You are on page 1of 30

2.

5 POROSITY OF MORTARS
M. Thomson1, J-E Lindqvist2, J. Elsen3, C.J.W.P. Groot4
1
Chemical Lime Company, Henderson, Nevada, USA
2
The Swedish National Testing and Research Institute, Sweden
3
Katholieke Universiteit Leuven, Belgium
4
Delft University of Technology, The Netherlands

1. Introduction

The role of porosity of a historic mortar is important in terms of moisture transport, mechanical
properties, durability and compatibility of the mortar to masonry as a whole. Methods to
determine porosity are as diverse as its influence. Any characterization of historic mortar
through porosity determination should not simply be done to obtain a value. Different
techniques for determining porosity may give different values for the same sample. It is the
interpretation of the value and its influence on the behaviour that is important.

Porosity refers to the volume that is not composed of solids. Among existing classifications of
pore sizes the most recognised is used by the IUPAC [1972] (Fig. 1.1) -- pores are classified as
micropores, mesopores and macropores.

Micro pores
Meso pores
Macro pores

Capillary pores
Sorption pores Coarse pores

1 nm 1 µm 1 mm
Figure 1.1. IUPAC pore size classification. The pore ranges of sorption, capillary and coarse
pores are illustrated for reference

An additional definition of capillary pore size in mortars related to the moisture transfer
properties is useful. Capillary pores range from approximately 0,1 to 100 micrometer and
contribute to the capillary water transfer. Water present in finer pores is bound more tightly to
the material and these pores provide no moisture transfer contribution. Pores coarser than 100
micrometer contribute to the water permeability through gravity or wind driven water ingress.
Porosity can be reported as total pore volume and further described by pore size distribution
and specific surface area. When measured by intrusive methods only those pores that are
interconnected are measured. This provides a relationship of porosity to permeability. When
measured by optical techniques, all pores, which can be observed, are measured. Some of
these pores may not be interconnected to each other. It becomes apparent that depending on
the nature of the measurement a different value might be obtained.
The objective of this chapter is to provide a basic understanding of porosity in historic mortars
with the focus on lime based mortars. This includes:
o how porosity is initially developed;
o how porosity changes over time;
o how porosity influences the properties and characteristics of a mortar

The various techniques for determining porosity are described with an emphasis placed on
their relevance to characterizing historic mortar with respect to repair.

The chapter is concluded with a suggested road map for the investigation of the porosity of
historic mortars.

2. Origin of Porosity in Mortars

2.1. Primary Porosity


2.1.1. Sorption-pores
Sorption pores less than 0,1 micrometer. Water is held to the surface of these pores and no
moisture is transported between these pores. Sorption pores in mortars occur where there is
the presence of hydrated hydraulic phases such as C-S-H (calcium-silicate-hydrate). Lime-
only mortars do not have sorption pores.

2.1.2. Capillary Pores


Capillary pores range in size from 0,1 to 100 µm and are found within the binder and at
aggregate/binder interface. When the mortar is first mixed it is a dense slurry, with the
aggregate and binder particles (ideally), uniformly distributed and surrounded by water.
Primary porosity is developed through the movement of water due to either absorption into the
surrounding masonry unit or evaporation to the air. These pores are highly interconnected and
moisture transport through these pores is by capillary transport. Once the moisture level
reaches an equilibrium level with the environment and masonry, the total porosity is further
influenced by the carbonation of Ca(OH)2 to CaCO3. As a simple replacement, carbonation
represents a 10% volume increase -- the molar volume of calcium hydroxide is 33.1 cm3 and
calcite is 36.9 cm3. Accommodation of the volume change is taken up by a reduction of total
porosity but no significant shift in pore size distribution (Papayanni and Stefanidou, 2001).
Carbonation is a long-term process, and can take many years to complete.

Hydrating hydraulic phases also contribute to primary capillary porosity. Intergranular spaces
are filled with water as the hydraulic minerals hydrate, forming pores once the water is either
consumed or evaporates out. Hydration reactions can continue over a long period of time, and
will result in a decrease in total porosity and a shift to finer pores (Papayanni, 2001).

The water/binder ratio and the binder/sand ratio will also influence the capillary porosity.
Capillary pores form at the contact of the binder to the sand. A higher water content and a
lower sand content results in a higher relative total porosity (Hildsdorf & Schaffer, 1996,
Hayden et al., 2001). In addition, to the initial water/binder ratio of the mortar the insitu
water/binder ratio will vary with the absorption behaviour of the contacting masonry unit
(Groot and Larbi, 1999; Brocken, 1998; Sugo et al., 1998). Given the same initial water to
binder content, mortar in contact with a highly absorptive masonry unit will tend have a lower
total porosity than mortar in contact with weakly absorptive masonry.

2.1.3. Coarse Pores


Coarse pores are greater than 100 micrometer in diameter. They are generally formed by:
• Entrapped Air. Entrapped air pores are irregular in shape and distribution and can
account for up to 8% total porosity. They form through the entrapment of air during
the mixing process. The water content and application of the mortar can influence the
total coarse porosity of the mortar. Mortars which are very dry tend to entrain air
more than wetter mortars. The placement and mechanical treatment of mortars such
as tooling will reduce the entrained air content near the surface by pressing the air out.
• Entrained Air. These are round voids (bubble-like) formed by the introduction of
admixtures which are surfactants. In historic mortars these are organic materials
containing proteins (Sickels, 1981; Sommer, 1987; Chandra and Aavik, 1987).
• Shrinkage cracks. Cracking due to drying shrinkage is not uncommon (Figure 2.1).
Shrinkage cracks increase with increasing water content of the mortar and the rate of
evaporation of the water from the mortar. Coarse shrinkage cracks will contribute to
water ingress and mechanical weakening.

Figure 2.1. Photomicrograph of thin section of a laboratory prepared lime-


aggregate mortar. Pores and cracks are yellow, aggregate is white-grey, lime
binder has a dark grey fuzzy appearance. The orange aggregate in the middle of
view is sphene. Width of photomicrograph is 5 mm.

2.2. Secondary Porosity


Secondary porosity results from the interaction of hardened mortar with the environment.
Chemical reactions, as well as mechanical processes, can create secondary porosity.
2.2.1. Dissolution/Re-precipitation
The transfer of moisture in the pores of a mortar also includes the transfer of the soluble ions.
Calcium is the most ubiquitous ion, coming from the calcium hydroxide and calcium carbonate
of the mortar binder. Wetting and drying cycles will cause the transfer of fluid through the
mortar. The calcium ions are transferred and may come in contact with fluids enriched in CO2
as carbonic acid and calcite will precipitate. Ring-like textures of calcite are common along
the interior of larger voids and at the exterior of mortar and are related to dissolution and re-
precipitation. Narrow cracks may heal by calcite precipitation, known as autogenous healing.
The influence of dissolution/re-precipitation on porosity is not completely predictable.
Dissolution will cause an increase in porosity, but re-precipitation, probably at a different
locality will reduce porosity. If the calcite precipitates on the mortar surface, the surface
becomes more impermeable and there is a loss of mass to the mortar.

2.2.2. Acid Leaching


The solubility of calcite in the mortar increases with decreasing pH. Acid leaching results in
the dissolution of the mortar binder, and carbonate aggregates if present. It rarely leads to re-
precipitation and therefore results in an increase of total porosity and the pore size is usually
coarse (see damage chapter, and petrography chapter). Acidic environments with pH values as
low as 3 are not uncommon to historic masonry. It can come from humic acids of soils or
lichens, or sulphates and nitrates from dry deposition on the masonry surface (Charola, 2000).
While the latter type of acid leaching is generally associated with industrial environments, it is
possible that localised dry deposition from a fire pit or fireplace can produce the same type of
deterioration of the mortar.

2.2.3. Mechanical Actions


Cracking due to mechanical actions occurs where resistance to different types of stresses is
exceeded. The cracks may be coarse or capillary sized. Frost damage is a special case and is
discussed more fully in section 2.3.3.2.

2.2.4. Secondary minerals


Secondary mineral formation in mortar after hardening may greatly influence the pore
structure of the mortar. Secondary mineral growth is the result of the dissolution of the
primary binder, and the formation of a new mineral not including calcite. The new mineral
formation usually results in cracking, and the cracks are typically coarse. Secondary minerals
such as calcium or magnesium sulphate may form if soluble calcium or magnesium and
sulphate are available. Ettringite or thaumasite may form if alumina and sulphate are available,
resulting in pore or crack filling and possibly expansion of the mortar. The chemical analysis
of a mortar indicating any of these elements does not automatically indicate the presence of the
minerals and mineralogical studies are necessary.

2.3. Role of Porosity in Mortars

2.3.1. Influence on mechanical properties


2.3.1.1. Compressive and Tensile Strength
The compressive strength of a material is greatly affected by porosity. In general, compressive
strength decreases with increase in porosity. The relation between strength and total porosity
can be expressed as (Powers and Brown, 1947):

β = β o (1 − v p ) n [2.1]
where βo is the compressive strength of the material without porosity, β is the actual
compressive strength and vp is the pore volume in volume % and n is a material specific
constant.

A large pore or other defect may act as a limiting flaw, however, and result in failure at a lower
stress then Equation 2.1 would predict. This can be expressed in a general way as:

EGc
σt =θ [2.2]
α

where σt is the critical tensile stress, α the dimension of the limiting flaw, E is the Young’s
modulus, Gc the surface energy per unit area and θ is a material specific constant. As the
constants n and θ are generally not known, these equations provide a qualitative rather than a
quantitative understanding.

Laboratory studies have investigated the mechanical properties of lime and hydraulic lime
mortars (Schäfer and Hilsdorf, 1992; Van Balen and Van Gemert, 1994; Hayen et al., 2001;
Papayianni et al., 2001). Table 2.1 is a compilation of values for total porosity for different
lime based mortars. Lime mortars have higher total porosities than hydraulic lime mortars.
Figure 2.2, taken from Schäfer and Hilsdorf (1992) illustrates that the higher the porosity the
lower the compressive strength. Hayen et al. (2001), Papayianni et al. (2001) further show that
progressive carbonation of lime mortars results in an increase of compressive strength and a
decrease of porosity.

Table 2.1. Summary of pore data orf mortars made of different binders

Schäfer & Hilsdorf, Hayen et al., 2001 Papayianni et al.,


1992 2001
Type of binder Total Mean Total Mean Total Mean
Porosity pore Porosity pore Porosity pore
[%] size [%] size % size
[µm] [µm] (µm)
Lime putty 19.2 28 3
Hydrated lime 18 20 1 20 1.0
Semi-hydraulic 16 17 0.5
lime
Hydraulic lime 14.7
ln βD βD
2.5 (MPa)
Lime Putty
Hydrated Lime 10.0
Semi-Hydraul. Lime
Hydraulic Lime
2

5.0
1.5

2.0
0.5

0 1.0
-2 -1.5 -1 -0.5 0
ln (1- vp)

0.8 0.6 0.4 0.2 0


vp

Figure 2.2. Relationship of compressive strength to porosity (Schäfer and Hilsdorf, 1992)

2.3.1.2. Adhesion or Bond


Adhesion of mortar to the masonry unit, or bond, occurs in a narrow interfacial zone (< 1.5
mm). Lawrence & Cao (1987), Sugo et al. (1996) and Brocken (1998), among others, have
demonstrated for hydraulic mortars that bond is determined at the narrow interfacial zone of
mortar to masonry unit. This is due to the transport of binder and aggregate fines to the
interfacial zone. It is the uniformity of the interfacial zone which contributes to bond. This
same model is assumed for lime mortars.

Disruption of the interfacial zone by coarse porosity (> 100 µm) will greatly reduce the
adhesion of mortar to masonry units. The formation of voids at the contact zone may be the
result of:

• Irregularities on the brick surface causing poor contact between brick and mortar
• Over sanding. Increased sand content will increase the porosity of the mortar. This will
increase the chance for larger (and more?) voids to develop, thus influencing the bond.
• Poor size distribution of aggregate. For example, the absence of an adequate amount of
fine fraction to be drawn into the contact zone, will result in increased porosity , and
decreased adhesion.
• Cracking or dissolution due to environmental influences (see damage chapter)
• Working and tooling of the mortar. Tooling of the mortar involves pushing on the plastic
mortar, reducing the size or eliminating the voids at the mortar / masonry unit interface.
2.3.2. Influence of Porosity on Moisture Transfer Properties
Moisture in pores occurs as adsorbed vapour on the surface and as condensed liquid, partially
or completely filling pores (Pel, 1995; Brocken, 1998; Fig. 2.3). Transport of moisture within
the pores of mortar, occurs only within interconnected pores. This is permeability. Relative
humidity (RH) of the air filling the pores has a direct influence on whether or not water is in
the form of a gas or a liquid (Fig. 2.4). At an RH of less than 65%, liquid water does not form
a continuous layer and vapour dominates. This is the so-called critical moisture content (wcrit).
The vapour may be as a single layer of molecules, or may build up to a multiplayer with
increasing RH. At relative humidity values between 65% and 95%, vapour and water exist
together, with water forming a continuous phase in the pores. This is the capillary transport
regime (wcap). At an RH >95% the pores are saturated with liquid water (wsat).

Different driving potentials govern the transfer process of moisture. Moisture transfer through
vapour diffusion is governed by a difference in vapour pressure, while the liquid moisture
transfer is governed by a pressure difference on the liquid. In coarse pores or cracks external
pressure such as wind pressure and gravity loads affect moisture transfer. Due to the high
moisture capacity of cracks and macro-pores, however, they can act as a reservoir drastically
increasing the available water to be moved into the mortar. Temperature gradients also
influence transport. In a material with a temperature gradient, water may be transported
through vapour diffusion in one part and then condense and be transported through capillary
suction in other parts. Condensation of vapour to liquid will occur at the dew point, and this is
often within a masonry wall. An increase in salinity of the pore fluid will increase the relative
humidity of pores and increase the dew point temperature.

Figure 2.3. Moisture transport mechanisms with different pore size (after Meng, 1996)
wsat all pores filled

Free Water
E
Moisture Content (wt.%)

wcap capillary suction

Absorbed Water
D D: Water in pores,
capillary suction

wcrit.

A B C Adsorbed water

0 20 40 60 80 100
Relative Humidity (%)

Region A: Single Layer of Region B: Multiple layers of Region C: Interconnected layers,


adsorbed water molecules adsorbed water molecules (Internal capillary condensation)

Figure 2.4. Moisture content of pores dependent of relative humidity.


Curves represent saturation curves (after Straube, 2002)

In masonry, moisture exchange also occurs between the mortar and different masonry
materials it bonds to. A simple rule of thumb is the direction of capillary transport will occur
from coarse porous material to fine porous material. This is generally considered to be a
fundamental premise to the concept of compatibility of mortar to masonry. The preferred
direction of moisture transport is from the masonry unit to the mortar. This is so the moisture
load is in the mortar which makes it sacrificial to the masonry unit. No thorough work has
been done to investigate the nature of the interfacial zone of lime mortars. Brocken (1998)
points out that the nature of the interfacial zone is important for the transport of moisture
across this boundary and it is not the same as the body of the mortar. He explains that for a
lime-cement mortar the lack of coarse aggregate and process of carbonation in the interfacial
zone may well create essentially a moisture barrier. How this relates to lime mortars needs to
be investigated.

2.3.3. Influence of Porosity on Durability


Durability is generally related to the resistance of a material in its environment. For masonry
and mortar the greatest influence on durability is the presence of pore fluid and the
composition of the pore fluid.

2.3.3.1. Salt Transportation and Salt Damage


Soluble salts in mortar or masonry units can present a serious durability risk to a mortar (see
damage chapter, and Fig. 2.5). Salts may come from the masonry materials themselves, or
supplied from exterior sources (salt laden wind, salt from soils through rising damp, residue
from combustion of wood, or biological residue). Damage occurs at or just below the surface
of the mortar, where there is a change in the degree of saturation of the salt laden solution
(Charola, 2000; Lewin, 1982; Selwitz et al., 2001).

Figure 2.5. Example of salt crystallization and mortar delamination in lime mortar

Wetting and drying cycles in the mortar are required to generate damage from salt. This
cycling occurs dominantly in capillary pores. An increase in relative humidity will cause some
salts to go into solution, but in the case of sodium sulphate (thenardite) there is a phase change
to sodium sulphate decahydrate. This phase change represents an increase in volume. If the
volume of the new phase cannot be accommodated the expansion, or if the displaced fluid
cannot move out of the pore the pore wall will crack. A decrease in relative humidity results in
the precipitation of the soluble salts, and the resulting crystallization also may cause damage
because of a change in volume and cracking. Wetting and drying cycles over time may result
in spalling of the surface of the masonry (Fig. 2.5, Fig 2.6). If a rendering or masonry has a
surface layer or coating that prevents capillary transport to the surface salt deposition can
occur at the contact zones. This is termed sub-florescence. The fluid becomes saturated with
salt, due to evaporation and precipitation and deposition occurs behind the render. This results
in delamination and bubbling of the render.
Figure 2.6. Example of efflorescence on brick surface and surface spalling of mortar

The influence of salts on composition of pozzolan or hydraulic mortars may also be


significant. A pore solution saturated in chloride will alter calcium-aluminum-hydrate phases
to calcium chloroaliminates, resulting in an increase of porosity. A sulphate rich solution may
combine with calcium from the pore solution of the mortar to form gypsum. Gypsum deposits
occur on the surface, often creating a coating, which may in turn reduce the surface
evaporation of moisture and enhance sub-florescence as described above.

2.3.3.2. Freeze-thaw durability


The porosity and the related moisture properties have a major influence on the freeze thaw
durability of a mortar. The degree of saturation, the total pore volume and the pore size and
distribution all contribute to freeze-thaw durability (Litvan, 1980; Litvan, 1981, Maage, 1984;
Robinson, 1984)

A simplistic view of freeze thaw damage is the volume increase 9% associated with the phase
transition from liquid to solid in water. If the pores are saturated with water, and transfer of
water is not possible the expanding volume will result in cracking. This is not a very important
process as water transport occurs during freezing. It can however affect porous aggregate in
the mortar.

T. C. Powers (1958) advanced the model of hydraulic pressure for frost damage in concrete.
Rapid freezing will cause the need for quick displacement of liquid water due to ice formation.
If moving the water to an unfilled pore does not accommodate the hydraulic pressure
generated from the confinement of the water, failure is likely to occur. The hydraulic pressure
increases with increasing average distance between the air-filled voids and higher degrees of
saturation.
It is clear that the porosity of a mortar is an important factor for frost resistance of a mortar.
Most microscopic techniques (see 3.1.2) investigate the total pore volume and specific surface
and calculated a spacing factor which is the maximum distance of any point in the binder from
the periphery of an air void (assuming that all voids are equal-sized spheres arranged in a cubic
lattice). Litvan (1981) makes the critical note: “Indirect methods for assessing frost resistance
are based on determination of parameters such as air content, spacing factor, critical degree of
saturation, pore size distribution and surface area. Each test gives valuable information on
some aspect of freeze-thaw durability. Unfortunately, none of these tests will provide clear-cut
answers, so that the need for careful analysis of the material properties and environmental
characteristics remains imperative”. Twenty years later, very little work has been completed on
freeze-thaw resistance of lime mortars. Field studies such as Waldum and Anda, (2000) and
laboratory studies such as by Marie-Victoire and Bromblet (2000) and Maurenbrecher et al.
(2000) have contributed some to our understanding of frost durability of lime mortars. All
point out that lime mortars, which are not fully carbonated, are more vulnerable to freeze-thaw
damage. Testing of fully carbonated laboratory mixed mortars has not been completed.

3. Methods for Assessing Porosity

The three values, which are fundamental to describe porous materials, are:

• Total Porosity (TP) is defined as the fraction of the total volume that is occupied by
pores. The porosity is further subdivided into open porosity and closed porosity. Open
porosity consists of pores which are connected to one another and is measured by
microscopy and intrusion techniques. Closed porosity consists of pores which are not
connected to each other and is measured by microscopy techniques.
• Pore Size Distribution (PSD) is defined as the fraction of total pore volume in which the
pores lie within a stated size range. It can be expressed either in integral or differential
form. Pore size distribution values representing total porosity must be obtained by using
different measurement techniques.
• Specific Surface (SS) is defined as the area or accessible surface area contained in unit
mass of volume of solid and can, in principle, be calculated using total porosity and pore
size distribution or indirectly from analytical techniques

Techniques for determining these values can be subdivided into direct methods, where the pore
structure is evaluated by direct observation; and indirect methods where the presence or size of
pores is inferred from measurements. It is not uncommon for two techniques to provide two
different values for the same sample. In describing the different techniques an effort is made
to give the reader an overview, not specifics. References detailing the technique are given.

3.1. Direct Methods for Total Porosity and Pore Size Distribution
3.1.1. Visual Assessment
This method uses the experience of the human eye to estimate the type and volume of porosity.
Visual assessment of porosity is one of the first steps in characterizing mortar (see sampling
chapter), but it must be realized that this method is a gross approximation of the total porosity
and pore size distribution. Describing shape and size as well as spatial distribution is
important. The unassisted eye can see pores larger than approximately 50 μm depending on the
surface, and using a 10X magnifying eye piece finer pores can be seen.
Comparison charts are a valuable tool for estimating percentages (Fig. 3.1). These are
commonly used in geological applications. They show different graphic representations of the
same percentages depending on shape and distribution. It is assumed that the surface is a cross
section of three-dimensional objects, which in this case are pores. Finding a flat surface
therefore is essential when using this technique.

3.1.2. Microscopy
A microscope provides a much more accurate method to visually assess the pore shape and
porosity of a mortar. The microscope can be a simple stereo binocular microscope, an optical
petrographic microscope, or as complex as a scanning electron microscope. The three types of
microscope may be required for a complete analysis.

Stereo binocular microscopes use incident light on the surface of the sample. The
magnification is from 10 to 40 times. This type of microscope has a relatively large working
distance and allows for samples up to 8 cm in height to be examined. The field of view is up to
30 mm depending on the magnification. The sample can be rough as taken from the field or
prepared for a more detailed porosity examination. The latter requires that the sample is sawn
to a flat upper and lower surface and polished.

Figure 3.1. Black dots represent 2 dimensional projections of pores. Percent volume of pores
progress 2% to 5 to 10% from left to right.

To prevent breaking the sample and ease of handling during preparation, the sample is usually
impregnated with a low viscosity resin. This resin may include a UV florescent dye which will
highlight the pores and cracks under UV light. Final preparation will involve polishing the
sample to a very flat and highly reflective surface. The sample is ready for both simple and
more advanced microscopic and point counting techniques.

The petrographic microscope provides the next level of magnification (500 times or more), but
the working distance is static. The optics cannot be moved up and down to accommodate
different sample dimensions. The field of view is limited from a few millimetres to tens of
micrometres depending on the magnification.

The preparation of a mortar sample for a petrographic microscope requires the mortar to be
sawn flat, glued on a glass slide and sliced and ground to a thickness of 30 µm. This is where
the term, “thin section” comes from. The thin section may be covered with a glass cover slip or
polished to a highly reflective surface (polished thin section). (It is the experience of the
authors that grinding the mortar sample thinner to 15µm greatly reduces the optical refraction
associated with the carbonate rich binder, thus allowing for easier viewing of pore-binder
interface. Mineral identification, however, at this thickness is very difficult.)

The petrographic microscope allows for the transmission of light through the thin section.
Binder and pores can be distinguished based on colour and texture. The use of coloured epoxy
for impregnating the mortar sample is very helpful to distinguish pore and binder. Training in
mineralogy and petrology is required to use this technique to its fullest. The reader is referred
to an optical mineralogy text such as Kerr (1977) for a fundamental overview of optical
mineralogy, and the use of the petrographic microscope.

The petrographic microscope can also be used with reflected light. In this case pores of the
sample are impregnated under a vacuum with a low viscosity resin, sawn flat and polished to a
highly reflective surface. The reflected light will distinguish the void spaces from the binder,
by the quality of the reflection. Resin filled voids are usually very reflective, compared with
binder and aggregate. The UV dye can enhance this examination, perhaps picking up pores
missed in other light conditions.

The scanning electron microscope (SEM) provides the highest magnification of the commonly
used microscopy tools. Detailing the science of scanning electron microscopes is beyond the
scope of this introduction and the reader is referred to such fundamental references as Reimer
(1998). Pores as small as 0.5 µm can usually be imaged easily. The ability to increase the
magnification depends on a number of factors, which include the composition of the sample
and the care in preparation. It is essential when determining porosity using a scanning electron
microscope that a flat, well polished sample is used.

3.1.2.1. Photomicrograph cut and weigh technique


A simple technique to grossly estimate the total pore volume is to take a photomicrograph of a
flat mortar surface. The area of the mortar in the photograph is cut out and weighed. The voids
are cut out and weighed. The difference in weight is an approximation of the total porosity.
This technique should not be used, except for the crudest of estimates of total porosity. It is
good introductory education tool. Point counting or digital imaging are the preferred optical
methods.

3.1.2.2. Point Counting


Point counting provides a good estimate of total porosity and pore size distribution within the
range of magnification of the technique. Interconnectivity of pores cannot be determined using
this method.

Point counting requires the use of either a stereo binocular microscope or optical microscope.
A manual or automated mechanical stage moves the sample the same distance in both the x and
y planes. It is conducted on a smooth, flat surface; usually a polished sample or thin section.
With each movement of the stage the observer assesses whether the point under the cross hair
is on a pore or other. If on a pore the diameter is measured manually and the count for pores is
increased by one. A specific cross-sectional area is examined in this manner to obtain a
statistical assessment of the pore volume.

The method is relatively inexpensive and an observer can be easily trained. The accuracy and
precision of the method is in the sample preparation. The observer must be able to clearly
distinguish a pore. It can be a tedious job, taking a technician several hours to complete a
sample. Digital techniques are a less time consuming approach (see 3.1.3).
3.1.2.3. ASTM C 457 Linear Traverse Method
This method can be used to calculate the air-void content, specific surface area and the average
distance between pores, or spacing factor.

The method requires stereo binocular microscope that provides magnification of x125, a
mechanical stage and a counting apparatus. This system can be automated in an image
analysis system. The traverse length required is dependent on the aggregate size. For historic
mortars this often requires the measurement of several samples in order to have a statistically
significant sampling.

3.1.3. Digital Image Analysis


Computerised digital image analysis as applied to optical light microscopy uses a digital
camera attached to the microscope. If a video camera is used the video signal is taken in to a
frame grabber in a computer where the analogue video signal is transferred to a digital image.
The image is composed of pixels, typically about “1024x1024”, giving over one million pixels
for each image. Each pixel defines a point in the image for which the position and grey scale
value or colour is known. It is then possible for the computer programming to classify and
group pixels in the image according to criteria established by the operator. There are many
commercially available imaging systems with proprietary computer algorithms. It is an area
where there has been tremendous technological advance with the progress of computing
capacity.

When applying image analysis for measurement of porosity is it possible to measure according
to linear traverse method or total sample area analysis. The total area method analyses the
complete image, providing complete coverage of the sample, and a more direct measurement
of the total porosity, pore size distribution and spacing factor than the linear traverse method.

It is also possible to use image analysis in combination with SEM in order to analyse smaller
diameter pores. The secondary image digital signal from the SEM can be taken in directly to
the image analysis system or alternatively it can be captured and stored for later analysis. Most
commonly, however, scanning electron microscopes using solid state Back-Scattered Electron
detector (BSE) are used (Fig. 3.1). Images acquired with a BSE-detector contain
compositional information because the grey level of the different phases observed in the image
is atomic-number dependent. The pores can easily be distinguished from other mineral phases
present in BSE-mode because of the low average atomic number of the epoxy-resin in the
pores. Automated image analysis techniques can be used to acquire a large number of images
and to measure the pore size distribution using area methods.
Figure 3.1.
Figure 3.1Examples of BSE-images
Examples of BSE- of polished
images ofsections
polished(porosity in black)
sections of
(porosity in
two mortars, one with low porosity (left) and relatively
black) of two mortars one with low porosity (left) and relatively high high porosity
(right)
porosity (right).

3.2. Total Porosity and Pore Size Distribution by Indirect Methods


Indirect methods can be thought of as black box techniques. The pores are not imaged, but
rather filled with a liquid and the amount of liquid introduced can be used to determine the
total porosity and, in some cases, pore size distribution.

3.2.1. Porosity by Immersion


This simple technique provides a good estimate of total porosity. It is based on the concept
that the weight of an object is due to the absolute mass of the solid. It follows therefore, that
by filling the pores of the object with a liquid medium of known density the total effective
porosity can be calculated from the change in weight. The term total effective porosity comes
from the fact that the pores must be connected for the measurement to be made. Pores, which
do not fill are not accounted for in this or any of the other similar methods.

Using water as the immersion liquid is the simplest approach. Simple immersion for a set
period of time, say 24 hours gives an indication of the pore volume occupied by the larger
capillaries. The water volume uptake by smaller capillaries takes much longer, but these pores
can be filled if left submerged in water. To speed up the test method a vacuum is drawn on the
sample and water sucked into the small pore when the vacuum is released. The total open
porosity of porous materials is most often determined by a simple measurement of the porosity
accessible to water under vacuum (RILEM CPC 11.3, 1984)

The porosity accessible to liquid is expressed as a percentage of the bulk volume and is
calculated as follows:

⎡ M 3 − M1 ⎤
Total Porosity = 100 x ⎢ [ 3.1]
⎣ M 3 − M 2 ⎥⎦

where
M1 = mass oven-dried,
M2 = mass in water,
M3 = mass in air of water-saturated sample after vacuum.
A problem often encountered with small mortar samples is the determination of M3 which is
the mass of the sample in air of the saturated sample. Before weighing the wet specimen in air,
excess water has to be taken away using damp chamois leather. Just the abrasion of the leather
could damage a friable sample. In addition, water may be lost from the larger pores. This
makes the method too inaccurate when dealing with very small specimens as samples of
historic pointing mortar.

Van der Klugt (1998) suggests the following method for smaller samples. The sample is dried
to a constant weight (m1) at 60°C and partially immersed in oligomeric siloxane dissolved in
mineral spirits until it is saturated. Then immerse the sample for one hour in the solution.
Allow the sample to dry in a fume hood or outside (if above 0°C), protected from rain for 72
hours. Dry the sample for 24 hours at 60 °C. Let it cool in a desiccator and weigh to closest
0.01 g (M2). Weigh the sample in water, to the closest 0.01 g (M3). Make sure that during the
weighing the sample lies completely below the water.

Total Porosity = 100(1 – SM/2650 kg/m3) [3.2]

where

Specific Mass = SM kg/m3= 1000 M1/(M2 - M3) [3.3]

and 2650 kg/m3 is assumed to be the theoretical absolute density of the mortar.

3.2.2. Gas Pychnometry


Gas pychnometry provides values for total effective porosity and specific gravity (if mass is
known). The principle is based on Boyle’s gas law: At a constant temperature the volume (V)
of a gas varies inversely with the pressure (P) to which the gas is subjected.

P1V1 = P2V2 [3.4]

A pychnometer is comprised of a known volume cell, into which is placed the test sample.
There is a reference volume cell, separated from the test sample by a valve with a pressure
transducer. The test specimen cell is pressurized (~ 0,12 MPa) to a preset pressure and the
separating valve is open. Equation 3.4 shows that the pressure ratio and reference volume
value will determine the sample cell volume. The difference of the empty cell to that with the
sample in it is the difference in volume of the solid component of the sample. Helium is
typically used to penetrate the pores of the sample because it is inert and has a small atomic
radius. The minimum pore size penetrated is 5 nm diameter.

The sample size required is 0.5 g to 4 g (depending on the model). The sample preparation is
the same as for Mercury Intrusion Porosimetry. Commercially available pychnometers are
used. Testing is relatively quick, and reproducible. This technique determines the total pore
volume. As with Mercury Intrusion Porosimetry there is concern over using high penetration
pressures, which may break wall of pores. The reference volume cells must be calibrated
routinely to ensure the volumes are known.
3.2.3. Mercury Intrusion Porosimetry (MIP)
Mercury Intrusion Porosimetry provides total porosity and pore size distribution values of the
sample. The technique consists of enclosing the sample in a chamber, surrounding it with
mercury, applying monitored increments of pressure and measuring the volume of mercury
being forced into the sample. From the pressure, the surface tension of mercury, and an
assumed wetting angle, the pore diameter of an assumed cylindrical pore can be calculated.

This measurement technique is described in RILEM TEST N° I.5, ASTM D 4404. The use of
this technique and difficulties in the interpretation of the measurement data obtained were
recently reviewed by Diamond (2000). The advantage of this technique is the wide range of
pore sizes that may be measured. It requires a small sample size (0.5 g). It is a fairly rapid test
and is reproducible. Problematic is interpreting the data as an absolute representation of pore
size distribution. Figure 3.2 illustrates pore size distribution curves for the same sample using
the techniques of SEM and MIP. The two distributions are different, especially in respect of
the coarsest pores. It is well known that the MIP method can give a false distribution when
bigger pore spaces have to be filled by mercury through narrower connecting channels. This
effect, known as the "inkbottle-pore effect", can also be observed for most porous building
materials.

40
cumulative porosity (vol%)

35
30
25 MIP
20
15
10
5 SEM

0
1000 100 10 1 0.1 0.01 0.001
pore diameter (µm)

Figure 3.2. Pore size distribution as determined by: - Mercury Intrusion


Porosimetry (MIP) and - Scanning Electron Microscopy (SEM)

3.2.4. Nitrogen Adsorption/Desorption


This technique provides data for total porosity and pore size distribution. It uses the principle
that the amount of gas in a pore depends on the volume of the pore and the partial pressure of
the gas relative to its saturation pressure. The test procedure starts out with the pores saturated
with nitrogen (adsorption) at a temperature of 77 °K. The temperature is maintained, and the
partial pressure of the gas is increased resulting in progressive desorption. The pore volume
and pore size distribution is determined according to the BJH model (Barret et al., 1951) which
uses the Kelvin equation:

ln p/p0 = -2γVL*cosθ/rRT [3.5]

where,
p is the equilibrium vapor pressure of a liquid in a pore of radius r,
p0 is the equilibrium pressure of the same liquid on a plane surface,
γ surface tension of the liquid,
VL molar volume of the liquid (nitrogen),
θ contact angle with which the liquid meets the pore wall,
R the gas constant,
and T absolute temperature (77K).

The pore size penetrated is 3 to 300 nm, in the micro- and meso-pore range. It is an order of
magnitude smaller than that of MIP. The application to historic mortars is limited because for
the most part pores are coarser.

3.2.5. Specific Surface Area (BET)


The technique is referred to as BET referring to the initials of Drs. Brunauer, Emmett and
Teller (1938) who developed the theory and pioneered the equipment. This method is based on
the principle that at very low temperatures a gas will condense on the surface of a solid as a
monolayer. Simplified the equation is:

Surface area (m2/g)= VmσN/M [3.6]

where Vm is the volume of the gas required to form a monolayer,


σ the area of an adsorbed molecule,
N Avogadro’s number,
and M the molar volume of the gas.

While other gases are used, for most applications nitrogen gas is typical. When nitrogen is
used the equation is simplified to

Surface area = 4.25 Vm [3.7]

ASTM D 3663 provides a good overview of the technique.


3.3. Summary of Techniques

Table 3.1. Summary of Direct and Indirect Methods for Total Porosity and Pore Size
Distribution Determination.

Technique Information Tools Sample Pros Cons


Obtained Prep
Visual Assessment - total coarse - unassisted Simple -quick assessment - inaccurate
porosity eye, hand lens - coarse pores
– shape of coarse only
pores
Total Porosity by - total coarse - camera, Simple - low skill - inaccurate
phototmicrograph cut and porosity balance
weigh -shape of coarse
pores
Point Counting -total porosity - stereo or Moderate - accurate value - time
-pore size petrographic within the range of consuming
distribution microscope the microscope - requires
- mechanical statistical
stage interpretation
ASTM C 457 Linear - total porosity - steroscopic Moderate - long history for - requires
Traverse Method, EN 411 - distance between microscope, concrete analysis skilled
pores mechanical technician
stage, counting
apparatus
Digital Image Analysis -total porosity - digital video Moderate - automated, -requires
(Bench top microscope) -pore shape camera, acquire large mathematical
analysis - stereoscopic amount of data algorithm to
-pore size or petrographic quickly calculate
distribution microscope
– mechanical
stage
Digital Image Analysis -total porosity -SEM with Moderate Automated, - requires
(SEM) - pore shape back scatter acquire large sophisticated
analysis capabilities amount of data equipment,
-pore size quickly, - < 500 µm
distribution - identify pores pores
0.5µm or larger analyzed
Immersion and Vacuum -total effective - weigh scale Simple - relatively simple
Saturation porosity - oven
- vacuum
chamber and
pump
- pychnometer
Gas Pychnometry -total effective - weigh scale Simple Relatively simple -requires
porosity - oven and fast to specialized
pychnometer complete test equipment
with limited
use.
Mercury Intrusion -total effective -weigh scale Simple - relatively quick - hazardous
Porosimetry porosity- -oven - long history of material
pore size -mercury technique - specialized
distribution porosimeter equipment
-specialized
training
Nitrogen - total effective -weigh scale Simple -relatively quick - specialized
Adsoprtion/Depsorption porosity - nitrogen gas equipment
-pore size - BET machine -specialized
BET distribution training
specific surface
4. Methods for Determining Moisture Uptake and Movement

4.1. Capillary Water Absorption (Water Absorption Coefficient)


Moisture transfer in mortars as in other porous building materials can be considered as a
special case of ‘Multiphase Flow Theory’ with two fluid phases: a wetting phase (water) and a
non-wetting phase (air) Dracos (1991), Elsen (2000). In this regime, capillary forces largely
dominate the water flow in the porous material. Secondly, water flow is caused by the action
of gravity and by external pressures. Capillary liquid transfer is caused by a capillary pressure
difference in the porous microstructure. The origin of this capillary pressure difference is a
surface tension at the interface water-air. At the interface a disequilibria is formed because of
an imbalance between cohesive and adhesive forces of molecules of different type. The system
will tend to a state of minimum free energy by establishing an equilibrium by changing the
interface curvature. In this way water is transferred in a capillary without any external pressure
gradient. A detailed description of capillary phenomena can be found in Dullien (1979) or in
Dracos (1991).

The capillary pressure pc over the air-water interface is given by the Washburn equation.

pc = 2 cos φ (σa,w /r) [4.1]

where
r = radius of capillary,
σ a,w = surface tension between air and water,
and
φ = contact angle assumed to be zero for mortars (surface is completely wet and the solid is
fully hydrophilic).

The capillary behaviour of a sample of a mortar can be characterised by a simple test method
'the capillary rise water absorption test' as described by RILEM TEST N°II.6 and by EN-1925.
After drying to constant mass, the base of the sample is immersed in water on one of its sides
and the increase in weight is measured as a function of time. The mass of water absorbed,
divided by the horizontal area of the base of the specimen (g/m2) (Y) is plotted on a chart
against the square root of time (X) until the rising water reaches the top of the sample. The
slope of this straight line (A) is relative to the velocity of the capillary absorption and is called
the water absorption coefficient by capillarity. The second line shows a very slow increase of
the sample's weight and corresponds to the filling of very small pores and to the diffusion of
the trapped air in the pore structure, which can lead to an increase of the saturation up to nearly
100% after a very long time. Mortars (as most inorganic building materials) obey the √t law,
so that the following equation holds:

Y = k + A √t [4.2]

Where
A = water absorption coefficient by capillarity (g/m2√s),
and
k = a finite generally small positive intercept.

Other available methods are ASTM C 1403 and EN 1015-18: 2002.


An easy and rapid test for measuring the water absorption by the mortars is the initial rate of
absorption (IRA). The IRA is measured by immersing a dry sample of known mass in water to
a depth of 3 mm. The sample is removed from the water after one minute and weighed. The
IRA is calculated as the difference in mass divided by the area of the absorbing surface (EN
772-11:2000). This technique is not as accurate as the water coefficient method described
above.

4.2. Suction Curves


The variation of moisture transport through different pore sizes may be determined by means
of suction tests. Three methods are necessary to cover the complete suction range of pore
diameters ranging from 3.0 mm to 3 nm:
• Suction plate method
• Pressure membrane method
• Vacuum desiccator method

The three methods are described by RILEM TEST N°I.4 .

4.3. Water Vapour Transmission, Permeance and Permeability


Water vapour transport will dominate in mortar when the liquid moisture content is lower than
the critical value. Vapour will condense to liquid below the dew point temperature.
Condensation can occur in a masonry wall or on its surface. The conditions for condensation
can cycle by the hour depending on the weather; by the day depending on the variation in day
and night temperature, and seasonally. Vapour transmission, is therefore a relative value, not
absolute.

Determination of water vapour transmission (WVT) uses the method of determining the weight
of vapour transferred through the mortar over a period of time through a specific surface area.

WVT (g/s·mm2)= G/t·A [4.3]

where G = weight change (g)


t = time during which G occurred (sec.)
A = test area through which vapour was transmitted (mm2)

Permeance is WVT over the difference in vapour pressure on either side of the test sample.

Permeance = WVT/Δp [4.4]

Permeability is permeance over a unit thickness

Permeability =Permeance /unit thickness [4.5]

ASTM E 96 and DIN 52615 provide several test methods to determine water vapour
transmission, permeance and permeability. They require the use of materials to define the
relative humidity below and above the sample. Two commonly used materials are CaCl2 and
water. Other salts are used for a wide range of relative humidity values.

These test methods provide, at best, relative vapour transfer values. Combined with capillary
water uptake and vapour transmission, it might be possible to determine the localized changes
due to such things as the presence of water repellents or a change in mortar material. The
problem with using this technique on historic mortar samples is that often the samples are very
small, too small to be able to conduct a meaningful test.

4.4. Fluid Flow Imaging


There is a need for studies of water transfer in masonry during initial setting of the mortar and
during wetting/drying cycles. Several methods can be used to image fluid flow in porous
media such as X-ray Computed Tomography (CT), Nuclear Magnetic Resonance (NMR;
Jeremy et al., 1987) and Dynamic Neutron Radiography (DNR, Jasti et al., 1987). The
methods are generally expensive and are mainly applied in research.

NMR is a powerful technique for determination of chemical structure. It has been used in
studies of hydration of cement paste. It can, however, also be used for analysis of porosity and
fluid moisture and ion flow (Pel, 1995) and (Kopinga and Pel, 1994)). The principle of NMR
is based on the interaction of nuclei of certain atoms exposed to a strong magnetic field and a
radio frequency field. When the spin quantum number of a nucleus is nonzero, the nucleus
possesses a magnetic moment. The equipment requires experienced staff for operation and the
interpretation of the spectra.

DNR images are obtained by placing the sample in front of a neutron beam. As the neutrons
pass through the sample they are attenuated and the intensity of the beam decreases. The
amount of attenuation depends on the thickness and composition of the sample. A pore filled
with fluids will thus decrease the intensity more than a pore filled with air. The intensity of the
attenuated beam contains information on porosity and water saturation. It is possible to study a
water flow in real time on a TV screen.

Computed tomography (CT) is most widely used for medical applications. The usage of
conventional film radiography has the disadvantage that a three-dimensional object is
compressed into a two-dimensional image. CT employs a collimated X-ray beam, a detector
package, and computer algorithms to inspect planes within the object. CT images give
reconstructions of cross-sectional profiles. Internal features can be viewed in three dimensions;
valuable information on material properties and dimensional characteristics of the object are
preserved. This method can give information about porosity distributions, fluid saturation
distributions, capillary pressure functions, and relative permeability functions.

These methods are not routine, and are typically used for high-level investigations. Their
results are applicable in broad application, rather than specific ones.
4.5. Summary of Techniques

Table 4.1. Summary of Moisture Transfer Determination Techniques.

Technique Information Tools Sample Pros Cons


Obtained Prep
Capillary Water Rate of water - weigh scale Simple - simple -difficult for
Absorption (IRA) absorption - oven - common small
- water pan data for samples
-timer bricks -consider
ed only
surface
pores
Suction Curves -moisture - suction Simple - relatively - requires
transport plate, simple multiple test
through membranes method to set-ups
different pore or estimate
sizes dessicator gross pore
size
distribution
Water Vapour - determines - weigh scale Simple - good - not an
Transmisssion/Perme vapour -oven comparative absolute
ance/ Permeability transmission -water pans data value
properties -desiccants -requires
sufficiently
large sample
Fluid Flow Imaging -images fluid -specific Complex - very
flow equipment sophisticated
technique
- routine
application
not
established
NMR Imaging Total -NMR Simple - tremen - highly
effective machine dous specialized
porosity, potential equipment,
specific application
surface to masonry
not complete

5. Discussion

The complexity of the pore structure of historic mortars, makes it difficult to interpret total
porosity and pore size distribution data. The mortar consists of a porous binder together with
different, often porous aggregates. These include different types of stone aggregate and lime
lumps, each component having itself a different pore structure. Only a combination of
techniques could characterize the porosity of such a mortar.

5.1. Potential and limitations of porosity characterisation


Three values are fundamental to describe porous materials:
• total porosity (TP)
• pore size distribution (PSD)
• and specific surface (SS)
It is these values that are used to assess the influence of porosity on mechanical properties,
moisture transport and durability. Deciding, however, the approach and technique appropriate
for the question at hand should be thoughtful. Table 5.1 organizes the techniques described in
this chapter: top to bottom represents an increase in complexity of analysis and detail of
information to be obtained. The most direct and the simplest tools include the naked eye, hand
lens and stereo binocular microscope. These allow for the examination of macro pores and
cracks. A petrographic or scanning electron microscope examination allows for more detailed
investigation.

Table 5.1. Organization chart for porosity determination techniques.

Porosity Investiagion of a Mortar

Mortar Sample

On site visual Inspection Microscopic Inspection Indirect Measurement

Naked eye stereo binocular microscope Total Volume

Hand lense petrogaphic thin section examination


vacumm saturation
scanning electron micrscope
pychnometry
tomography
BET

Pore size distribution

Mercury intrustion

Ni adsorption/desortion

Primary and secondary porosity can be viewed and values for total porosity, and pore size
distribution, using image analysis can be determined
The simplest indirect porosity measurement is total pore volume determination by vacuum
saturation. Intrusive techniques using pychnometry and MIP, will also provide total porosity
and pore size distribution, but equipment is required. Provisions for mercury in the work
environment must also be considered. Specific surface measurements are best completed by
BET methods. Tomography, which includes NMR is in its developmental stage, and is
showing itself as an excellent research tool for moisture transport, but application to
characterization is limited.

In any analysis and interpretation of porosity of mortar the following must be taken into
consideration:

o There is no analytical method available, which will give the precise nature of porosity
of a material.
o One analytical method may not give the same results as the other.
o Without a complete statistical sampling the analytical results are estimates. Other
data such as chemical analysis or petrographic examination must be used to
understand if the sample is representative of the original mortar, replacement mortar
or damaged mortar.
5.2. The practical significance of porosity characterisation
The porosity of a repair or replacement mortar must be compatible with the surrounding
masonry unit and the masonry as a whole in the following areas:

o Mechanical properties,
o Moisture transport properties
o Durability

5.2.1. Mechanical Properties


Determining the mechanical properties of a historic mortar can be difficult because of the size
of the sample, or the condition of the sample. Porosity analysis can be helpful to gain some
comparative information on compressive strength, because of the inverse relationship of a
decrease in compressive strength with a gain in total porosity. There should be no expectation
of absolute values.

Where a decrease in compressive strength might be suspected a non-damaged sample should


be compared against a damaged sample. An increase of total porosity is an indication of
decreased compressive strength. This would also influence the critical tensile strength. Pore
Size Distribution would indicate if the gain in total porosity represents a coarsening of pores or
an increase in the amount pores. In most damage cases the increase is due to an increase in
pore size.

5.2.2. Moisture Transport Properties


Characterizing moisture transport properties must be done in context of knowing where the
mortar is from and its function. The location of the mortar allows the investigator to consider
its importance to moisture transport. Mortar in contact with the surface, masonry units or deep
in the interior to the structure will have different moisture loads and therefore have different
roles for moisture transport.

Surface mortar, which would include exterior renders or the first 2 cm from the exterior
surface of a joint can be either the wick for moisture or the reservoir. Rain on the surface can
be driven into the wall by rain, or drawn in by capillary transport. Near surface cracks or voids
filled with water can act as ponds, supply a source of water for capillary transport into the
interior of the mortar. IRA tests of the removed mortar can provide a good indication of the
nature of the pore structure at the surface. Large cracks or voids cannot draw water into them,
and the IRA values will be relatively lower than a mortar that has numerous capillary pores at
the surface.

Mortar in immediate contact with the masonry unit should be investigated for the nature of the
pore structure. This is particularly important where there is concern in a possible change of the
direction of the moisture transport. Recall that moisture will move from coarse pores to smaller
pores by capillary action. Petrographic examination to determine the composition of the
interfacial zone and PSD are the best tools to investigate this. Knowing the PSD and IRA of
the contacting masonry unit is also necessary. A relatively non-porous stone will not have the
impact on moisture transport across the joint as a very porous stone.
Mortar in the interior of the masonry structure should be investigated to determine the nature
of the primary mortar. Typically this mortar will not have secondary overprints associated with
exterior and contact mortar. The total porosity and pore size distribution will be an indication
of the bearing capacity of the mortar and the nature of the materials. High total porosity (>
20%) without small pores (<1 µm) most likely indicates a lime-sand mortar. Higher values
still of both total porosity and coarsening of the pores would indicate that the mortar is
probably altered and friable. Low values of both total porosity and more fine pores might
indicate the presence of a hydraulic lime phase. This would be confirmed using other
analytical techniques.

5.2.3. Durability
For durability the ability of moisture to move within the mortar and the accessibility of a
mortar to damaging agents is an important parameter. In these respect porosity plays a major
role. The pore distribution and changes of pore distributions over time can be evaluated using
microscopic techniques. This includes the estimation of a binder-moisture ratio at the time of
production of a mortar, the establishing of the use of additives such as air-entraining agents,
the formation of cracks, the change of porosity as a result of leaching, effects of transport of
soluble material resulting in reprecipitation or recrystallisation. Also microscopic techniques
are a powerful tool to study interfaces between old repair material and, between aggregates and
pastes. These interfaces are often the places from where durability failure may originate.

Another important durability parameter is frost resistance of mortars. For lime mortars the
susceptibility to frost damage especially of young uncarbonated mortars is notorious.
Microscopic techniques can be deployed to evaluate carbonation progress as a function of time
and composition (porosity) of the mortar resulting in recommendations for the application of
lime mortars in frost prone regions. Although carbonated lime mortars show high frost
resistance it is known as well that some restoration measures, like the application of very dense
repair jointing mortars or the use of water repellent may influence the moisture conditions in
masonry in an unfavourable way increasing the frost susceptibility of lime mortars ( see for
instance chapter damage cases: case 3.1).

The durability of structures may be essentially impaired by expansive reactions of salts; these
are crystallisation and hydration. As damage caused by salt often looks similar to that of frost
damage a distinction between the two of them is needed to take considered repair measures.
Using petrographical methods and eventually SEM-techniques it is possible determine the
nature of the salts if present.
It can be concluded that microscopic techniques may considerably contribute to the
understanding of durability problems and damage cases in masonry structures.
6. References

1. ASTM C 457-98 Standard Test Method for Microscopic Determination of Parameters


of the Air-Void System in Hardened Concrete Linear Traverse Method, ASTM
International, West Conshohocken, PA, USA.
2. ASTM C1403-00 Test method for rate of water absorption of masonry mortars,
ASTM International, West Conshohocken, PA, USA.
3. ASTM D 3663-03 Standard Test Method for Surface Area of Catalysts. ASTM
International, West Conshohocken, PA, USA.
4. ASTM D 4404-00 Standard Test Method for Determination of Pore Volume and Pore
Volume Distribution of Soil and Rock by Mercury Intrusion Porosimetry, ASTM
International, West Conshohocken, PA, USA.
5. ASTM E 96-00 Standard Test Methods for Water Vapor Transmission of Materials.
ASTM International, West Conshohocken, PA, USA.
6. Balen, van K. and Gemert, van D. (1994), Modelling lime mortar carbonation”,
Material and Structures, Vol 27, 393-398.
7. Barrett, E. P., Joyner, L.G., and Halenda, P.P. (1951), The determination of pore
volume and area distribution in Porous substances. I. Computation from Nitrogen
Isotherms. Journal of the American Chemical Society, 73, 373-380.
8. Brocken, H.J.P. (1998), Moisture transport in brick masonry: a grey area between
bricks. PhD-Thesis, TU, Eindhoven , The Netherlands, ISBN 90-6814-087-6.
9. Brunauer, S., Emmett, P. H. and Teller, E. (1938), The Adsorption of Gases in Multi-
molecular Layers, Journal of the American Chemical Society, 60, 309-19.
10. Chandra, S. and Aavik, J. (1987), Influence of proteins on some properties of portland
cement mortar. Int. Journal of Cement Composites and Lightweight Concrete, Vol 9,
2, 1987, 91-94.
11. Charola, A.E. (2000), “Salts in the Deterioration of Porous Materials: An Overview”.
JAIC 39 (2000): pp. 327-343.
12. Diamond, S. (2000), Mercury Porosimetry. An inappropriate method for the
measurement of pore size distributions in cement-based materials. Cement and
Concrete Research, 30, 2000,pp 1517-1525. 22.
13. DIN 52615 Determination of the water vapour (moisture) permeability of
construction and insulating materials.
14. Dracos, Th. (1991), Multiphase flow in porous media, in Modelling and Applications
of transport phenomena in porous media, J. Bear and J.M. Buchlin, Editors, Kluwer
Academic Publishers, London.
15. Dullien, F. (1979), Porous Media. Fluid Transport and Pore Structure, Academic
Press Limited, London.
16. Elsen, J. and Barquin de, F. (2000) , Modelling of the Capillary Water Absorption of
Porous Building Materials, Int. Zeitschrift für Bauinstandsetzen und
Baudenkmalpflege, 6. Jahrgang, A edification Publishers, Heft 3, 2000, 293-306.
17. EN 772-11: 2000 Methods of test for masonry units. Determination of water
absorption of clay, aggregate concrete, autoclaved aerated concrete, manufactured
stone and natural stone masonry units due to capillary action.
18. EN 1015-18: 2002 Method of test for mortar for masonry – Part 18: Determination of
water absorption coefficient due to capillary action of hardened mortar.
19. EN 772-11: 2000 Methods of test for masonry units Part 11: Determination of water
absorption of aggregate concrete, manufactured stones and natural stone masonry nits
due to capillary action and the initial rate of absorption of clay masonry units.
20. EN 1925: 1999 Natural stone test methods. Determination of water absorption
coefficient by capillarity.
21. Groot, C. and Larbi, J. (1999), The influence of water flow (reversal) on bond
strength development in young masonry. HERON, Vol.44, N°2, (1999) ISSN 0046-
7316,63-78.
22. Hayen, R., Balen van, K. and Gemert van, D. (2001), The Influence Of Production
Processes And Mortar Compositions On The Properties Of Historical Mortars, Eds.
P.H. Bischoff et al., Proc. 9th Canadian Masonry Symposium, New Brunswick,
Canada. 2001 (CD ROM).
23. IUPAC Manual of Symbols and Terminology, Appendix 2, Pt. 1 (1972), Colloid and
Surface Chemistry, Pure Appl. Chem., 31, 1972, 578.
24. Jasti, J.K., Lindsay, J.T. and Fogler, H.S. (1987), Flow imaging in porous media
using neutron radiography. Society of Petroleum Engineers, 16950, p 175-182).
25. Jeremyn, K.M.S. and Hunter, B.K. (1987), Modern NMR Spectroscopy” Oxford
University Press)
26. Kerr, P. F. (1977), Optical Mineralogy, McGraw Hill, 442 pp
27. Klugt van der, L.J.A.R. (1983), SBR/CUR-publication 299/- 98-03 The quality of
masonry pointing (Annex F), Rotterdam, 1983. [in Dutch] [new edition]. Porosity
determined on basis of water displacement of a sealed mortar sample and assumed
absolute density of 2650 kg/m3 .
28. Kopinga K. and Pel L. (1994), “One-dimensional scanning of moisture in porous
materials with NMR,” Rev. Sci. Instrum., vol. 65, 1994, 3673-3681.
29. Lawrence, S.J. and Cao H.T. (1987), Microstructure of the Interface Between Brick
and Mortar. Proc. 8th International Brick/Block Masonry Conference, Dublin. Ed. J.
W. DeCourcy. Vol. 1. pp.194-204.
30. Lawrence, S.J. and Cao, H.T. (1986), An Experimental Study of the Interface
between Brick and Mortar. 4th North American Masonry Conference, Los Angeles,
48.1 – 48.14.
31. Lewin, S.Z. (1982), The mechanism of masonry decay through crystallisation. In
Conservation of historic stone buildings and monuments. Washington, D.C.: National
Academy Press. 120-44
32. Litvan, G.G. (1980), Freeze-thaw durability of porous building materials. ASTM STP
691 Durability of Building Materials and Components, p 455-463
33. Litvan, G.G. (1981), Frost action in porous systems. Séminar: Durabilité des Betons
et des Pierrres. Séminaire organisé avec la collaboration de l’UNESCO par le Collège
international des sciences de la construction. Saint-Rémy-lès-Chevreuse (France) pp
95-108.
34. Maage (1984), Frost resistance and pore size distribution in bricks, Matériaux et
Constructions, Vol 17, No 101, p 345-350.
35. Marie-Victoire, E. and Bromblet, P. (2000), A new generation of cement based
renderings: an alternative to traditional lime based mortars. In: P. Bartos, C. Groot,
and J.J. Hughes, editors, International RILEM Workshop on Historic mortars:
Characteristics and Tests. ACM University of Paisley, May 1999, RILEM.
36. Maurenbrecher, A.H.P, Suter, G.T., Trischuk, K., and Fontaine, L. (2000),
Contribution to Pointing Mortar Durability. In: P. Bartos, C. Groot, and J.J. Hughes,
editors, International RILEM Workshop on Historic mortars: Characteristics and
Tests. ACM University of Paisley, May 1999, RILEM.
37. Neville, A.M. (1981), Properties of Concrete, Pitman Publising, Inc., Marshfield,
Mass, pp 271, 1981
38. Papayianni, I. and. Stefanidou, M. (2001), The Evolution of Porosity in Lime Based
Mortars. Proc. 8th Euroseminar on Microscopy Applied to Buildig Materials, Athens,
Greece.
39. Pel, L. (1995), Moisture Transport in Porous Building Materials, Ph.D. thesis,
Eindhoven University of Technology, 1995
40. Powers, T.C. (1958), Journal of American Ceramic Society, Vol.42, no. 1, 1958, 1-
6.].
41. Powers, T.C. and Brown T. L. (1947), Studies of the physical properties of hardened
cement paste – Part 6. Relation of physical characteristics of the paste to compressive
strength. Journal of the American Concrete Institute, 43, 845-865.].
42. Reimer, L, (1998), Scanning Electron Microscopy: Physics of Image Formation and
Microanalysis (2nd Edition). T. Tamir and A.L. Schawlow (Eds.) Springer Series in
Optical Sciences, Vol. 45.
43. RILEM Technical Recommendation CPC 11.3 (1994), Absorption of water by
concrete by immersion under vacuum. RILEM technical recommendations for the
testing and use of construction materials. E & FN Spon. 1994.
44. RILEM Test I.5 (Essai nº I.5) (1980), Porosity accessible to water. Matériaux et
Constructions, 13 [75] (1980).
45. RILEM Test II.4 (Essai nº II.4) (1980), Pore-size distribution (suction). Matériaux et
Constructions, 13 [75] (1980).
46. RILEM Test II.6 (Essai nº II.6) (1980), Water absorption coefficient (capillarity).
Matériaux et Constructions, 13 [75] (1980).
47. Robinson, G. C. 1984. The relationship between pore structure and durability of
brick. Ceramic Bulletin. Vol 63, No 2. p 295-300
48. Scanning Electron Microscopy: Physics of Image Formation and Microanalysis (2nd
Ed)(Springer Series in Optical Sciences, Vol 45 by Ludwig Reimer, T. Tamir
(Editor), A. L. Schawlow (Editor)
49. Schäfer, J. and Hilsdorf H.K. (1992), Ancient and New Lime Mortars – the
correlation between their composition, structure and properties. In Conservation of
Stone and Other Materials. Proceedings of the International RILEM/UNESCO
Congress, Volume Two, Prevention and Treatments, Proceedings 21, 1992, 605-612.
50. Sickels, L-B. (1981), Organic additives in mortars. Edinburgh Architecture Research,
8, 7-20.
51. Sommer, H. (1987), Beton aus der Römerzeit , Zement und Beton 32 Jahrgang , Heft
2, 1987, 62-66.
52. Sugo, H.O., Page, A.W. and Lawrence, S.J. (2001), The development of mortar/unit
bond Eds. P.H. Bischoff et al., Proc. 9th Canadian Masonry Symposium, New
Brunswick, Canada. 2001 (CD ROM).
53. Sugo, H.O, Page, A.W. and Lawrence, S.J. (1996), Influence of the macro- and
micro-constituents of air entrained mortars on masonry bond strengths. Seventh North
American Masonry Conference, South Bend, USA, 1996, 230-241.
54. Sugo, H.O., Page, A.W. and Lawrence, S.J. (1998), Influence of lime and methyl
cellulose on the microstructure and bond strength of mortars in combination with
calcium silicate units. Proceedings 8th Canadian Masonry Symposium, Jasper,
Canada, 1998, 348-359.
55. Waldum, A.H. and Anda, O. (2000), Durability of lime-based mortars in a severe
climate. Results from field and artificial ageing tests. In: P. Bartos, C. Groot, and J.J.
Hughes, editors, International RILEM Workshop on Historic mortars: Characteristics
and Tests. ACM University of Paisley, May 1999, RILEM.