Remediation and protection of masonry structures with crystallising moisture blocking treatment

Mazen J. Al-Kheetan (Department of Civil Engineering, Brunel University, Uxbridge, UK)
Mujib M. Rahman (Department of Civil Engineering, Brunel University, Uxbridge, UK)
Denis A. Chamberlain (DAC Consulting, Newcastle Upon Tyne, UK) (Department of Civil Engineering, Brunel University, Uxbridge, UK)

International Journal of Building Pathology and Adaptation

ISSN: 2398-4708

Publication date: 9 April 2018

Abstract

Purpose

The purpose of this paper is to investigate the performance of new and innovative crystallising materials, so-called moisture blockers, in protecting masonry structures from water ingress.

Design/methodology/approach

Two masonry wells were constructed: one with lime mortar and the other with cement-based mortar in order to hold water inside, and then a moisture blocking product was applied at dry and wet conditions to the negative hydrostatic pressure side. The moisture levels of both, the surfaces and the substrate, were then observed for 14 days.

Findings

Results demonstrated that moisture blocking materials are effective methods in reducing the levels of surface moisture for bricks, mortar-brick interface and mortar.

Originality/value

Moisture blockers use the available water in the masonry to block the passage of water to the surface of the masonry, filling pores, cracks and spaces at the interface between mortar and bricks. This approach will deliver a wider understanding of how water-based moisture blockers work and the scenarios in which they are best applied. The pursuit of possible environmentally friendly and sustainable materials for use in the construction industry is the key driver of this research.

Keywords

Citation

Al-Kheetan, M., Rahman, M. and Chamberlain, D. (2018), "Remediation and protection of masonry structures with crystallising moisture blocking treatment", International Journal of Building Pathology and Adaptation, Vol. 36 No. 1, pp. 77-92. https://doi.org/10.1108/IJBPA-02-2017-0011

Download as .RIS

Publisher

:

Emerald Publishing Limited

Copyright © 2018, Emerald Publishing Limited


Background

The UK transportation network operates some 40,000 masonry highways and 33,000 railway arch bridges. Many of these structures are over 100 year old and are subjected to increased traffic loads, speeds and progressive weathering resulting in structural deterioration. Water ingress in such structures is a widespread problem, traditionally addressed by time consuming and expensive injection of resins, with unpredictable success. Water ingress problem areas in masonry are often deliberately obscured by the introduction of canopies and sheeting, possibly with integral water collection. For aesthetic reasons, and because of the possibility of concealing destructive deterioration, this tends to be unacceptable, especially in the case of heritage structures.

Portland cement was not invented until 1828, thus masonry structures prior to the mid-nineteenth century were constructed using so-called natural mortars, with lime as the binding constituent. Masonry structures later than this generally included Portland cement as the binding constituent (Grimm, 1982; McKee, 1973). Under the action of water ingress, there is a tendency for the lime component to be washed out in the case of natural cement masonry, potentially causing hidden voids within the layers of masonry. This is often referred to as “lime leaching”. However, high porosity and other small-scale water transmitting defects could initiate a lot of problems with both types of mortar.

In mortar, there are two types of pores: interlayer space or very small capillary pores and larger capillary pores that contribute to permeability within a hydrated mortar mix. The former is approximately 0.015 µm in diameter and accounts for 28 per cent of the hydrated mortar volume. The larger capillary pores have a diameter of approximately 5 µm and constitute 0-40 per cent of the hydrated mortar volume (Ghosh and Melander, 1991; Luping and Nilsson, 1992; Mehta and Manmohan, 1988). Neville (2011) reported that pores larger than 5 µm are entrapped air and are unrelated to permeability. Cement fineness has been reported to influence water ingress through mortar (Powers et al., 1954; Nyame and Illston, 1981). However, researchers reported that the factors central to water ingress are the water-cement ratio and curing conditions (McMillan and Lyse, 1930; Wiley and Coulson, 1938).

Bricks that create masonry structures are formed mainly from clay (Hendry, 2001). This type of bricks may reach a compressive strength of 100 N/mm2, although bricks with an average of 30 N/mm2 are enough for the construction of houses and walls. Another constituent that takes part in the formation of masonry structures is mortar (Hendry, 2001). Mortars contributes in less than 10 per cent of the total volume of the masonry structure; however, it has an enormous effect on the stability and durability of the structure.

Moisture when exists in bricks it is considered the main reason for the deterioration of masonry structures, and when different types of salts coexist in these structures the condition gets worse (Franzen and Mirwald, 2009). D’Agostino (2013) reported that the process in which moisture contributes in the deterioration of these structures is hard to be fully cognized, as there are multiple mechanical and chemical activities weigh in its movement. It was found that a brick material would take longer to become fully saturated if it had a small threshold diameter (maximum diameter of continuous pores) as this exhibits a lower permeability. Also, it has been found that differences in pore size between the interior and exterior parts of the brick will contribute to brick deterioration. With a surface that has a low porosity obstructing the flow of water through the brick, and a centre that has a high porosity, the centre can become saturated leading to problems of freeze-thaw deterioration (Hansen and Kung, 1988; McBurney, 1929). As a direct result of this phenomenon, spalling brickwork can develop; water “entrapped” behind the face of the brick can freeze and expand which in turn will mechanically deface the brick. Continual cycles of wetting and drying might also contribute to gradual deterioration of masonry as it causes contraction and expansion in the bricks, and may result in cracks at the brick/mortar interface. This leads to an increase in the number of entry points for water ingress to take place.

Newman and Whiteside (1981) conducted a study in which it was found that just 17 per cent of total leakage was due to water ingress at the brick and mortar alone, which suggests that a large amount of water had leaked at the brick-mortar interface. The bond strength between the brick and mortar is undoubtedly a highly influencing factor of crack initiation and development at the brick-mortar interface. The surface texture of the brick and mortar consistency also play a part in bond integrity; however, the main influence of bond strength is the water retention capacity of the mortar and rate of absorption of the brick (Palmer, 1931).

A research made by Pazderka and Zigler (2013) on protecting heritage masonry buildings advised to use some systems like bitumen and plastic coatings to protect masonry structures from water attacks, as they found that these systems offer the most effective solutions in terms of sustainability.

Many types of protective materials do exist and they are widely used for protecting concrete (Christodoulou et al., 2014; Bubalo, et al. 2014; Al-Kheetan et al., 2017b, c) and, on a narrower level, protecting masonry structures. Cementitious coating materials have been tested to treat water tanks and swimming pools, and they have shown some promising results in terms of protecting the structures from deterioration and in terms of keeping the quality of water within very good limits (Al-Kheetan, et al. 2017a). However, these kinds of materials were only tested on concrete structures without considering masonry structures. Protection of porous materials by means of impregnation, like dual crystalline technology, is increasingly gaining popularity in civil engineering industry as it improves waterproofing and durability of porous media by filling voids with a non-soluble, highly resistant crystalline formation (Perkins, 2002). A schematic diagram of the dual crystalline material and their interaction with water was produced by authors, and it is shown in Figure 1.

The use of crystallisation materials, so-called moisture blockers applied to the negative water pressure side of masonry, presents an alternative approach to remediating water ingress in masonry. Their possible use is an attractive prospect because drilling is not required and the appearance of the masonry is unaffected after the treatment. The central claim is that they use the available water in the masonry to block the passage of water to the surface of the masonry, filling pores, cracks and spaces at the interface between mortar and bricks.

Research objective and scope

The aim of this research is to develop an understanding on moisture blocking techniques in reducing water leaking in the masonry structures. A quantitative and qualitative analysis is provided on the effectiveness and differences of two moisture blocking materials on cement and lime mortars, as well as on clay bricks. This approach will deliver a wider understanding of how water-based moisture blockers work and the scenarios in which they are best applied. The pursuit of possible environmentally friendly and sustainable materials for use in the construction industry is the key driver of this research.

Two water retaining wells are constructed from clay bricks: one using a lime-based mortar and the other a cement-based mortar. Two different moisture blocking materials are used, these separately applied to two different sides of the water wells, with one side of each water well left untreated as a control surface. The moisture blockers are applied to the negative hydrostatic pressure side, i.e. the exterior of the water wells, for the determination of their ability to dry the exterior surface. The performance of the moisture blockers when applied on brickwork with cement mortar and brickwork with lime mortar are investigated. The remaining side of each water well has a moisture blocking material applied while the surface is dry. This allows a comparison of the effectiveness of the moisture blockers when applied to initially wet or initially dry substrates.

Experimentation

Specification

To determine the effectiveness of moisture blockers, two water reservoirs were constructed (as shown in Figure 2). The dimensions of which are 550 mm×550 mm×675 mm. They are constructed using the following materials:

  1. clay bricks bonded with Natural Hydraulic Lime (NHL)-based mortar, and

  2. clay bricks bonded with Portland cement-based mortar.

Class B bricks were used in this test. The NHL mortar mix consisted of grade 3.5 NHL, which complies with the related British Standard BS EN 459-1: 2015 (British Standards Institution: BS EN 459-1, 2015), with sharp sand and water ratio of 1:3:0.45. The cement mortar consisted of Portland cement, sand and water in the ratio 1:3:0.45. These ratios were measured by volume as specified in Eurocode 6 (British Standards Institution: BS EN 1996-2, 2006; British Standards Institution: BS EN 1996-3, 2006; Roberts and Brooker, 2014). Great care was taken when mixing the mortars to achieve the best workmanship level in the water wells. A conventional cement mixer was used to achieve a consistent and thoroughly mixed mortar, and each batch of mortar was mixed for ten minutes in the mixer.

A moisture tolerant copolymer sealant was used to seal any visible defects in the water well, for example gaps in the brick/mortar interface. The water wells were cured for a period of two weeks to allow time for the mortars to properly harden and form bonds with the bricks. This is important as majority of water ingress in masonry occurs at the brick/mortar interface.

Two different moisture blocking materials were tested: MB1 and MB2. The technical specifications for both materials are given in Table I.

The sequence of material application is shown in Figure 3. MB1 was tested on two sides of the water wells under wet and dry application conditions. MB2 was tested on two water well sides under wet surface conditions. One side of each water well was left untreated as a control surface.

Both internal and surface moisture readings were taken over the course of 14 days, at three different heights on all sides of the water wells, 230, 380 and 530 mm from the base of the water well as shown in Figure 4. These are referred to as the low, middle and high points, respectively, in this paper. These points were free from efflorescence. It should be noted that the presence of some efflorescence on the surface does not have any impact on the moisture reading.

A Protimeter MMS2 moisture meter, with surface and internal moisture reading capabilities was employed. This device gives surface moisture readings by using electrical resistance values between two pin electrodes, which are placed on the test surface. Buried moisture is determined according to the dielectric constant measured by radio frequency waves projected into the solid. Readings are measured as percentage wood moisture equivalent (%WME) and relative wood moisture equivalent (RWME), respectively. WME is the moisture level in any material as if it were in close contact and in moisture equilibrium with wood expressed as a per cent moisture content of wood. %WME values below 17 are deemed dry, between 17 and 20 are at risk, and anything above 20 is wet. RWME values below 170 are deemed to be dry, values between 170 and 200 are deemed to be at risk and finally values above 200 are deemed to be wet.

UV dye was added to the water inside the wells with the intention of making leaking water more easily detectable on well surfaces. A portable illuminated digital microscope was employed to capture image of the well surfaces.

Experimental procedures

First, the initial surface and internal moisture readings were obtained at the three different heights on the exterior faces of each well. MB1 and MB2 were then applied to the allocated dry condition sides. On day 1, the wells were filled with water and the UV dye added to the contained water. The UV torch was then used to detect any major leakages. These leakages were then sealed with waterproof sealant to stop leaking. On day 2, the moisture readings were taken at the same three points on each side of each well. Once the wells had been refilled on day 2, MB1 and MB2 were applied to the wet sides, in the same fashion as before. On day 3, internal and surface moisture readings were then taken at the same three points on each face of each well. The wells were then refilled with water. The process was repeated until day 14, when the final readings were taken.

Results and analysis

Crystallisation of moisture blocker

Images for moisture blocker crystallisation were captured using a digital microscope, viewing the brick and mortar surfaces at approximately ×550 magnification. This was carried out on every face of each well, including the faces that were left untreated with the aim of obtaining a point of reference. Images were captured at the end of the experiment and they are shown in Figure 5(a)-(p). Irrespective of the type of moisture blocker, crystals are formed on most of the treated surfaces. The crystals are smooth and transparent on the low textured surfaces of cement mortar and smooth faces, with greater build on lime mortar and coarser face bricks. It appears that the surface texture of the substrate plays a great influence on the development of crystals on the surface.

Internal moisture at the brick-mortar interface

The internal moisture readings for both brick lime-based mortar and brick cement-based mortar interface are shown in Figure 6(a)-(f).

As expected, the level of surface moisture increases with depth under the action of increased water pressure with depth. Results analysed according to the depth reading and are discussed in the following sections.

High sampling point

At the high sample point (Figure 6(a) and (b)), each face of the well starts at similar levels of moisture in the range of 180-230 RWME. The level is marginally lower with cement-based mortar than the lime mortar. By day 4, the untreated face of the well starts to show increased readings of the internal moisture at a relatively fast rate; however, the rate starts to flatten around day 12, with a value around 340 RWME. The treated sides’ moisture levels also increase by day 4, albeit at a much slower rate and, by day 6, started to steadily decline, presumably marking the point in which the moisture blockers take effect. Interestingly, by day 12 the internal moisture levels on the treated sides of the well have decreased beyond the initial levels of moisture, and then continue to decrease. The face that was treated with MB1, while dry, reached internal moisture levels less than the other faces, but with marginal differences.

Unlike lime mortar, the moisture level for cement-based mortar increases from the outset, but settles by day 3 to similar moisture levels to those recorded for the lime mortar (about 210-230 RWME). The relative dryness of the cement mortar before the wells were filled is likely to be the cause for this, absorbing moisture relatively fast at the beginning. Like the lime mortar, internal moisture for the treated faces of the cement mortar well starts to steadily decrease by days 3 and 4, with no significant difference in performance between the two moisture blocker materials or their application conditions. From this point, internal moisture for the untreated face steadily increases for the remainder of the experiment. The marginal discrepancies in moisture levels, taken from day 3-4, are probably due to varying degrees of workmanship.

Middle and low sampling points

The middle (Figure 6(c) and (d)) and lower sampling points of the NHL well (Figure 6(e) and (f)) behave similarly in terms of internal moisture levels, all of which start at just over 200 RWME. The wet treated and untreated sides of the well have a relatively rapid increase in the internal moisture levels compared with the high sampling points, reaching their peak values by day 2-3. At the middle sample point, the two wet treated sides peaked at around 420 RWME with MB1-dry peaking at 377 RWME, whilst at the low sample point the wet treated sides and MB1-dry peaked at around 500 and 420 RWME, respectively. This increase in moisture is more contrasting with the low sample point relative to the middle sample point; the reason could be summarised in the following two points:

  • there is larger hydrostatic pressure that acts at the base of the well, driving greater moisture transmission, and

  • water levels in the well dropped due to leakage after refilling, causing the water pressure to be reduced at the top sampling point.

By day 4, internal moisture levels of the wet treated sides start to decline steadily towards approximately 250 RWME at the end of the experiment. Among the wet treated sides, MB1 seems to have a faster influence on the internal moisture levels; however, by day 7 at the middle sample point and day 5 at the lower sample point, the RWME values are very similar. The internal moisture levels of the untreated face show readings that follow the trend of the other sides until day 3, where levels continue to increase to approximately 500 RWME at the middle sample point and 600 RWME at the low sample point.

MB1-dry side shows an increase in the internal moisture that is less pronounced, with a significantly lower peak moisture level than the other sides. This is probably attributable to the opportunity that the masonry could absorb the moisture blocker into the substrate before the water ingress had commenced, allowing it to initially penetrate deeper into the core of the structure than the wet treated sides. This would also explain the reason behind forming much less crystals on the dry treated surface than that on the wet sides (explained in earlier section). By day 4-5 the moisture levels on the MB1-dry side at the lower and middle sample points, respectively, have peaked and converged with the other sides. Levels of moisture in all treated sides steadily decline at similar rates once converged, with no side significantly drier than another.

In terms of mortar type, the internal moisture levels in the cement mortar behaved very similarly to the lime mortar wet treated sides incurring a faster rise and peak in moisture levels than the dry treated side. The untreated sides increase as the treated sides, before settling and plateauing at the end of the experiment. There is a small but significant rise in the moisture levels with MB2-wet and MB1-dry sides at the middle point on day 4 (Figure 6(c) and (d)). This is probably attributable to small leaks at the brick-mortar interface.

Treatment of initially wet vs dry substrate

The internal moisture readings at the low sample point peak on day 2 at 500 RWME on the wet treated faces, whereas MB1-dry moisture levels peak from day 5-7 at 448 RWME. The readings for each treated face, as with the NHL well, converge around day 7, with MB1-dry levels marginally but consistently smaller than the other two treated faces. The internal moisture readings in the first four days of the dry treated face of both wells show a trend, the levels of moisture increases from day 1-2, then they set constant, before ageing increases. It is probable that water initially travels through the pores in the substrate where it reaches the moisture blocker, where it then activates the hydrophilic crystals leading to the settlement of the levels. The water then finds a route around this hydrophilic crystallisation through the brick/mortar interface leading to a secondary rise in moisture levels. Crystallisation then starts to form at the interface causing the moisture levels to reduce and the substrate to dry. Moisture levels at the middle and low sample points of both wells did not decrease all the way to the starting levels; however, their steady and consistent decline suggests that there is a large possibility that they would reach the starting levels if they were given more time.

Effectiveness of moisture blockers

The low sample point data are used in Table II to illustrate the difference in moisture levels between the two wells and their faces. The internal moisture levels of the NHL mortar were significantly higher than those of the cement mortar prior to the filling the wells, with the difference in RWME in the range of 15-27 between the corresponding external faces of each well.

Interestingly, the two water wells have very similar peak moisture values at each face with differences in RWME of 6. By the end of the experiment, the NHL mortar internal moisture levels have been reduced below that of the cement mortar (marginal reduction). The difference in moisture levels between all treated faces of both wells was kind of small with the largest and smallest values of 369 and 333 RWME, respectively. Should the experiment have continued, there is a possibility that this difference in moisture levels at this depth may have increased, as lime mortar has the ability to allow water vapour transmission, allowing faster drying of the substrate as MB1-dry has a lower peak internal moisture level.

Surface moisture for mortar

The %WME profiles at high, medium and low sample points are similar for the mortar-brick interface, although the magnitudes are significantly lower. Therefore, for brevity, only key measurements are shown in Tables III and IV.

Surface moisture at high, medium and low sample points

At the high sample point, the surface moisture on each face of the lime mortar started at similar levels in the range of 31-37%WME, although the rate of increase for each face was found to be different. The surface moisture of side MB2-wet peaks by day 4 at 47.1%WME, whilst sides MB1 (wet and dry) peak on day 5 at 47.5 and 46.1%WME, respectively. The untreated side, as expected, continues to increase throughout the experiment. The sides MB1-dry and MB2-wet decrease in surface moisture from their peak in a similar rate and trend.

The surface moisture of the cement mortar well at the highest sample point was similar to that of lime mortar, but there are some differences in moisture levels between the two. The MB1-dry and MB2 surface moisture increase in similar rates, initially faster than the untreated and MB1-wet sides. The moisture levels of the different sides converge and are similar by day 5. MB1-wet and MB2-wet moisture levels peaked by day 6 at 42.1 and 45.8%WME, respectively, before declining in the following days. This declination increases by the ninth day of the experiment and the surface drying rate increases, with MB1 consistently dryer than MB2.

Both the middle and lower sample point moisture readings for the lime mortar start at similar levels around the range of 32-40%WME, where they all increase at similar rates apart from MB1-dry that is slightly slower. As expected, the low sample point had a higher rate of moisture level increase. The untreated surface moisture continues to increase and plateaus at about 65 and 73%WME at the middle and low sample points, respectively.

The mortar surface moisture values of all treated sides reached their peak by days 5-6, at the middle height sample point, with a %WME range of between 55 and 58 per cent. Low sample points peaked at days 4-5 with a range of 64-68 per cent. At the low sample point MB1-dry has a slower initial rate of surface moisture build up, although it peaks slightly higher than the other treated sides. This initial rate is probably due to the activated moisture blocker material in the substrate, holding off some of the water ingress. The secondary rise is likely to be caused by water that has travelled around this crystallisation through the brick/mortar interface.

Initially wet vs dry treated substrate

At high and middle sample points on the lime mortar, MB1 is slightly more effective at drying the surface of the mortar than MB2, and it would remain to be seen whether this difference is significant on the long run or not. The same applies when comparing MB1’s effectiveness when applied on wet or dry conditions; MB1-dry appears to have been marginally more effective but the significance of this small difference is not determined.

At the low sample point, the wet treated sides and not treated side have a fast-initial increase in mortar surface moisture, before slowing down, the wet treated sides continue to increase until day 5 where they peaks at 51-54%WME. Not treated side continues to rise steadily at 64%WME. On the other hand, MB1-dry moisture levels increase slower than the other sides at the beginning, but with a steady increase and peaks at 55% WME on day 5, along with the other treated sides. By day 6, moisture levels on the mortar for all treated sides were declining, and by day 9 of the experiment the declination of moisture levels at all sides had strengthened. This increase in declination of mortar surface moisture has been observed at every sample point in both wells, except the sample points where some leakages took place. This is due to seepages at the brick/mortar interface; the seepage inhibits the proper drying of the surface, until enough crystals are formed at the interface (through the hygroscopic and hydrophilic nature of the material). This, then, reduces seepages and allows faster drying of the masonry surface.

Lime vs cement mortar

The surface moisture of lime mortar starts at higher levels than cement and remains consistently higher throughout the experiment, so at peak moisture, the difference decreases by the end to testing. The moisture difference between the two wells on their MB1-wet sides is shown in Table V.

This suggests that, given a longer period of monitoring, the surface moisture of the lime mortar might have been reduced more than the cement mortar. On the other hand, the difference in surface moisture at the MB2-wet sides increases between the peak –11.1%WME difference, and end –18.1%WME difference. MB2 applied to the wet substrate seems to have a significantly more effect on reducing surface moisture of cement mortar than lime mortar. It was observed that MB1-dry did not reach internal moisture readings as high as the other treated sides. This was not the case for the moisture levels on the surface of the mortar. This is because of water travelling through small brick/mortar interface gaps and gather on the surface, instead of travelling through the capillaries of the mortar matrix.

Brick surface moisture

The %WME profiles at high, medium and low sample points are similar to those of mortar and brick-mortar interface although the magnitudes are significantly lower. For brevity, only key measurements are shown in Tables VI and VII. Similar to mortar surface moisture readings, surface moisture levels on bricks show a large increase in moisture levels at day 13 and maintain at this level of moisture at day 14. This could be due to humidity observed at the end of experiment. It should be noted that although day 14 was taken as final reading, the high surface humidity may have affected the results at this stage.

For lime-based well (Table VI), treated sides commence at similar moisture levels of approximately 15%WME, and NT at 11.8%WME. The NT values increases steadily and levels at 30.5%WME by day 12. MB1-dry peaks on day 5 at 25.3%WME, earlier than the other treated sides, since the MB material had time to be absorbed into the brick and it was crystallising in the substrate, inhibiting water ingress. MB1-wet and MB2-wet surface moisture steadily increased; MB1 increased at a significantly slower rate and peaked at 21.6 and 26.4% WME, respectively, on day 6. After peaking, the surface moisture of the bricks on all treated sides then steadily declined, until day 10 where an increase in humidity occurred and thus, condensation on the surface caused an increase in moisture readings (see Figure 6(a)).

The surface moisture on the untreated side of the cement well (Table VII) commenced at 19.8%WME where it then steadily increased throughout the experiment, and started to settle by day 12 at 36.7%WME. All treated sides had a similar steady increase in surface moisture, with MB1-wet and MB2-wet sides reaching the peak moisture in the fifth day at 23.1 and 27.7%WME, respectively. MB1-dry surface moisture peaked later on day 6 at 27.1%WME, possibly due to crystal formation within the substrate, slowing, but not stopping, the water ingress through the brick capillary structure.

Brick surface moisture at middle and low sample points shows that all sides of the NHL commenced similarly, ranging from 13-16%WME, where moisture then steadily increased. As expected at both the middle and low sample heights, the brick surface moisture on the untreated side continues to rise and ends at 37.6 and 48.4%WME. By day 5 at the middle sample height MB1-wet and dry have both peaked at 31.3 and 28.9%WME, respectively. They then slowly started to decrease in surface moisture, while MB1-dry remained consistently dryer. Moisture at the MB2-wet brick surface continued to rise and did not peak until day 8 at 34.6%WME. At the low sample point both MB1-wet and MB2-wet reached their peak surface moisture by day 5 at 38.9 and 41.3%WME, where both started to decline. However, the rate of declination is significantly faster in MB1-wet side than MB2-wet.

Lime vs cement mortar

At both sample heights, all surface readings on the cement mortar well (Table VII) commenced at approximately 20%WME. When the surface moisture of the treated sides reached peak value, the untreated sides at the middle and lower sample points continue to increase and by the end of the experiment were 42.1 and 49.8%WME, respectively. The bricks in the cement mortar well actually started significantly wetter than the bricks in the NHL well. This difference decreases largely by the end of the experiment. The biggest change was on the MB1-dry sides of the wells, as seen in Table VIII.

The fact that every side of the cement well had wetter bricks suggests that the observed trend cannot be solely attributed to the conditions of the bricks pre-construction. This was observed due to NHL mortars ability to allow the transmission of water vapour. This allows the moisture in the NHL mortar to attempt to diffuse via the mortar, whereas the moisture in the cement mortar is absorbed, in part, by the bricks surrounding it. This would also explain why the cement mortar started dryer than the NHL mortar.

The final surface moisture on the bricks at the low sample point on the sides treated with MB1 is significantly lower than that of MB2. Therefore, the MB1 material is more effective at drying out the surface of clay bricks.

Conclusions and further work

Key conclusions drawn from this investigation are:

  • In general, the moisture blocker treated sides of the water wells were effective at reducing the levels of moisture transmission for bricks, mortar-brick interface and mortar. The NHL mortar surface moisture peaks in general a day before cement mortar. This is a result of the higher porosity of NHL mortar, which allows water vapour transmission, than cement, thus leading to a quicker impregnation of moisture blocker crystals within the substrate.

  • Both moisture blocking materials proved their effectiveness, similarly, at drying out the exterior of the water well masonry. However, MB1 did have faster initial influence on reducing moisture levels. When applied to either a wet or dry surface, MB1 is similarly effective at drying out the exterior masonry substrate. However, where possible, application before exposure to water ingress would substantially reduce initial buried moisture presence (by 68 and 50 RWME for the cement and NHL wells, respectively).

  • It appears that MB1 is more effective at drying and preventing NHL surface moisture than MB2. However, the difference is small and in real world applications it is possible to be insignificant. MB1, when applied in both wet and dry conditions, appears to be more effective than MB2 applied in wet conditions, with MB1-dry marginally out-performing MB1-wet. Again, this difference between the wet and dry applications of MB1 is possibly insignificant in real world applications, over long periods of time. Overall, MB1 is significantly more effective at drying the surface of the clay brick than MB2.

  • Readings taken from the mortar and brick suggest that the general porosity and permeability of the clay bricks is lower than that of the mortar matrix. This is apparent from the observation that surface moisture on the brick takes one to two days longer to reach peak level values (which is smaller than that of the mortar).

  • With no relevant standard approach to water well based testing exists, the approach in this research was improvised. In the on-going investigations, a standard water well format is being determined, with the use of multiple water wells that will facilitate statistical analysis of experimental data.

Figures

The function and interaction of dual crystalline materials

Figure 1

The function and interaction of dual crystalline materials

Test water wells

Figure 2

Test water wells

Plan view of water wells showing treatment of each external face

Figure 3

Plan view of water wells showing treatment of each external face

Diagram illustrating points where the readings by MMS2 were taken

Figure 4

Diagram illustrating points where the readings by MMS2 were taken

Microscopic images

Figure 5

Microscopic images

Internal moisture readings

Figure 6

Internal moisture readings

Technical data for MB1 and MB2

Property MB1 MB2
Specific gravity 1.10 1.09
Viscosity (centipoise) 2.2 2.4
Freezing point −5°C −4°C
Boiling point 110°C 105°C
Colour Clear Clear
Odour None None
Toxicity None None
Fumes None None
Flammability None None
Cleaning Water Water
Drying time (25°C) 2-3 h 2-3 h

Internal moisture levels at the low sample point of each face of wells prior to filling, peak and final readings

Depth readings (relative WME) Brick/Cement Brick/Lime
Face of well NT MB2 wet MB1 wet MB1 dry NT MB2 wet MB1 wet MB1 dry
Prior to filling 187 185 181 177 202 208 208 198
Peak moisture 592 504 499 431 597 510 504 454
End moisture (day 14) 592 369 356 344 597 339 342 336

The surface moisture levels of the lime mortar at each face of wells prior to filling, peak and final readings

Lime mortar
High sample point Middle sample point Low sample point
Surface readings (%WME) NT MB1 dry MB1 wet MB2 wet NT MB1 dry MB1 wet MB2 wet NT MB1 dry MB1 wet MB2 wet
Prior to filling 31.9 36.7 34.3 36.3 32.3 38.6 34.1 34.9 39.3 38.9 36.6 34.3
Peak moisture 56.6 46.1 47.5 47.1 65.1 57.3 55.1 56.1 73.7 68.7 68.8 64.6
End moisture 56.6 39.7 43.6 42.3 64.7 51.1 48.4 48.8 73.7 55.6 52.1 54.2

The surface moisture levels of the cement mortar at each face of wells prior to filling, peak and final readings

Cement-based mortar
High sample point Middle sample point Low sample point
Surface readings (%WME) NT MB1 dry MB1 wet MB2 wet NT MB1 dry MB1 wet MB2 wet NT MB1 dry MB1 wet MB2 wet
Prior to filling 17.1 15.6 22.4 18.1 20.1 18.2 23.2 21.3 18.3 20.5 25.4 22.9
Peak moisture 51.1 43 43.4 45.8 53.9 54.5 46.4 54.3 64.8 54.9 51.2 53.5
End moisture 51.1 41.9 43.4 45.6 53.9 53.8 46.4 50.8 64.3 46.1 50.1 42.9

Moisture difference for the wells on the MB1-wet side

Moisture level WME difference (%)
Prior to filling 11.2
Peak moisture levels 17.6
End moisture levels 14

The surface moisture levels on the lime mortar for each face of wells prior to filling, peak and final readings

Lime-based mortar
High sample point Middle sample point Low sample point
Surface readings (%WME) NT MB1 dry MB1 wet MB2 wet NT MB1 dry MB1 wet MB2 wet NT MB1 dry MB1 wet MB2 wet
Prior to filling 11.8 15.3 14.7 15.3 12.4 14.8 12.5 16.2 12.3 15.4 8.9 15.4
Peak moisture 34 33.2 28.4 29.9 38.9 38.6 29.7 34.8 49.6 42.6 41.3 38.9
End moisture 34 33.1 28.3 27.6 38.9 36.3 29.7 34.8 49.6 42.6 37.1 35.5

The surface moisture levels on the cement mortar for each face of wells prior to filling, the peak and final reading

Cement-based mortar
High sample point Middle sample point Low sample point
Surface readings (%WME) NT MB1 dry MB1 wet MB2 wet NT MB1 dry MB1 wet MB2 wet NT MB1 dry MB1 wet MB2 wet
Prior to filling 20.9 19.8 16.9 17.5 19.7 20.7 18.5 19.4 20.1 21.9 21.8 19.3
Peak moisture 37.9 32.3 36.4 34.3 44.9 37.2 43 35.7 51.4 48.4 42.9 40.7
End moisture 36.2 31.3 35.3 28.3 44.9 37.2 41.8 34.8 51.4 48.4 38 36.1

Moisture difference for the wells on the MB1-dry side

Moisture co WME difference (%)
Prior to filling 12.9
Peak moisture levels 1.6
End moisture levels 1

References

Al-Kheetan, M.J., Rahman, M.M. and Chamberlain, D.A. (2017a), “Influence of early water exposure on modified cementitious coating”, Construction and Building Materials, Vol. 141, pp. 64-71.

Al-Kheetan, M.J., Rahman, M.M. and Chamberlain, D.A. (2017b), “Influence of crystalline admixture on fresh concrete to develop hydrophobicity”, Transportation Research Board 96th Annual Meeting , Washington, DC, 8-12 January.

Al-Kheetan, M.J., Rahman, M.M. and Chamberlain, D.A. (2017c), “Influence of hydrophobic admixture and curing agent on water-resistant concrete”, Transportation Research Board 96th Annual Meeting , Washington, DC, 8-12 January.

British Standards Institution: BS EN 1996-2 (2006), “Eurocode 6. Design of masonry structures. Design considerations, selection of materials and execution of masonry”, British Standards Institution, London.

British Standards Institution: BS EN 1996-3 (2006), “Eurocode 6. Design of masonry structures. Simplified Calculation methods for unreinforced masonry structures”, British Standards Institution, London.

British Standards Institution: BS EN 459-1 (2015), “Building Lime Part 1: definitions, specifications and conformity criteria”, British Standards Institution, London.

Bubalo, M.C., Radošević, K., Redovniković, I.R., Halambek, J. and Srček, V.G. (2014), “A brief overview of the potential environmental hazards of ionic liquids”, Ecotoxicology and Environmental Safety, Vol. 99, pp. 1-12.

Christodoulou, C., Tiplady, H., Goodier, C.I. and Austin, S.A. (2014), “Performance of silane impregnants for the protection of reinforced concrete”, in Grantham, M., Basheer, P.A., Magee, B. and Soutsos, M. (Eds), Concrete Solutions 2014, Proceedings of Concrete Solutions, the 5th International Conference on Concrete Repair, Taylor & Francis Group, Boca Raton, FL, pp. 385-392.

D’Agostino, D. (2013), “Moisture dynamics in an historical masonry structure: the cathedral of lecce (South Italy)”, Building and Environment, Vol. 63, pp. 122-133.

Franzen, C. and Mirwald, P.W. (2009), “Moisture sorption behaviour of salt mixtures in porous stone”, Chemie Der Erde-Geochemistry, Vol. 69 No. 1, pp. 91-98.

Ghosh, S.K. and Melander, J.M. (1991), Air Content of Mortar and Water Penetration of Masonry Walls, Portland Cement Association, Skokie, IL.

Grimm, C.T. (1982), “Water permeance of masonry walls: a review of the literature”, in Borchelt, J.G. (Ed.), Masonry: Materials, Properties, and Performance, ASTM STP 778, American Society For Testing and Materials, Philadelphia, PA, pp. 178-199.

Hansen, W. and Kung, J.H. (1988), “Pore structure and frost durability of clay bricks”, Materials and Structures, Vol. 21 No. 126, pp. 442-447.

Hendry, E.A.W. (2001), “Masonry walls: materials and construction”, Construction and Building Materials, Vol. 15 No. 8, pp. 323-330.

Luping, T. and Nilsson, A. (1992), “A study of the quantitative relationship between permeability and pore size distribution of hardened cement paste”, Cement and Concrete Research, Vol. 22 No. 4, pp. 541-550.

McBurney, J.W. (1929), “The water absorption and penetrability of brick”, Proc. of the American Society for Testing Materials, Vol. 29 No. Part 2, pp. 711-739.

McKee, H.J. (1973), “Introduction to early American masonry – stone, brick, mortar, and plaster”, National Trust for Historic Preservation, Columbia University, Washington, DC, p. 61.

McMillan, F.R. and Lyse, I. (1930), “Some permeability studies of concrete”, Proceedings of American Concrete Institution, May, pp. 101-142.

Mehta, P.K. and Manmohan, D. (1988), “Pore size distribution and permeability of hardened cement pastes”, The International Congress of Chemistry, Paris, pp. 1-5.

Neville, A.M. (2011), Properties of Concrete, 5th ed., Pearson Education Limited, Harlow.

Newman, A.J. and Whiteside, D. (1981), “Water and air penetration through brick walls – a theoretical and experimental study”, Trans. of the Journal of the British Ceramics Society, Vol. 80 No. 1, pp. 27-36.

Nyame, B.K. and Illston, S.M. (1981), “Relationship between permeability and pore structure of hardened cement paste”, Magazine of Concrete Research, Vol. 33 No. 116, pp. 139-146.

Palmer, L.A. (1931), “Water penetration through brick – mortar assemblages”, Journal of the Clats Institute, Vol. 34, No. 21, pp. 19-31.

Pazderka, J. and Zigler, R. (2013), “Refurbishment of moist building masonry in terms of sustainable building”, Central Europe Towards Sustainable Building: Sustainable Building and Refurbishment for Next Generations, Sustainable Refurbishment of Existing Building Stock, Prague, 26-28 June.

Perkins, P.H. (2002), Repair, Protection and Waterproofing of Concrete Structures, 3rd ed., CRC Press, London and Madras.

Powers, T.C., Copeland, L.E., Hayes, J.C. and Mann, H.M. (1954), “Permeability of portland cement paste”, Proc. American Concrete Institute, Vol. 51 No. 11, November, pp. 285-298.

Roberts, J.J. and Brooker, O. (2014), “How to design masonry structures using Eurocode 6”, The Concrete Centre, available at: www.eurocode6.org (accessed 24 March 2014).

Wiley, G. and Coulson, D.C. (1938), “A simple test for water permeability”, Journal of the American Concrete Institute, Vol. 34 No. 9, pp. 65-75.

Further reading

Tait, C.J., Khalaf, F.M. and Kermani, A. (2014), “Review of developments in water ingress through stressed masonry”, p. 5, available at: www.globalskm.com/ (accessed 3 March 2014).

Corresponding author

Mazen J. Al-Kheetan is the corresponding author and can be contacted at: mazen.al-kheetan@brunel.ac.uk

About the authors

Mazen J. Al-Kheetan is a PhD Researcher in Civil Engineering at Brunel University, London. He has four years’ experience as a Project Engineer in two of the most important projects in Jordan: Queen Alia International Airport project – Amman, and Abdali Mall project – Amman. He has a Master’s Degree in Civil Engineering: Highways and Transportation from The University of Nottingham (2015). He gained his Bachelor Degree in Civil Engineering from the Mu’tah University – Jordan (2011). His research focuses on concrete pavement protection and remedy using hydrophobic materials. He is also interested in the performance and efficacy of protective materials applied under adverse climatic conditions.

Mujib M. Rahman is a Chartered Engineer and a Senior Lecturer in Civil Engineering at Brunel University, London. He has more than 15 years professional experience and has gained extensive research experience on the fundamental characterisation of asphalt and concrete materials, non-destructive-based evaluation of civil engineering infrastructures and protection of porous construction materials. He is author/co-author of over 70 highly regarded international journals and conference articles and has also written over 100 design and consultancy reports for government departments and private organisations. His research is sponsored by EPSRC, EC, Dft, Highways Agency, Institution of Civil Engineers and private organisations.

Denis A. Chamberlain, Fellow Member of the Institution of Civil Engineers, has 21 years’ experience as a Civil Engineering Contractor and Consultant in the UK and abroad. This combines with 27 years’ service as a University Professor, building research groups addressing a range of industrially relevant problems. He has five Patents for his inventions and has authored and co-authored numerous papers dealing with construction materials and their application. The effect of application time adverse climatic conditions on the life performance of materials is a particular interest of his. He is currently supervising research with his pot-hole repair heater at Brunel University, London.