The use of excessive water in concrete mixtures is the single most common cause of damage to concrete. Excessive water reduces strength, increases curing and drying shrinkage, increases porosity, increases creep, and reduces the abrasion resistance of concrete. In fact, high durability is associated with low water-cement ratio and the use of entrained air.
Damage caused by excessive mix water can be difficult to correctly diagnose because it is usually masked by damage from other causes. Freezing and thawing cracking, abrasion erosion deterioration, or drying shrinkage cracking, for example, is often blamed for damage to concrete when, in reality, excessive mix water caused the low durability that allowed these other causes to attack the concrete. During petrographic examination, extreme cases of excessive mix water in hardened concrete can sometimes be detected by the presence of bleed water channels or water pockets under large aggregate. More commonly, examination of the batch sheets, mix records, and field inspection reports will provide confirmation of the use of excessive mix water in damaged concrete. It should be recognized, however, that water added to transit truck mixes at the construction site or applied to concrete surfaces during finishing operations often goes undocumented.
The only permanent repair of concrete damaged by excessive mix water is removal and replacement. However, depending on the extent and nature of damage, a number of maintenance or repair methods can be useful in extending the service life of such concrete.
Design faults can create many types of concrete damage. Discussion of all the types of damage that can result from faulty design is beyond the scope of this guide. However, one type of design fault that is somewhat common is positioning embedded metal such as electrical conduits or outlet boxes too near the exterior surfaces of concrete structures. Cracks form in the concrete over and around such metal features and allow accelerated freeze-thaw deterioration to occur. Bases of handrails or guardrails are placed too near the exterior corners of walls, walkways, and parapets with similar results. These bases or intrusions into the concrete expand and contract with temperature changes at a rate different from the concrete. Tensile stresses, created in the concrete by expanding metal cause cracking and subsequent freeze thaw damage. Long guardrails or handrails can create another problem. The pipe used for such rails also undergoes thermal expansion and contraction. If sufficient slip joints are not provided in the rails, the expansion and contraction cause cracking at the points where the rail attachment bases enter the concrete. This cracking also allows accelerated damage to the concrete from freezing and thawing.
Insufficient concrete cover over reinforcing steel is a common cause of damage to highway bridge structures. This can also be a problem in hydroelectric and irrigation structures. Insufficient cover allows corrosion of the reinforcing steel to begin. The iron oxide byproducts of this corrosion require more space in the concrete than the reinforcing steel and result in cracking and delaminating in the concrete.
Failure to provide adequate contraction joints or failure to make expansion joints wide enough to accommodate temperature expansion in concrete slabs will result in damage. Concrete with inadequate contraction joints will crack and make a joint wherever a joint was needed but not pro-vided. Unfortunately, such cracks will not be as visually attractive as a formed or sawed joint. Formation of the cracks relieves the tensile stresses and, though unsightly, seldom requires repair. Concrete slabs constructed with insufficient or too narrow expansion joints can cause serious damage to bridge deck surfaces, dam roadways, and the floors of long, steeply sloping, south facing spillways. Such concrete experiences large daily and seasonal temperature changes resulting from solar radiation. The resulting concrete expansion is greater in the top surfaces of the slabs, where the concrete temperatures are higher, and less in the cooler bottom edges. Such expansion can cause the upper portions of concrete in adjacent slabs to butt against one another at the joints between the slabs. The only possible direction of relief movement in such slabs is upward, which causes delamination to form in the concrete. These delaminations are commonly located at the top mat of reinforcing steel. In temperate climates, the formation of delaminations relieves the expansion strains, and further damage will usually cease. In cold climates, however, water can enter the delaminations where it undergoes a daily cycle of freezing and thawing.
Repair of damage caused by faulty design is futile until the design faults have been mitigated. Embedded metal features can be removed, handrails can be provided with slip joints, and guardrail attachment bases can be moved to locations with sufficient concrete to withstand the tensile forces. Mitigation of insufficient concrete cover over reinforcing steel is very difficult, but repair materials resistant to those particular types of corrosion can be selected for the repair. Repairs can also be protected by concrete sealing compounds or coatings to reduce water penetration. Slabs containing inadequate expansion joints can be saw cut to increase the number of joints and/or to widen the joints to provide sufficient room for the expected thermal expansion.
Some of the more common types of damage to concrete caused by construction defects are rock pockets and honeycombing, form failures, dimensional errors, and finishing defects.
Honeycomb and rock pockets are areas of concrete where voids are left due to failure of the cement mortar to fill the spaces around and among coarse aggregate particles. These defects, if minor, can be repaired with cement mortar if less than 24 hours has passed since form removal. If repair is delayed longer than 24 hours after form removal, or if the rock pocket is extensive, the area must be prepared and the defective concrete must be removed and replaced with dry pack and proprietary products such as cementitious prepackaged materials; polymer-cement grouts; polymer grouts; and mortars or replacement concrete. Some minor defects resulting from form movement or failure can be repaired with surface grinding.
There are many opportunities to create dimensional errors in concrete construction. Whenever possible, it usually is best to accept the resulting deficiency rather than attempt to repair it. If the nature of the deficiency is such that it cannot be accepted, then complete removal and reconstruction is probably the best course of action. Occasionally, dimensional errors can be corrected by removing the defective concrete and replacing it with proprietary products such as cementitious prepackaged materials; polymer-cement grouts; polymer grouts; and mortars or replacement concrete.
Finishing defects usually involve overfinishing or the addition of water and/or cement to the surface during the finishing procedures. In each instance, the resulting surface is porous and permeable and has low durability. Poorly finished surfaces exhibit surface spalling early in their service life. Repair of surface spalling involves removal of the weakened concrete and replacement with epoxy-bonded concrete. If the deterioration is detected early, the service life of the surface can be extended through the use of concrete sealing compounds.
Sodium, magnesium, and calcium sulfates are salts commonly found in the alkali soils and ground waters. These sulfates react chemically with the hydrated lime and hydrated aluminate in cement paste and form calcium sulfate and calcium sulfoaluminate. The volume of these reaction byproducts is greater than the volume of the cement paste from which they are formed, causing disruption of the concrete from expansion. Type V portland cement, which has a low calcium aluminate content, is highly resistant to sulfate reaction and attack and should be specified when it is recognized that concrete must be exposed to soil and groundwater sulfates.
Concrete that is undergoing active deterioration and damage due to sulfate exposure can sometimes be helped by application of a thin polymer concrete overlay or concrete sealing compounds. Alternate wetting and drying cycles accelerate sulfate deterioration, and some slowing of the rate of deterioration can be accomplished by interrupting the cyclic wetting and drying. Procedures for eliminating or removing waterborne sulfates are also helpful if this is the source of the sulfates. Otherwise, the deteriorating concrete should be monitored for removal and replacement with concrete constructed of type V cement, when appropriate.
Certain types of sand and aggregate, such as opal, chert, and flint, or volcanics with high silica content, are reactive with the calcium, sodium, and potassium hydroxide alkalies in portland cement concrete. Some concrete containing alkali reactive aggregate shows immediate evidence of destructive expansion and deterioration. Other concrete might remain undisturbed for many years. Petrographic examination of reactive concrete shows that a gel is formed around the reactive aggregate. This gel undergoes extensive expansion in the presence of water or water vapor (a relative humidity of 80 to 85 percent is all the water required), creating tension cracks around the aggregate and expansion of the concrete. If unconfined, the expansion within the concrete is first apparent by pattern cracking on the surface. Usually, some type of whitish exudation will be evident in and around the cracked concrete. In extreme instances, these cracks have opened 40 to 50 mm. It is common for such expansion to cause significant offsets in the concrete and binding or seizure of control gates on dams. In large concrete structures, alkali-aggregate reaction may occur only in certain areas of the structure. Until it is recognized that multiple aggregate sources are commonly used to construct large concrete structures, this might be confusing. Only portions of the structure constructed with concrete containing alkali reactive sand and/or aggregate will exhibit expansion due to alkaliaggregate reaction.
In new construction, low alkali Portland cements and fly ash pozzolan can be used to eliminate or greatly reduce the deterioration of reactive aggregates. In existing concrete structures, deterioration due to reactive aggregate is virtually impossible to mediate. There are no proven methods of eliminating the deterioration of alkali-aggregate reaction, although the rate of expansion can sometimes be reduced by taking steps to maintain the concrete in as dry a condition as possible. It is usually futile to attempt repair of concrete actively undergoing alkali-aggregate reaction. The continuing expansion within the concrete will simply disrupt and destroy the repair material. Structures undergoing active deterioration should be monitored for rate of expansion and movement, and only the repairs necessary to maintain safe operation of the facility should be made.
Freeze-thaw deterioration is a common cause of damage to concrete constructed in the colder climates. For freeze-thaw damage to occur, the following conditions must exist:
Water experiences about 15 percent volumetric expansion during freezing. If the pores and capillaries in concrete are nearly saturated during freezing, the expansion exerts tensile forces that fracture the cement mortar matrix. This deterioration occurs from the outer surfaces inward in almost a layering manner. The rate of progression of freeze-thaw deterioration depends on the number of cycles of freezing and thawing, the degree of saturation during freezing, the porosity of the concrete, and the exposure conditions. The tops of walls exposed to snowmelt or water spray, horizontal slabs exposed to water, and vertical walls at the water line are the locations most commonly damaged by freeze-thaw deterioration.
Damage caused by cyclic freezing and thawing of concrete occurs only when the concrete is nearly saturated. Successful mitigation of freeze-thaw deterioration, therefore, involves reducing or eliminating the cycles of freezing and thawing or reducing absorption of water into the concrete. It usually is not practical to protect or insulate concrete from cycles of freezing and thawing temperatures, but concrete sealing compounds can be applied to exposed concrete surfaces to prevent or reduce water absorption. The sealing compounds are not effective in protecting inundated concrete, but they can provide protection to concrete exposed to rain, windblown spray, or snow melt water.
Repair of concrete damaged by freeze-thaw deterioration is most often accomplished with replacement concrete if the damage is 150 mm or deeper, or with epoxy- bonded replacement concrete or polymer concrete if the damage is between 40 and 150 mm deep.
Concrete structures that transport water containing silt, sand, and rock or water at high velocities are subject to abrasion damage. Dam stilling basins experience abrasion damage if the flows do not sweep debris from the basins. Some stilling basins have faulty flow patterns that cause downstream sand and rock to be pulled upstream into the basins. This material is retained in the basins where it produces significant damage during periods of high flow. Abrasion damage results from the grinding action of silt, sand, and rock. Concrete surfaces damaged in this way usually have a polished appearance. The coarse aggregate often is exposed and somewhat polished due to the action of the silt and sand on the cement mortar matrix. The extent of abrasion-erosion damage is a function of so many variables—duration of exposure, shape of the concrete surfaces, flow velocity and pattern, flow direction, and aggregate loading—that it is difficult to develop general theories to predict concrete performance under these conditions. Consequently, hydraulic model studies are often required to define the flow conditions and patterns that exist in damaged basins and to evaluate required modifications. If the conditions that caused abrasion-erosion damage are not addressed, the best repair materials will suffer damage and short service life.
It is generally understood that high quality concrete is far more resistant to abrasion damage than low quality concrete, and a number of studies (Smoak, 1991) clearly indicate that the resistance of concrete increases as the compressive strength of the concrete increases.
Abrasion damage is best repaired with silica fume concrete or polymer concrete. These materials have shown the highest resistance to abrasion damage in laboratory and field tests. If the damage does not extend behind reinforcing steel or at least 150 mm into the concrete, the silica fume concrete should be placed over a fresh epoxy bond coat.
Cavitation damage occurs when high velocity water flows encounter discontinuities on the flow surface. Discontinuities in the flow path cause the water to lift off the flow surface, creating negative pressure zones and resulting bubbles of water vapor. These bubbles travel downstream and collapse. If the bubbles collapse against a concrete surface, a zone of very high pressure impact occurs over an infinitely small area of the surface. Such high impacts can remove particles of concrete, forming another discontinuity which then can create more extensive cavitation damage.
Cavitation damage is common on and around water control gates and gate frames. Very high velocity flows occur when control gates are first being opened or at small gate openings. Such flows cause cavitation damage just downstream from the gates or gate frames.
The cavitation resistance of many different repair materials has been tested by the laboratories of Reclamation, the U.S. Army Corps of Engineers, and others. To date, no material, including stainless steel and cast iron, has been found capable of withstanding fully developed instances of cavitation. Successful repairs must first include mediation of the causes of cavitation.
Corrosion of reinforcing steel is usually a symptom of damage to concrete rather than a cause of damage. That is, some other cause weakens the concrete and allows steel corrosion to occur. However, corroded reinforcing steel is so commonly found in damaged concrete that the purposes of this guide will best be served by discussing it as if it were a cause of damage.
The alkalinity of the portland cement used in concrete normally creates a passive, basic environment (pH of about 12) around the reinforcing steel which protects it from corrosion. When that passivity is lost or destroyed, or when the concrete is cracked or delaminated sufficiently to allow free entrance of water, corrosion can occur. The iron oxides formed during steel corrosion require more space in the concrete than the original reinforcing steel. This creates tensile stresses within the concrete and results in additional cracking and/or delamination which accelerate the corrosion process.
Some of the more common causes of corrosion of reinforcing steel are cracking associated with freeze-thaw deterioration, sulfate exposure, and alkali-aggregate reaction, acid exposure, loss of alkalinity due to carbonation, lack of sufficient depth of concrete cover, and exposure to chlorides.
Exposure to chlorides greatly accelerates the rates of corrosion and can occur in several manners. The application of deicing salts (sodium chloride) to concrete to accelerate thawing of snow and ice is a common source of chlorides. Chlorides can also occur in the sand, aggregate, and mixing water used to prepare concrete mixtures. Concrete structures located in marine environments experience chloride exposure from the sea water or from windblown spray. Finally, it was once a somewhat common practice to use concrete admixtures containing chlorides to accelerate the hydration of concrete placed during winter conditions.
The occurrence of corroding reinforcing steel can usually, but not always, be detected by the presence of rust stains on the exterior surfaces and by the hollow sounds that result from tapping the affected concrete with a hammer. It can also be detected by measuring the half-cell potentials of the affected concrete using special electronic devices manufactured specifically for this purpose. When the presence of corroding steel has been confirmed, it is important to define what actually caused the corrosion because the cause(s) of corrosion will usually determine which repair procedure should be used. Once the cause of damage has been defined and mitigated, if necessary, proper preparation of the corroded steel exposed during removal of the deteriorated concrete becomes important. Steel that has been reduced to less than half its original cross section by the corrosion process should be removed and replaced. The remaining steel must then be cleaned to remove all loose rust, scale, and corrosion byproducts that would interfere with the bond to the repair material. Corroded reinforcing steel may extend from areas of obviously deteriorated concrete well into areas of apparently sound concrete. Care must be taken to remove sufficient concrete to include all the corroded steel.
The more common sources of acidic exposure involving concrete structures occur in the vicinity of under-ground mines. Drainage waters exiting from such mines can contain acids of sometimes unexpectedly low pH value. A pH value of 7 is defined as neutral. Values higher than 7 are defined as basic, while pH values lower than 7 are acidic. A 15- to 20-percent solution of sulfuric acid will have a pH value of about 1. Such a solution will damage concrete very rapidly. Acidic waters having pH values of 5 to 6 will also damage concrete, but only after long exposure.
Concrete damaged by acids is very easy to detect. The acid reacts with the Portland cement mortar matrix of concrete and converts the cement into calcium salts that slough off or are washed away by flowing waters. The coarse aggregate is usually undamaged but left exposed. The appearance of acid-damaged concrete is somewhat like that of abrasion damage, but the exposure of the coarse aggregate is more pronounced and does not appear polished. Acid damage begins, and is most pronounced, on the exposed surface of concrete but always extends, to a diminishing extent, into the core of the structure. The acid is most concentrated at the surface. As it penetrates into the concrete, it is neutralized by reaction with the portland cement. The cement at depth inside the structure, however, is weakened by the reaction. Preparation of acid-damaged concrete, therefore, always involves removal of more concrete than would otherwise be expected. Failure to remove all the concrete affected and weakened by the acid will result in bond failure of the repair material. Acid washes were once permitted as a method of cleaning concrete surfaces in preparation for repairs. It has been learned, however, that bond failures would occur unless extensive efforts were expended to remove all traces of acid from the concrete. Reclamation specifications no longer permit the use of acid washes to prepare concrete for repair or to clean cracks subject to resin injection repairs.
As with all causes of damage to concrete, it is generally necessary to remove the source of damage prior to repair. The most common technique used with acid damage is to dilute the acid with water. Low pH acid solutions can be converted to higher pH solutions having far less potential for damage in this manner. Alternately, if the pH of the acid solution is relatively high, coatings such as the thin polymer concrete coating system can be applied over repair materials to prevent the acid from redamaging the surfaces. Laboratory tests have revealed very few economical coatings capable of protecting repair materials from low pH solutions. Repairs to acid-damaged concrete can be made using epoxy-bonded replacement concrete, replacement concrete, polymer concrete, and, in some instances, epoxy-bonded epoxy mortar. Polymer concrete and epoxy mortar, which do not contain portland cement, offer the most resistance to acid exposure conditions.
Cracking, like corrosion of reinforcing steel, is not commonly a cause of damage to concrete. Instead, cracking is a symptom of damage created by some other cause.
All portland cement concrete undergoes some degree of shrinkage during hydration. This shrinkage produces multidirectional drying shrinkage and curing shrinkage cracking having a somewhat circular pattern. Such cracks seldom extend very deeply into the concrete and can generally be ignored.
Plastic shrinkage cracking occurs when the fresh concrete is exposed to high rates of evaporative water loss which causes shrinkage while the concrete is still plastic. Plastic shrinkage cracks are usually somewhat deeper than drying or curing shrinkage cracks and may exhibit a parallel orientation that is visually unattractive.
Thermal cracking is caused by the normal expansion and contraction of concrete during changes of ambient temperature. Concrete has a linear coefficient of thermal expansion of about 5.5 millionths inch per inch per degree Fahrenheit (°F). This can cause concrete to undergo length changes of about 0.5 inch per 100 linear feet for an 80 °F temperature change. If sufficient joints are not provided by the design of the structure to accommodate this length change, the concrete will simply crack and provide the joints where needed. This type of cracking will normally extend entirely through the member and create a source of leakage in water retaining structures. Thermal cracking can also be caused by using Portland cements developing high heats of hydration during curing. Such concrete develops exothermic heat and hardens while at elevated temperatures. Subsequent contraction upon cooling develops internal tensile stresses and resulting cracks at or across points of restraint.
Inadequate foundation support is another common cause of cracking in concrete structures. The tensile strength of concrete is usually only about 1.4 to 2.0 MPa. Foundation settlement can easily create displacement conditions where the tensile strength of concrete is exceeded with resulting cracking.
Cracking is also caused by alkali-aggregate reaction, sulfate exposure, and exposure to cyclic freeze-thaw conditions, as has been discussed in previous sections, and by structural overloads as discussed in the following section.
Successful repair of cracking is often very difficult to attain. It is better to leave most types of concrete cracking unrepaired than to attempt inadequate or improper repairs. The selection of methods for repairing cracked concrete depends on the cause of the cracking. First, it is necessary to determine if the cracks are "live" or "dead." If the cracks are cyclicly opening and closing, or progressively widening, structural repair becomes very complicated and is often futile. Such cracking will simply reestablish in the repair material or adjacent concrete. For this reason, it is normal procedure to install crack gages across the cracks to monitor their movement prior to attempting repair. The gages should be monitored for extended periods to determine if the cracks are simply opening and closing as a result of daily or seasonal temperature changes or if there is a continued or progressive widening of the cracks resulting from foundation or load conditions. Repairs should be attempted only after the cause and behavior of the cracking is understood.
If it is determined that the cracks are "dead" or static, epoxy resin injection can be used to structurally rebond the concrete. If the objective of the repair is to seal water leakage rather than to accomplish structural rebonding, the cracks should be injected with polyurethane resin. Epoxy resin injection can sometimes be used to seal low volume water leakage and structurally rebond cracked concrete members. Epoxy resins cure to form hard, brittle materials that will not withstand movement of the injected cracks. Poly-urethane resins cure to a flexible, low tensile strength, closed cell foam that is effective in sealing water leakage but cannot normally be used for structural rebonding. (Some two component polyurethane resin systems cure to form flexible solids that may be useful for structural rebonding.) These flexible foams can experience 300- to 400-percent elongation due to crack movement. It is not uncommon to find that damaged concrete contains cracking not related to the cause of the primary damage. If the depth of removal of the damaged or deteriorated concrete does not extend below the depth and extent of the existing cracks, it should be expected that the cracking will ultimately reflect through the new repair materials. Such reflective cracking is common in bonded overlay repairs to bridge decks, spillways, and canal linings. If reflective cracking is intolerable, the repairs must be designed as separate structural members not bonded to the old existing concrete.
Concrete damage caused by structural overloading is usually very obvious and easy to detect. Frequently, the event causing overloading has been noted and is a matter of record. The stresses created by overloads result in distinctive patterns of cracking that indicate the source and cause of excessive loading and the point(s) of load application. Normally, structural overloads are one-time events and, once defined, the resulting damage can be repaired with the expectation that the cause of the damage will not reoccur to create damage in the repaired concrete.
It should be expected that the assistance of a knowledgeable structural engineer will be required to perform the structural analysis needed to fully define and evaluate the cause and resulting damage of most structural overloads and to assist in determining the extent of repair required. This analysis should include determination of the loads the structure was designed to carry and the extent the overload exceeded design capacity. A thorough inspection of the damaged concrete must be performed to determine the entire effect of the overload on the structure. Displacements must be discovered and the secondary damage, if any, located. Care should be taken to ensure that some other cause of damage did not first weaken the concrete and make it incapable of carrying the design loads. The repair of damage caused by overloading can, most likely, best be performed with conventional replacement concrete. The need for repair and/or replacement of damaged reinforcing steel should be anticipated and included in the repair procedures.