of lowering small volumes of concrete in bonom-opening canvas bags. The bags are reusable and are sealed at the top wllh a chain or rope and secured with a toggle. When the bag is located over the repair area, the bottom is released to discharge the concrete . Placing concrete with toggle bags involves the same procedures as for using bottom-openlng skips with regards to preventlng cement washout.



ln this method, fresh concrete IS placed Into bags and th en placed under water by divers. This method IS normally used to repair scour, renew ballast,2 or as a temporary repalr measure to se al holes and construct expendable formwork. 33 The bags are made of a strong fabnc (usually hessian) and are usually of 10 to 20 Iiters (2-% to 5 gal.) capaclty when the work IS perform~d by divers. The concrete used has a slump of between 19 and 50 mm (31, and 2 in.) and the maximum aggregate size is approxlmately 40 mm (‘\-% in.). If smaller bags are used (5 to 7 Iiters/1.28 to 1.8 gal.), maximum aggregate size should not exceed about 10 mm (o/lln.) . .IO

The bags are usually placed in a bnck bond fashlon and are half-filled ta ensure good Interlocking to form a solid structure. The cement paste which seeps out from between the weave of the fabric provides adheslon between the Indivldual bags.33

Grout bags were recently used ta change the downstream slope of a small dam to elimlnate a dangerous undertow.54 The bags were made out of polyester and measured 1.8 m (6 ft.) wide by 0.6 m (2 ft.) thick. The length of the bags vaned between 2.1 m (7 ft.) and 7.3 m (24 ft.). To Interlock the bags together 0.76 m (2-% ft.) long epoxy coated bars were forced through them at about 1.8 m (6 ft.) spaclngs .



The maintenance or repalr of the submerged ponion of a wharf or bridge substructure requires a clear understanding of what caused or is causing distress or deterioration. As described in Chapter 2, the causes are many and the sequence in which the various processes occurred are difficult ta determine. Frequently, underwater damage or deterioration originates tram design errors and poor construction practices. Relntorced and prestressed concrete piles supporting marine faCilities or offshore structures are often subJected to severe loadings from winds, waves, currents, Ice, chemlcal and biological attack, and ship impact. Bridge piers and abutments are often underminetJ due to scour caused by floodwaters or changed channel flow conditions.

Evaluating these structures requires consideration of the natural phenomena which are present in a marine environment and the effects these have on the substructure.1 Distress or deterioration which continues may require an on-golng investigation process in order ta obtain a complete understanding of the type, cause, extent, and rate of deterioration that is occurring. This may require a senes of inspection and testing techniques at regular intervals which can provide guidance to the engineer for determlning the optimum repair solutian.

There are not as many underwater repair procedures as there are for work above the water. Cleaning and surface preparation, which 15 essential for a successful repair, is more difficult under water. Underwater construction work is more difficult, slower, and its quallty is less certain than for work above water. Declding on a repair alternative requires expen engineering knowledge, expenence and judgement. Sorne of the procedures available for underwatef repalr and maintenance of wharf and pier substructures are shown in Table 6.1. The basic steps in any underwater concrete repalr are:2

(a) Cleaning the deteriorated concrete surface of a” marine growth

(b) Removal of a” deteriarated cancrete and badly corroded reinforcing steel

(c) Replacing the removed reinfarcing steel with new rebars

(d) Sealing any cracks by epoxy injection

(e) Replacing the concrete section with new concrete

(f) Applying protective surface coatings to the concrete

The following sections provide a summary of some of the repalr methods listed in Table 6.1. The material in this chapter has been adapted from different avallable references. especially 3, 6 and 9.



There are several methods available for pile maintenance and repair and many of them are propnetary in nature and are variations of the same principle. These repair and maintenance methods can be grouped into six general categories: epoxy patchlng/injection, protectlve coatings, encapsulatlon (or wrapping), reinforced concrete Jacketing, partial replacement, and cathodic protection. The epoxy patching/inJection techniques presented ln Chapter 5 are applicable to concrete pile repalr, so they Will not be discussed here.



Surface coatings are usually applied to concrete surfaces to act as a barrier against further deterioration. They prevent the Ingress of corrosive chemicals but are also useful for resisting abrasion and freeze-thaw damage. For a coating system to be effective, proper surface preparation IS Important and the coatlng should have the following minimum charactenstics:3

  • It should have an adhesive strength greater than the tensile strength of the concrete
  • The coefficient of expansion should closely match that of the concrete
  • It should be fairly elastic and reslst creep
  • It should have a long durabllity


There are a great number of surface treatments available today for use in rehabilitating concrete surfaces. Most of these treatments can be categorized as penetrating sealers, coatings, or membranes. 4 It should be noted that some of these coating systems must be applied in dry conditions. In these cases, cofferdams may be constructed around the structure being repaired, sa that the water can be pumped out to m.’llntain dry worklng conditions. Table 6.2, which was developed by Bruner,4 provides a selection guide to concrete surface treatments along with their performance characteristics. The followlng is a summary of the most commonly used surface coatings for protecting concrete piles in a marine environment.



Asphalt coatlngs are highly reslstant to acids and oxidants and can be applied cold with the use of a solvent. Tar is often used for repainng concrete in a marine environment, but it is not very reslstant to acids and bases. It also does not offer a high degree of protection against abrasive action. Both coatings are applied in two coats; the second coat containing a silica filler compound for added stiffness.3 Rubberized asphalt tape has also been used for protecting concrete piles.



Epoxy coatings have gained wldespread use in the past decade pnmanly due 10 thelr good adhesive properties, low shnnkage, and hlgh compressive strength. The epoxy IS a two or three component system wlth 100 percent solids, conslstlng of a hardener and a base reSin Epoxy coatings are generally chemically Inert, Impervlous to water vapor, and mOlsture reslslant. 4 Epoxies are relatlvely easy ta apply (usually wlth a brush or roller) , but reqUire stnct quahty control during mlxing.

Epoxy compounds are temperature-sensitive and, once mlxed, must be applied immediately due to their rapid settlng charactenstics. Epoxy coatings have an Inherent tendency to creep, are impact sensitive, and have a strain incompatibility with the underlylng concrete surface. Since epoxies are impermeable, any moisture trapped beneath the coating may cause Il to blister and peel off if the coating is exposed to direct sunlight or freeze-thaw cycles. Accordingly, epoxles should not be used unies::. the concrete IS capable of wlthstandlng freeze-thaw cycles on ItS own.



This type of coatlng has been used successfully for protectlng concrete surfaces against relnforcement corrosion by preventlng chloride ion penetration and carbonatlon of the concrete. The coating IS a highly elastlc rubber whlch reportedly performs two functions ln chloride contamlnated concrete: It acts as a barner aga.nst further detenoration, and it allows chloride Ions to move freely wlthln the concrete pore matrix to prevent peak concentrations from occurnng at the surface of the rebar. Further details of ItS configuration, behavior, and method of application have been reported by Swaml and Tanlkawa.



Encapsulatlon or wrapplng techniques are often used for repairing concrete piles that have minor surface detenoration and no significant loss of structural capacity. They act primarily as a protectlve barner against funher detenoration and Isolate the pile from the various aggresslve agents causlng the deterioratlon. These repalrs are generally performed as a maintenance Item to extend the service life of the structure. 1 There are several proprietary repalr systems available for encapsulatlng piles which can be grouped Into two basIc categories: Impermeable plastic surface wrapping using polyvinylchloride (PVC) sheets, and molded glass fiber-reinforced plastiC (FRP) Jackets wlth epoxy graut. These are described below.



This type of protection technique IS widely used for timber pile repair but can be easily adapted for repainng or protecting concrete piles. Two commonly used proprietary systoms, one consistlng of a single-unit and the other a two-Unlt barrier wrap are iIIustrated in Figures 6.1 and 6.2, respectively. Commerclally available PVC pile wraps can be purchased in prefabricated sizes to fit many pile sizes and lengths. Both systems require cleaning of the plie to remove ail marine growth and soft surface concrete.

The two-Unlt system conslsts of an upper intertidal unit which starts at least one foot above mearn high water (MHW) level and extends to at least 1 m (3 ft.) below mean low watet (MLW). The bottom unit overlaps the intertidal Unit a minimum of 300 mm (12 ln.) and extends to below the mud line. The closure seam ot the lower unit IS rotated 90° trom the upper unit seam. Polyurethane foam seals are Installed at each end of the InterMal Unit ta prevent the Ingress of water and alr.9 The basic Installation procedure IS as folloW5:

(a) The PVC wrapper 15 placed around the pile and tlghtened by rolllng the ends of the vertical seam uSlng poles and a ratchet wrench

(b) The wrapper is then fastened wlth regularly spaced alumlnum alloy bands alon9 the pile

(c) After the PVC barner 15 installed, the area around the base of the plie IS backfllied wlth bagged concrete. When the pile is surrounded by stones, the base can be backfilled wlth hydraulic cement.


With the single-unit system, the full length of the pile to be protected 15 wrap;:>ed wlth a single jacket. Polyurethane foam seals are wrapped around the pile at the top and bottom end of the jacket. The Installation procedure for installing the single unit system IS simllar ta that of the twoUnit system and IS as follows: 9

(a) The Jacket IS placed around the pile, and the closlng zipper IS started from the top.

(b) After Installing the polyurethane seal at the top, the top nylon strap IS Installed and tightened to secure the jacket in place.

(c) The zipper is then closed contlnuously to the bottom, wlth nylon straps belng Installed at every 1 m (3 ft.) on center.

(d) If the base around the pile has been previously ellcavated, It can be backfllied uSIng the same methods as for the two-unlt system.


Another proprietary system (RETROWRAPS) developed by Cathodic Systems Inc., makes use of geo-synthetic technology and 15 shawn ln Figure 6.3. The system 15 speclfically developed to protect piles in the splash zone and can be used on piles of any geometrical configuration. It 15 reportedly designed to wlthstand the deterioratlng effects of ultraviolet radiation, envlronmental, ozone, and temperature variations for the design life of the system. The system can also be provlded wlth an outer coatlng to prevent bUlldup of marine growth. The system 15 modular and is capable of encapsulating any length of pile. The butt jOints of multiple Unlts are sealed wlth a cummerbund unit to provlde contlnuous encapsulation. The unlts can be removed for the purpose of inspecting the substrate and relnstalled without damage to the system Itself.

Each unit is comprised of an outer geo-membrane bonded ta an inner layer of geotextlle fabric . The outer membrane is constructed of a nylon fabric encapsulated in polyurethane or polyether. The inner fabnc is Impregnated with a thlxotroplc gel that can be used to carry corrosion inhlbltors, biocides or cathodic protection anodes. Once tensloned, the elastlc properties of the membrane generate sufflclent forces to push out oxygen and water from the substrate interface allowlng the gel to form a !10mogeneous contact wlth the pile surface. These forces permit the system to ‘self heal’, If punctured, by forcing the Impregnated gel into the damaged area.

The units are Installed ln a slmllar manner as other types of wraps. At the leading edges of tlle fabric are pockets for Inserting stiffeners used for sealing of the unit. A semi-ngid polypropylene Inner sealing flap ensures a 3600 seal at the loading edges. To facilitate proper installation, tensloning cali pers (Figure 6.4) can be obtalned. Typical specifications for the RETROWRAP pile encapsulation system are included in Appendix M.



Polymer pile encapsulation IS the state-of-the-art method for protecting and resurfacing concrete piles. The system consists of pumping epoxy grout Into rigid encapsulation jackets that are custom fabricated to precisely fit the pile for each job. The encapsulation is highly corrosion reslstant, has a very low permeability, and possesses high compressive, tensile, and impact strengths. They are relatively easy to Install and, if properly Installed, they cali provide a long service IIfe.6

typical polymer pile encapsulallorl system conslsts of two syll1lllC’tncal flbel glnss r ClflfolGl’d polyester or vlnylester Jacket urlltS (Figure 6.5) each wltll a rTllrlllllUm tllll’kllm~s l,f :1 rllfll (l,,, l’) The unlts can be elther rectangular or clrcular ln shape and me SILed silonlly lai DOl ln titi 11-” 1 ;1\l11 than tlle pile belng repalred The Iwo unlts ::Ire JOlnad togetller élrouml tlll’ plie ta fO(lll ,1 l,r~lIi space between the pile and the Jacket Eacr mdlvldual Ulllt Ih frHecl V;lfll pt”‘:ymcl . ,;1 ~wd ,111:-; bOllded to the Intenar of the Jackets ta provlde a ufllform nnnulLs bor_veen th) olle Hl:! ~~l’_’ j. H hl’t The stand-offs are cone-shaped ta mlnlmlze pile conlact area

Once Installed, the bottom of the Jacket IS sealed and the cpoxy Urolll l’, purnfJPd IIno”!J” ail injection port located al the bottom of the Jacket. l he principal compormnts and IIH~ vaIlOll’; phases of a successful pile encapsulatlon system are shawn Ir! FIOUl cc; fj.f) ilnd fi 7, rosPI!ctlVt’ly A standard trealment of the r erTlatntng portion of thp pde abovo ti-lU FRP pekot, wtllcll l’, nft(!11 used by the Flonda State Department of Transportation, 15 Illustrtlted 111 Figure Cl El

Polymer pile encapsulatlOn was successfully used tCi r8fJwr rock borer da: naue to 1.~37 di I(J 1 I,H m (4.5 ft. and 5.5 ft.) outslde dlameter precast, post tcrlsloneu COrlee ete cyllnde’ piles of oJ li l”,!ln located ln the Arabian Gulf 7 Several alternative repalr schemcs W8U: (J’/alu;llr.’r! an,l Il ‘/lf).’-, determlned thal FRP encapsulatlOn (Figure 6.9) was the me st ‘1labla OrIllOn arid “”a~ :::11:.I”üll duu to the followlng reasons: they are sUltable for Installation ln OpfJl1 sua. Ihüy oro Ifllp(:(rn(~dLJII! Ir)

rock borers. they have a high reslstance to mechamcallmpact and abrasion, and their potentially long service life. Although polymer encapsulations have several advantages over other systems, a wide range of problems have been encountered on a number of proJects. Basad on field observations of a large number of marine concrete structures, comb!ned wlth vanous field and laboratory tests, Snow6 has developed a list of causes Influellclng performance and has provlded possible remedies for mlnlmlzing the problems. Below IS a brlef summary of the flndlngs.



Discontinulty of graut was C’bserved at or near the original water elevation at the time of construction. In some Instances, vOids developed in areas at lower elevations. The phenomenon was attnbuted pnmanly to entrapped air as the grout was poured into the jacket from the top. Another principal cause of grout dlscontlnuity IS not provlding an adequate number of stand-offs to malntaln the proper spaclng between the jacket and the pile, thereby rGstricting grout flow. These deflclencles can be easlly corrected by:

(a) Pumplng the epoxy grout Into the jacket from the bottom.

(b) USlng translucent Jackets sa that graut flow can be monitored and the necessary correctIOns can be made before the grout sets.



The most common cause for bond failure between the grout and the substrate was attributed to Improper surface preparation, and the presence of biofilms on the :”ubmerged surfaces whlch begln ta develop Imffit~diately after cleaning. As a result, the fo”owlng precautionary measures are suggested:

(a) Use of proper surface preparation techniques including a suitable pile “anchor profile”. This is usua”y accomphshed by using sand blasting or abrasive ratary tools.

(b) Preparation of substrate, installation of jackets, and pumping of grout !nta the jackets withln the shortest tlme-frame possible, preferably less than 36 hours.

(c) Pumping the epoxy graut !nto the translucent jacket from the bottom up through injection ports provided ln the Jacket.



Bond failure between the grout and the FRP jacket is slmilar to the lack of bond between the grout and the pile. The most probable causes are reported ta be improper preparation of the Inside surface of the jacket, the presence of mold release agent resldue on the jackets, manne biofilms, or inadequate compactlon of the graut dunng placement. In addition to Implementlng preventive measures (h) and (c) ln Section (2), the followlng are also suggested.

(a) The Inslde of the jackets should be roughened at the site Immedlately pnor to Installation, preferably by IIght grlt blastlng.

(b) Use of jackets wlth protectlve liners. Sorne jacket manufactures provlde Jackets wlth a liner that leaves a rough finish on the Inslde surface of the Jacket when It IS peeled oH



The field observations showed that several of the encapsulatlOns falled as a result of Improper mlxing of the grout components. This resulted ln soft spots of uncured matenals whlch easlly peeled off the substrate. To address thls prablem, tile followlng solutions were suggested:

(a) Selection of different color (i.e., black and white) epoxy components. If the components are properly mlxed, they will produce a l!nlform dlHerent coler (grey) wlthout any streaks of the original colors.

(b) The grout should be mlxed and pumped by the ·plural component method· uSlng commerclally avallable dispensing machines.



Polymer encapsulation materials have a much hlgher coeHiclent of expansion than the concrete substrate. As a result, the materials become parllally debonded and ln sorne cases, the FRP jacket is removed Dy wave action. However, when polymer encapsulatlons are properly installed, debondlng caused by thermal incompatlblilty IS apparently ehmlnated. Therefore, to mlnlmlze distress caused by temperature fluctuatIons, the followlng procedures should be adopted:

(a) Implementlng ail the procedures prevlously outhned.

(b) Performlng in-situ tests on the completed encapsulation to ensure that good bond has been achieved.

(c) Installing encapsulatlons dunng the summer season 50 that the polyrner materials will be placed ln a more expanded state and thelr subsequent cooling will create a tightly bonded system. If the encapsulation IS Installed during col der periods, the grout should be heated and warm water should be pumped Into the Installed jacket prior to grout InJection. This will slmulate installation dunng summer months.



Most of the jackets that falled as a result of UV detenoratlon appeared ta be fabricated with glass ‘Ibers. However, ln more recent applications, jackets ‘abncated using a combinatlOn of woven roving, mat ‘Ibers, and an outslde gel coat, have shawn Increased reslstance ta UV detenoration. Translucent jackets 3 mm (Va ln.) thlCk, consistlng of one woven rovlng and two mats plus a gel coat, have passed SOO-hour accelerated weathenng tests wlthout any sigmficaril damage. The following suggestions Will help to mlnimize the effects of UV detenoratlon:

(a) USlng “hand laid up or pultruded jackets· wlth adequate resin caver over the glass fibers.

(b) Add UV screening agents ta the jacket reslns at the tlme of fabrication

(c) For severe UV exposure, coat the completed encapsulatlon with a compatible polyurethane paint to block the UV rays.



As prevlously stated, the most prevalent cause of distress ln polymer encapsulations is debonding of the matenals. To ensure that good bond is aChieved, periodic in-situ bond tests, bath above and below the water IIne can be conducted uSlng the modified Elcometer Bond Strength ïester (Figure 6.10). It IS a field test device that determlnes direct tenslle bond on an isolated section of a completed encapsulatlOn. The tester applles a callbrated tenslle load on a 80 mm (3-Va in.) diameter test “dolly” tt …. t IS glued ta the outslde of the FRP jacket. Prior ta applying the load, the section belng tested is isolated from the rest of the encapsulation by cutting a circular groove around the dolly down ta the substrate surface.

Results of several hundred tests performed wlth the device indicate that the bond between polymer mater/ais 15 usually greater than the bond between the grout and the pile. Hence, a weil bonded system IS achleved when fallure occurs at or below the grout-to-sub5trate interface .



This repalr method IS used fm piles that have undergone slQrllflcanl loss ln cros,> ‘ieCIIOIl and CcHI no longer support the design service load Relnforced eoncmlo Jilckets aro usnd 10 provont further detenoratlon and restore the servlco load capaclty 01 Hw pllos 1 DopefldlllU 1)11 tilt) cJoqllJo of detenoratlon, the steel relnforcement t.sed ln the repalr can bo Wlro Illesh, slancJarcl delorllloci bars (epaxy coated or uncoated), or a cornblnatlon Ttle basIc tllchrllqull IcqUlm<; c!o;ulInn tllo pile and removlng ail delenorated cancrete, Instalhng steel reinfOrCC11TlPIlI drolHHJ IIlc (lalTl”!jlHl areas, placlng a Jacket around the prie, and fllling the annulm spaco IJfJIWUPfl IIlfl IclckPI “fiel tlln pile wlth concrete.

Several propnetary systems for repamng submergeel pllos hava büor! dO’Jülopou but ;IIIIJ<!’,lcéJlly involve the techmque outhned above Most commerclally tlvd!lrJlJio ~3ystems IJSU flUXlblo falme: Jackets, wh Ile ngld farms have also been used extenslvely ‘j For cach syslmn, ca~,tllllJ ‘ … 1 ldorwa!e, concrete around the piles and relnlorclng requlres speeldl toctlP;qum; and propCJ! rlIdtG! lat proportlonlng and selection. Submerged pites can al~o bfl mpr.llreu LJ31rlg ft ln rJl'(J!)/;!·.~ud aggregate and bagged concrete methods.u



An example of a commonly used proprietary system (SeaForms) uSlng flexible fabric Jackets IS IIlustrated ln Figure 6.11. After the pile is cleaned, a wlre reinforcmg mesh IS placed around the pile using 75 mm (3 ln.) PVC spacers to provide a grout space between the pile and the form as shown ln Figure 6.12. The fabnc IS then placed around the pile, the zipper 15 closed, and the form IS secured to the pile at the !OP and bottom wlth mechanlcal fasteners so that It does not sllde down dunng concrete placement. The concrete 15 pumped Into the torm tram the top through openmgs (seacocks) suppt1led in the fabnc usmg a sUltable hose whlch IS extended down to the lowest pOint ln the jacket Dunng concrete placement, the form should be Jostled ta make sure the co.1crete settles unlformly ln the farm. When the farm IS full, the pump hose is remaved and the seacacks are sealed.9 Figure 6.13 illustrates the vanous steps involved in the procedure .



Many types of ngld form systl:::ms have been used for repalring submerged piles. The most common type used by manne contractors IS the spht flberglass-relnforced polyester jacket,ll Illustrated in Figure E\.14. The Jackets are Installed around the pile and locked with a “z-boad” closure. A minimum spaclng of 38 mm (1-‘/2 ln.) IS malntalned between the pile and the wlre mesh relnforclng, and between the relnforclng and the jacket. Relnforclng bands Installed at regular spacings along the length of the pile stablhze the jacket dunng concrete placement. If the repalr area extends below the mud line, a water jet or alrhft (Chapter 7) can be used to excavate the reqUired cavlty for Instalhng a base seal. The concrete can elther be placed tram the top by the tremle method or pumped thraugh a valve at the base of the Jacket. The graut flilis topped off wlth an epoxy cap traweled at a 45° angle. The area around the base of the pile should be backfilled 9

A ngld form system L:Slng split flberboard can also be used. This system IS very slmllar ta the hberglass system descnbed above, and is Illustrated ln Figure 6.15. The forms can be fltted wlth a closure at the lower end and suspended trom the top, or the end closure may be Installed flrst and suspende … from above to support the forms. 9

For piles whlch have deteriorated to the pOint where the structural Intp.grrty IS ln question. the jacket IS relnforced wlth standard deformed relnforclng bars. The Flonda State Department of Transportation has used concrete Jackets rel’lforced wlth epoxy coated rebars. The Jackets are cast by the convention al method uSlng fabncated plywood forms 2 A typlcal relnforclng design is Illustrated ln Figure 6 16.

ln a recent pilot test program at the Port of Oakland ln Cahfornla, a seml-ngld fiberglass tubuli:.1r jacket wlth one longitudinal seam was used to repalr test piles. B The relnforcing steel consisted of 150 mm x 150 mm (6 ln x 6 ln.) welded wlre fabnc and 20 M (15 mm) longitudinal rebars at approxlmately 230 mm (9 In.) spaclng wrapped around the repalr area. The seml-flgld Jacket was flHed around the pile, and posltioned with top and bottom centenng devices. A bottom seal was installed and the Jacket was tightened with steel bolts placed ln closely spaced holes along the longitudinal seam. The concrete was placed from the top uSlng a 50 mm (2 in.) dlameter PVC tremie pipe extending down to the bottom of the repair area. The added advantage ln using thls syst(;m IS that the jackets can be easlly removed and reposltioned for reuse, permlttlng repairs to be do ne ln multiple lifts.



Precast concrete half-cylinders wlth a wall thickness of 75 mm (3 ln.) and relnforcing mesh projecting from the sides and ends of each modular unit have been used successfully for protection of piles.2 Although it was used for protecting tlmber piles, the method can be easlly adapted for protecting concrete plies. The concrete cylinder halves are placed around the pile above the water and the projecting relnforcing mesh IS twlsted together to make a complete unit. The end and side joints are then sprayed wlth concrete (gunlted) and the complete unit IS lowered into the water. A second unit IS made in the same manner and placed on the first one. This procedure is contlnued until the concrete Jacket is Jetted to the deslred depth below the mud line. The annulus which IS formed between the jacket and the pile is th en ~illed with grout.



This method was successfully used for many years by the Port of Oakland as a standard methc j of repair for tlmber piles, but can also be used to repair concrete plles.2 The procedure inv’llves replacing a major portion of the length of the plie with new concrete. A 190 liter (50 gal.) steel oil drum with a hole the size of the plie is eut in the bottom and is fitted around the pile and filled with concrete. Relnforcing can be Installed, If required, and the oil drum is filled with tremie concrete. Polyethylene sheets are usually wrapped around the pile before placing the concrete to obtain a tight, oxygen-free seal adjacent to the pile surface.



This repair alternative is needed when the piles have deteriorated ta the point where they can no longer support any load. In this type of repair, the deteriarated portion of the pile would be removed and replaced with a new load transferring mechanism to restore ils : … 11 ~ervice load capacity. A common technique used by contractors Involves installation of steel pipe jacks between sound portions of the pile and encaslng the member wlth a reinforced concrete jacket, 1 as shawn in Figure 6.17.

ln rare cases, replacing the deteriorated pile wlth a new one may be more economical. One methad which is useful for concrete deck structures consists of cutting a hale in the deck between existing pile locations and adjacent to the deteriorated pile. A new concrete pile is driven through the hole and eut off below the top of the deck, and a concrete cap 15 poured under the deck around the new pile to ensure adequate load transfer. This method is iIIustrated in Figure 6.18.



Cathodic protection (CP) is an electrochemical method used to stop or decrease the rate of steel corrosion, It is frequently used to protect concrete located in seawater by making the embedded reinforclng steel cathodic with respect to the concrete. A direct current is applied between the reinforcement, which aets as the cathode, and a permanent anode mounted into or on the concrete surface, eliminating electnc potential differences along the steel surface. Cathodic protection in a marine environment can be applied in IWO basIc ways: the galvanlc (sacriflcial) anode system, and the Impressed current (inert anode) system.2 These systems are shown schematically in Figure 6.19, and a brief description of each follows. Further background to these systems IS available ln Reference 10.



A galvanic anode system consists of eleetrically connecting a sacrificial anode to th() reinforcing steel and immersing it in an electrolyte (in this case seawater). The potential difference which is created between the anode and the structure cathode, consumes the anode to produce the electric currem which keeps the structure in a cathodic state. Metals with high potemials such as zinc, magneslum, and aluminum have ail been used effeetively as sacrificial anodes. 10



An impressed current system is similar to a battery in which the anodes, made of high-sillcon cast iron or graphite, are conneeted to an external OC power source to produce the electrlc current . The anodes are installed in the electrolyte and connected to the positive terminal of the OC source, while the structure being protected IS always connected to the negative terminal. la This method IS also termed the ‘rectifier type,.2 The resulting impressed current slowly consumes the anodes and is often accompamed by gaseous reaction products. lO



The selection of the speciflc cathodic protection system to be used is often a difficult process which requires careful economic consideration. Besides economic consideration, other determlnlng factors include cathodic interference, lack of power, or current requirements. For instance, in seawater, the current reqUirement is usually 54 to 108 mA/m2 (5 to 10 mA/ft. ~ of exposed area while that of freshwater IS in the range of 11 to 32 mA/m2 (1 to 3 mA/ft.~. Because seawater is a very low resistivlty environ ment, the voltage required to produce the current is also low.2

The galvanic anode system is usually used in lower resistlvity envlronments, whereas the impressed current system can be used in almost any resistlvity environment. Due to concrete characteristics, the impressed current system is usually preferred for cathodic protection of reinforced concrete structures. However, when maintenance or access is difficult, or when OC power is unavailable, the use of sacrificial anodes may be the best solution. la Betore a decision can be made on which type of CP system to use, a complete engineering and economic analysis must be made. The following IS a list of sorne of the advantages and disadvantages of the two systems as reported by Lamberton et al.:2

(a) Galvanlc Anode System


  • It does not require an external power source
  • Adjustment is not required after the proper current drain is determined
  • It is easy to install
  • Cathodic interference is minimal
  • It requlres very little maintenance during the life of the anode
  • The current can be delivered uniformly over a long structure
  • Overprotection at drainage points is minimized
  • It is easy to estimate the cost


  • It has a limited current output
  • It is not economlcal in high reslstivity environments
  • It requires numerous anodes to protect a large structure

(b) Impressed Current System


  • The system can be designed for a wlde range of applied voltage.
  • The system can be designed for a wlde range of current requirements.
  • A simple installation can prote ct a large area.
  • The applied voltage and the current output can be varied.
  • The current drain can be easily monrtored at the rectifier.


  • There is a risk of cathodic interference current from other structures.
  • Its operation is affected by power failures.
  • Electrical inspection and maintenance are required.
  • It is very difficult to estimate cost because of the numerous possible design variations.



Cathodic protection of concrete piles damaged by relnforcement corrosion has baen used extenslvely by the Florida State Department of Transportation. Galvanlc zinc anodes have been successfully used ta protect reinforced concrete piles ln salt or brackish water. The method requires cleaning a I …. :ge enough area on the exposed relnforclng bar to accommodate the zinc anode assembly (Figure 6.20). One anode IS clamped to the exposed rebar for each 2 m (6 ft.) of pile in contact with water (Figure 6.21). The spacing Will vary with the welght of the zinc anode chosen. The normal application consists of using 3.2 kg (7 lb.) anodes.2



Aspects such as the service environment, installation constraints, and any long-term maintenance concerns are ail factors which must be consldered in selecting the type of anode to be used .

Cathodlc protection by sacrificlal anodes using magnesium, zinc, or alumlnum alloys have been commonly used. However, a recent study by de Rinc6n et al.” concluded that magnesium and zinc sacnflcial anodes are not suitable for embedment ln concre~e. The study found that magnesium produce~ a large volume of oxidatlon products whlch crack the concrete ln a short period of time, and that ZinC does not adequately polarize the steel. Alumlnum anodes produce a much sm aller volume of oXldatlon products and protect the relnforclng steel more effectlvely because of therr bener diffUSion propertles. Accordingly, the report concluded that, cathodlc protection uSlng alumlnum anodes, erther embedded in concrete or immersed in water, 15 a feasible method to control the corrosion of chloride contamlnated relnforced concrete in the splash zone.



Recent developments with the impressed current system have made cathodic protection more efficient, durable, and cast effective for protecting reinforced concrete structures. In the late 19505, Impre5sed current systems, whlch wcr~ developed for the protection of reinforeed conerete bridge decks, used hlgh silicon cast iron anodes ln a conductlve :1sphalt overlay. Ta reduce the thickness and thus the welght of the overlay, anodes conslstlng of platlnum-ctad nroblum wlre and graphite fibers were ptaced Into grooves and set ln a eonductlve potympr grout. 12