How to Seal Your Driveway
Clean your substrate surface well. Clear all loose debris. Sealed stains will become significantly harder to remove. Loose debris may prevent sealer from proper contact with the substrate.
Always test a small area. Roll on with roller, or spray on with low pressure garden-style sprayer (preferably with a back-roll to assist adhesion and minimize streaks and puddles).
Penetrating sealers may be applied to slightly damp substrates. Penetrating sealers if over-applied may leave white residue, which will not affect the performance of the sealer and will dissipate with normal wear. Penetrating sealers generally do not require a second coat. Re-sealing is as needed consult your product label.
Topical sealers require very dry, clean substrates. Topical sealer applied over damp substrates will result in white hazy film under the sealer which may, or may not, eventually cure away. Topical sealers if over-applied will result in a flaky, bubbled appearance. To remove this condition clean damaged area with mild solvent and re-apply new sealer. Topical sealers will have re-coat times of between 2-24 hours (consult your product label). To apply second coat after 24 hours may require a light sanding as preparation.
How to Seal Your Brick
Clean your substrate surface well. Clear all loose debris. Sealed stains will become significantly harder to remove. Loose debris may prevent sealer from proper contact with the substrate.
Always test a small area. Roll on with roller, or spray on with low pressure garden-style sprayer (preferably with a back-roll to assist adhesion and minimize streaks and puddles).
Penetrating sealers may be applied to slightly damp substrates. Penetrating sealers if over-applied may leave white residue, which will not affect the performance of the sealer and will dissipate with normal wear. Penetrating sealers generally do not require a second coat. Re-sealing is as needed consult your product label.
Topical sealers require very dry, clean substrates. Topical sealer applied over damp substrates will result in white hazy film under the sealer which may, or may not, eventually cure away. Topical sealers if over-applied will result in a flaky, bubbled appearance. To remove this condition clean damaged area with mild solvent and re-apply new sealer. Topical sealers will have re-coat times of between 2-24 hours (consult your product label). To apply second coat after 24 hours may require a light sanding as preparation.
How to Seal Your Patio
Clean your substrate surface well. Clear all loose debris. Sealed stains will become significantly harder to remove. Loose debris may prevent sealer from proper contact with the substrate.
Always test a small area. Roll on with roller, or spray on with low pressure garden-style sprayer (preferably with a back-roll to assist adhesion and minimize streaks and puddles).
Penetrating sealers may be applied to slightly damp substrates. Penetrating sealers if over-applied may leave white residue, which will not affect the performance of the sealer and will dissipate with normal wear. Penetrating sealers generally do not require a second coat. Re-sealing is as needed consult your product label.
Topical sealers require very dry, clean substrates. Topical sealer applied over damp substrates will result in white hazy film under the sealer which may, or may not, eventually cure away. Topical sealers if over-applied will result in a flaky, bubbled appearance. To remove this condition clean damaged area with mild solvent and re-apply new sealer. Topical sealers will have re-coat times of between 2-24 hours (consult your product label). To apply second coat after 24 hours may require a light sanding as preparation.
How to Seal Your Pavers
Clean your substrate surface well. Clear all loose debris. Sealed stains will become significantly harder to remove. Loose debris may prevent sealer from proper contact with the substrate.
Always test a small area. Roll on with roller, or spray on with low pressure garden-style sprayer (preferably with a back-roll to assist adhesion and minimize streaks and puddles).
Penetrating sealers may be applied to slightly damp substrates. Penetrating sealers if over-applied may leave white residue, which will not affect the performance of the sealer and will dissipate with normal wear. Penetrating sealers generally do not require a second coat. Re-sealing is as needed consult your product label.
Topical sealers require very dry, clean substrates. Topical sealer applied over damp substrates will result in white hazy film under the sealer which may, or may not, eventually cure away. Topical sealers if over-applied will result in a flaky, bubbled appearance. To remove this condition clean damaged area with mild solvent and re-apply new sealer. Topical sealers will have re-coat times of between 2-24 hours (consult your product label). To apply second coat after 24 hours may require a light sanding as preparation.
How to Seal Your Pool Deck
Clean your substrate surface well. Clear all loose debris. Sealed stains will become significantly harder to remove. Loose debris may prevent sealer from proper contact with the substrate.
Always test a small area. Roll on with roller, or spray on with low pressure garden-style sprayer (preferably with a back-roll to assist adhesion and minimize streaks and puddles).
Penetrating sealers may be applied to slightly damp substrates. Penetrating sealers if over-applied may leave white residue, which will not affect the performance of the sealer and will dissipate with normal wear. Penetrating sealers generally do not require a second coat. Re-sealing is as needed consult your product label.
Topical sealers require very dry, clean substrates. Topical sealer applied over damp substrates will result in white hazy film under the sealer which may, or may not, eventually cure away. Topical sealers if over-applied will result in a flaky, bubbled appearance. To remove this condition clean damaged area with mild solvent and re-apply new sealer. Topical sealers will have re-coat times of between 2-24 hours (consult your product label). To apply second coat after 24 hours may require a light sanding as preparation.
How to Seal Your Sidewalk
Clean your substrate surface well. Clear all loose debris. Sealed stains will become significantly harder to remove. Loose debris may prevent sealer from proper contact with the substrate.
Always test a small area. Roll on with roller, or spray on with low pressure garden-style sprayer (preferably with a back-roll to assist adhesion and minimize streaks and puddles).
Penetrating sealers may be applied to slightly damp substrates. Penetrating sealers if over-applied may leave white residue, which will not affect the performance of the sealer and will dissipate with normal wear. Penetrating sealers generally do not require a second coat. Re-sealing is as needed consult your product label.
Topical sealers require very dry, clean substrates. Topical sealer applied over damp substrates will result in white hazy film under the sealer which may, or may not, eventually cure away. Topical sealers if over-applied will result in a flaky, bubbled appearance. To remove this condition clean damaged area with mild solvent and re-apply new sealer. Topical sealers will have re-coat times of between 2-24 hours (consult your product label). To apply second coat after 24 hours may require a light sanding as preparation.
How to Seal Your Stone
Clean your substrate surface well. Clear all loose debris. Sealed stains will become significantly harder to remove. Loose debris may prevent sealer from proper contact with the substrate.
Always test a small area. Roll on with roller, or spray on with low pressure garden-style sprayer (preferably with a back-roll to assist adhesion and minimize streaks and puddles).
Penetrating sealers may be applied to slightly damp substrates. Penetrating sealers if over-applied may leave white residue, which will not affect the performance of the sealer and will dissipate with normal wear. Penetrating sealers generally do not require a second coat. Re-sealing is as needed consult your product label.
Topical sealers require very dry, clean substrates. Topical sealer applied over damp substrates will result in white hazy film under the sealer which may, or may not, eventually cure away. Topical sealers if over-applied will result in a flaky, bubbled appearance. To remove this condition clean damaged area with mild solvent and re-apply new sealer. Topical sealers will have re-coat times of between 2-24 hours (consult your product label). To apply second coat after 24 hours may require a light sanding as preparation.
Penetrating Sealer Chemistry
Silane
Inorganic (does not contain carbon)
Very low viscosity, 4 angstroms (deep penetration)
Hydrophobic (repels water)
Oliophobic (repels oil)
Chemical Process – forms covalent bond with minerals (non-film forming)
Cross-Linking – No
Common Function – sealing dense surfaces such as high performance concrete, clay bricks, dense stone.
Silicate (Silicate Salts)
Inorganic (does not contain carbon)
Low viscosity, 8 angstroms (deep penetration)
Hydrophobic (reduces porosity)
Oliophobic (reduces porosity)
Chemical Process – forms calcium-silicate-hydrate (non-film forming)
Cross-Linking – No
Common Function – densifying machine trowel concrete
Siliconate
Organic (contains carbon)
Medium viscosity, 20 angstroms (moderate penetration)
Hydrophobic (repels water)
Oliophobic (repels oil)
Chemical Process – forms cross linked methyl-silicone resin (internal membrane)
Cross-Linking – Yes
Common Function – sealing porous concrete, brick and stone and curing new concrete
Siloxane
Organic (contains hydro-carbon)
High viscosity, 100 angstroms (shallow penetration)
Hydrophobic (repels water)
Oliophilic (bonds with oil, Hydrocarbon Bond)
Chemical Process – forms covalent bond with minerals (generally non-film forming)
Cross-Linking – No
Common Function – sealing very porous substrates such as block
Summary
Each silicon-based penetrating sealer chemistry is unique. Understanding the unique characteristics forms the basis for proper application.
Table
FLATWORK SCALING TODAY: AIR ENTRAINMENT ISN’T ENOUGH
BY
DAVID LANKARD
PRESIDENT AND PETROGRAPHER
LANKARD MATERIALS LABORATORY, INC
400 FRANK ROAD
COLUMBUS, OHIO 43207
A PAPER FOR PRESENTATION AT
THE 4TH ANNUAL CONCRETE TECHNOLOGY FORUM
FOCUS ON PERFORMANCE PREDICTION
CINCINNATI, OHIO
MAY 13-15, 2009
ORGANIZED BY
NATIONAL READY MIXED CONCRETE ASSOCIATION
SILVER SPRING, MARYLAND 20910
2009 Concrete Technology Forum © 2009 National Ready Mixed Concrete Association
FLATWORK SCALING TODAY: AIR ENTRAINMENT ISN’T ENOUGH ABSTRACT
Problems with the scaling of concrete can be traced to the early years of the last century.
In the early to mid-decades of the 1900s scaling was a problem not only with exposed
flatwork (sidewalks, walkways, driveways, aprons, parking lots, and curbs), but also with
highway pavements and bridge decks. The discovery and use of air-entrainment in the
1930s subsequently played a significant role in reducing the incidences of scaling in all of
these applications. Despite the increased use of deicing salts on the nation’s highways in
the 1960s, scaling eventually has become less of a problem in pavements and bridge
decks as these constructions began to use better quality concretes with a w/cm typically
under 0.45. The present paper is focused on residential and commercial flatwork, where,
despite the advances that have benefited the highway concretes, scaling continues as a
frequent, costly, and annoying problem. And, (bad news here) the majority of the
flatwork concrete that is scaling today is air-entrained; much of it properly so. This paper
discusses the material, construction, and environmental factors that influence and control
scaling when it occurs in air-entrained flatwork concrete. Steps that can be taken to
reduce the frequency and severity of scaling in concrete flatwork are discussed.
BACKGROUND
The scaling of concrete flatwork involves the loss of a portion or of the entire finished
wearing surface of a flatwork slab. Most commonly scaling occurs when water-saturated
flatwork experiences episodes of freezing and thawing. In most cases, scaling is a
cosmetic issue. Seldom is the structural integrity or the functionality of the slab
unacceptably impaired (exceptions here are instances where pedestrian safety is an issue).
This fact notwithstanding, some scaled residential and commercial flatwork is removed
and replaced every year. These cases of scaling which require remedial action represent
only a small fraction of the exposed flatwork in service today that show some degree of
scaling distress (as an inspection drive along the streets of a sub-division that is more
than a few years old will prove). This situation obviously does not create a favorable
image for concrete in these applications.
At least 90 percent of the scaled flatwork concrete that I have examined petrographically
in the past 20 years has been air-entrained. In the remainder of this paper I present my
views and opinions on (1) why air-entrainment has not solved the scaling problem in
these applications, (2) the factors controlling scaling in air-entrained concrete flatwork,
and (3) what can be done to improve the situation. My opinions on the subject are based
primarily upon petrographic examinations of scaled concrete flatwork from residential
and commercial applications in Ohio, Pennsylvania, Indiana, Kentucky, Michigan, New
York, New Jersey, and Utah. These examinations are conducted in accordance with the
relevant guidelines of ASTM C 856 (Standard Practice for the Petrographic Examination
of Hardened Concrete) and ASTM C 457 (Standard Practice for Microscopical
Determination of Parameters of the Air-Void System in Hardened Concrete).
2009 Concrete Technology Forum © 2009 National Ready Mixed Concrete Association
2
THE NATURE OF SCALING IN AIR-ENTRAINED CONCRETE
The nature of scaling in air-entrained concrete will be depicted with the aid of Figure 1,
which is an illustration of a section view of a flatwork slab, perpendicular to the plane of
the finished wearing surface. The wearing surface is at the top in the drawing. Consider
that this is a properly air-entrained concrete, which contains a good quality 1 inch
maximum size coarse aggregate. The original finished wearing surface is a thin mortar
layer (in yellow), which is in contact with the topmost coarse aggregate particles (colored
red). In a flatwork slab this mortar layer typically has a minimum thickness of
1 mm to 3 mm. In subsequent text this mortar layer is referred to simply as the wearing
surface mortar.
Figure 1. Representation of a flatwork slab showing the thin mortar layer, which
comprises the original finished wearing surface layer, and which is in contact
with the topmost coarse aggregate particles (A).
When scaling occurs in air-entrained flatwork the most frequent mode of failure is the
loss of this thin mortar layer overlying the topmost coarse aggregate particles. Although
the minimum thickness of the mortar layer is typically 1 mm to 3 mm, it can be as thick
as 4 mm to 5 mm. In mild to moderate cases of scaling the distress is confined to the
loss of that portion of the mortar that lies directly over a coarse aggregate particle. An
example of this phenomenon, which is referred to as mortar liftoff is shown in Figures
2 and 3. The arrows in Figure 2 and 3 point to sites where the mortar liftoff distress has
occurred. In this example the mortar liftoff features are quite common and they are
present as discrete islands of distress. In this paper the ranking of scaling distress as
mild, moderate and severe is subjectively based upon the percent of the total slab
wearing surface area that is affected, not on the depth of scaling.
2009 Concrete Technology Forum © 2009 National Ready Mixed Concrete Association
3
Figure 2. Photograph of a flatwork slab showing the mortar liftoff type of scaling
distress. The arrows point to a few of the numerous mortar liftoff sites.
Figure 3. Enlarged view of the Mortar Liftoff scaling distress in the slab shown in
Figure 2. The loss of the wearing surface mortar is confined to the area directly
over the topmost coarse aggregate particles (A). The aggregate particles are not
fractured. The scale is in inches.
2009 Concrete Technology Forum © 2009 National Ready Mixed Concrete Association
4
In more advanced cases of scaling the individual mortar liftoff features join together to
form larger areas where the wearing surface mortar is lost. An example of this feature is
shown in Figure 4.
Figure 4. Photograph of another area of the wearing surface of the flatwork slab shown
in Figures 2 and 3 at a site where the individual mortar liftoffs have joined
together to form larger areas (arrows) in which the wearing surface mortar
layer has been lost. Scale in inches.
In extreme cases the entire original finished wearing surface mortar layer of a flatwork
slab can be lost through scaling in a given area. However, even in these cases, if the
concrete is air-entrained, the scaling fractures are typically confined to the top few
millimeters of the flatwork slab (even after long periods of exposure to the elements).
In contrast, the cracking distress associated with scaling of a non air-entrained flatwork
slab extends to depths well below the level of the topmost coarse aggregate particles. An
example is shown in Figure 5, which is a lapped section view of a non air-entrained
scaled flatwork core perpendicular to the wearing surface. The entire wearing surface
mortar layer has been lost, exposing the topmost coarse aggregate particles (A). Subsurface
cracks oriented parallel to the wearing surface that resulted from freeze/thaw
cycling in the core have been marked with a black marking pen. These cracks extend
into the core to a depth of 30 mm below the wearing surface. This is the nature of
freeze/thaw-related cracking distress that is expected in a non air-entrained concrete slab.
It is expected that with time and continuing exposure to the elements, the freeze-thaw
related fractures will continue to advance downward into the slab from which this core
was taken.
2009 Concrete Technology Forum © 2009 National Ready Mixed Concrete Association
5
In air-entrained flatwork the scaling fractures typically do not extend below the level of
the topmost coarse aggregate particles because the air-entrainment protects the cement
paste from the effects of freezing and thawing, even if the concrete is saturated.
Figure 5. Lapped section view of a 4 inch diameter core taken from a scaled non airentrained
concrete flatwork slab. The entire original finished mortar layer
overlying the topmost coarse aggregate particles (A) has been lost. Other
cracks in the core are marked with a black marking pen.
THE MECHANISM OF SCALING IN AIR-ENTRAINED CONCRETE
In the air-entrained flatwork cores that I have examined the most common form of
scaling is various degrees of the mortar liftoff distress feature. The mechanism of
mortar liftoff can be described with the help of the illustration in Figure 1 (section view
of an air-entrained flatwork concrete slab).
During exposure to winter weather conditions, moisture accumulates on the slab surface.
Water in liquid form moves downward into the slab and passes into and through the
topmost 1 mm to 3 mm thick mortar layer. As shown in Figure 1 this thin surface mortar
2009 Concrete Technology Forum © 2009 National Ready Mixed Concrete Association
6
layer is in direct contact with the topmost coarse aggregate particles. Water passing
through the mortar layer eventually enters these aggregate particles. Given enough time
and water availability, both the wearing surface mortar layer and the aggregate particles
can become critically saturated. As the temperature in the saturated slab surface falls
below the freezing point, the various constituents of the concrete (cement paste and
aggregates) experience different forms of dimensional change. These include contractive
strains (shrinkage) in the air-entrained cement paste and expansive strains in the saturated
coarse aggregate particles. The factors involved in these dimensional changes are
discussed below.
Dimensional Changes in the Water-Saturated Cement Paste on Cooling
Figure 6 plots the strains that occur in water-saturated air-entrained and non air-entrained
hardened portland cement paste as a function of decreasing temperatures below the
freezing point of water.
With no air-entrainment the saturated cement paste experiences an expansion as a result
of the unaccommodated increase in volume accompanying the effects of the water to ice
transformation.
In contrast, the air-entrained cement paste shrinks on cooling at below freezing
temperatures and the cumulative contractive strains increase as a function of the
decreasing temperature. The colder it gets below freezing the greater the magnitude of
the shrinkage strain in the air-entrained cement paste.
Dimensional Changes in Water-Saturated Aggregates on Cooling
Like saturated non air-entrained cement paste, water-saturated aggregate particles also
expand when they are subjected to temperatures below freezing due to the volume
increases accompanying the water to ice transformation. This type of aggregate
expansion is responsible for the freeze/thaw-related distress condition in concrete
pavements known as D-Cracking (Klieger 1974).
Figure 7 shows the expansions experienced by concrete beam specimens when subjected
to freezing and thawing cycles. The coarse aggregates in the test beams (primarily
limestones) have known histories with regard to the D-Cracking issue. For the concretes
containing aggregates with a history of early D-Cracking distress in service, the
expansions in the beam specimens begin early in the freeze/thaw environment and
increase sharply with increasing numbers of freeze/thaw cycles. For aggregates with a
satisfactory service record the beam expansions are significantly lower in rate and
magnitude.
2009 Concrete Technology Forum © 2009 National Ready Mixed Concrete Association
7
Figure 6. The effect of air entrainment on dimensional changes in saturated hardened
portland cement paste as a function of decreasing temperatures below the
onset of freezing. (Powers 1953).
The data shown in Figure 7 make three important points as regards the mortar liftoff
phenomenon in a flatwork scaling situation.
1. Since the concrete in the test beam specimens is air-entrained, the expansions of
the beams can be attributed to the expansions that accompany the water-ice
transformations in the saturated coarse aggregate particles on freezing.
2. With each new freeze/thaw cycle the amount of damage that occurs within the
concrete increases and is cumulative.
3. There are large differences in the magnitude of expansions depending on the
coarse aggregate source. However, all of the beam specimens have expanded,
confirming that all of the aggregate sources have experienced expansions to some
degree in the test.
2009 Concrete Technology Forum © 2009 National Ready Mixed Concrete Association
8
Figure 7. Relationship between coarse aggregate service record and the expansion of airentrained
concrete beam specimens containing the aggregates in laboratory
freezing and thawing tests. (Klieger 1974).
The consequences of the differential strains in the air-entrained cementitious phase and
the aggregate constituents of the saturated flatwork concrete on cooling below the
freezing point are discussed in the following section as they relate to the scaling of airentrained
concrete flatwork in service.
Mortar Liftoff Scaling in Service
In this discussion the focus is on the dimensional changes that occur concurrently on
cooling below the freezing point in (1) the 1 mm to 3 mm thick wearing surface mortar
layer and in (2) the topmost coarse aggregate particles immediately underlying this thin
mortar layer. Again, reference is made to Figure 1. Consider that both the cementitious
2009 Concrete Technology Forum © 2009 National Ready Mixed Concrete Association
9
phase in the wearing surface mortar and the topmost coarse aggregate particles are
critically saturated. The aggregates are most commonly sedimentary rocks. The
consequences of the dimensional changes that occur within the wearing surface layer of
the slab as the temperature falls below the freezing point are summarized below.
1. The air-entrained cementitious phase in the mortar layer experiences a uniform
and continual contractive strain (shrinkage) on cooling.
2. Upon reaching the freezing point the topmost aggregates particles (A) expand, as
indicated by the arrows in Figure 1. The particles begin to increase in volume
relative to the pre-freezing condition in the concrete.
3. With the aggregate particles expanding and the air-entrained cement paste
contracting, tensile stresses are created within the mortar layer overlying the
aggregate particles.
4. The magnitude of the contractive strain in the cement paste increases as a function
of decreasing temperature below the freezing point. This results in an increase in
the stresses within the shrinking mortar phase as it is constrained in this
movement by the aggregate particles. The lower the temperature falls below
freezing, the greater the magnitude of stresses in the mortar layer. It does matter
how cold it gets!
5. Eventually the tensile stresses in the mortar layer exceed the tensile strength of
the mortar. At this point the mortar layer overlying the aggregate particles
fractures, lifts-off, and is lost (scaled).
6. In some instances the mortar lifts cleanly off the surface of the coarse aggregate
particle. This commonly occurs when the aggregate particles have smooth
surfaces (gravels) and/or if the cementitious phase in the mortar has a
significantly elevated w/cm. An example of this phenomenon is shown in
Figure 8.
7. In those instances where the coarse aggregate particle has a rough surface
(crushed aggregates) and/or the wearing surface mortar has a more moderate
w/cm, the liftoff fracture occurs primarily or completely within the air-entrained
cementitious phase as shown in Figure 9.
8. Typically, the great majority of the coarse aggregate particles involved in the
fracture and liftoff of the overlying mortar layer do not fracture themselves.
These are not sub-standard or poor quality aggregates. These are not pop-outs,
which are attributed to weak, highly absorptive aggregate particles.
9. Once the thin wearing surface mortar layer overlying the topmost coarse
aggregate particles (the original finished surface of the slab) has scaled, there
typically is no further scaling distress at these sites at lower depths in the slab.
2009 Concrete Technology Forum © 2009 National Ready Mixed Concrete Association
10
This is because (1) the cementitious phase of the concrete is air-entrained and
may have a lower w/cm relative to the wearing surface mortar, (2) the coarse
aggregate particles that have been exposed in the scaled wearing surface are free
to expand, (3) sub-surface aggregate particles have less chance of becoming
saturated, and (4) sub-surface aggregate particles do not have a free surface to
work against (i.e. the slab wearing surface).
Figure 8. Lapped section view (10X) of an air-entrained concrete core taken from a
scaled flatwork construction. This example shows a clean liftoff of the wearing
surface mortar (arrows) from the coarse aggregate particle (A).
Figure 9. Lapped section view (7X) of an air-entrained concrete core taken from a
scaled flatwork construction. In this example of mortar liftoff much of the airentrained
wearing surface mortar remains bonded to the aggregate particle.
2009 Concrete Technology Forum © 2009 National Ready Mixed Concrete Association
11
In the next section the factors that control the onset and extent of the scaling of airentrained
concrete flatwork are identified and discussed. There is an involvement of
material, construction, environmental, and operational variables.
FACTORS AFFECTING SCALING OF AIR-ENTRAINED CONCRETE
The potential for scaling distress to occur and the potential severity of the scaling distress
are dependent primarily on two factors, which are (1) the severity of the exposure
conditions and (2) the quality of the mortar layer that comprises the finished wearing
surface of the flatwork slab.
Severity of the Slab Exposure Condition
The severity of the exposure condition regarding the potential for scaling is controlled by
a number of material, environmental, and operational factors, which include,
1. The opportunity for water to accumulate on the surface of the concrete.
2. The duration of the presence of liquid water on the slab surface prior to the onset
of freezing conditions.
3. The degree of saturation of the slab surface at the time of freezing (as influenced
by Items 1 and 2). There are also material factors involved here.
4. The presence of deicing salts on the slab surface from intentional applications and
roadway snow (rendering ice into liquid water at below freezing temperatures).
5. The number of freezing and thawing cycles.
6. The lowest temperature experienced by the slab in the freezing cycles.
There is of course nothing to be done about Items 2, 5, and 6, which are simply the
weather. Often, as well, there is not much that can be done to keep deicing salts off of
the flatwork. It is however, important to acknowledge that air-entrained flatwork
concrete that can survive mild or moderate freeze/thaw exposure conditions, may scale
when these factors are significantly in play. This is especially important when it comes
time to decide who is responsible for flatwork that scales.
Saturation of the Concrete
The overriding factor controlling the onset and the severity of scaling of air-entrained
concrete flatwork slabs is the amount of moisture in the wearing surface mortar layer of
the slab at the time that freezing occurs. If the wearing surface layer is dry or if the
moisture content is below critical saturation in both the cement paste and the aggregates,
no scaling will occur regardless of the number of freezing and thawing cycles. As PCA
2009 Concrete Technology Forum © 2009 National Ready Mixed Concrete Association
12
researcher Hubert Woods put it over 50 years ago, Dry hardened concrete or concrete in
which the hardened cement paste and the aggregate are both below some critical degree
of saturation by water, is highly resistant to freezing and thawing in nature (Woods
1956).
Many commercial buildings or stores that front on a street or parking area have sidewalks
and entrance walkways that are partially under roof. These situations provide excellent
examples of the effect that the degree of moisture saturation has on the occurrence and
the severity of flatwork scaling. One such example is given in Figure 10.
Figure 10. The sidewalks fronting these shops are partially under roof. The sidewalk
panels that are not under roof (3) show moderate to severe mortar liftoff
scaling. The sidewalk panels that are fully covered by the roof (1) show
virtually no scaling.
In the example of Figure 10 the sidewalk is 150 feet long and the width is 7 feet. The
two foot wide panels that are not under roof (3) show moderate to severe mortar liftoff
scaling, with some liftoff areas that have coalesced. These sidewalk panels abut the
parking area and are subject to full contact with rain and snow, as well as deicing salts
from the parking area. The middle three foot wide panels (2) which are under roof, but
partially exposed to the elements show mild scaling, with a low number of discrete
mortar liftoffs. The two foot wide sidewalk panels that abut the store fronts (1) are fully
under roof and show virtually no scaling defects. In this example concrete from the same
supplier was placed at the same time by the same crew. Within a distance of only 7 feet
the performance of the concrete ranges from no scaling to severe scaling; affected only
by the degree of saturation of the wearing surface mortar at the time of freezing. There
are literally thousands of examples of the type just shown in shopping malls in severe
weather areas of the country.
The point of this discussion is to emphasize the fact that to both understand and to control
scaling in air-entrained concrete it is necessary to keep in mind the dominant role that the
2009 Concrete Technology Forum © 2009 National Ready Mixed Concrete Association
13
degree of saturation of the concrete at the time of freezing plays in the process. And,
short of installing a roof over all of our exposed flatwork, it is the quality of the finished
wearing surface mortar layer that controls both the rate of saturation of the slab and the
saturation level in the surface of the slab at any given time.
Quality of the Mortar Layer Comprising the Slab Wearing Surface
The properties and features of the wearing surface mortar layer that control its quality
regarding its impact on the scaling resistance of the flatwork are those of the cementitious
phase of the mortar and include (1) permeability, (2) porosity, (3) tensile strength, and
(4) the presence and quality of the entrained air void system.
Permeability is the property of the cementitious phase of the mortar that has the greatest
effect on its ability to minimize the ingress of water into the slab. The porosity of the
cementitious phase controls the amount of freezable water in the mortar layer at the time
of freezing. The tensile strength of the cementitious phase controls the ability of the
mortar to resist the stresses developed during freezing. The most important material
property controlling these properties of the cementitious phase is the water to
cementitious material ratio (w/cm).
Compromised Quality of the Wearing Surface Mortar
In the majority of flatwork concretes that require remedial action the quality of the
wearing surface mortar layer has been compromised relative to the concrete at lower
depths in the slab. This situation is affected by a number of factors including,
1. The water to cementitious material ratio (w/cm) of the cementitious phase of the
mortar as affected by the starting water content of the concrete.
2. The water to cementitious material ratio (w/cm) of the cementitious phase of the
mortar as affected by the slab finishing practice.
3. The presence of supplementary cementitious materials (SCM) in the cementitious
phase.
4. The level of maturity of the cementitious phase at the time that freezing first
occurs as affected by both the duration and effectiveness of curing.
5. The quality of the entrained air void system in the cementitious phase as affected
by the slab finishing practice.
Water to Cementitious Material Ratio (w/cm)
The importance of w/cm cannot be overemphasized. It is instructive to think of the
wearing surface mortar layer as a lid on a flatwork slab. The tighter the lid, the less
2009 Concrete Technology Forum © 2009 National Ready Mixed Concrete Association
14
water that gets into the slab. The lower the permeability of the cementitious phase of the
wearing surface mortar, the less likely it is that critical saturation levels will be reached in
the topmost portions of the flatwork slab, including the topmost coarse aggregate
particles. The recommendation (by the American Concrete Institute and others) of a
maximum w/cm of 0.45 for exposed flatwork concrete has a sound scientific basis as is
shown in Figure 11.
Figure 11. The relationship between permeability and water-cement ratio of hardened
portland cement paste (Powers 1954).
Below a w/cm of 0.45, a well cured paste has both a low amount of capillary porosity and
a low permeability. The coefficient of permeability rises exponentially above a w/cm of
around 0.55. Below a w/cm of 0.40 a mature sample of hardened portland cement paste
is virtually impermeable.
In the majority of the scaled concrete samples submitted for petrographic examination,
the w/cm of the original finished wearing surface mortar layer is in excess of 0.45, often
significantly so. An example of a wearing surface mortar that has been compromised in
this manner is shown in Figure 12. The 0.5 mm to 1 mm thick wearing surface mortar
layer lying above the arrows in Figure 12 (H) is lighter in color due to an elevated w/cm
relative to the underlying concrete (L). The w/cm in the compromised wearing surface
mortar is estimated at 0.65. Relative to the underlying concrete, which has a w/cm
estimated at 0.50, the cementitious phase in the wearing surface mortar is softer and
weaker, and is more porous and permeable. At this site on the core the mortar has not yet
lifted off of the coarse aggregate particle (A) but the paste/aggregate bond has been
disrupted. A second example of this microstructural feature is shown in Figure 9, where
the surface mortar layer lying between the red and blue arrows has a higher w/cm relative
to the underlying concrete.
2009 Concrete Technology Forum © 2009 National Ready Mixed Concrete Association
15
Figure 12. Lapped section view (20X) of an air-entrained concrete flatwork core that
scaled. The cementitious phase of the wearing surface mortar layer above the
arrows (H) has an elevated w/cm relative to the underlying concrete (L).
The conditions that can lead to an elevated w/cm in the wearing surface mortar layer
(relative to the concrete as-placed) include,
1. Finishing the slab prior to the complete reabsorption of bleed water.
2. Finishing the slab prior to the cessation of bleeding.
3. The intentional placement of water on the slab surface to facilitate the finishing
operation.
4. Working rain water into the slab surface during finishing.
The issues listed in Items 1 through 4 are examples of improper finishing procedures. It
is also possible for the slab to be properly finished and cured and still have an elevated
w/cm in the wearing surface mortar if the concrete as supplied has a w/cm in excess of
the recommended maximum w/cm of 0.45. Even here the risk of scaling is increased.
The presence of an elevated w/cm in the wearing surface mortar layer is the most
common way in which the quality of the mortar layer comprising the finished slab
2009 Concrete Technology Forum © 2009 National Ready Mixed Concrete Association
16
surface is compromised. But, there are also other compromising factors that can result in
a reduced strength and an elevated porosity and permeability in the wearing surface
mortar at the time of first freezing. They include,
1. An inadequate period of curing prior to first freezing.
2. Ineffective curing or no curing.
3. Early carbonation of the wearing surface.
A final issue to be addressed is the effect that the use of supplementary cementitious
materials has on the scaling of concrete flatwork. The focus here is on the effect that
SCMs can have on the quality of the wearing surface mortar layer.
Supplementary Cementitious Materials (SCM)
Supplementary cementitious materials (SCM) include silica fume, ground granulated
blast furnace slag (slag cement), fly ash, and natural pozzolans. SCMs are widely used in today’s concretes, including many of the high performance concretes. In these latter applications the concretes typically have a low water to cementitious material ratio (typically under 0.45). Relative to straight portland cement concretes the SCM blends provide (1) improved long term strength, (2) improved resistance to alkali-silica attack, (3) improved durability, and (4) reduced long term porosity and chloride permeability. Despite these proven benefits there is a considerable body of literature that raises questions regarding the effect that SCMs have on the scaling resistance of concrete. While many laboratory studies have shown a potential negative effect of SCMs on scaling, this finding has not been fully borne out by field observations. Most of the studies of field installations have been concerned with pavements and bridge decks. Reflecting the fact that this issue has not been satisfactorily resolved, the concern regarding the potential adverse effect of SCMs on scaling is current and widespread. A number of Highway Departments in the U.S. and Canada place restrictions on the amount of SCMs that can be used in the cementitious material in concretes. The U.S. Department of Transportation has a User Guideline website concerning the use of slag cement as an SCM in portland cement concrete (U.S. Department of Transportation 2009). In their most recent posting they state regarding salt scaling, Concrete containing high concentrations of GGBF slab may be susceptible to salt scaling. Due to this problem, some agencies limit the amount of slag in a portland cement concrete mix to 25 percent of total cementitious weight. They site a 1985 paper for this caveat. 2009 Concrete Technology Forum © 2009 National Ready Mixed Concrete Association
17
The Iowa State Center for Portland Cement Concrete Pavement Technology
recently completed Phase I of a Research Study entitled Deicer Scaling
Resistance of Concrete Pavements, Bridge Decks, and other Structures
Containing Slag Cement (Schlorholtz 2008). This pooled fund research program
is sponsored by the Federal Highway Administration and the Slag Cement
Association along with DOTs from 5 states.
I share concerns on this matter based upon my own studies. It is not the intent of the present paper to fully review the literature on this subject but a few summary points will be made (1) as they relate overall to the effect of SCMs on scaling and (2) as they relate to the condition and microstructure of the top few millimeters of thickness of the finished wearing surface mortar in a flatwork slab. Two of the recent scaling projects I worked on involved cores which represented two seasons of construction and which involved a number of different finishing personnel.
One of the projects was a municipal sidewalk project requiring 1400 cubic yards of
concrete which were placed on two successive years (2005 and 2006). The project
concrete contained 30 percent of slag cement (by weight) in the cementitious phase. The sidewalks front a major commercial roadway (6 lanes) in the city. Snow removed from the roadway accumulates on the leading edge of the sidewalks. The three mile sidewalk section on the north side of the road was placed first and went through the winter of 2005 2006 with no scaling distress. The three mile section on the south side of the roadway was placed in the summer of 2006. Following the winter of 2006 – 2007 most of the sidewalk constructions on both sides of the roadway showed mortar liftoff scaling.
The winter of 2006 – 2007 was a particularly severe weather year in our area regarding the factors that influence scaling. The petrographic examination of cores from the project revealed an elevated w/cm in the top 1 mm to 3 mm thickness of the original broom finished wearing surface mortar layer.
A second project involved the scaling of driveway and apron slabs in residential home developments. The concrete used in the constructions was an air-entrained concrete containing a ternary cement (portland cement slag cement and fly ash). The SCM replacement was 39 percent (by weight) of total cementitious. Over a two year period over half of the 50 or so constructions showed some level of mortar liftoff scaling. The cores that I examined from the project showed an elevated w/cm in the wearing surface mortar.
Considering the size of these projects and the number of different finishers involved it is not reasonable to expect that all of the compromised wearing surface layers can be attributed to poor finishing techniques. Inspectors were on the job for the municipal sidewalk project.
Laboratory studies looking into this situation have also identified the presence of an
elevated w/cm in the finished wearing surface of laboratory specimens of concretes
containing SCMs. 2009 Concrete Technology Forum © 2009 National Ready Mixed Concrete Association
18
In a 2002 laboratory and field study (Bleszynski 2002) Canadian researchers, using
concretes with a w/cm of 0.42 report the following; Concrete mixtures containing blast furnace slag often exhibit a delayed set time resulting in an increased period of bleeding. Finishing the concrete prior to the cessation of bleeding can trap the escaping water in the surface layer leading to a concrete skin with a higher w/cm than the interior concrete, and hence greater porosity. Other Canadian researchers (Pigeon 1996) report similar microstructural observations regarding the quality of the wearing surface in concrete containing SCMs. These workers conducted laboratory studies on plain and fly ash concretes produced at a w/cm of 0.40, and they offer the following commentary on this microstructural feature. The scanning electron microscope observations carried out clearly indicate that the first millimeters
below the surface of troweled laboratory concrete specimens can have a microstructure different than that of the bulk of the concrete. In all of the concretes tested, an extremely porous layer (i.e. with a very high water/binder ratio) was observed at the surface. The scaling test results show that the higher porosity of the surface layers tends to markedly reduce the deicer salt scaling durability of wood troweled laboratory samples during the first cycles of freezing and thawing. The use of fly ash (20 % and 40 %) was found to increase the thickness and porosity of the surface layer. The effect is probably related to the fact that the fly ash used in this series of tests was found to increase the bleeding of concrete.
In their studies the cited authors measured the weight of scaled material obtained from trowel finished surfaces as well as on sawed surfaces. They found the following as reported in a later article (Talbot et. al. (2000). Even taking into account the fact that sawing exposes aggregate as well as paste, the amount of scaling residues from the sawed surfaces is in all cases quite small compared to that collected from the troweled surfaces. The scaling process is relatively linear for the sawed surfaces, but not for the troweled surfaces. For these surfaces the scaling process is always very rapid during the first few cycles (probably due to the loss of a weak and porous layer), but much slower afterwards. It is also very clear that the use of SCM increases the deterioration due to scaling, not only on the troweled surfaces, but also on the sawed surfaces. Many of the studies conducted to date on this research topic have used concretes with a w/cm under 0.45. The great majority of the concretes containing SCMs that are used for bridge deck and pavement construction also have a w/cm under 0.45, and not uncommonly, at or under 0.40. These latter concretes are also usually not hand finished. In contrast the concretes used for residential and even commercial flatwork are often placed at a w/cm above 0.45. In addition, the cosmetic appearance of a residential or commercial flatwork slab is definitely under more scrutiny than that of a bridge deck or pavement.
All three of the commonly used SCMs have a finer particle size relative to portland
cement. Inasmuch as bleeding is a sedimentation phenomenon, it is reasonable to assume that the bleeding period of the SCM-containing concretes will be prolonged relative to a straight portland cement concrete due to, 2009 Concrete Technology Forum © 2009 National Ready Mixed Concrete Association
19
The finer particles size of the SCMs. The greater viscosity of the water phase in the fresh portland cement concretes relative to those diluted with the slower reacting SCMs. A longer open time relative to the straight portland cement concretes.
Concrete finishers following what they believe to be suitable finishing procedures for portland cement concretes may be unknowingly conducting the finishing step while the SCM concretes are still bleeding. It is expected that this occurrence would be more likely in the flatwork concretes relative to highway concretes due to the use of higher starting w/cms in the former. A different but related matter dealing with the use of SCMs in flatwork concrete has to do with curing and carbonation. SCMs that participate in the pozzolanic reaction require the presence of calcium hydroxide as the other reactant. If curing is inadequate and early carbonation occurs in the wearing surface of the flatwork, a portion of the added SCM can end up simply as an unreacted diluent, with a resultant adverse effect on strength, permeability, and porosity of the cementitious phase of the wearing surface mortar.
SUMMARY
The focus of this paper is on the continuing problem of scaling distress in the wearing surface of residential and commercial flatwork constructions (sidewalks, walkways, aprons, driveways, parking lots, and curbs). My opinions on the subject are based primarily upon petrographic examinations of scaled concrete flatwork from residential and commercial applications in Ohio, Pennsylvania, Indiana, Kentucky, Michigan, New York, New Jersey, and Utah.
The primary points made in the paper include the following,
A distinction is made between the scaling of highway concretes (pavements and
bridge decks) and the scaling of residential and commercial flatwork. Concretes
in the highway applications typically have a lower w/cm than flatwork concretes
and commonly are more closely supervised during the construction, finishing, and
curing operations.
In areas of the country that experience freezing and thawing the percentage of
concrete flatwork that shows some degree of scaling is quite high. While not
necessarily affecting the functionality of the constructions, the presence of the
unsightly distress does not provide a good image of concrete in these applications.
Most of the concrete flatwork scaling today is air-entrained; much of it properly
so.
2009 Concrete Technology Forum © 2009 National Ready Mixed Concrete Association
20
The nature of scaling is different in air-entrained concrete relative to non airentrained
concrete. Air-entrainment reduces the damage to flatwork concrete in
severe weather environments. Even low levels of intentional air entrainment are
beneficial.
The scaling distress in air-entrained concretes is commonly confined to the top
few millimeter of the finished wearing surface mortar layer of the slab. The form
of scaling that is most common in air-entrained flatwork slabs is various degrees
of the phenomenon referred to as mortar liftoff. The involvement of the nearsurface
coarse aggregate particles in this form of scaling is discussed.
In those instances where the degree of scaling has been extensive enough to
warrant a petrographic examination it is often found that the top few millimeters
thickness of the finished wearing surface has been compromised and is of lower
quality relative to the underlying concrete. This condition contributes to the risk
of scaling and considerable discussion is devoted to this issue in the paper.
There are currently questions regarding the effect that the use of supplementary
cementitious materials has on the scaling resistance of concrete. This issue is
addressed in the paper.
It is a point of emphasis in the paper that to both understand and to control scaling
in air-entrained flatwork it is necessary to acknowledge the dominant role that the
degree of water saturation of the concrete at the time of freezing plays in the
process. Although the significant influence of deicing salts in this matter is
acknowledged, scaling can not be blamed on deicing salts.
It is a further point of emphasis that once the concrete is saturated, the potential
for scaling distress increases with both the number of freezing and thawing cycles
and the minimum temperatures reached in the process.
Eliminating the Flatwork Scaling Problem
In a paper that I authored in 2001 I offered the opinion that the most important step that
could be taken to reduce the scaling problem was to assure that concretes provided for
use in flatwork applications adhere to the maximum recommended w/cm guideline of
0.45. I still believe this to be the case. It is still the exception rather than the rule for
water to cementitious material ratio to be specified for a flatwork project, particularly in
residential work.
The arguments against the use of concretes having a w/cm at or under 0.45 in flatwork
applications seem to be that (1) they are too expensive, and (2) they are difficult to work
with. At the very least the use of these low w/cm concretes could be specified in
applications that are known to represent the most severe exposure conditions in the field.
2009 Concrete Technology Forum © 2009 National Ready Mixed Concrete Association
21
That is, in applications where contact with both moisture accessibility and deicing salts
will be at high levels.
The long-standing industry guidelines for air-entrainment, finishing procedures, and
curing procedures still must be followed. The use of certified flatwork finishers on
residential and commercial flatwork projects should be considered.
The questions regarding the effect that supplementary cementitious materials have on
scaling have not been satisfactorily resolved. The fact that flatwork concretes are often
installed at a w/cm above 0.45 places added emphasis on this matter. It is prudent to
consider not using SCMs in flatwork until the issue is resolved.
Given all of the variables that can affect the scaling of concrete flatwork it is likely that
scaling can never be eliminated. However, the tools and knowledge are at hand to reduce
the problem if we would just use them.
REFERENCES
Bleszynski, R., Hooton, R.D., Thomas, M.D.A., and Rogers, C.A., (2002). “Durability of
Ternary Blend Concrete with Silica Fume and Blast-Furnace Slag: Laboratory and
Outdoor Exposure Site Studies. American Concrete Institute Materials Journal,
September-October, 2002. 499-508.
Klieger, P., and Stark, D. (1974). D-Cracking of Concrete Pavements in Ohio,
Portland Cement Association R & D Bulletin RD047.01P. October 1974. 1-97.
Lankard, D. (2001). Scaling Revisited, American Concrete Institute Concrete
International, May 2001. 43-49.
Pigeon, M., Talbot, C., Marchand, J., and Hornain, H., (1996). Surface Microstructure
and Scaling Resistance of Concrete. Cement and Concrete Research, Volume 26, No.
10. 1555-1565.
Powers, T.C., and Helmuth, R.A., (1953). Theory of Volume Changes in Hardened
Cement Paste During Freezing. Proceedings of the Highway Research Board, Thirty-
Second Annual Meeting (1953). 285-297. As presented in Concrete Second Edition
by Mindess, S.L. et. al. (page 500).
Schlorholtz, S. and Hooton, R.D., (September 2008). Deicer Scaling Resistance of
Concrete Pavements, Bridge Decks, and Other Structures Containing Slag Cement: Site
Selection and Analysis of Field Cores. Report No. CTRE Project 05-202, Pooled Fund
Project TPF-5(100). National Concrete Pavement Technology Center, Iowa State
University. (1-122)
Talbot, C., Pigeon, M., and Marchand, J., (2000) Influence of Fly Ash and Slag on
Deicer Salt Scaling Resistance of Concrete. American Concrete Institute Special
2009 Concrete Technology Forum © 2009 National Ready Mixed Concrete Association
22
Publication SP-192, Durability of Concrete, Proceedings of the Fifth International
Conference, Barcelona, Spain, 2000. Volume II. Paper SP 192-39. 645-657.
U.S. Department of Transportation (2009). Blast Furnace Slag. Turner-Fairbank
Highway Research Center. <http://www.tfhrc.gov/hnr20/recycle/waste/bfs3.htm>
Woods, H., (1956) Observations on the Resistance of Concrete to Freezing and
Thawing. Portland Cement Association Research Department Bulletin 67, February,
1956. 345-349.
This paper has benefited from my discussions with Nick Scaglione, President and
Petrographer of Concrete Research and Testing, Columbus, Ohio, and his thoughtful
comments and suggestions are greatly appreciated.
2009 Concrete Technology Forum © 2009 National Ready Mixed Concrete Association
