Documenti di Didattica
Documenti di Professioni
Documenti di Cultura
ENTERPRISE IRELAND
INNOVATION PARTNERSHIP PROGRAMME PROJECT CODE IP/2008/0540
ECOCEM IRELAND
Executive summary
The case for research into the deterioration associated with water and wastewater infrastructure has been clearly made. For example in the US alone, the Congressional Budget Office has estimated that annual investments of up to $20 billion and $21 billion is required to provide adequate infrastructure for drinking water and wastewater respectively. It is also estimated that the annual operation and maintenance costs associated with drinking water and wastewater infrastructure to be in excess of $31 billion and $25 billion respectively. In Ireland, the Government have previously announced that under the Water Services Investment Plan, 5.8 billion will be spent on this sector. As such, this represents a potentially lucrative market that exists both in Ireland and internationally. Research was conducted into the performance of concrete samples manufactured using a range of binder combinations which were subjected to aggressive environments selected to represent the most extreme of conditions they will encounter in service. These included attack from sulfates and from biogenic sulfuric acid. The test results found that: In sodium sulfate expansion tests, traditional CEM I cements performed significantly poorer than all other binder combinations. Samples manufactured using the newer CEM II/A-L binders displayed increased resistance to sulfate attack but still failed to meet standard performance criteria. The samples with the highest resistance to sulfate attack were those that contained 50% or 70% GGSB as a cement replacement level, which displayed very low expansion levels. This also included superior resistance levels than that obtained using sulfate-resisting Portland cement. In all cases deterioration was primarily due to bulging, spalling and warping, most likely as a result of the formation of gypsum. When subjected to a 1% sulfuric acid solution (pH 1.3), significant surface corrosion was observed for all binder combinations. Very little distinction was observed between the various binder combinations. The main deterioration mechanism consisted of the formation of gypsum on the external surfaces of the concrete specimens. This was followed by surface delamination, some spalling. In the long term a widespread lack of cohesion leads to a failure mechanism that spreads directly to the core.
The results of this investigation have clearly outlined the cause of concrete deterioration in wastewater treatment systems and distinguished between degradation due to sulfate attack and that due to a sulfuric acid attack in this environment. There is a need to train concrete specifiers in understanding the harsh conditions that these structures will encounter in service. However for this to be fully realised, the range of aggressive environments associated with wastewater applications needs quantification.
TABLE OF CONTENTS 1 2 3 3.1 3.2 INTRODUCTION .................................................................................................................... 1 LITERATURE REVIEW ......................................................................................................... 3 EXPERIMENTAL PROGRAMME: MATERIALS AND METHODS ................................. 4 Experimental Overview .................................................................................................... 4 Materials and methods ...................................................................................................... 5 3.2.1 3.2.2 3.3 3.3.1 3.3.2 3.4 4 4.1 4.2 4.3 Sodium sulfate expansion tests .................................................................................. 5 Sulfuric acid tests ....................................................................................................... 7 Review of current research ....................................................................................... 10 Experimental setup ................................................................................................... 11
Permeability tests ............................................................................................................ 13 EXPERIMENTAL PROGRAMME: RESULTS ................................................................... 14 Overview ........................................................................................................................ 14 Sodium sulfate expansion results ................................................................................... 14 Discussion of sodium sulfate expansion results ............................................................. 23 4.3.1 4.3.2 4.3.3 4.3.4 4.3.5 Dilution effect .......................................................................................................... 23 Permeability and porosity......................................................................................... 24 The Influence of C3A ............................................................................................... 25 Influence of C3S and C2S ......................................................................................... 27 Sulfate resisting capabilities of CEM II/A-L and GGBS concretes ......................... 27 Mass loss results ....................................................................................................... 29 Discussion of deterioration mechanism ................................................................... 34 Cube strength tests ................................................................................................... 36 Sulfuric acid expansion tests .................................................................................... 37 Stiffness loss in due to sulfuric acid testing ............................................................. 38 Stiffness loss in concrete exposed to a 1% sulfuric acid solution ............................ 41 Discussion of ultrasonic results ................................................................................ 46 Microstructural Effects ............................................................................................. 46
4.4
4.5
Permeability, absorption and sorptivity results .............................................................. 49 DISCUSSION ........................................................................................................................ 51 Sulfate experimental programme.................................................................................... 51 Sulfuric acid programme ................................................................................................ 52 Ultrasonic analysis .......................................................................................................... 54 CONCLUSIONS .................................................................................................................... 56 Sodium sulfate tests ........................................................................................................ 56 Sulfuric acid tests............................................................................................................ 57 Summary......................................................................................................................... 58
1 INTRODUCTION
The deterioration of wastewater infrastructure has long been a cause for concern but the issues surrounding the problem remained unknown for many years. Traditionally designed to resist high levels of sulfate attack, wastewater treatment systems are subjected to a considerably more aggressive form of deterioration biogenic sulfuric acid corrosion. To this day the true mechanisms of attack have yet to be fully grasped by concrete specifiers and this is evident in the poor state of infrastructure used in this environment. There is clearly a need to distinguish between the two forms of attack and the research conducted in this investigation shows that both the nature and severity of attack is drastically different. Existing methods aimed at counteracting the corrosive forces focus on remedial work, periodic maintenance and replacement of defective components. This is clearly not a cost effective practice especially considering that this form of attack can manifest itself within a relatively short period of time (under ten years). In examining current European standards one discovers it is not necessarily the fault of the specifier that integrity of wastewater infrastructure is being compromised at such an early age, but rather a combination of a lack of understanding of the true mechanisms at work (owing to largely disjointed research efforts) and a concrete standard that does not adequately distinguish between the effects of sulfate and sulfuric acid attack. Further research reveals that the current ability to model sulfate degradation is lacking in both accuracy and reliability meaning adopting the recommendations of such work would be an unwise course of action. Much of this uncertainty is reflected by the very complex nature of sulfate attack, and to an extent sulfuric acid attack. Both mechanisms require an accelerated experimental analysis to monitor their destructive effects in a reasonable timeframe but by doing so this in turn modifies the nature and severity of the attack. In what would appear to be a somewhat Catch 22 scenario, existing recommendations for these aggressive environments resort to limiting the exposure of concrete to such environments or increasing cement content while lowering the w/c ratio without truly examining the mechanisms at work. The need to affect a change in such practice has become more important with the increased use of secondary cementitious materials in modern concrete practice. The effect of such additions often modify the chemical composition of a hydrated cement paste making some more
vulnerable than others to aggressive chemical environments. Simply referring to these constituents as cementitious materials may no longer be adequate terminology. Taking all of this into consideration, it is clear that the design of wastewater infrastructure is both imprecise and most likely inaccurate yet it is through the standards that change must come in order for a greater understanding to be achieved. Using guidelines such as increasing the cement content cannot justifiably claim to prevent the occurrence of complex corrosion mechanisms like those of sulfate, sulfuric acid or biogenic sulfuric acid attack. It would appear that a more complex and precise means of assessing concrete durability for long term exposures must be thought of. The challenge remains then to transform such a method into a usable tool for concrete specification. Existing research efforts on wastewater infrastructure tend to focus on either the cause or effect of the corrosion phenomenon while paying little attention to forming a solution to the problem. On the other hand, those examining possible solutions to sulfate or acidic corrosion fail to understand the biological aspects sometimes at work and the variability of the bacterial attack. These disjointed efforts remain the primary hurdle in establishing an authoritative direction in which to lead future research and it is the goal of this investigation to adequately contrast the separate corrosive phenomenon and to provide the guidance required to affect such a path.
2 LITERATURE REVIEW
A thorough review of the literature was conducted and a review paper was written entitled Biochemical Attack on Concrete in Wastewater Applications: a State of the Art Review. This paper was published in the peer-reviewed journal Cement and Concrete Composites and the full citation for the paper is: O'Connell, M, McNally, C & Richardson, MG (2010) 'Biochemical attack on concrete in wastewater applications: a state of the art review'. Cement and Concrete Composites, 32 (7):479-485. The paper is attached as an Appendix to this report. This study found that international research to date has focussed on three distinct topics in the study of sulfate/sulfuric acid effects on concrete. These are: Studies of the biological processes behind the corrosion of wastewater infrastructure, with particular reference to the role of sulfate-reducing and sulfur-oxidising bacteria. Studies of the chemical effects of sulfates and sulfuric acid on concrete mixes Laboratory-based research methodologies, especially those incorporating the biological effect on concrete. It was concluded that chemical tests alone do not fully represent the microbial effects on concrete, although they may help in assessing the types of damage that can occur. Some researchers have carried out full-scale laboratory analysis, but it is worth noting that the equipment necessary to adequately mimic in-situ conditions is invariably complicated, cumbersome and custom built. The realisation of resources required to undertake such research continues to be an obstacle to addressing this topic. The use of such complex research apparatus in routine performance-based specification is impractical. Although there exists significant quantities of data on the topics of sulfate, sulfuric acid and biogenic corrosion of concrete, little has been achieved in the way of formulating an accepted mathematical model of deterioration that incorporates agreed parameters of significance. This represents a significant knowledge gap and acts as a technical barrier towards using material design as a means of controlling corrosion due to biochemical attack. This continues to inhibit the design of durable concrete wastewater infrastructure and has significant implications for public expenditure in this area. The need to consider the interaction of biological and chemical processes may hold the key to achieving greater progress and allow practitioners to use concrete mix design as a means of delivering intended service lives. 3
CEM I, CEM II/A-L, SRPC GGBS (0%, 50% & 70%) None
Coarse Aggregate
Water
Tap water
Tap water
% GGBS
0 50 70 0 70 0
The standard exposure solution used in this test method contains 50g of sodium sulfate (Na2SO4) per litre of distilled water. Each solution was prepared with 4.5l of distilled water and mechanically stirred until fully dissolved. The solution was then topped up with distilled water until a volume of 5l was achieved. It was then placed into a standard domestic polyethylene container ensuring the quantity was sufficient to cover the prisms by a minimum of 5mm.
Figure 3.1 Example of the ASTM C1012 standard mortar prisms The prisms were stored for a period in excess of one year and the solution was refreshed on a monthly basis. This consisted of the disposal of the spent solution, the storage tank rinsed with water and a fresh sodium sulfate solution prepared as previously detailed. Comparator readings were taken every four weeks. The readings consisted of taking an initial reference measurement for each prism and a standard reference invar bar prior to immersion in the sodium sulfate solution. The change in length of the prism is recorded with reference to the initial reading and is then calculated according to: =
100
Eqn. 1
L = change of length at age x (%) Lx = comparator reading of specimen at age x reference bar comparator reading at age x Li = initial comparator reading of specimen reference bar comparator reading at the same time Lg = nominal gauge length or 250mm as applicable
The percentage change of length of each prism was expressed to an accuracy of 0.001% and the average of the four test specimens was recorded. A photographic record of the prisms was 6
carried out at eight week intervals to document any change in the physical appearance of the specimens.
3.2.2 Sulfuric acid tests The aim of the experimental programme was to assess the performance of six different concrete mixes exposed to environmental conditions similar to those found in wastewater treatment plants. The cubes were immersed in a 1% sulfuric acid solution for six months and monitored for mass loss, expansion, visual appearance and compressive strength. Three of the mixes consisted of CEM II-A/L limestone cement with 0%, 50% and 70% additions of GGBS as a cement replacement, two contained CEM I Portland cement with 0% and 70% GGBS while the final mix was a standard sulfate resisting SRPC mix. The performance of GGBS in an acidic environment was of particular interest while the performance of limestone cement remains largely undocumented in this type of aggressive environment. 7
Fourteen concrete cubes measuring 100mm were cast for each mix with varying percentages of GGBS as a cement replacement indicated in Table 3.2. The water / binder ratio was specified at 0.45 and binder content at 360kg/m3 to represent the limitations permitted under exposure class XA3 (>3000mg/l and 6000 mg/l SO42-; pH 4.5 and >4) as detailed in IS/EN 206-1. Simultaneously, four prisms measuring 250mm x 75mm x 75mm were also cast with each of the different mixes. The mix proportions for each set of cubes are outlined in Table 3.3.
Notes: F = Fine aggregate; C = Coarse aggregate Volume = Volume fraction of cube space occupied by solid matter SRPC = Sulfate resisting Portland cement
All of the cubes were stored in a curing tank at 20C for twenty-eight days following which the strength of cubes 1 and 2 for each mix was tested under compression until failure. The remaining twelve cubes were now divided into two sets of six split between continued storage in the curing tank and immersion in a 1% sulfuric acid solution at pH1.5 for up to 168 days. The dimension of each cube was also recorded prior to commencing the analysis. Table 3.4 shows the experimental schedule adopted for the test programme. The samples were brushed with a wire brush every seven days under running water which resulted in a milky-white runoff. Brushing was ceased when the runoff colour reverted to clear water. All cubes were weighed every seven 8
days for the first month and every twenty-eight days thereafter to record any mass loss or mass gain. For those stored in the sulfuric acid solution loosely adhering corrosion products were brushed away prior to recording mass.
Table 3.4 Storage and test conditions adopted for the brushed programme
Cube No.
1, 2
Storage Condition
28 days curing water @20C ONLY
Mass Readings
7, 14, 21 & 28 days
Compression Test
After 28 days curing water @20C
3, 4
9, 12
6, 7
10, 13
5, 8
11, 14
The sulfuric acid solution for all experimental schedules was maintained at a pH of approximately 1.5 by titrating the solution with a more concentrated sulfuric acid to keep the pH within a margin of 0.3. The solution was refreshed once a month to avoid prolonged contamination associated with the corrosion products of degraded concrete. A second set of mass loss experiments was undertaken over a twelve-month period exploring the effect that brushing the specimens, to remove loosely adhering corrosion particles, has on the attack rate. To this end, a further twelve specimens were cast (two for each respective mix) and their mass recorded every seven days for the first twenty-eight days and again every twenty-eight days thereafter until one hundred and sixty eight days exposure was achieved. The 9
cube strengths were also determined for each of the samples after one hundred and sixty eight days. The technique in the first programme has already been outlined while the second technique involved no weekly brushing. The specimens were briefly held under running water once a month, however, before being immersed in a fresh solution of sulfuric acid. The principle behind each brushing technique was to not only assess the behaviour in two extreme abrasion conditions for both stagnant and flowing water but also to attempt to mimic the severity of the bacterial acid attack. Finally, a parallel experimental regime was carried out to investigate the change of length of the concrete prisms exposed to the same sulfuric acid solution while adopting an identical weighing and weekly brushing technique as in Table 3.4. For this, following twenty-eight days curing, two prisms from each mix were immersed in the sulfuric acid solution while a further two remained stored in water. An initial reading for each prism was performed prior to immersion with a standard reference invar bar and subsequent readings taken at twenty-eight day intervals. The technique and calculations were performed according to those outlined for the sodium sulfate expansion tests.
Storage Condition
Mass Readings
Comparator Readings
3,4
Note: All prisms adopted an identical curing regime of 28 days in water at 20C following which they were exposed to the storage condition outlined above
of promoting high levels of cement replacement materials, the advantages and disadvantages should be acknowledged. While high replacement levels will have inherent benefits such as a denser microstructure or the ability to reduce the carbon footprint of a building, they also have some drawbacks, the most prominent of which is slower initial gain of strength. The development of a reliable ultrasonic technique to monitor the development of stiffness in early age concrete would allow us to work with this material property. This was the basis of a research paper that was presented at a Geophysics conference in Dublin in 2009. This paper described how a method was developed at UCD that would allow us to use ultrasonic testing to detect the rapidly occurring changes in the modulus of early age concrete. The full citation for this paper is: O'Connell, M, McNally, C, Donohue, S, Bonal, J & Richardson, MG (2009) Assessment of ultrasonic signals to determine the early age properties of concretes incorporating secondary cementitious materials. In: Proceedings of the 15th European Meeting of Environmental and Engineering Geophysics, Dublin, 7- 9 September. A copy of the paper is included in Appendix B of this report.
3.3.2 Experimental setup The benefit of this work is that while a proven ultrasonic analysis technique can be used to chart stiffness development in hydrating concrete, it may also be used in monitoring stiffness loss in concrete subjected to a corrosive environment. During a chemical attack concrete undergoes a significant alteration of its internal microstructure leading to a loss of strength and ultimately failure. Often the effects of a sulfate attack, for instance, can only be realised when the phenomenon has reached a substantial stage of progression. Ultrasonic analysis has the potential to detect degradation at an early stage and its benefits could be exploited in the monitoring of critical transport and water infrastructure whether it bridge foundations exposed to aggressive ground sulfate levels or biogenic sulfuric acid corrosion in wastewater treatment systems. The ultrasonic investigation of each of the cube specimens required S-wave (transverse shear waves) and P-wave (longitudinal waves) ultrasonic readings to be taken every twenty-eight days following immersion in acid. This was maintained for one hundred and sixty eight days. Squareshaped plastic caps measuring 50mm were attached to two sides of the cubes subjected to ultrasonic testing and the edges sealed with a mastic glue. This ensured a region protected from exposure to acid / corrosion on which to place the ultrasonic transducers and obtain a reading. 11
Two sets of transducers were used for this purpose in conjunction with a square wave pulserreceiver in through-transmission mode (Figure 3.3). The S-wave transducers had a frequency of 100 kHz, the P-wave transducers 50 kHz.
Figure 3.3 Ultrasonic testing equipment (L-R): Laptop, digital oscilliscope, signal generator, 100mm cube with ultrasonic transducers and ultrasonic couplant
For both S-wave and P-wave measurements each transducer was coupled directly opposite the other using a shear wave couplant gel as this was found to give excellent acoustic properties for both wave forms. It was essential to ensure a good connection between the couplant and the concrete as this was found to affect the amplitude of the signal. A 100V amplitude pulse was emitted into the specimen and the signal allowed to settle. The signals were recorded using a digital oscilloscope connected to a laptop computer and the results analysed with a Python shear wave detector algorithm to aid in the detection of the first arrival point (Bonal et al., 2010). An identical procedure was applied to the specimens used to investigate the effects of brushing with regular measurements taken at twenty-eight day intervals.
12
Figure 3.4 Autoclam permeability system with controller on the left and the unit (in blue) on top of the specimen
13
14
Days Exposure
Figure 4.1 Mortar prisms expansions following immersion in a 5% sodium sulfate solution Table 4.1 American Concrete Institute (ACI) performance guidelines for concrete exposed to sulfate when tested according to ASTM C1012.
Exposure Level Moderate Severe Very severe Exposure Class S1 S2 S3 Dissolved sulfate in water (ppm) 150 SO42- 1500 (Seawater) 1500 SO42- 10000 >10000 Max Expansion When Tested Using ASTM AS C1012 At 6 months 0.10% 0.05% 0.10%* 0.10% At 18 months At 18 months
*The 12-month month expansion limit applies only when the measured expansion exceeds the 6-month month expansion limit
Mix ME is a 30% Portland CEM I cement with 70% GGBS while SR is a Portland CEM I sulfate resisting cement. Mixes MB and MC are both CEM II-A/L A/L limestone cement mortars with 50% and 70% replacement levels of GGBS respect respectively. ively. At three hundred and eight days exposure,
15
the sulfate resisting mortar SR has now begun to expand beyond the level of mix ME leaving the three mortars with GGBS as a cement replacement showing the least expansion. There were some marked differences in visual deterioration of the mixes varying from corrosion related deposits and discoloration to cracking and warping of the specimens, or a combination of each. The severity of the visual deterioration corresponds well to the degree of expansion observed. Examination of each of the test mortar prisms indicated the formation of longitudinal cracks around 0.035% - 0.04% and this was common to all mixes having reached this expansion level (Figure 4.3, Figure 4.4 & Figure 4.6). Figure 4.3 and Figure 4.4 show these longitudinal white-filled cracks along the edges of both the SRPC and CEM II 100% specimens. The CEM I specimens (both MD and ME) show a similar crack formation although lacking in any white substance (Figure 4.5 & Figure 4.6). The remaining specimens containing GGBS (MB and MC) have not yet shown any crack formation. Longitudinal cracking along the length of the specimens was not an exclusive mechanism with radial cracking observed on one of the MA specimens along the boundary of the reference stud (Figure 4.2a). One of the other visual distinctions between MA and MD were notable depositions of a white substance occurring in blotches at random intervals on one of the prisms which can be seen in Figure 4.2b. These deposits seemed to be an integral part of the paste and were not soft to touch, nor had they the ability to be removed by scratching the surface.
(a)
16
(b) Figure 4.2 Radial cracking observed around the reference studs (a) and white blotches appearing on the surface of the prism (b) Generally however, the appearance of white deposits was less locally concentrated than in Figure 4.2b and consisted primarily of an intermittent speckled pattern throughout the prisms. This was applicable to all specimens and mixes after one year except those containing 70% GGBS as a cement replacement. Of the specimens whose deterioration has reached an advanced stage (approximately 0.1% expansion or greater - MA and MD), an increase in the crack density / length appears to be a common occurrence combined with the earnest commencement of corner spalling. Figure 4.7 (a) & (b) below show missing corner edges on specimens from mix MA and MD at an approximate expansion value of 0.1%. There is some evidence of rounding of the edges on specimens SR and ME however they have not reached the extent of spalling observed below.
17
Figure 4.3 White filled crack drawn parallel to yellow line and indicated by red arrows (SR)
Figure 4.4 White filled crack drawn parallel to yellow line and indicated by red arrows (MA)
18
Figure 4.5 Longitudinal crack drawn parallel to yellow line and indicated by red arrows (ME)
(b) Figure 4.7 Spalling of MA (a) and MD (b) prisms at 0.1% expansion
Simultaneous to this process, increased radial expansion could be observed to begin on some specimens between approximately 0.2% expansion and 0.45% expansion around the reference stud (Figure 4.10). Although this stage of deterioration was particularly visible for mix MD, the initiation of radial cracking was also observed on one on the specimens from mix MA, as seen in Figure 4.2a, suggesting that this process may be replicated once expansion has reached an equivalent level.
Figure 4.8 Spalling of the corners and loss of cohesion of the CEM I mix (MD) after one years exposure Continuing corrosion could be observed on specimens MD as the acceleration of the degradation process had significantly advanced with this mix after sixty-four weeks. At fifty-two weeks exposure and an expansion of approximately 0.3%, a slight warping of one of the prisms was observed along with the propagation of a crack perpendicular to the surface. At sixty-four 20
weeks the crack had significantly developed and extensive warping was visible (Figure 4.9). This perpendicular cracking occurred for prisms MD3 and MD4 as these individually had the most advanced level of expansion.
(b)
21
Visually, both mixes MA and MD appear to have suffered the most. Cracking and spalling was widespread after one year for both with white deposits seemingly a prominent feature for the attack on MA while MD exhibited almost exponential expansion. Mixes MB and MC, the CEM II/A-L limestone cements with 50% and 70% GGBS respectively, have shown comparatively little expansion and little differentiation. MB has shown some minor discoloration while MC has shown no visual evidence of attack. The sulfate resisting cement specimens (SR) have been outperformed by all mortars containing GGBS either with CEM I or CEM II/A-L after one year. Visually it has begun to exhibit the same common degradation phenomenon when approaching 0.030% 0.035% expansion, namely cracking, spalling and a white speckled appearance (the latter, however, generally beginning to appear within eight weeks). It can be concluded that within the given testing parameters, there is a clear benefit to the addition of a high percentage ( 50%) of GGBS to mortar. This would seem to be the case regardless of the cement type used although it would appear to have a greater contribution in increasing the resistance of CEM I mixtures to sulfate attack. The CEM II/A-L limestone cement used in this experimental programme exhibits an inherent sulfate resisting capability although may not be sulfate resistant. When combined with a percentage of GGBS greater than or equal to 50% the resulting mixture exhibits a superior level of sulfate resistance to that of all others tested, including a standard sulfate resisting Portland cement.
22
4.3.1 Dilution effect The dilution effect is simply used to describe the replacement of clinker with a non-cementitious filler material such as limestone. The latter shows no pozzolanic properties and thus does not produce C-S-H gel, the main binding component of a hydrated cement paste (Ramezanianpour et al., 2009). The effect reduces the amount of hydration products reacting during a sulfate attack and has been attributed by some authors as contributing to the improved resistance in cements containing minor limestone additions (Hooton, 1990, Gonzlez and Irassar, 1998). This improvement can be further enhanced by ensuring the C3A content of the cement is minimized. 23
In the sulfate resistance testing described in the previous section, the C3A values for the CEM I and CEM II cements were 12.3% and 8.4% respectively as derived from Bogues calculations. While neither value constitutes levels associated with a sulfate-resisting specification, the reduced percentage may partly account for the better performance of CEM II mixes in the experimental programme. Nonetheless, several authors have noted the poor performance of limestone cements even with moderate C3A levels. With this in mind, and given the results presented, the effect of the percentage limestone added should be given considerable attention. Table 4.2 Cement chemical analysis Composition (% by oxides) Compound CaO SiO2 Al2O3 Fe2O3 MgO Mn3O4 TiO Na2O K2O P2O5 SO3 Bogue Equations: CEM I: CEM II/A-L: C3 A = C3 A = = C3 A = = CEM I 60.29 18.24 6.19 2.45 3.55 0.26 0.00 1.17 1.32 4.15 2.38 CEM II/A-L 61.18 18.05 5.46 3.61 3.66 1.15 1.28 1.02 1.60 0.00 2.99
4.3.2 Permeability and porosity There are clearly other questions regarding the nature of the reaction of limestone with both the cement paste and sulfate ions. Issues surrounding the effect on permeability, porosity and 24
tortuosity are all topics that have been discussed by several authors (Hornain et al., 1995, Tsivilis et al., 2003, Pipilikaki et al., 2009). It should also be noted that the effective w/c ratio also increases with an increasing percentage of filler used and this can be attributed to the lack of pozzolanic properties of limestone (hence it cannot be considered a cementitious material). According to Gonzalez and Irassar (1998), the capillary porosity depends on the w/c ratio and the hydration degree and thus this affects the overall porosity of the cement. Tsivilis et al. (2003) showed that a Portland limestone cement exhibited lower water permeability values when compared to an ordinary Portland cement; they added however that permeability is not simply a function of porosity but also of the size, distribution, shape, tortuosity, and continuity of the pores. Their tests were carried out on specimens that contained a range of between 0% and 35% limestone. Pipilikaki et al. (2009) however conducted tests that showed Portland limestone cement containing 35% limestone had a higher porosity than an ordinary Portland cement. The authors then suggested that this may indicate a higher permeability, thus contradicting Tsivilis et al. (2003). They make the observation however, that limestone cements have an absence of large capillaries which may delay the ingress of sulfates and lower initial expansion but stress that the mechanism by which limestone affects the sulfate resistance of cement is far from being well understood. Whether it is due to a decrease in permeability, the absence of large capillaries or simply the resultant of the dilution effect, what is clear is that the results presented in section 4.2 indicate a benefit of using CEM II limestone cement with a filler content of between 5 and 10%. The addition of 50% or 70% GGBS as a cement replacement further enhances the sulfate resisting abilities and can more than likely be attributed to not only a decrease in permeability associated with GGBS but also the reduction in calcium hydroxide in the hydrated cement paste.
4.3.3 The Influence of C3A The level of C3A in the cement is clearly an important characteristic of its sulfate resistance. It is a compound that contributes little or nothing to cement strength except at early ages where it is responsible for flash set. It does however, act as a flux and reduces the temperature of burning clinker and allows the combination of lime and silica (Neville, 1995). Limiting the amount of C3A in the system reduces the monosulfate (Afm) phase that can lead to the formation of 25
ettringite. This compound greatly contributes to the expansion of cementitious materials and, along with gypsum formation, is one of the principal sulfate attack mechanisms. Minimising the amount of C3A forms the basis of a sulfate resisting Portland cement. However, given that all mixes containing GGBS as a cement replacement exceeded the performance of SRPC mortar specimens in the experimental programme, it is clear that reducing the potential for ettringite formation should not be the only preventative measure taken against a sulfate attack. Examining CEM II cements, the presence of limestone alters the hydration reactions of C3A according to several authors. Firstly, ettringite formation is accelerated by the presence of CaCO3 (Bonavetti et al., 2001) and secondly the reaction between the limestone and C3A forms carboaluminates (Gonzlez and Irassar, 1998) which compete with monosulfoaluminate stability and ettringite transformation. With this in mind it is evident that the addition of limestone further complicates the already complex process of sulfate attack. Gonzalez and Irassar (1998) also highlight that a high level of C3A extends the interaction of sulfate ions with unstable hydrates in the mortar. As previously discussed, the level of C3A in both the CEM I and CEM II mixes in the experimental programme were 12.3% and 8.4% respectively. The 100% CEM I specimen (MD) showed expansion at one year that was more than three times that of a 100% CEM II specimen (MA). The inclusion of GGBS as a cement replacement has demonstrably affected the sulfate performance as evidenced by the low expansion of mixes MB, MC and ME. It has been claimed that a higher replacement level of a pozzolanic material dilutes the C3A, reduces the aluminate phases and decreases ettringite formation (Al-Dulaijan et al., 2003, Gonzlez and Irassar, 1998). Deterioration in limestone cements has primarily been attributed to gypsum formation while the mechanism in ordinary Portland cements is dominated by ettringite formation (Pipilikaki et al., 2009). Irassar et al. (2003) further supports this claim by observing greater gypsum formation in moderate C3A limestone cements subjected to a 5% sodium sulfate solution as specified in ASTM C1012. The concentration of the sodium sulfate solution is also important in deciphering the dominant form of attack. In solutions with sulfate levels exceeding 8g/l the mechanism is said to be primarily due to gypsum formation (Hekal et al., 2002, Tosun et al., 2009). The tests carried out in the sulfate expansion programme contained a sulfate strength of 33.8g/l, more than exceeding the accepted threshold. This may be further supported by the visual observations of substantial white deposits on the CEM II limestone specimens. With the CEM I mortars however, 26
the situation may be less clear. Although the solution strength would dictate a gypsum dominated attack, the deposits of the white substance were considerably less than on the limestone mortars while cracks tended not to be characterised by white veins.
4.3.4 Influence of C3S and C2S Both the C3S (alite) and the C2S (belite) contents of the cements are also important indicators of performance under a sustained sulfate attack. The presence of a limestone additive, such as that in CEM II-A/L, also modifies the Ca/Si ratio of the C-S-H phase, with the interaction of CaCO3 accelerating the hydration of the C3S content (Ramezanianpour et al., 2009). Alite and belite play important roles in characterising the strength of a hydrated cement paste, the former contributing much towards strength at early ages, the latter developing those characteristics as time progresses. However with an increasing percentage of C3S, the quantity of calcium hydroxide formed during hydration also increases raising a cements vulnerability to gypsum formation during a sulfate attack (Ramyar and Inan, 2007). Research conducted on pure C3S cement pastes also demonstrated that even gypsum formation caused considerable expansion (Tian and Cohen, 2000), although this mechanism of attack for cements used in practice is, by the authors own admission, still a cause for considerable debate . Tosun et al. (2009) observed that cement with a high percentage of limestone and a high C3S/C2S ratio was more prone to attack by sulfates, however their use of an extraordinarily high sulfate solution (200g/l SO42-) means their conclusions must be greeted with some scepticism. It is generally regarded that a cement low in C3S and C3A will perform well during a sulfate attack; nonetheless this may be difficult as it has been pointed out that reducing the C3A content raises the C3S/C2S ratio (Al-Dulaijan et al., 2003).
4.3.5 Sulfate resisting capabilities of CEM II/A-L and GGBS concretes Portland limestone cements suffer similar chemical reactions from a traditional sulfate attack as ordinary Portland cements resulting primarily in the formation of gypsum and ettringite. The formation of thaumasite is of particular concern given the high level of carbonate in the system; however there is often a precondition of low temperatures before this becomes a favourable attack mechanism (Higgins and Crammond, 2003), although this is debatable (Irassar et al., 2005). Nonetheless the chemical interactions involved in the production of gypsum, ettringite and 27
thaumasite, whether during a sulfate attack or during cement hydration, can be modified in the presence of limestone. According to Gonzalez and Irassar (1998), during cement hydration carbonate ions from the limestone filler compete with sulfate ions from gypsum to react with aluminate ions from C3A forming monocarboaluminate, monosulfoaluminate and ettringite. Irassar et al. (2003) detailed the sequence of a sulfate attack concluding that diffusion of sulfate ions is followed by calcium hydroxide leaching, ettringite formation, gypsum formation and depletion of CH. The latter stages involve the decalcification of C-S-H followed by thaumasite formation. The sequence of attack would corroborate well with what was noted in the experimental results in section 4.2. As observed with CEM II/A-L limestone mix MA, there was an initial low level expansion detected with very little visual deterioration which could indicate the onset of ettringite or early gypsum formation. As the attack progressed white deposits began forming on the exterior of each prism, followed by a lack of cohesion and spalling at the edges, possibly indicating the decalcification of the C-S-H phase. Irassar et al. (2003) also describe corrosion of edges and corners and attribute it to gypsum formation in parallel veins to the sulfate attack front. As seen in Figure 4.4 cracks / veins with a white deposit can be observed at the elapsed exposure time when the specimen began to shed mortar particles. Almost identical attack sequences were observed in all mortar mixes undergoing observable degradation suggesting that none of these phenomena are particularly unique to limestone cements. The CEM I specimens of MD however, exhibited less visual evidence of white deposits and more physical effects in the form of cracking, warping, spalling and surface delamination. This could perhaps be attributable to a more dominant stage in the attack sequence compared to those observed in limestone cements. In examining the results of the sulfate expansion experimental programme, it is clear that the addition of a relatively high percentage of GGBS to both CEM I and CEM II mortars has had a significant effect on their resistance to a 5% sodium sulfate solution. Both the CEM II mortars (MB and MC) have shown the least amount of expansion and visual deterioration while the CEMI I mortar with 70% GGBS (ME) has shown a resistance equalling SRPC for the majority of the exposure period but eventually exceeding it. The sulfate resisting capabilities of GGBS have been discussed on many occasions (BRE, 2003, Higgins and Crammond, 2003) with much of the benefit being attributed to a denser matrix, decreased permeability and a reduction in calcium hydroxide present in the hydrated system (Pava and Condren, 2008, Al-Dulaijan et al., 2003, 28
Osborne, 1999). Furthermore, with the formation of a secondary C-S-H phase attributable to the interaction between calcium hydroxide and GGBS, much of the alumina in the system becomes locked up in this product and is not available to form ettringite during a sulfate attack (Gollop and Taylor, 1996). The authors also claim that as the percentage replacement of cement with GGBS increases the proportion which reacts decreases with cement and this limits the quantity of alumina released at high slag contents. The combination of a reduction in calcium hydroxide, decreased permeability and the locking up of potentially reactive alumina may account for the behaviour of the CEM I mortars but it is essential to investigate any further effects of CEM II cements with a limestone addition. The results presented above have indicated the potential increased benefit of using this with at least 50% GGBS as a cement replacement. As previously discussed, limestone cements have an absence of large capillaries which may delay the ingress of sulfates (Pipilikaki et al., 2009). When combining this with the more impermeable matrix from GGBS cements, the opportunity for sulfates to interact with the cement compounds is being severely limited. Visually, all specimens containing GGBS have exhibited almost no evidence of a lack of cohesion from exposure to the sulfate solution. Researchers (Brown and Taylor, 1999) have attempted to account for this effect in limestone cements and have put forward a plausible explanation. With an increase in GGBS levels, there is also an increase in hydrated C-S-H in the system. Correspondingly, there is a decrease in the level of calcium hydroxide. Ettringite and gypsum preferentially obtain their calcium from this phase but in the absence of a sufficient quantity available, calcium from the C-S-H phase will serve as a source. This phase constitutes the primary binding capability of a cement matrix and its degradation leads to a major loss in cohesion. The authors claim that by adding calcium carbonate (limestone) as an additive, this in turn will serve as the source of calcium for ettringite and gypsum thus preserving the integrity of the C-S-H phase.
4.4
4.4.1 Mass loss results The measurement of a loss of mass of the concrete specimens was considered an acceptable means of assessing the performance of each of the mixes in a sulfuric acid environment and was 29
previously used by Chang et al. (2005). The results of the first technique, using a wire brush, are presented in Figure 4.11. The results of this procedure indicate that there may be a slight increase in mass over the first twenty-eight days of exposure or very little mass loss. The concrete made from CEM II-A/L limestone cement with no addition of GGBS showed a higher initial gain in mass compared to the five other mixes, although the amount could be regarded as not significant. After the initial month of exposure the decrease in mass remained constant for the most part with little divergence from this trend. The performance of each of the six mixes remained largely unchanged following completion of the testing programme, regardless of GGBS content or cement type. Although the figures indicate that a 70% GGBS content, regardless of cement type, performed the best throughout the testing period, the difference between the mix which performed the worst (SRPC) was considered to be not significant. This is confirmation of the aggressive nature of the sulfuric acid solution and the inherent difficulties in exposing cementitious materials to this type of environment. The graph in Figure 4.12 shows the results from exposure to the acid without using the brushing technique. It can be seen that there is a slight increase in mass for all specimens within the first twenty-eight days following which there is a steady increase in mass loss. This continues at a similar rate until eighty-four days at which point there appears to be a small decrease in the rate of mass loss for the unbrushed specimens compared with those that have been brushed at weekly intervals. This point can be illustrated by Figure 4.13. The phenomenon appears to coincide with the dissolution of the outer layer of the cement matrix and the protrusion of the 10mm and 20mm limestone aggregate although this apparent relationship does not seem applicable to the brushed experiment.
30
Brushed Concrete
2000 0 Mass Loss (g/m^2) -2000 -4000 -6000 -8000 -10000 Days Exposure 0 56 112 168 CEM II 100% CEM II + 50% GGBS CEM II + 70% GBS CEM I 100% CEM I + 70% GGBS SRPC
Unbrushed Concrete
2000 0 Mass Loss (g/m^2) -2000 -4000 -6000 -8000 -10000 Days Exposure 0 56 112 168 CEM II 100% CEM II + 50% GGBS CEM II + 70% GBS CEM I 100% CEM I + 70% GGBS SRPC
31
CEM II 100%
2000 0
56
84
112
140
168
28
56
84
112
140
168
-10000
. (b)
CEM II + 70% GGBS
2000 0 0 28 56 84 112 140 168 Mass Loss (g/m2) -2000 -4000 -6000 Unbrushed -8000 -10000 Brushed Days Exposure 0 28 56 84 112 140 168
CEM I 100%
(c)
CEM I + 70% GGBS
2000 0 Mass Loss (g/m2) Mass Loss (g/m2) -2000 -4000 -6000 -8000 -10000 Days Exposure Unbrushed Brushed 0 28 56 84 112 140 168 2000 0 -2000 0 -4000 -6000 -8000 -10000 28 56
(d)
SRPC 100%
84
112
140
168
(e)
(f)
32
(a)
(d)
(b)
(e)
(c)
(f)
Figure 4.14 Typical corrosion levels for brushed concrete at 28, 56 & 84 days (a-c) and unbrushed concrete over the same time intervals (d-f). Note differing degree of aggregate exposure. 33
For both experiments there is very little between the performances of each of the six different mixes in terms of mass loss or compressive strength. Although the data seems to suggest that an addition of 70% GGBS improves the resistance to mass loss after six months, this must be seen in the context of the compressive strength losing at least 60% of its 28-day value regardless of the cement type or brushing technique used (Figure 4.17). Several useful observations can be gleaned from the obtained data. It is clear that the use of brushing to mimic abrasive behaviour (e.g. flowing water) has an effect on the loss of material from the surface of concrete. (Figure 4.14 a-f). Allowing build-up to occur may slow down some aspects of the corrosive effects of acid, albeit on a superficial basis, as clearly there are internal chemical transformations occurring. This is again demonstrated by Figure 4.19 showing the failure mechanism of a cube exposed to acid for six months. It is evident that a total loss of cohesion has occurred throughout the specimen proving that while much of the ongoing reactions are surface oriented there are more sinister forces at work beneath the cube surface. Furthermore, there may be a useful relationship between the initial visibility of exposed aggregate and the underlying state of the concrete integrity exposed to acidic environments as the cube strength data reveals.
4.4.2 Discussion of deterioration mechanism For both the brushed and unbrushed experimental programmes the primary mechanism of deterioration was disintegration of the cement matrix along with some secondary spalling. There was no evidence of cracking at any point over the six month exposure period. The manifestation of the deterioration consisted of the formation on the surface of the cube specimens of a white mushy substance (most likely gypsum) that was soft to touch and easy to remove. This was visible after approximately one week of exposure to the acid and built up following each successive removal of loosely adhering corrosion products. Regular examination of the unbrushed concrete cubes also revealed the complete loss of cohesion of the surface layer of the cement matrix after twelve weeks.
34
Figure 4.15 Complete surface delamination of one side of a 100mm cube exposed to 1% sulfuric acid (unbrushed)
Figure 4.16 Concrete material losing cohesion and falling off the specimens
The type of deterioration visible showed the entire side of a cube fall away as a single piece and this can be seen in the highlighted section of Figure 4.15. Regular examination of the brushed 35
specimens confirmed the mechanism of deterioration with material falling off the sides and corners of the specimen and gathering at the bottom of the tank. This can be seen in Figure 4.16 and was a characteristic present throughout the duration of the experimental programme (including that of the unbrushed specimens).
4.4.3 Cube strength tests The experimental programme also examined the cube strengths of each of the mixes at twentyeight, fifty-six and one hundred and sixty eight days.
80 70 60 50 40 30 20 10 0 0
80 70 60 50 40 30 20 10 0
28
56
84 112 Days
140
168
28
56
84 Days
112
140
168
80 70 60 50 40 30 20 10 0 0
MD - 100% CEM I
28
56
84 112 Days
140
168
28
56
84 Days
112
140
168
36
80 70 60 50 40 30 20 10 0 0
SR - 100% SRPC
28
56
84 112 Days
140
168
28
56
84 112 Days
140
168
Figure 4.17 Comparison between cube strength exposed to water only and 1% sulfuric acid only (brushed)
4.4.4 Sulfuric acid expansion tests Simultaneous to tests conducted on concrete cubes exposed to sulfuric acid, two prisms from each mix were immersed in a 1% sulfuric acid solution while two were immersed in water over a six-month period. The solution pH was kept at approximately 1.5 and refreshed at monthly intervals where the acid was renewed and the containers cleaned of any debris. The change of length of the prism from the initial reading is then calculated according to Eqn. 1. The change in length of each of the specimens was measured twice for accuracy and the results obtained are available in Appendix H. The data showed that no appreciable expansion occurred and in fact minor contractions were observed across all six mixes. The change in length of each of the specimens was measured twice for accuracy, however it was found that at times there were differences between readings from the same specimen, sometimes significant. It was concluded that the observed changes in length may be due to experimental error and in fact relatively little movement was observed. The procedure for measuring the specimens differed from those of the sulfate expansion tests in that a metal ball had to be placed between the reference studs therefore minor deformities could not be accounted for across the sphere. The reading on the comparator was also observed to vary considerably while the specimen settled into the base groves. Again this behaviour was not observed for the sulfate prism tests. According to Monteiro et al. (2008), an expansion of 0.5% for concrete exposed to elevated sulfate levels was deemed as the failure point. As can be seen 37
from the experimental results, following six months of exposure to acid the movement in the prisms was exceptionally far removed from this figure. The determination was that change of length was not considered a significant contributor to concrete degradation when exposed to sulfuric acid.
4.5
Ultrasonic results
4.5.1 Stiffness loss in due to sulfuric acid testing The results in the previous section have shown the potential use for the application of ultrasonic analysis of concrete, in conjunction with the shear wave detector algorithm developed by Bonal et al. (2008), to monitor stiffness development in mixes containing varying percentages of GGBS. In theory the same process has the potential to be used in recording the loss of stiffness in concrete exposed to aggressive chemical environments. The technique is now applied to monitor the degradation of 100mm cubes made with five concrete mixes with varying percentages of GGBS, and one sulfate resisting SRPC mix, exposed to a 1% solution of sulfuric acid (H2SO4) over a six month period. The cement type and percentage of GGBS used as a cement replacement is indicated in the table below.
Photographic observation of the cube specimens indicated the acid attack was primarily focused on the cement matrix with the limestone aggregate largely avoiding the corrosive effects 38
(Figure 4.18). Failure of the specimen was by complete loss in cohesion of the binding properties (Figure 4.19). The ultrasonic analysis therefore takes two approaches: monitoring the degradation of concrete as a single inhomogeneous material and then again by considering it as a two-phase model consisting of mortar and coarse aggregate. By doing this, it can be assumed that the cause of a loss in stiffness in concrete is attributed to complete disintegration of the cement matrix leaving the aggregate assumed to be unaffected by the acid.
Figure 4.18 Condition of a cube following six months exposure to sulfuric acid. The aggregate remains largely unaffected while the cement matrix has suffered spalling and disintegration.
Figure 4.19 The same cube tested under compression until failure. The mode of failure clearly indicates a loss of cohesion in the binding properties of the cement matrix.
39
In order to separate the concrete into a two two-phase phase system, a model was developed that permits the velocity of an ultrasonic wave through concrete to be written in terms of the velocity of the mortar component and the velocity of the coarse aggregate component (Lin et al., 2003). 2003)
Eqn. 2
Where C is the velocity in concrete, M is the velocity in mortar, CA is the velocity in coarse aggregate, VM is the volume of mortar, VCA is the volume of coarse aggregate and VC is the volume of concrete. By calculating the volume of coarse aggregate per 100mm cube specimen, the phase was then considered as a single layer of limestone 46mm deep, the remaining 54mm being the mortar phase. With volume parameters and the velocity of concrete known, using a large specimen of limestone the ultrasonic velocities were also determined for the coarse aggregate phase. The transmission of the signal through air voids and pores and the possible effects of this were not considered.
Amplitude (micro-Volts)
Time (ns)
Figure 4.20 S-wave wave signal from the solid limestone showing the first arrival point as determined by the shear wave detector algorithm Re-arranging arranging the equation yields an expression for the velocity of an ultrasonic wave in mortar, where CA is now the velocity through the solid limestone layer:
40
Eqn 3
4.5.2 Stiffness loss in concrete exposed to a 1% sulfuric acid solution Ultrasonic testing was carried out on two samples stored in water and two samples stored in a 1% sulfuric acid solution at twenty-eight day intervals for each of the six concrete mixes. The results for those specimens made from CEM II-A/L with 0%, 50% and 70% GGBS used as a cement replacement are presented in Figure 4.21, Figure 4.22 and Figure 4.23 respectively. It can be seen that determining the small strain Youngs modulus using the shear wave detector algorithm has the ability to monitor the degradation of concrete exposed to a 1% sulfuric acid solution. There is, however, a difficulty in distinguishing between the six-month performances of each mix individually, perhaps reflecting the aggressive nature of the solution in which they are stored. Cubes containing both 50% and 70% GGBS, though, show a small resistance to a drop in the small strain Youngs modulus up to approximately eighty-four days with the 70% GGBS apparently showing almost complete resistance up to this point. This apparent beneficial performance of GGBS may be somewhat misleading however, as the fifty-six day cube strengths (for those exposed to acid) actually show a 46% drop (70% GGBS) and 42% (50% GGBS) drop compared to those stored in water (Figure 4.17). The mix containing 0% GGBS showed the least drop in strength recording a loss of only 30% in comparison. This brings up an interesting possibility as to how GGBS may be affecting ultrasonic velocities in limestone concrete. As can be seen from the results documenting early age strength gain, the small strain Youngs modulus follows the trends of both the cube strengths and that of the static Youngs modulus much more closely. At this early stage in hydration many of the chemical compounds giving concrete its strength (and in particular GGBS concretes) have yet to be formed. The results also show that by treating the concrete as a two-phase system, degradation of the mortar phase closely mimics the trend of the system as a whole giving confidence in the fact that the disintegration of concrete can be attributed to the loss in cohesion of the cement matrix.
41
Table 4.4 Sample data set obtained from the ultrasonic experimental programme. Values are obtained up to week 24 for two samples stored in water and two stored in 1% sulfuric acid for each time step.
Mix: MB
Week Cube TP-wave (sec) 0 0 4 4 6 12 21.00 20.40 VP-wave (m/sec) 4975 5128 TS-wave (sec) 31.0 31.6 VS-wave (m/sec) 3279 3215 Density (kg/m ) 2400 2400 2400 2400
3
E (GPa) 58 58
Figure 4.21 Ultrasonic analysis of small strain Youngs modulus for mix MA immersed in 1% sulfuric acid
42
Figure 4.22 Ultrasonic analysis of small strain Youngs modulus for mix MB immersed in 1% sulfuric acid
Figure 4.23 Ultrasonic analysis of small strain Youngs modulus for mix MC immersed in 1% sulfuric acid
The results of the second set of three mixes detail the performance of CEM I with 0% and 70% GGBS as a cement replacement (Figure 4.24 and Figure 4.25). The performance of a sulfate-resisting SRPC mix is also considered as a benchmark (Figure 4.26). Again it is clear that the ultrasonic analysis has the ability to detect concrete degradation in the latter stages of the 43
experimental programme. The performance of the three mixes is somewhat ambiguous in the first fifty-six days with little differences being recorded between those specimens stored in water and those stored in acid. This again is somewhat misleading as the recorded cube strengths (see Figure 4.17) begin to show a substantial loss in compressive resistance by this stage in the programme. Furthermore, the ability to distinguish between the performance of a CEM I mix, a CEM II-A/L mix and those containing GGBS is minimal. It is also clear that up to the time frame between fifty-six days and eighty-four days it may be difficult for this method to detect any degree of loss in stiffness. The loss in concrete strength shown in Table 4.5 above indicates that regardless of the cement type used the performance of each of the mixes after six months exposure to the acidic solution is poor showing very little difference in the ability to resist the aggressive environment. Given that the results of the ultrasonic tests also show few differences after six months, it could be hypothesised that the method is indeed a reliable indicator in the latter stages of attack, although seemingly less sensitive than the behaviour shown in the early age stiffness development tests.
Table 4.5 Percentage loss in strength after six months exposure to a 1% sulfuric acid solution Mix Designation MA MB MC MD ME SR % loss in strength 65% 74% 66% 76% 74% 72%
44
Figure 4.24 Ultrasonic analysis of small strain Youngs modulus for mix MD immersed in 1% sulfuric acid
Figure 4.25 Ultrasonic analysis of small strain Youngs modulus for mix ME immersed in 1% sulfuric acid
45
Figure 4.26 Ultrasonic analysis of small strain Youngs modulus for mix SR immersed in 1% sulfuric acid
4.5.3 Discussion of ultrasonic results There remains a distinct lack of in-depth knowledge into the relationship between ultrasonic velocities and the development or degradation of stiffness in concrete. Much of the discussion centres around two aspects of concrete: the microstructural make-up of the mix and the influence of hydration products (and consequently corrosion products). The ultrasonic results presented show that the shear wave detector algorithm appears to be more sensitive to a change in stiffness in the early days of hydration but less so when monitoring the effects of corrosion due to sulfuric acid attack. To account for these differences it is essential to outline the factors affecting both processes.
4.5.4 Microstructural Effects Three parameters within the microstructural make-up of a concrete mix have been attributed to variations in ultrasonic pulse velocities: entrapped air voids, the water/cement ratio and the arrangement of aggregate in the mix. For concrete in the early stages of hydration it has been shown that the velocity of p-waves are slow to develop (Robeyst et al., 2008) but, as can be seen from the results achieved in the early age tests in section 4.4.1, this begins to approach an asymptotic value after a few days of hydration for both P-waves and S-waves. 46
Velocity (m/s)
P-wave velocity
4000 3500 3000 2500 2000 0 5 10 15 20 25
S-wave velocity
Time (Days)
Figure 4.27 Development of P-wave and S-wave velocities in early age stiffness development
Robeyst et al. (2008) showed that it was in the first twenty-four hours that the greatest gain in P-wave velocity was observed. With regard to the ultrasonic tests carried out on specimens exposed to sulfuric acid, there was almost no effect on the P-wave velocity over the six-month experimental period with it generally remaining at a consistent level for all mixes. This is interesting as it has been claimed that P-wave velocity is directly related to the dynamic Youngs modulus (Voigt et al., 2005). Clearly however, the observed drop in cube strengths (Figure 4.17) show that the stiffness decreased over the exposure period. This would seem to indicate that Pwave velocities may solely be an indicator of cement particle interconnectivity and not necessarily of stiffness (i.e. there may be bonds but they might be weak). In a hydrating mix, for instance, cement particles remain in suspension for a period and the wave paths thus remain elongated due to the presence of entrapped air voids (Lee et al., 2004, Chaix et al., 2006, Robeyst et al., 2008), thus increasing the time from transmitter to receiver. With regard to a specimen degrading from exposure to sulfuric acid, the cement particles are not in a fluidic suspension and remain bonded, although perhaps with increasing weakness as time progresses. It may also be hypothesised that the corrosion products formed in the acid attack (e.g. gypsum) may fill the air voids providing an unobstructed wave path and masking the effects of a loss of stiffness on Pwave velocities. This type of behaviour occurs during hydration when primary ettringite serves as a medium through which p-waves can travel, giving a false indication of stiffness development. 47
The effect of the water to cement ratio also has a bearing on the velocity of ultrasonic waves in concrete. Although for this experimental programme the value was constant at 0.45, it is still worthwhile to mention the effect it has on the cementitious system. According to one group of authors, for concrete with a w/c ratio greater than 0.5 an increase in aggregate content will lead to an increase in ultrasonic velocity but with little increase in strength. Conversely, for a high w/c ratio an increase in cement content will lead to a decrease in ultrasonic velocity while for a low w/c ratio the velocity doesnt change as the cement matrix is already dense. The increase in w/c ratio effectively increases the distance between cement particles, complicating wave transmission by the non-direct path from transmitter to receiver while the degree of tortuosity of the paste also has a similar effect (Lee et al., 2004). Furthermore as concrete is not a homogeneous material, it cannot be said with certainty how the transmission of waves through aggregate-paste boundaries affects the received signal. While the influence of air voids and the interconnectivity of cement particles obviously affect ultrasonic velocities, this merely takes into account the microscopic variabilites. Those particles on a much larger scale, however, cannot be ignored. The internal settling of aggregate leads to an increase in ultrasonic velocity while the downward movement during compaction must also be taken into account. Lin et al (2003) also point out that the distribution of aggregate within a specimen may have an unexpected effect on the ultrasonic velocity. If the aggregate is concentrated along the direct line from the ultrasonic transducer transmitting the signal to the receiver then naturally the wave will travel through more aggregate and the velocity of the wave will be high. However, if the aggregate is sparse around this line then the velocity will be comparatively low. This may be an issue for laboratory prepared concrete due to individual compaction techniques and could differ from that found on site.
4.5.5 Chemical effects The lack of a substantial difference in performance of concrete exposed to sulfuric acid means it is quite difficult to pinpoint any effects rising from the presence of limestone in the cement or indeed that of GGBS. As a result much of what can be discussed arises from the early age tests but nonetheless may contribute to the performance of the sulfuric acid programme. Both the formation of primary ettringite during hydration and the effect of the CSH phase subsequently 48
have been found to contribute to variations in ultrasonic velocities. Concretes made with GGBS form a greater degree of CSH due to the interaction between the former and hydrated calcium hydroxide. The resulting linkages between cement grains and aggregates provide a path for wave propagation increasing the velocity. The CSH phase is the main binding compound in concrete but in concretes made with GGBS this takes time to develop. In early ages tests the 70% GGBS mix showed the slow development of both velocity and small strain stiffness reflecting the lack of appreciable CSH formed. After the cubes are immersed in sulfuric acid, the effect may not be so clear. Both mixes with 70% GGBS (MC and ME) show an early reluctance to a decrease in small strain stiffness up to eighty-four days, thereafter however, the trend is downward. The mixes without GGBS (MA and MD) show a drop in small strain Youngs modulus relative to an initial value. However the final result for all mixes is rather uniform possibly reflecting the aggressiveness of the environment. With regard to ettringite formation, this may be more applicable to the early age tests. Ettringite is unstable in pH values below approximately 11 after which it decomposes. The sulfuric acid tests are carried out in a solution that has a pH of approximately 1.5 and it could be assumed that the presence of ettringite is neglected when assessing the ultrasonic results of concrete in this environment. Nevertheless it has been pointed out that the formation of primary ettringite during hydration contributes to an increase in p-wave velocity but this has no bearing on the strength or stiffness of concrete. The ettringite needles fill the pore space previously occupied by water with a solid product decreasing the space filled with air voids and allowing the transmission of ultrasonic waves. It could be hypothesised that when concrete is attacked by the sulfuric acid, forming gypsum and filling the voids of the surrounding spaces, the same behaviour may occur and affect an accurate assessment of the materials stiffness.
49
Table 4.6 Permeability and absorption results Mix Air permeability (ln(m )/s) Sample a MA MB MC MD ME SR -0.054 -0.047 -0.055 -0.073 -0.034 -0.082 Sample b -0.028 -0.013 -0.018 -0.015 -0.014 -0.006
2
Water permeability (m /min ) Sample a 3.87E-08 2.35E-08 3.12E-08 7.57E-08 4.07E-08 6.44E-08 Sample b 1.78E-08 1.30E-08 1.96E-08 2.68E-08 2.15E-08 2.25E-08
3 0.5
Absorption (m3/min) Sample a 1.56E-08 5.07E-09 2.59E-08 3.97E-08 2.44E-08 5.35E-08 Sample b 2.27E-08 4.04E-09 1.59E-08 1.45E-08 2.56E-08 1.70E-08
The goal of the permeability and sorption testing was to ascertain a performance ranking of the six concrete mixes. The results of the investigation proved somewhat inconclusive with obtained values at times differing significantly for identical mix specifications. The results may be due to experimental error, as the Autoclam requires a completely sealed testing surface, or due to the effect that aggregate has on the diffusion characteristics of the concrete.
50
5 DISCUSSION
5.1 Sulfate experimental programme
While it has been stated that testing through sulfate exposure to assess concrete tolerance to wastewater applications is largely an inaccurate method (Monteny et al., 2000), it is clear that this particularly important fact has not been accepted or realised within the engineering community. The sulfate testing procedure in this experimental programme was designed to highlight any variations in the performance of CEMI I, CEM II-A/L and various combinations of GGBS when compared to SRPC, the standard specification cement for wastewater applications. The results of the investigation highlighted some very significant differences in the attack mechanism compared with results obtained in sulfuric acid exposure tests and the conditions observed in two wastewater treatment plants at Swords in North Co. Dublin and Kilkenny in the midlands. While the condition of the concrete in both plants resembled the effects of the acid tests the sulfate testing programme yielded starkly differing results. Several significant behavioural differences were monitored over the course of the investigation; while the specimens submersed in a 50g/l sodium sulfate solution exhibited expansion, cracking and warping those in acid showed only mass loss and disintegration of the cement matrix. The manifestation of attack also progressed at different paces. While the prisms in the accelerated sulfate tests took several weeks, sometimes months, to show appreciable expansion or appearance of corrosion products (other than the CEM I 100% specimens), the cubes submersed in the 1% sulfuric acid solution began to deteriorate within days. These cubes rapidly began to show build-up of a white mushy material followed then by mass loss. This clearly shows that exposure to sulfuric acid is a more aggressive form of attack and may result in substantially decreased service life estimates. The sulfate tests also demonstrated the ability to significantly distinguish between the performances of CEM I, CEM II-A/L, SRPC and various additions of GGBS. Throughout the experimental programme the cement paste of each prism largely remained intact. Minimal build up of any corrosion products was observed and only the extreme corners of the specimens showed any mass loss. These behavioural differences call into question the differences in the chemical makeup of both the solutions used. While each contains a sulfate ion, traditionally considered the main aggressor, the presence of a H+ ion from the sulfuric acid solution is clearly having the most detrimental effect resulting in a lack of cohesion and an 51
effective dissolution of the cement matrix. With regard to the sulfate testing procedures, further investigations into the mechanism of attack may be investigated by modifying the solution by using magnesium sulfate for example. Research has suggested that this attacks primarily the CSH phase of cement rather than calcium hydroxide in order to obtain the calcium used in its degradation process. It is unclear whether this may be a more beneficial test to conduct in conjunction with sulfuric acid tests; however sodium sulfate has generally been recognised as the standardised solution for sulfate exposure and is used by both the ASTM C1012 and the Dutch CUR 48 tests. It can be concluded that in the context of concrete in wastewater applications, standard tests to investigate the sulfate performance of a variety of mixes cannot be considered as a reliable indicator for representing in-service conditions. The tests carried out in this experimental investigation, however, have highlighted the necessity to affect a change in practice when specifying concrete for aggressive wastewater applications. The stark contrast in both the manifestation of corrosion and the physical effects on the cementitious system serve as proof that this is an urgent requirement that needs to be accounted for. Whats most alarming, however, is that there appears to be an inability for any concrete to survive this acidic environment despite the mix being designed to EN 206 XA3 class, deemed the most resistant specification against chemical attack. It should be noted however that the test solution of pH 1.5 represents the most severe conditions to be expected in service. The pH may vary in reality on account of environmental conditions, including temperature and humidity, which undoubtedly affect the activity of the sulfuric acid producing bacteria. Sulfate tests may draw a concrete specifier into a false sense of security and while this may satisfy the requirements of current standards, it is clear that it does not satisfy the true nature of the problem.
al. (2005) showed reductions in strength not exceeding 30% for five out of six cases over an identical time period and using a similar experimental regime, although only two of those were mixes similar to this programme. A contributing factor toward the authors result however may be to what extent the specimens are brushed. Chang et al. (2005) recorded rinsing the specimens under flowing water and gently brushing them of loose material with a wire brush. The brushing conducted in this investigation varied from extensive removal of loose material with regular wire brushing and then again using only flowing tap water to rid the cube of excess build-up. A small decrease in the rate of mass loss was noted after eighty-four days using the rinsing method although the differences may be regarded as not significant. The extent of brushing was an important part of the regime as the idea was to create conditions representative of the bacterial environment detailed in the literature review. It is unclear whether the assumption of a 1% sulfuric acid solution is representative of the in-service conditions but it may, however, constitute an acceptable component in an accelerated test method. The effect of various brushing techniques yields results that are also far from conclusive and may confirm the suspected intolerance of concrete to aggressive sulfuric acid environments regardless of the effects of abrasion. The acid testing programme also raised concerns over the use of sulfate resisting Portland cement in wastewater applications. It now appears there is a common misconception that specifying a SRPC mix will sufficiently resist the expected aggressive sulfate environment. While this is not entirely untrue (some industries such as the brewing industry discharge sulfateladen waters from their production processes), it is clear that some form of extra protection is required for these treatment systems to adequately resist the acidic conditions over a long period of time. The results presented in this investigation show almost no benefit in using GGBS. Despite the reduced amount of calcium hydroxide (and the resulting lesser quantities of gypsum formed as a result of sulfuric acid attack), this does not account for any improvements in the performance of GGBS concrete. This calls into question other factors such as the possible increased capillary effect of the GGBS mixes, from decreased pore sizes, or more likely the dissolution effect of the hydrogen ion of the sulfuric acid. The results from the sulfate prism tests further highlight the differences in attack mechanisms despite many of the same corrosion products (i.e. gypsum) being present in both instances. The sulfuric acid tests in this investigation, while highlighting the vulnerabilities to acidic conditions, suffer from some inaccuracies inherent in accelerated testing. As has been mentioned, 53
the attack present in wastewater systems is bacterial in nature (not purely chemical), evolving over a period of time far exceeding the six months used in this investigation. Environmental conditions also play a role; in-service infrastructure will experience temperature fluctuations from night to day and from season to season. Water levels may also rise and fall in reality, creating wetting and drying conditions which are known to be more detrimental to concrete corosion. When this is combined with varying water flows, the experimental programme cannot claim to represent each and every environmental condition that will be present over the service life of a component. It can however, serve as an indicator in the performance of various concrete mixes to this highly aggressive environment. While more accurate methods exist of replicating the bacterial environment, the complexity of such an investigation is both time consuming and cost prohibitive. The experimental programme carried out in this instance may therefore represent the most accurate method available at minimal cost with the least constraints on time.
small strain stiffness in concrete at early ages and distinguish between the different binders used. The method appeared less accurate however, in monitoring a loss in stiffness following exposure to a 1% sulfuric acid solution for six months. At the same it was less clear to distinguish between the different binders used which may possibly reflect the very harsh nature of the aggressive solution to calcium-based cementitious materials.
55
6 CONCLUSIONS
The most important aspect of this research has been highlighting the significant differences between sulfate attack and sulfuric acid attack. Based on the obtained data the former is clearly an expansion dominated phenomenon with a defined point of corrosion acceleration. The latter, however, appears to be a surface dissolution mechanism combined with an extremely destructive / rapid diffusion process that immediately leads to an attack on the binding capabilities of cement This has implications with respect to the fact that current design specifications for water and wastewater treatment facilities call for the widespread use of sulfate-resisting cement,. The specifications fail to account for the presence of an extremely corrosive acidic environment that can ultimately affect whole-life costs. Furthermore, the current European EN206 standards also fail to account for pH environments below 4 which literature has suggested is far higher than what is commonly encountered in wastewater facilities. It has already been highlighted that there is a noticeable change between pH 4 and pH 1 in terms of mass loss acceleration, indicating an increase in the severity of attack.
CEM II/A-L limestone cements appear to possess and inherent sulfate resisting capability that is superior to a CEM I cement. When combined with 50% or 70% GGBS it represented the best performing binder combination. 100% CEM I was the worst performing cement
All conclusions are based on findings from exposing mortar prisms to the aggressive solution. The behaviour of concrete may differ from those observed in the mortar tests.
57
Mass loss may not be an accurate performance indicator of the deterioration level. Despite a difference between the brushed and unbrushed specimens, cube strengths revealed almost no change in performance. Expansion was not deemed to be an important parameter in sulfuric acid based degradation. The use of GGBS appeared to have little or no effect on improving resistance. Similarly SRPC had no effect on the performance of concrete in this environment. The rate of visual deterioration of a 1% solution of sulfuric acid attack greatly exceeded that of a 5% sodium sulfate solution. The 1% sulfuric acid solution (pH1.5) represents the most severe conditions to be expected in service. Actual pH levels may vary according to time, temperature and bacterial activity. Sulfate deterioration differs from that of sulfuric acid deterioration. It may be possible upon visual examination to confuse the two mechanisms on account of the presence of gypsum, common to both.
6.3 Summary
The results of this investigation have clearly outlined the cause of concrete deterioration in wastewater treatment systems. Consequently, a clear distinction has been drawn between degradation due to sulfate attack and that due to a sulfuric acid attack in this environment. It is evident that neither the concrete standards nor concrete specifiers are taking into account the harsh nature of this form of attack by suitably distinguishing between the two corrosion phenomena. The laboratory programme has also failed to highlight a concrete specification that is capable of withstanding biodeterioration. For this to be fully addressed, the range of aggressive environments associated with wastewater applications needs to be quantified and used as an input for future research work.
58
REFERENCES
AL-DULAIJAN, S. U., MASLEHUDDIN, M., AL-ZAHRANI, M. M., SHARIF, A. M., SHAMEEM, M. & IBRAHIM, M. (2003) Sulfate resistance of plain and blended cements exposed to varying concentrations of sodium sulfate. Cement and Concrete Composites, 25, 429437. ASTM (2004) C1012 - 04 Standard Test Method for Length Change of Hydraulic-Cement Mortars Exposed to a Sulfate Solution. BONAL, J., DONOHUE, S. & MCNALLY, C. (2008) Examination of a novel wavelet approach for bender element testing. In: CANNON, E., WEST, R. & FANNING, P eds. Bridge and Infrastructure Research in Ireland, BRI 2008 Galway, Ireland, 4-5 December. BONAL, J., DONOHUE, S. & MCNALLY, C. (2010) Wavelet analysis of bender element signals, Geotechnique, submitted for publication. BONAVETTI, V. L., RAHHAL, V. F. & IRASSAR, E. F. (2001) Studies on the carbo-aluminate formation in limestone filler-blended cements. Cement and Concrete Research, 31, 853-859. BRE (2003) BRE Special Digest 1, Concrete in Aggressive Ground. Part 1: Assessing the aggressive chemical environment. 2nd Edition ed. Watford, England. BROWN, P. W. & TAYLOR, H. F. W. (1999) The role of ettringite in external sulfate attack. In: MARCHAND, J. & SKALNY, J. (Eds.) Proceedings of Seminar on sulfate attack mechanisms. Quebec, Canada. CHAIX, J.-F., GARNIER, V. & CORNELOUP, G. (2006) Ultrasonic wave propagation in heterogeneous solid media: Theoretical analysis and experimental validation. Ultrasonics, 44, 200-210. CHANG, Z.-T., SONG, X.-J., MUNN, R. & MAROSSZEKY, M. (2005) Using limestone aggregates and different cements for enhancing resistance of concrete to sulfuric acid attack. Cement and Concrete Research, 35, 1486-1494. GOLLOP, R. S. & TAYLOR, H. F. W. (1996) Microstructural and microanalytical studies of sulfate attack. V. Comparison of different slag blends. Cement and Concrete Research, 26, 10291044. GONZLEZ, M. A. & IRASSAR, E. F. (1998) Effect of limestone filler on the sulfate resistance of low C3A portland cement. Cement and Concrete Research, 28, 1655-1667. HEKAL, E. E., KISHAR, E. & MOSTAFA, H. (2002) Magnesium sulfate attack on hardened blended cement pastes under different circumstances. Cement and Concrete Research, 32, 14211427. HIGGINS, D. D. & CRAMMOND, N. J. (2003) Resistance of concrete containing ggbs to the thaumasite form of sulfate attack. Cement and Concrete Composites, 25, 921-929. HOOTON, R. D. (1990) Effects of carbonate additions on heat of hydration and sulfate resistance of Portland cements. IN: KLIEGER, P. & HOOTON, R.D. (Eds) Carbonate Additions to Cement, ASTM Special Technical Publication.
59
HORNAIN, H., MARCHAND, J., DUHOT, V. & MORANVILLE-REGOURD, M. (1995) Diffusion of chloride ions in limestone filler blended cement pastes and mortars. Cement and Concrete Research, 25, 1667-1678. IRASSAR, E. F., BONAVETTI, V. L. & GONZLEZ, M. (2003) Microstructural study of sulfate attack on ordinary and limestone Portland cements at ambient temperature. Cement and Concrete Research, 33, 31-41. IRASSAR, E. F., BONAVETTI, V. L., TREZZA, M. A. & GONZLEZ, M. A. (2005) Thaumasite formation in limestone filler cements exposed to sodium sulphate solution at 20 C. Cement and Concrete Composites, 27, 77-84. IRASSAR, E. F., GONZLEZ, M. & RAHHAL, V. (2000) Sulphate resistance of type V cements with limestone filler and natural pozzolana. Cement and Concrete Composites, 22, 361368. LEE, H. K., LEE, K. M., KIM, Y. H., YIM, H. & BAE, D. B. (2004) Ultrasonic in-situ monitoring of setting process of high-performance concrete. Cement and Concrete Research, 34, 631-640. MONTEIRO, P. J. M. & KURTIS, K. E. (2008) Experimental Asymptotic Analysis of Expansion of Concrete Exposed to Sulfate Attack. ACI Materials Journal, 105, 62-71. NEVILLE, A. (1995) Properties of Concrete, Harlow, England, Longman. O'CONNELL, M., MCNALLY, C. & RICHARDSON, M.G. (2010) 'Biochemical attack on concrete in wastewater applications: a state of the art review'. Cement and Concrete Composites, 32, 479-485 O'CONNELL, M., MCNALLY, C., DONOHUE, S. & RICHARDSON, M.G. (2009) Assessment of ultrasonic signals to determine the early age properties of concretes incorporating secondary cementitious materials. In: Proceedings of the 15th European Meeting of Environmental and Engineering Geophysics, Dublin, 7- 9 September. OSBORNE, G. J. (1999) Durability of Portland blast-furnace slag cement concrete. Cement and Concrete Composites, 21, 11-21. PAVA, S. & CONDREN, E. (2008) Study of the Durability of OPC versus GGBS Concrete on Exposure to Silage Effluent. Journal of Materials in Civil Engineering, 20, 313-320. PIPILIKAKI, P., KATSIOTI, M. & GALLIAS, J. L. (2009) Performance of limestone cement mortars in a high sulfates environment. Construction and Building Materials, 23, 1042-1049. RAMEZANIANPOUR, A. A., GHIASVAND, E., NICKSERESHT, I., MAHDIKHANI, M. & MOODI, F. (2009) Influence of various amounts of limestone powder on performance of Portland limestone cement concretes. Cement and Concrete Composites, 31, 715-720. RAMYAR, K. & INAN, G. (2007) Sodium sulfate attack on plain and blended cements. Building and Environment, 42, 1368-1372. ROBEYST, N., GRUYAERT, E., GROSSE, C. U. & DE BELIE, N. (2008) Monitoring the setting of concrete containing blast-furnace slag by measuring the ultrasonic p-wave velocity. Cement and Concrete Research, 38, 1169-1176.
60
TIAN, B. & COHEN, M. D. (2000) Does gypsum formation during sulfate attack on concrete lead to expansion? Cement and Concrete Research, 30, 117-123. TOSUN, K., FELEKOGLU, B., BARADAN, B. & AKIN ALTUN, I. (2009) Effects of limestone replacement ratio on the sulfate resistance of Portland limestone cement mortars exposed to extraordinary high sulfate concentrations. Construction and Building Materials, 23, 2534-2544. TSIVILIS, S., TSANTILAS, J., KAKALI, G., CHANIOTAKIS, E. & SAKELLARIOU, A. (2003) The permeability of Portland limestone cement concrete. Cement and Concrete Research, 33, 1465-1471. VOIGT, T., GROSSE, C. U., SUN, Z., SHAH, S. P. & REINHARDT, H. W. (2005) Comparison of ultrasonic wave transmission and reflection measurements with P-and S-waves on early age mortar and concrete. Materials and Structures, 38, 729-738.
61
Days Exposure
Days 0 28 56 84 112 140 175 196 224 252 280 308 336 364 392 420 448 476 504
L (%) P1 0 0 0.006 0.012 0.025 0.027 0.029 0.035 0.043 0.045 0.051 0.06 0.066 0.073 0.081 0.101 0.105 0.119 0.141
L (%)P2 0.000 0.000 0.003 0.007 0.018 0.031 0.029 0.034 0.046 0.053 0.062 0.077 0.090 0.109 0.128 0.154 0.182 0.222 0.268
L (%)P3 0 0.01 0.004 0.01 0.019 0.024 0.026 0.031 0.041 0.046 0.045 0.053 0.058 0.069 0.077 0.093 0.102 0.119 0.138
L (%)P4 0 0.01 0.01 0.016 0.027 0.029 0.031 0.037 0.046 0.047 0.05 0.058 0.062 0.072 0.078 0.092 0.097 0.114 0.131
Days Exposure
Days 0 28 56 84 112 140 175 196 224 252 280 308 336 364 392 420 448 476 504
L (%)P1 0 0.004 0.006 0.012 0.019 0.015 0.013 0.015 0.017 0.018 0.017 0.018 0.018 0.018 0.021 0.021 0.022 0.026 0.027
L (%)P2 0.000 0.006 0.007 0.012 0.018 0.018 0.016 0.017 0.017 0.020 0.022 0.022 0.024 0.024 0.026 0.028 0.030 0.030 0.032
L (%)P3 0 0.004 0.008 0.009 0.013 0.016 0.015 0.016 0.017 0.017 0.020 0.021 0.022 0.022 0.026 0.025 0.027 0.030 0.030
L (%)P4 0 0.006 0.005 0.011 0.012 0.015 0.019 0.018 0.020 0.021 0.024 0.026 0.026 0.027 0.03 0.031 0.032 0.033 0.035
Days Exposure
Days 0 28 56 84 112 140 175 196 224 252 280 308 336 364 392 420 448 476 504
L (%)P1 0 0.005 0.004 0.008 0.010 0.013 0.017 0.02 0.019 0.022 0.024 0.026 0.026 0.026 0.029 0.028 0.034 0.033 0.033
L (%)P2 0.000 0.002 0.004 0.010 0.010 0.015 0.012 0.014 0.014 0.017 0.018 0.021 0.020 0.021 0.024 0.026 0.030 0.030 0.031
L (%)P3 0 0.008 0.002 0.012 0.014 0.016 0.015 0.017 0.017 0.020 0.022 0.023 0.023 0.024 0.027 0.027 0.029 0.031 0.035
L (%)P4 0 0.005 0.004 0.012 0.010 0.014 0.014 0.016 0.017 0.019 0.02 0.025 0.023 0.024 0.029 0.027 0.03 0.032 0.036
CEM I 100%
0.8 0.7 0.6 Expansion % 0.5 0.4 0.3 0.2 0.1 0 0 56 112 168 224 280 336 392 448 504 Prism 1 Prism 2 Prism 3 Prism 4
Days Exposure
Days 0 28 56 84 119 140 168 196 224 252 280 308 336 364 392 420 448
L (%)P1 0 0.013 0.025 0.029 0.037 0.048 0.059 0.072 0.088 0.109 0.123 0.153 0.185 0.249 0.309 0.407 0.546
L (%)P2 0.000 0.003 0.015 0.024 0.031 0.043 0.055 0.068 0.084 0.104 0.122 0.153 0.184 0.239 0.297 0.377 0.502
L (%)P3 0 0.008 0.022 0.031 0.041 0.056 0.071 0.086 0.107 0.134 0.162 0.209 0.258 0.341 0.445 0.571 0.725
L (%)P4 0 0.014 0.026 0.034 0.046 0.058 0.072 0.089 0.109 0.132 0.155 0.196 0.236 0.315 0.404 0.52 0.664
Days Exposure
Days 0 28 56 84 119 140 168 196 224 252 280 308 336 364 392 420 448
L (%)P1 0 0.003 0.006 0.015 0.013 0.018 0.016 0.019 0.02 0.021 0.023 0.023 0.024 0.028 0.03 0.031 0.033
L (%)P2 0.000 0.008 0.012 0.016 0.017 0.021 0.024 0.026 0.028 0.033 0.034 0.034 0.039 0.041 0.042 0.046 0.049
L (%)P3 0 0 0.006 0.009 0.014 0.017 0.018 0.021 0.023 0.025 0.026 0.028 0.029 0.034 0.034 0.035 0.037
L (%)P4 0 0.002 0.008 0.013 0.021 0.024 0.026 0.028 0.031 0.034 0.036 0.038 0.039 0.042 0.044 0.046 0.047
SRPC 100%
0.08 0.07 0.06 Expansion % 0.05 0.04 0.03 0.02 0.01 0 0 56 112 168 224 280 336 392 448 Prism 1 Prism 2 Prism 3 Prism 4
Days Exposure
Days 0 28 56 84 112 140 168 196 224 252 280 308 336 364 392 420
L (%)P1 0 0 0 0.003 0.008 0.012 0.015 0.022 0.024 0.024 0.032 0.041 0.048 0.051 0.064 0.071
L (%)P2 0 0 0.003 0.007 0.012 0.016 0.02 0.024 0.029 0.031 0.036 0.044 0.053 0.059 0.068 0.075
L (%)P3 0 0 0.002 0.004 0.01 0.014 0.018 0.023 0.027 0.031 0.036 0.044 0.051 0.058 0.067 0.073
L (%)P4 0 0.001 0.006 0.009 0.012 0.014 0.019 0.023 0.026 0.028 0.028 0.037 0.042 0.045 0.054 0.06
CEM II 100%
0.18 0.16 0.14 Expansion (%) 0.12 0.1 0.08 0.06 0.04 0.02 0 0 5 10 15 Time ^ 0.5(Days) 20 25
Time ^ 0.5(Days)
CEM I 100%
0.4 0.35 0.3 Expansion (%) 0.25 0.2 0.15 0.1 0.05 0 0 5 10 15 Time ^ 0.5(Days) 20 25
SRPC 100%
0.08 0.07 0.06 Expansion (%) 0.05 0.04 0.03 0.02 0.01 0 0 5 10 15 Time ^ 0.5(Days) 20 25
MA53
MA16
y = -0.0283x + 6.1883
10 Time (mins)
15
20
6 5.9 Ln(Pressure)
MB15
Ln(Pressure)
MB16
y = -0.013x + 6.200
10 Time (mins)
15
20
6 5.9 Ln(Pressure)
MC15
Ln(Pressure)
MC16
5.8 5.7 5.6 5.5 5.4 5.3 0 5 10 Time (mins) 15 20 y = -0.0555x + 6.2162
y = -0.018x + 6.221
10 Time (mins)
15
20
MD15
Ln(Pressure)
MD16
y = -0.015x + 6.224
10 Time (mins)
15
20
6.05 6 5.95 5.9 5.85 5.8 5.75 y = -0.0345x + 6.1994 5.7 5.65 0 5 10 Time (mins) 15 20
ME15
ME16
Ln(Pressure)
Ln(Pressure)
y = -0.014x + 6.235
10 Time (mins)
15
20
SR15
Ln(Pressure)
SR16
y = -0.082x + 6.1198
y = -0.006x + 6.223
10 Time (mins)
15
20
10 Time (mins)
15
20
MA53
MA16
y = 1.78E-08x - 9.05E-09
2.5
3 Time ^0.5
3.5
MB15
MB16
y = 1.30E-08x + 1.40E-08
MC15
Water (m3)
MC16
1.00E-07 8.00E-08 6.00E-08 4.00E-08 2.00E-08 0.00E+00 2 2.5 3 Time ^0.5 3.5 4 y = 3.12E-08x - 2.85E-09
y = 1.96E-08x + 3.96E-08
2.5
3 Time ^0.5
3.5
MD15
1.40E-07 1.20E-07 Water (m3) 1.00E-07 8.00E-08 6.00E-08 4.00E-08 2.00E-08 0.00E+00 2
MD16
2.50E-07 2.00E-07 1.50E-07 1.00E-07 5.00E-08 0.00E+00 2 2.5 3 Time ^0.5 3.5 4 y = 7.57E-08x + 1.96E-08
y = 2.68E-08x + 1.26E-08
2.5
3 Time ^0.5
3.5
ME15
ME16
SR15
Water (m3)
SR16
y = 6.44E-08x + 2.96E-08
2.5
3 Time ^0.5
3.5
2.5
3 Time ^0.5
3.5
Water sorptivity
6E-08 5E-08 Water (m3) Water (m3) 4E-08 3E-08 2E-08 1E-08 0 2 2.5 3 Time ^0.5 3.5 4 y = 1.56E-08x - 6.25E-09 2.00E-08 0.00E+00 2 2.5 3 Time ^0.5 3.5 4
MA53
MA16
y = 2.27E-08x + 1.59E-08
MB15
Water (m3)
y = 5.07E-09x + 6.55E-09
MB16
y = 4.04E-09x + 1.76E-08
2.5
3 Time ^0.5
3.5
2.5
3 Time ^0.5
3.5
MC15
Water (m3)
MC16
8.00E-08 6.00E-08 4.00E-08 y = 2.59E-08x - 4.23E-09 2.00E-08 0.00E+00 2 2.5 3 Time ^0.5 3.5 4
y = 1.59E-08x - 3.35E-08
2.5
3 Time ^0.5
3.5
MD15
Water (m3)
MD16
1.00E-07 5.00E-08 0.00E+00 2 2.5 3 Time ^0.5 8E-08 7E-08 6E-08 3.5 4 y = 3.97E-08x + 2.36E-08
ME15
ME16
Water (m3)
5E-08 4E-08 3E-08 2E-08 1E-08 0 2 2.5 3 Time ^0.5 2.50E-07 2.00E-07 3.5 4 y = 2.44E-08x - 2.35E-08
y = 2.66E-08x + 5.75E-09
2.5
3 Time ^0.5
3.5
SR15
SR16
Water (m3)
1.50E-07 1.00E-07 5.00E-08 0.00E+00 2 2.5 3 Time ^0.5 3.5 4 y = 5.35E-08x + 2.76E-08
2.5
3 Time ^0.5
3.5
1 2449 2437 2445 2438 2451 2451 2461 2464 2462 2511 2487 2500 3550 3527 3582 3611
2 2448 2436 2444 2438 2451 2451 2468 2473 2472 2517 2493 2502 3548 3526 3593 3623
3 2449 2435 2443 2439 2451 2451 2446 2449 2448 2498 2475 2486 3550 3528 3569 3588
4 2453 2435 2445 2439 2455 2452 2442 2446 2444 2497 2473 2485 3544 3528 3570 3582
8 2436 2446 2453 2453 2357 2344 2406 2415 3550 3527 3395 3422
12
16
2446
2447
2446
2446
2454
2454
2454
2454
2238
2147
2071
1999
WEEK 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
2 17 33 17 33 33 50 250 283 317 545 564 527 -12 -104 243 313
4 100 17 33 50 100 67 -183 -167 -150 182 200 218 -58 -81 -23 -162
12
16
20
50 0 67 83
50
67
50
50
100
100
100
100
-1650 -1817 -3583 -5100 -6367 -7567 -1018 -1055 -2818 -4200 -5509 -6709 12 -93 -2052 -2017 23 -81 -4000 -3942 23 -81 -5461 -5426 12 -70 -6678 -6701 35 -70 -7896 -7849
MASS (g) CUBE No. 3 4 5 6 7 8 9 10 11 12 13 14 PRISM No. 1 2 3 4 MASS (g/m2) CUBE No. 3 4 5 6 7 8 9 10 11 12 13 14 PRISM No. 1 2 3 4
WEEK 0 2423 2434 2421 2424 2423 2436 2445 2429 2442 2485 2485 2456 3485 3468 3494 3498
1 2426 2436 2422 2426 2426 2437 2441 2425 2440 2484 2485 2452 3486 3470 3483 3496
2 2426 2436 2422 2427 2424 2438 2438 2426 2434 2488 2487 2455 3487 3470 3488 3504
3 2426 2437 2423 2426 2425 2436 2422 2405 2417 2467 2472 2435 3484 3468 3466 3480
4 2424 2436 2422 2425 2425 2437 2411 2394 2406 2458 2463 2427 3485 3470 3448 3463
12
16
MIX 20
MB 24
2422 2425 2425 2437 2273 2281 2361 2325 3485 3470 3281 3303
2423
2423
2423
2424
2438
2438
2438
2438
2175
2099
2016
1956
WEEK 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
12
16
20
17 17 33 17
33
33
33
50
33
33
33
33
-2600 -2683 -4450 -5717 -7100 -8100 -2255 -2382 -3964 -5364 -6800 -7945 0 23 -2470 -2261 12 23 -4128 -3965 23 23 -5472 -5264 12 35 -6852 -6574 23 58 -7884 -7571
MASS (g) CUBE No. 3 4 5 6 7 8 9 10 11 12 13 14 PRISM No. 1 2 3 4 MASS (g/m2) CUBE No. 3 4 5 6 7 8 9 10 11 12 13 14 PRISM No. 1 2 3 4
WEEK 0 2437 2434 2439 2421 2439 2438 2407 2435 2447 2478 2494 2472 3449 3472 3515 3475
1 2438 2437 2441 2424 2440 2440 2404 2433 2443 2478 2498 2475 3451 3477 3512 3476
2 2439 2436 2441 2425 2441 2440 2410 2439 2451 2485 2504 2481 3452 3476 3520 3486
3 2438 2436 2441 2423 2441 2439 2394 2427 2438 2471 2491 2470 3452 3475 3507 3471
4 2437 2435 2440 2423 2440 2439 2387 2420 2424 2461 2482 2457 3450 3476 3494 3452
12
16
MIX MC 20 24
2441 2442 2442 2440 2442 2424 2441 2439 2440 2440 2438 2441 2311 2311 2205 2122 2051 1986 2379 2353 2272 2198 2135 2078 3450 3475 3370 3334 3452 3476 3228 3193 3451 3475 3107 3064 3450 3475 2998 2957 MIX 3453 3478 2912 2868 MC 24
WEEK 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
12
16
20
33 50 33 17
50
50
17
50
33
33
50
-2067 -2267 -4033 -5417 -6600 -7683 -2091 -2164 -3636 -4982 -6127 -7164 12 35 -1681 -1635 35 46 -3328 -3270 23 35 -4730 -4765 12 35 -5994 -6006 46 70 -6991 -7038
MASS (g) CUBE No. 3 4 5 6 7 8 9 10 11 12 13 14 PRISM No. 1 2 3 4 MASS (g/m2) CUBE No. 3 4 5 6 7 8 9 10 11 12 13 14 Prism No. 1 2 3 4
WEEK 0 2519 2516 2516 2527 2515 2500 2450 2446 2446 2502 2496 2481 3527 3499 3512 3503
1 2522 2518 2519 2529 2517 2504 2435 2433 2432 2496 2496 2480 3528 3504 3497 3484
2 2522 2518 2518 2529 2517 2502 2436 2434 2432 2497 2496 2481 3530 3505 3497 3489
3 2521 2518 2518 2529 2516 2503 2419 2419 2414 2482 2481 2468 3530 3503 3477 3461
4 2520 2517 2518 2528 2517 2502 2409 2411 2404 2475 2474 2457 3528 3502 3464 3444
12
16
MIX MD 20 24
2518 2519 2519 2519 2518 2529 2517 2503 2504 2503 2503 2502 2298 2297 2191 2104 2020 1957 2377 2364 2270 2199 2124 2063 3529 3503 3325 3297 3531 3503 3174 3136 3530 3504 3061 3024 3532 3504 2950 2916 MIX 3531 3504 2867 2838 MD 24
WEEK 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
12
16
20
33 33 33 50
50
50
50
33
67
50
50
33
-2467 -2483 -4250 -5700 -7100 -8150 -2164 -2127 -3836 -5127 -6491 -7600 23 46 -2168 -2388 46 46 -3919 -4255 35 58 -5229 -5554 58 58 -6516 -6806 46 58 -7478 -7710
MASS (g) CUBE No. 3 4 5 6 7 8 9 10 11 12 13 14 PRISM No. 1 2 3 4 MASS (g/m2) CUBE No. 3 4 5 6 7 8 9 10 11 12 13 14 PRISM No. 1 2 3 4
WEEK 0 2447 2437 2423 2427 2411 2446 2432 2420 2433 2471 2486 2470 3532 3471 3494 3469
1 2450 2437 2424 2427 2412 2449 2432 2422 2435 2483 2493 2480 3536 3476 3468 3489
2 2448 2440 2424 2427 2413 2448 2437 2425 2440 2486 2498 2484 3537 3476 3492 3478
3 2448 2438 2425 2428 2413 2448 2416 2407 2418 2460 2481 2464 3536 3475 3473 3451
4 2448 2438 2425 2427 2413 2448 2408 2396 2409 2458 2472 2455 3536 3473 3460 3440
12
16
MIX ME 20 24
2425 2426 2425 2425 2426 2428 2413 2448 2449 2448 2449 2449 2303 2313 2210 2127 2042 1982 2386 2372 2284 2215 2144 2080 3536 3474 3331 3329 3536 3475 3198 3199 3535 3475 3084 3076 3535 3476 2974 2952 MIX 3535 3476 2896 2870 ME 24
WEEK 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
12
16
20
33 17 33 33
50
33
33
50
50
33
50
50
-1950 -2000 -3717 -5100 -6517 -7517 -1818 -1782 -3382 -4636 -5927 -7091 46 35 -1890 -1623 46 46 -3432 -3130 35 46 -4754 -4557 35 58 -6029 -5994 35 58 -6933 -6945
MASS (g) CUBE No. 3 4 5 6 7 8 9 10 11 12 13 14 PRISM No. 1 2 3 4 MASS (g/m2) CUBE No. 3 4 5 6 7 8 9 10 11 12 13 14 PRISM No. 1 2 3 4
WEEK 0 2481 2459 2470 2478 2482 2458 2461 2478 2464 2520 2509 2522 3543 3550 3542 3555
1 2482 2461 2471 2480 2483 2461 2451 2472 2452 2526 2514 2530 3546 3551 3541 3556
2 2482 2462 2472 2479 2483 2460 2448 2469 2450 2524 2512 2529 3546 3549 3537 3552
3 2482 2460 2471 2480 2483 2460 2431 2447 2434 2508 2498 2512 3546 3550 3512 3524
4 2482 2460 2471 2480 2483 2460 2423 2437 2425 2500 2490 2507 3545 3550 3504 3514
12
16
MIX SR 20 24
2461 2472 2472 2472 2472 2471 2484 2461 2462 2461 2462 2460 2337 2326 2190 2105 2028 1962 2415 2421 2309 2232 2159 2092 3545 3550 3395 3395 3546 3552 3211 3205 3546 3553 3075 3077 3547 3553 2965 2968 MIX 3548 3553 2875 2881 SR 24
WEEK 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
8 33 33 33 50
12
16
20
33
33
33
17
67
50
67
33
-2350 -2300 -4567 -5983 -7267 -8367 -1709 -1836 -3873 -5273 -6600 -7818 23 0 -1704 -1855 35 23 -3838 -4058 35 35 -5414 -5542 46 35 -6690 -6806 58 35 -7733 -7814
Week MA 0 1 2 3 4 8 12 16 20 24 0 261 380 29 11 1601 3586 5047 6314 7505 MB 0 51 25 335 513 2442 4127 5454 6832 7875
Mass Loss (g/m2) MC 0 6 105 107 285 1984 3567 4974 6182 7219 MD 0 153 137 408 567 2300 4065 5402 6728 7735 ME 0 66 139 198 337 1844 3415 4762 6117 7121 SR 0 15 55 353 483 1959 4084 5553 6841 7933
MIX
WEEK 0 1 2 3 4 8 12 16 20 24 0 91 218 255 182 -1127 -2982 -3818 -4527 -5091 0 91 255 291 273 -836 -2800 -3582 -4473 -4927
WEEK 0 1 2 3 4 8 12 16 2469 2484 2498 2516 2512 2390 2294 2258 2494 2511 2526 2542 2548 2409 2310 2272
MIX WEEK 0 1 2 3 4 8 12 16 20 0 273 527 855 782 -1436 -3182 -3836 -4527 0 309 582 873 982 -1545 -3345 -4036 -4636
WEEK 0 1 2 3 4 8 12 16 2459 2482 2496 2516 2529 2394 2299 2255 2484 2477 2517 2534 2549 2426 2329 2291
MC 24 2231 2257 MC
WEEK 0 1 2 3 4 8 12 16 20 24 0 418 673 1036 1273 -1182 -2909 -3709 -4273 -4145 0 -127 600 909 1182 -1055 -2818 -3509 -4036 -4127
WEEK 0 1 2 3 4 8 12 16 2553 2563 2575 2580 2581 2471 2378 2331 2544 2557 2569 2580 2578 2453 2366 2325
MD 24 2266 2267 MD
WEEK 0 1 2 3 4 8 12 16 20 24 0 182 400 491 509 -1491 -3182 -4036 -4836 -5218 0 236 455 655 618 -1655 -3236 -3982 -4691 -5036
WEEK 0 1 2 3 4 8 12 16 2478 2498 2512 2525 2534 2424 2326 2281 2483 2503 2513 2531 2541 2424 2338 2288
MIX WEEK 0 1 2 3 4 8 12 16 20 0 364 618 855 1018 -982 -2764 -3582 -4000 0 364 545 873 1055 -1073 -2636 -3545 -4055
WEEK 0 1 2 3 4 8 12 16 2510 2526 2540 2549 2540 2435 2337 2291 2529 2549 2556 2557 2552 2459 2351 2303
SR 24 2220 2229 SR
MASS (g/m2) WEEK CUBE No. 0 1 2 3 4 8 12 16 0 291 545 709 545 -1364 -3145 -3982 0 364 491 509 418 -1273 -3236 -4109
Week MA 0 1 2 3 4 8 12 16 20 24 0 61 158 183 153 982 2891 3700 4500 5009 MB 0 194 370 577 589 1491 3264 3936 4582 4727
Mass Loss (g/m2) MC 0 97 425 649 820 1118 2864 3609 4155 4136 MD 0 140 286 377 383 1573 3209 4009 4764 5127 ME 0 243 389 577 692 1027 2700 3564 4027 3991 SR 0 219 346 407 323 1318 3191 4045 4882 5364
Compressive Strength: Mix MA Days 0 28 56 168 Water 57 57 58 60 Acid 57 50 35 21 % Loss 0% 12% 40% 65%
Cube strength (MPa) 80 70 60 50 40 30 20 10 0 0
28
56
84 Days
112
140
168
Compressive Strength: Mix MB Days 0 28 56 168 Water 61 65 65 70 Acid 61 53 35 19 % Loss 0% 18% 46% 74%
Cube strength (MPa) 80 70 60 50 40 30 20 10 0 0
28
56
84 Days
112
140
168
Days 0 28 56 168
Water 58 58 63 65
Acid 58 51 37 22
80 70 60 50 40 30 20 10 0 0
28
56
84 Days
112
140
168
Days 0 28 56 168
Water 65 71 75 77
Acid 65 56 43 19
80 70 60 50 40 30 20 10 0 0 28
MD - 100% CEM I
56
84 Days
112
140
168
Days 0 28 56 168
Water 54 57 59 69
Acid 54 49 34 18
80 70 60 50 40 30 20 10 0 0
28
56
84 Days
112
140
168
Days 0 28 56 168
Water 60 58 68 75
Acid 60 59 41 21
80 70 60 50 40 30 20 10 0 0
SR - 100% SRPC
28
56
84 Days
112
140
168
Ref Bar (mm) 5.981 5.989 6.618 6.851 6.189 6.192 6.436 6.424 6.229 6.249 6.596 6.621 6.433 6.439 Ref Bar (mm) 5.975 5.997 6.802 6.882 6.191 6.192 6.258 6.47 6.243 6.252 6.61 6.621 6.438 6.438
Reading (mm) 2.621 2.625 3.345 3.476 2.807 2.823 3.138 3.047 2.887 2.861 3.213 3.23 3.046 3.047 Reading (mm) 2.884 2.874 3.685 3.725 3.031 3.039 3.22 3.328 3.056 3.082 3.427 3.424 3.311 3.222
L (%) 0.036 -0.004 -0.007 -0.002 0.026 -0.005 0.009 -0.010 -0.008 -0.011 -0.009 -0.011 L (%) 0.002 -0.014 -0.015 -0.012 0.034 -0.008 -0.026 -0.019 -0.024 -0.030 -0.002 -0.037
6.608 Ref Bar (mm) 5.956 5.994 6.664 6.869 6.19 6.192 6.234 6.448 6.232 6.251 6.605 6.62 6.437 6.439 Ref Bar (mm) 5.985 6 6.848 6.883 6.192 6.192 6.396 6.484 6.245 6.254 6.624 6.618 6.438 6.431
3.347 Reading (mm) 2.75 2.761 3.422 3.631 2.942 2.937 3.014 3.209 2.99 3.005 3.352 3.357 3.176 3.178 Reading (mm) 2.851 2.822 3.625 3.665 2.981 2.976 3.188 3.246 3.013 3.019 3.354 3.372 3.177 3.171
-0.016 L (%) -0.004 -0.002 -0.006 -0.009 0.005 -0.002 -0.004 -0.005 -0.008 -0.012 -0.011 -0.011 L (%) -0.018 -0.016 -0.013 -0.015 -0.012 -0.024 -0.022 -0.023 -0.037 -0.027 -0.033 -0.033
Ref Bar (mm) 6.135 6.132 6.034 6.038 6.214 6.244 6.05 6.058 6.25 6.248 6.612 6.62 6.434 6.434 Ref Bar (mm) 6.133 6.133 6.035 6.039 6.233 6.247 6.058 6.056 6.251 6.253 6.614 6.619 6.437 6.438
Reading (mm) 3.182 3.172 3.07 3.064 3.224 3.277 3.075 3.075 3.276 3.273 3.635 3.635 3.451 3.451 Reading (mm) 2.949 2.951 2.833 2.83 3.07 3.037 2.829 2.831 3.015 3.016 3.385 3.379 3.196 3.194
L (%) -0.002 -0.006 -0.012 -0.003 -0.006 -0.009 -0.006 -0.006 -0.007 -0.010 -0.009 -0.009 L (%) -0.007 -0.010 0.008 -0.010 -0.018 -0.016 -0.021 -0.021 -0.018 -0.022 -0.023 -0.024
Ref Bar (mm) 6.134 6.133 6.035 6.039 6.212 6.247 6.057 6.059 6.25 6.252 6.612 6.619 6.435 6.436 Ref Bar (mm) 6.132 6.132 6.033 6.039 6.241 6.247 6.054 6.059 6.244 6.251 6.616 6.62 6.433 6.437
Reading (mm) 3.085 3.076 2.885 2.893 3.157 3.142 2.895 2.833 3.027 3.045 3.386 3.381 3.203 3.202 Reading (mm) 3.017 2.881 2.721 2.724 2.959 2.93 2.74 2.73 2.917 2.918 3.386 3.29 3.088 3.091
L (%) -0.037 -0.036 0.001 -0.019 -0.042 -0.068 -0.066 -0.060 -0.068 -0.072 -0.070 -0.071 L (%) -0.024 -0.026 -0.012 -0.026 -0.025 -0.031 -0.030 -0.033 0.008 -0.032 -0.038 -0.038
Ref Bar (mm) 6.147 6.15 6.463 6.468 6.242 6.249 6.054 6.056 6.249 6.253 6.61 6.612 6.444 6.446 Ref Bar (mm) 6.149 6.15 6.466 6.464 6.248 6.25 6.056 6.059 6.252 6.255 6.611 6.612 6.443 6.446
Reading (mm) 3.134 3.1 3.374 3.396 3.141 3.145 2.951 2.953 3.145 3.148 3.502 3.495 3.32 3.323 Reading (mm) 3.042 3.013 3.235 3.226 2.995 3.01 2.8 2.803 2.986 2.982 3.338 3.331 3.151 3.153
L (%) -0.016 -0.009 -0.020 -0.022 -0.021 -0.021 -0.022 -0.022 -0.023 -0.027 -0.030 -0.029 L (%) -0.038 -0.040 -0.046 -0.041 -0.048 -0.048 -0.052 -0.054 -0.054 -0.058 -0.062 -0.062
Ref Bar (mm) 6.152 6.15 6.466 6.466 6.246 6.25 6.055 6.057 6.25 6.255 6.611 6.612 6.443 6.446 Ref Bar (mm) 6.151 6.15 6.464 6.464 6.246 6.25 6.054 6.056 6.253 6.257 6.611 6.611 6.444 6.447
Reading (mm) 2.98 2.966 3.298 3.274 3.219 3.108 2.98 2.884 3.114 3.07 3.432 3.427 2.253 3.254 Reading (mm) 3.132 3.162 3.402 3.399 3.172 3.163 2.984 2.969 3.169 3.162 3.514 3.517 3.334 3.334
L (%) 0.006 -0.003 0.063 0.017 0.044 0.004 0.019 0.000 0.002 0.000 -0.402 -0.003 L (%) -0.017 -0.018 -0.022 -0.027 -0.020 -0.027 -0.026 -0.030 -0.031 -0.030 -0.036 -0.038
Ref Bar (mm) 5.962 5.819 6.464 6.444 6.148 6.153 6.049 6.053 6.253 6.254 6.614 6.616 6.44 6.443 Ref Bar (mm) 5.964 5.716 6.467 6.46 6.153 6.153 6.052 6.054 6.253 6.254 6.615 6.616 6.443 6.443
Reading (mm) 2.717 2.559 3.215 3.187 2.891 2.883 2.794 2.794 2.989 2.985 3.348 3.347 3.174 3.174 Reading (mm) 2.662 2.574 3.209 3.259 2.988 2.981 2.882 2.87 3.075 3.062 3.405 3.402 3.222 3.213
L (%) 0.004 0.001 0.001 -0.004 0.002 0.000 -0.002 -0.004 -0.002 -0.004 -0.002 -0.004 L (%) 0.018 0.040 0.055 0.052 0.053 0.047 0.050 0.044 0.037 0.035 0.032 0.029
Ref Bar (mm) 5.964 5.825 6.466 6.434 6.152 6.153 6.051 6.055 6.253 6.254 6.615 6.616 6.441 6.442 Ref Bar (mm) 5.797 5.732 6.466 6.466 6.151 6.153 6.054 6.055 6.25 6.257 6.615 6.616 6.442 6.443
Reading (mm) 2.876 2.732 3.372 3.337 3.047 3.043 2.995 2.959 3.148 3.147 3.508 3.506 3.332 3.33 Reading (mm) 2.582 2.51 3.254 3.257 2.893 2.907 2.787 2.789 2.982 2.981 3.34 3.339 3.158 3.156
L (%) 0.000 -0.002 -0.005 -0.007 0.015 -0.001 -0.005 -0.006 -0.006 -0.007 -0.006 -0.008 L (%) 0.004 0.005 -0.014 -0.010 -0.018 -0.018 -0.018 -0.022 -0.021 -0.022 -0.025 -0.026
Ref Bar (mm) 5.683 5.955 6.458 6.464 6.146 6.15 6.054 6.054 6.255 6.229 6.609 6.61 6.442 6.445 Ref Bar (mm) 5.937 5.956 6.463 6.465 6.15 6.149 6.058 6.056 6.255 6.239 6.611 6.597 6.444 6.445
Reading (mm) 2.672 2.889 3.224 3.185 2.869 2.853 2.748 2.743 2.942 2.905 3.297 3.297 3.123 3.119 Reading (mm) 2.681 2.689 3.129 3.119 2.806 2.804 2.696 2.69 2.893 2.87 3.226 3.201 3.053 3.047
L (%) -0.067 -0.085 -0.084 -0.092 -0.096 -0.098 -0.099 -0.103 -0.098 -0.099 -0.101 -0.104 L (%) -0.027 -0.032 -0.031 -0.031 -0.038 -0.040 -0.038 -0.041 -0.047 -0.052 -0.050 -0.052
Ref Bar (mm) 5.927 5.955 6.461 6.46 6.15 6.152 6.055 6.058 6.255 6.236 6.61 6.608 6.444 6.445 Ref Bar (mm) 5.95 5.957 6.462 6.461 6.151 6.154 6.059 6.058 6.255 6.245 6.61 6.6 6.444 6.446
Reading (mm) 2.592 2.631 3.131 3.083 2.817 2.758 2.711 2.678 2.875 2.841 3.222 3.213 3.045 3.044 Reading (mm) 2.715 2.72 3.162 3.156 2.827 2.825 2.734 2.718 2.88 2.923 3.246 3.228 3.067 3.061
L (%) 0.002 -0.017 0.001 -0.024 -0.004 -0.018 -0.018 -0.024 -0.021 -0.024 -0.026 -0.026 L (%) -0.025 -0.027 -0.035 -0.037 -0.035 -0.041 -0.055 -0.034 -0.051 -0.054 -0.056 -0.059
a r t i c l e
i n f o
a b s t r a c t
The costs associated with the provision and maintenance of drinking water and wastewater infrastructure represents a signicant nancial demand worldwide. Maintenance costs are disproportionately high, indicating a lack of adequate durability. There remains a lack of consensus on degradation mechanisms, the performance of various cement types, the role of bacteria in the corrosion process associated with wastewater applications and testing methodologies. This paper presents a review of the literature, outlining the various research approaches undertaken in an effort to address this problem. The ndings of these varying approaches are compared, and the different strategies employed are compiled and discussed. It is proposed that a key step in advancing the understanding of the associated deterioration mechanism is a combined approach that considers the interaction between biological and chemical processes. If this can be achieved then steps can be taken to establishing a performance-based approach for specifying concrete in these harsh service conditions. 2010 Elsevier Ltd. All rights reserved.
Article history: Received 14 July 2009 Received in revised form 20 April 2010 Accepted 1 May 2010 Available online 7 May 2010 Keywords: Sulfate attack Sulfuric acid Thiobacillus Secondary cementitious materials Performance specications
1. Introduction The provision of high quality water and wastewater infrastructure requires signicant international expenditure on concrete with consequent expectations of lengthy service lives. For example in the US alone, it is estimated that annual investments of up to $20 billion and $21 billion is required to provide adequate infrastructure for drinking water and wastewater respectively [1]. It is also estimated that the annual operation and maintenance costs associated with drinking water and wastewater infrastructure to be in excess of $31 billion and $25 billion respectively. Against this backdrop, it is surprising to note that the corrosion of water and wastewater infrastructure has been a topic of debate for decades, with little consensus on the methods for designing and specifying this infrastructure to optimally meet the harsh environmental demands it will meet in service [27]. The majority of studies to date have focused on the deterioration of concrete in sewer systems and pipelines [5,8,9]. However little detailed research has been conducted into the effect of corrosion on the vital treatment facilities that are processing our wastewater. Concrete pipes in sewer systems tend to be an off-the-shelf product with little input by the specier into specication of mix design. As a result the performance of the product is largely dependent on the manufacturers mix design which is inuenced by local factors. In treatment plants the concrete may be specied by the engineer, but a lack of in-depth research into the deterioration of these structures has
meant little change in professional practice concerning concrete mix design. Existing evidence has shown that corrosion is present in many concrete structures associated with water and wastewater treatment. The alarming fact is that some of these facilities are deteriorating signicantly after less than a decade in service (Fig. 1). In this context it is clear that current design practices based on prescriptive approaches to concrete specication may not be appropriate to deal with the aggressive nature of wastewater, and in some cases, the treatment processes involved in drinking water purication [10]. Existing research ndings are not yet inuencing current construction practice. The lack of widely quoted durability design formulae illustrates that the deterioration mechanisms associated with this critical infrastructural application are not yet widely accepted or understood. This paper will assist in bridging this gap by considering the role of key parameters such as environmental conditions, the nature of the attack and the physical results of the attack on the concrete. This will promote increased understanding of the deterioration mechanism and facilitate the introduction of a performance-based design approach. 2. Characterising the wastewater environment The deterioration of sewer systems has long been a topic under considerable scrutiny and in the mid 1940s a comprehensive scientic evaluation was undertaken in an attempt to understand the corrosion process [11]. Current research has continued to focus on the deterioration of concrete sewer pipes and case studies have taken place throughout the world, including comprehensive
* Corresponding author. Tel.: +353 1 716 3202; fax: +353 1 716 3297. E-mail address: ciaran.mcnally@ucd.ie (C. McNally). 0958-9465/$ - see front matter 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.cemconcomp.2010.05.001
480
acid reacts with a cement matrix, the rst step involves a reaction between the acid and the calcium hydroxide (Ca(OH)2) forming calcium sulfate according to the following equation:
This is subsequently hydrated to form gypsum (CaSO42H2O), the appearance of which on the surface of concrete pipes takes the form of a white, mushy substance which has no cohesive properties and has, the consistency of cottage cheese [23]. In the continuing attack, the gypsum would react with the calcium aluminate hydrate (C3A) to form ettringite, an expansive product:
reviews on current infrastructure [8]. In the latter, the condition of the sewer system in four cities in the Lebanon was evaluated whereby certain contributory factors in corrosion were outlined: Biological Oxygen Demand (BOD) levels, high sulfate and dissolved sulde concentration, high temperatures, high H2S gas concentration, high turbulence and long detention times, low dissolved oxygen levels, low water velocity and low wastewater pH. These and other criteria have been outlined in several publications, all of which detail the conditions leading to corrosion in sewer environments [4,5,1214]. The contributory factors outlined above are not only limited to sewer piping they are also found in wastewater treatment plants. Occurrences of concrete degradation in these structures have been recorded in a limited fashion in aeration tanks [15], in septic tanks and pumping stations [16] and the underside of concrete slabs and in primary inuent channels [17]. The latter two sources both make reference to the fact that corrosion has been observed just above the waterline. This is signicant in that prior experimental research [5,18] into understanding degradation of concrete in sewer pipes has proven that optimum corrosion levels also occur just above the waterline. Work carried out into determining depth proles of sulfate ingress into concrete noted that core samples were taken from the walls surrounding the spiral pump of a sewage treatment plant as well as the concrete walls of a clarier which has been damaged by sulfates originating from the sewage waters [19]. Evidence thus far has identied bacterial manifestation of the genus Thiobacillus as a major contributor to the deterioration process of concrete sewer pipelines [5,11]. The product of their metabolism results in sulfuric acid being formed which attacks the cementitious matrix of the concrete causing loss of strength and cohesion. Thiobacillus however, plays only a part of a much broader and complicated corrosion process. In the often anaerobic conditions which develop in raw sewage inuent, sulfate-reducing bacteria convert sulfates into suldes such as hydrogen sulde (H2S) gas. In favourable conditions this diffuses into the atmosphere and, in the presence of oxygen, is further reduced to elemental sulfur or partially reduced sulfur compounds. In turn, they provide the catalyst necessary for the aerobic Thiobacillus bacteria to begin producing sulfuric acid; a more detailed explanation of the corrosion process is presented in a subsequent section. Sulfuric acid has been identied as a corrosive agent not only in corroding sewers but also in wastewater treatment plants [20,21]. An attack by sulfuric acid however is a combined acidsulfate reaction with the hydrogen ion causing a dissolution effect, coupled with corrosive role played by the sulfate ion [2,22]. When sulfuric
According to Skalny et al. [22], the ettringite can be located in deeper sections of concrete as long as the pH is high enough for it to form and the gypsum can migrate into these regions. The evidence gathered by Davis et al. [23] in their analysis of piping, however, showed that little ettringite was discovered in the corroding front and that the thermodynamics of the conversion to gypsum may be so fast that ettringite is a short-lived intermediate. From the evidence discussed above there seems to be a distinct relationship between the corrosion occurring in concrete sewers and that in wastewater treatment facilities. Common variables include environmental conditions, the nature of the attack and the physical results of the attack on the concrete. Mehta and Burrows [24] have discussed how a paradigm shift is required in concrete design, moving away from the traditional prescriptive approach to one that promotes a performance-based design. However for such an approach to succeed, it is imperative that the deterioration mechanism is fully understood. In this light it is necessary to account for the severe environments that wastewater infrastructures will encounter in service, and to take an in-depth look at the current state of research into sulfate and sulfuric acid corrosion in a wastewater environment. 3. Biodegradation aspects 3.1. Providing resistance to biochemical attack When assessing the available scientic research, it is important to consider sulfate attack, sulfuric acid attack and how they are both relevant in determining the resistance of current concrete design specications to such attacks as biogenic sulfuric acid (BSA) corrosion. As expected there is much conicting data available on the subject, a scenario which is eloquently detailed in one practicing engineers publication on the topic [6]. Also of interest is the performance of cements containing additions of ground granulated blast-furnace slag (GGBS) which, when mixed with Portland cement, has been proven to possess an inherent sulfate resisting capability [25,26]. GGBS is being used in increasing quantities in concrete practice today along with other secondary cementitious materials (SCMs) such as pulverised fuel ash (PFA). With the high CO2 emissions associated with the production of Portland cement, these SCMs have the advantage of being by-products from other industrial processes and as such can help reduce the CO2 footprint of a construction project. While concretes produced using these binders are more dense and durable in the long term, they are also prone to reduced early age strengths and require particular attention when curing [2729]. In assessing experimental test methods previously used by researchers many contrasting opinions exist [4,30], including the proposed inadequacy of sulfate testing as a method to analyse biological corrosion in a wastewater environment while others stipulate simultaneous biological and chemical sulfuric acid testing as
481
the only true methodology [4]. The participation of the sulfate ion in sulfuric acid (H2SO4) corrosion and that of residual sulfates present in wastewater (found in efuent from food and beverage industries [31]) cannot be ignored however. Reviewing in situ and simulated experimental test methods also provides a valuable insight into the aggressive nature of the environment that sewers and wastewater treatment plants are exposed to. This also helps to characterise the environmental conditions favourable to the initiation of biogenic sulfuric acid corrosion, allowing scope for investigating the role played by both sulfate-reducing and sulfuroxidising bacteria.
3.3. Sulfur-oxidising bacteria The formation of sulfur is perhaps the critical link in the chain of events leading to the corrosion of concrete in a wastewater environment. In microbiological experiments carried out in 1945 to investigate why sewer pipes were corroding, Parker [11] discovered ve strains of the species Thiobacillus on the surface of concrete which oxidise sulfur, or some partially reduced form of sulfur, to form sulfuric acid. More recent research suggests some of the Thiobacillus strains involved in concrete corrosion as being Thiobacillus thiooxidans, Thiobacillus intermedius, Thiobacillus perometabolis, Thiobacillus novellus, Thiobacillus thioparus, Thiobacillus neapolitanus and Thiobacillus versutus all of which are known to oxidize and grow with reduced inorganic sulfur compounds [14,34]. Research has also identied iron-oxidising bacteria, such as Thiobacillus ferrooxidans, as being involved in the production of sulfuric acid in pyritic ground and in sewage treatment plants [35,36]. Bacteria of the genus Thiobacillus do not attach themselves to the surface of concrete under any conditions. Roberts et al. [14] state that the pH of the concrete has to be reduced to 9 and assuming sufcient moisture, nutrients and oxygen are present only then will the Thiobacillus bacteria colonise. Several theories for the lowering of the pH of the concrete to around 9 have been put forward, including the involvement of the dissociation process of hydrogen sulde as discussed above. However the most widely assumed theory is that the pH will be lowered due to the effects of carbonation [13,14,23]. As a result of in situ tests conducted in a sewage system with high concentrations of hydrogen sulde (>600 ppm) an alternative theory has been put forward into determining the conditions necessary for bacterial colonisation [7]. The authors claim that the generally accepted role of carbonation in lowering the pH of a concretes surface does not hold for their experiments. Instead they theorise that in the thin moisture layer itself, the bacteria oxidise the hydrogen sulde gas to form sulfuric acid, thereby reducing its pH. They further claim that the bacteria will grow in the layer even when the pH of the concrete itself ranges from pH 1113. Parker [37] noted in his experimental observations however, that Thiobacillus concretivorus (as he termed the strain of Thiobacillus found to attack concrete) did not convert the hydrogen sulde directly into sulfuric acid but only free sulfur or other forms of utilisable sulfur compounds including thiosulfate [38]. In characterising the strain T. thiooxidans, Waksman and Joffe [39] and Nica et al. [38] also stated that hydrogen sulde and other suldes are not used directly by the sulfur-oxidising organism. In recognising the wide range of Thiobacillus strains that take part in sulfuric acid production, it must be noted that not all thrive in an identical environment. Some of these strains are categorised into acid-preferring acidophilic sulfur-oxidising microorganisms (ASOM), such as T. thiooxidans and neutral-preferring neutrophilic sulfur-oxidising microorganisms (NSOM), such as T. intermedius [14,38]. It is proposed that different strains of neutrophilic bacteria colonise the surface of the concrete as its pH depresses from approximately a value of 8 to around a value of 6 through their production of sulfuric acid [23]. It was also found that microbial succession is a surface phenomenon and that the ASOM move into the corroding concrete with the corroding layer whereas NSOM do not. Parker [11] observed that the bacteria which he was cultivating survived to a pH of approximately 6.5 above which none was capable of growth. At these slightly acidic pH values the acidophilic sulfur-oxidising bacteria colonise and further depress the pH of the concrete surface to as low as 2 at which level the strain T. thiooxidans can be found thriving [18,40]. The optimum temperature at which the acid was produced in its highest quantities after 50 days was found to be 30 C while Barbosa et al. [15] noted in their research that sulde oxidation by the strain T. dentricans decreased
3.2. Sulfate-reducing bacteria Initiating the bacterial processes, sulfates present in the raw sewage in sewer system are converted into suldes by sulfatereducing anaerobic bacteria such as Desulfovibrio [4]. In partially lled sewers, anaerobic conditions can only occur in the slime layer on the walls of the pipe above the water line. Some of the essential environmental conditions necessary in the wastewater environment for these bacteria to function and grow are dissolved oxygen levels approaching zero and sufcient carbon and sulfate concentrations in the wastewater itself [32]. When this occurs they utilise the sulfates present in the wastewater to obtain the oxygen they require and in turn release sulfur ions [33]. According to research aimed at quantifying microbial-induced deterioration of concrete, the bacteria derive the energy required for the reduction of sulfate by the oxidation of organic compounds and H2 [14]. In their assessment of Lebanons sewer network, Ayoub et al. [8] claim that sulfate to sulde reduction takes place when the bacteria derive their oxygen from dissolved oxygen and nitrates in the wastewater. They state however, that corrosion in Lebanons sewers was not observed to have occurred in areas where dissolved oxygen levels were greater than zero. If sulfate-reducing bacteria require dissolved oxygen to induce the corrosion cycle, as they suggest, one must ask why is it that no corrosion was found where dissolved oxygen exists. Their own search of existing literature suggested that sulde build-up could not occur with dissolved oxygen levels greater than 0.5 mg/l whereas Hewayde et al. [33] set a level of 0.1 mg/l above which corrosion will not occur. The nal process in the initial stage of concrete deterioration involves the sulfur ions released by the bacteria. These in turn react with dissolved hydrogen in the wastewater to form an essential contributory product in the corrosion process, hydrogen sulde (H2S) [33]. The hydrogen sulde initially formed is found in its dissolved liquid form but for this poorly soluble compound to contribute to the concrete deterioration process it must leave the wastewater and enter a gaseous phase. The normal pH of sewage is slightly acidic and in the range pH 56 but when this begins to lower in conjunction with turbulent water (often found in sewer pipes or associated with some wastewater treatment processes), the H2S escapes and collects in the atmosphere above the water level [4,16,33]. A thin layer of moisture exists on the surface of the concrete pipe exposed to the atmosphere and it is into here the hydrogen sulde is gas is dissolved. The condensate layer has a high pH attributed to the alkalinity of the concrete (which can have a pH of between 11 and 13). It also serves as the driving force behind the gas dissolution. At high pH levels the hydrogen sulde is separated into HS or S2 ions which attract more H2S into the moisture layer [14]. Research has also shown that the concentration of H2S in the moisture lm increases as the pH of the mortar lining of the concrete pipe decreases [13]. In the presence of oxygen the H2S reacts to form elemental sulfur or partially oxidised sulfur species [4,9,14,21], which can sometimes be seen in the corrosion products deposited on the concrete surface [18].
482
at low temperature and was inhibited at 15.6 C. Parker [11] also discovered that the rate of acid production in his bacteria increased with increasing nitrogen levels up to a concentration of 50 ppm, above which there appeared to be a slight inhibition. 3.4. Other acids and organisms In a departure from the accepted role of the species Thiobacillus in lowering the surface pH of concrete from approximately 8 to 4, some authors have also attributed the initial reduction to that of fungus growth [5,41]. Mori et al. [5] found an unidentied green fungus which grew at high pH levels and was capable of reducing the pH to levels suitable for colonisation and growth of T. thiooxidans. Gu et al. [41] go further in their explanation and identied the fungus they observed as Fusarium. They claim that this has a more detrimental effect on the concrete that that of the neutrophilic bacteria T. intermedius. In their research they described the latter as being able to etch the surface of the concrete while the fungus Fusarium was able to penetrate the material. They also state that a wide range of acids are produced by fungi including acetic, oxalic and glucuronic acids. A further set of experiments [42] using mortar inoculated with bacteria including T. intermedius was conducted independently of Mori et al. [5]. In these experiments the deterioration of concrete was thought to be caused by the sulfuric acid produced by the bacteria; however the authors noted little gypsum and limited change in the sulfate ion concentration of the culture medium. They concluded that primary deterioration of the concrete was caused by carbonic and organic acids, which include acetic acid which are all metabolites produced by bacteria. 4. The role of biogenic sulfuric acid corrosion 4.1. Attack mechanisms Only limited work has been carried out in assessing the performance of concrete mixes in a biological environment [4] a surprising fact considering several researchers have claimed that biogenic sulfuric acid corrosion found in wastewater systems is more severe than chemical sulfuric acid and sulfate attack [43,44]. This represents a key knowledge gap in the development of a material based performance specication. While research has identied gypsum, ettringite and even thaumasite as the end-product of the corrosion product the debate centres on the order of their formation, their quantities and specic effects on the cement matrix. The corrosive nature of a sulfuric acid attack has been well documented from both in situ observations and chemical testing on concrete [4,5,22,30,4547]. The dissolution effect of the hydrogen ion and the separate effect of the sulfate ion combine to create an aggressive set of chemical reactions, threatening the stability of a cement matrix. Debate exists however regarding the mechanisms behind chemical and biological sulfuric acid attacks, and resistance to the former does not necessarily result in resistance to the latter [4,9,48]. Explanations centre on the involvement of the sulfuric acid producing bacteria Thiobacillus where Monteny et al. [4] claim that it is the moist conditions in the gypsum corrosion front that constitute an excellent breeding ground for the bacteria to thrive. They then migrate into the concrete producing acid much closer to the corrosion front although Yamanka et al. [7] dispute this claiming it is the acid itself moving inward. In a chemical attack however, the poor penetration of sulfuric acid limits the effects of corrosion to the surface [49]. The acid must negotiate its way through this corrosion layer in order for the attack to continue. It is generally assumed that this results in less severe consequences relative to a biological attack, as the corroded surface acts as a barrier for further penetration. Hence regular brushing of
loosely adhering particles may be important in any attempt to mimic biological activity with chemical testing [4,50]. 4.2. Types of sulfuric acid attack In 1945, C.D. Parker described a sulfuric acid attack on concrete sewers as producing a white putty-like deposit, moist, aky and easily removed from the surface [11,37]. The calcium sulfate (gypsum) formed was a result of a reaction between the hydration products in the cement matrix and the sulfuric acid [22], as previously described in Eq. (1). Experimental and in situ analysis of both mortar and concrete has conrmed that gypsum formation is one of the primary corrosion mechanisms involved in the deterioration of the cement matrix leading to a loss of cohesion in cementitious calcium compounds [5,23,33,51]. The degradation of concrete foundations of an Italian building exposed to sewage waters however was attributed to the growth of gypsum crystals at the aggregate-paste interface causing a loss of strength [52]. The build-up of gypsum though can also act as a barrier to further penetration, slowing an attack [4,53] however it has also been claimed that the rougher surface area leads to a greater surface area to be attacked [18]. The relative resistance of various binder combinations to sulfuric acid attack has been discussed by some researchers. Experiments exposing 100 mm Portland cement concrete cubes with a binder of 35% ordinary Portland cement (OPC)/65% GGBS to an H2SO4 solution for 5 months, as described in BRE Digest 363 [54], reported higher performance than for binders of 100% sulfate resisting Portland cement (SRPC) or 75% OPC/25% PFA [47]. This improved performance of concrete in acidic conditions has been attributed to either lower porosity, lower levels of calcium hydroxide or both [4,45,48] while Saricimen et al. [16] determined that in a 3% owing H2SO4 solution neither SRPC nor OPC showed any difference in resisting attack, a conclusion supported by [7]. Monteny et al. [43] suggest that a rened pore structure will increase the capillary action of the cement matrix and act as a mechanism for the aggressive solution to nd its way deeper into the concrete. In experiments to assess commercially available piping, De Belie et al. [9] prepared non-standard cylindrical specimens of concrete with CEM I and CEM III high sulfate resisting cement and exposed them to a 0.5% H2SO4 solution in alternating wet/dry cycles. They concluded the limestone aggregate, acting as a sacricial medium to reduce the rate of acid attack, played a more crucial role against attack that the cement type. The sulfate resisting cements also performed better than the blast-furnace slag cements, an observation similarly supported by other experimental results [29] (conducting experiments with 60% GGBS cylinders in a 1% H2SO4 for 168 days). This is however contradicted by research from Monteny et al. [4] who performed experiments in 15% H2SO4 solutions. However it is noteworthy that the signicance of the role played by the limestone aggregate has been emphasised [25,29]. Ettringite is a crystalline compound and its formation can be observed in the process of cement hydration (primary ettringite) and in the effects of an external sulfate attack (secondary ettringite). Some of the reactions associated with its formation involve calcium aluminates, such as C3A, and gypsum but may also incorporate an external sulfate attack on the calcium aluminate hydrates and monosulfate hydrate phases [4,54]. According to Skalny et al. [22] under sulfuric acid attack only limited amounts of ettringite will form in deeper sections of the concrete as long as the pH is high enough to maintain its stability and enough of the gypsum formed in the initial stages of attack can move into the concrete. This assessment concurs with other researchers who have also stated ettringites inability to survive in an acidic environment [44] and even in alkaline environments with pHs as high as 10.6 [49]. In contrast, Monteny et al. [4] stress in their assessments the importance of ettringite and its more devastating
483
effect on concrete than gypsum, while its formation from a sulfuric acid attack was also documented by others assessing the inuence of fungi on concrete corrosion [42] and simulated biogenic sulfuric acid corrosion [5]. It is noteworthy that the presence of such key compounds such as gypsum and ettringite is accepted as being a function of mix design and the binder combinations used. However the notion of using this material design to control the presence of these expansive compounds is not at this stage well developed. 4.3. Inuence of the sulfate ion As an attack by sulfuric acid is a combined acidsulfate reaction, many researchers have deemed it prudent to assess concrete susceptibility in standard sulfate testing solutions including sodium sulfate (Na2SO4), magnesium sulfate (MgSO4) or a combination of both. The validity of this method to assess attack in a wastewater environment has however drawn some uncertainty based on discrepancies in chemical and biological tests [4]. Sodium and magnesium-based sulfate solutions have substantially different effects on concrete. With the former, calcium hydroxide primarily undergoes decomposition to gypsum and subsequently ettringite. When there is an insufcient source of calcium for the reaction to continue only then will the solution begin to attack the CSH phase [22,51]. Magnesium solutions attack all phases simultaneously in the cement matrix preferring calcium hydroxide rst followed by the calciumsilicatehydrate (C SH) phase to obtain its reactive calcium. The products from a magnesium sulfate reaction include gypsum, ettringite, a magnesiumsilicatehydrate (which lacks cohesive properties) and the mineral form of magnesium hydroxide, brucite [55]. In sodium sulfate, ettringite can be associated mainly with the reaction between the AFm monosulfate phase and the sulfate ions migrating into the concrete. At low concentrations of sulfate solu tions (<1000 mg SO2 4 /l) ettringite will be the primary cause of deterioration [4] whereas at higher concentrations (>8000 mg SO2 4 /l) gypsum will dominate in a sulfate attack [56]. It is important, therefore, to use a concentration of sulfates that accurately represents the corrosion mechanism in the desired environment. In magnesium sulfate solutions the deterioration mechanism is primarily a result of the loss of cohesion and disintegration with the formation of gypsum and magnesium hydroxide [4,22]. The saturated solution pH of magnesium hydroxide is approximately 10.5 and consequently this causes the destabilisation of ettringite. As a result the circumstances favourable in the formation of ettringite from a magnesium sulfate attack are signicantly impeded [57]. Skalny et al. [22] do note however that a limited amount may form when the pH remains high enough in the concrete for a sufcient period of time while research on slag cements [58] attributed ettringite formation as substantially contributing to the damage produced by MgSO4 solutions. Gollop and Taylor [51,58] concluded in their analysis that the resistance of GGBS concretes to attack by sulfates increases with decreasing levels of Al2O3. Lower levels of C3A were noted by other researchers [16,29] in reducing the harmful effects of exposure to sodium sulfate. In using cement pastes in their analysis however, Gollop and Taylor neglected the effects of the aggregate-paste interface previously considered important in analysing a sulfate/ sulfuric acid attack [49,5961]. Their addition of increasing levels of GGBS up to a level of 92% increased resistance to attack by sodium sulfate solutions but had the opposite effect when exposed to magnesium sulfate. In an assessment of 150 mm 75 mm rein forced concrete cylinder specimens exposed to a 2.1% SO2 4 sulfate solution, Al-Amoudi [61] indicated that for a 60% GGBS replacement level, deterioration in the mixed magnesium/sodium based solution was considered signicant. It was concluded that GGBS
mixes fared poorest when compared to other cement replacement materials including silica fume (10% replacement) and y ash (20% replacement). In assessing results from a study by the BRE [25,46], Osborne [29] also came to similar conclusions regarding the effects of magnesium and sodium sulfate solutions and the use of high percentages of GGBS as a cement replacement. The TEG one-year review [62] also noted the benet of a 70% GGBS replacement level with limestone cement and good quality carbonate aggregate against conventional forms of sulfate attack. 4.4. Simulation of the biological corrosion process In a simulated wet/dry 17-day attack cycle Vincke et al. [21] exposed 2 2 5 cm specimens of concrete to a biological sulfur solution containing Thiobacilli bacteria following an incubation period in an H2S environment. After a total of 51 days and three cycles, the specimens made with a CEM I Portland cement and CEM III blast-furnace slag cement were analysed in terms of weight loss. Results indicated that both mixes performed similarly. De Belie et al. [9] used an almost identical process to the above [21], using specimens that were 80 mm diameter and 15 mm deep, which were subjected to a fourth cycle of 17 days. In their experiments they observed the sulfate ion concentration of their solution to increase from 2 g/l to 4 g/l which the authors cite as evidence for the production of sulfuric acid by the sulfur-oxidising bacteria. The results of their experiments concluded that Portland cement performed better than CEM III blast-furnace slag cement. In this instance they theorise that owing to the greater surface area of CEM III the bacteria are able to colonise the surface of the cement more rapidly than the Portland cement. Further investigation has revealed other methods of modelling biological corrosion in the wastewater environment. A simulation chamber was developed by researchers in Hamburg, described by Monteny et al. [4], which allowed the corrosion process to be modelled at eight times the in situ level could be reached through the optimisation of the corrosive environment. Test blocks of 60 11 7 cm were immersed in 10 cm of water at 30 C and sprayed with Thiobacilli bacteria. H2S gas at 10 ppmv was pumped into the chamber and acted as a substrate for the bacteria. The number of bacteria on the surface of the specimens was counted and it was found that the rate of corrosion was dependent on the levels of T. thiooxidans detected. Experimental work was also carried out into the corrosion mechanism involved in the deterioration of concrete constructing a simulated sewer pipe 20 m long and a diameter of 15 cm [5]. Test specimens of mortar bars 4 4 16 cm were made with the bottom half of these bars placed in sewage and exposed to H2S gas not exceeding 300 ppm. Identical mortar bars were placed half submersed into a sewage medium, an autotrophic basal growth culture medium without thiosulfate and distilled water. These were inoculated every two weeks with T. thiooxidans. Corrosion just above the waterline was observed on bars in the sewage and autotrophic basal media. Those in water remained unaffected while the sewage samples displayed the greatest corrosion rate. The authors concluded that based on these results the bacteria required a supply of moisture and nutrients to initiate the corrosion process while the corrosion products formed were determined to be gypsum and secondary ettringite. As with the issue of sulfuric acid attack, the possibility of restricting the formation of corrosion products through appropriate mix design is not developed. 5. Conclusions Three research foci were evident in the study of sulfate/sulfuric acid effects on concrete. These are:
484
M. OConnell et al. / Cement & Concrete Composites 32 (2010) 479485 [13] Jana D, Lewis RA. Acid attack in a concrete sewer pipe a petrographic and chemical investigation. In: Proceedings of 27th international conference on cement microscopy. Victoria, Canada; 2005. [14] Roberts DJ, Nica D, Zuo G, Davis JL. Quantifying microbially induced deterioration of concrete: initial studies. Int Biodeter Biodegr 2002;49(4):22734. [15] Barbosa VL, Burgess JE, Darke K, Stuetz RM. Activated sludge biotreatment of sulphurous waste emissions. Rev Env Sci Biotech 2002;1(4):34562. [16] Saricimen H, Shameem M, Barry MS, Ibrahim M, Abbasi TA. Durability of proprietary cementitious materials for use in wastewater transport systems. Cem Concr Compos 2003;25(4):4217. [17] Stufebean J. Report on bids and award of contract for the San Jos/Santa Clara water pollution control plant, FY 2007/2008 CIP, east primary inuent channel repair project. San Jos, California; 2007. [18] Vollertsen J, Nielsen AH, Jensen HS, Wium-Andersen T, Hvitved-Jacobsen T. Corrosion of concrete sewers the kinetics of hydrogen sulde oxidation. Sci Total Environ 2008;394(1):16270. [19] Weritz F, Taffe A, Schaurich D, Wilsch G. Detailed depth proles of sulfate ingress into concrete measured with laser induced breakdown spectroscopy. Constr Build Mater 2009;23(1):27583. [20] Idriss AF, Negi SC, Jofriet JC, Hayward GL. Corrosion of steel reinforcement in mortar specimens exposed to hydrogen sulde, part 1: impressed voltage and electrochemical potential tests. J Agr Eng Res 2001;79(3):22330. [21] Vincke E, Verstichel S, Monteny J, Verstraete W. A new test procedure for biogenic sulfuric acid corrosion of concrete. Biodegradation 1999;10(6): 4218. [22] Skalny J, Marchand J, Odler I. Sulfate attack on concrete. London: Spon Press; 2002. [23] Davis JL, Nica D, Shields K, Roberts DJ. Analysis of concrete from corroded sewer pipe. Int Biodeter Biodegr 1998;42(1):7584. [24] Mehta PK, Burrows RW. Building durable structures in the 21st century. Concr Int 2001;23(3):5763. [25] Building Research Establishment. BRE special digest 1, concrete in aggressive ground: part 1: assessing the aggressive chemical environment, 2nd ed.; 2003. [26] Higgins DD. Increased sulfate resistance of GGBS concrete in the presence of carbonate. Cem Concr Compos 2003;25(8):9139. [27] Chidiac SE, Panesar DK. Evolution of mechanical properties of concrete containing ground granulated blast furnace slag and effects on the scaling resistance test at 28 days. Cem Concr Compos 2008;30(2):6371. [28] McNally C, Richardson MG, Evans C, Callanan T. Determination of chloride diffusion coefcients for use with performance-based specications. In: Dhir RK, Newlands MD, Whyte A, editors. Proceedings of 6th international congress global construction: ultimate concrete opportunities; application of codes, designs and regulations. Dundee; 2005. p. 3217. [29] Osborne GJ. Durability of Portland blast-furnace slag cement concrete. Cem Concr Compos 1999;21(1):1121. [30] Chang ZT, Song XJ, Munn R, Marosszeky M. Using limestone aggregates and different cements for enhancing resistance of concrete to sulfuric acid attack. Cem Concr Res 2005;35(8):148694. [31] Percheron G, Bernet N, Moletta R. Start-up of anaerobic digestion of sulfate wastewater. Bioresour Technol 1997;61(1):217. [32] Mohan SV, Rao NC, Prasad KK, Sarma PN. Bioaugmentation of an anaerobic sequencing batch biolm reactor (AnSBBR) with immobilized sulphate reducing bacteria (SRB) for the treatment of sulphate bearing chemical wastewater. Process Biochem 2005;40(8):284957. [33] Hewayde E, Nehdi M, Allouche E, Nakhla G. Effect of geopolymer cement on microstructure, compressive strength and sulfuric acid resistance of concrete. Mag Concr Res 2006;58(5):32131. [34] Emtiazi G, Habibi MH, Setareh M. Isolation of some new sulphur bacteria from activated sludge. J Appl Microbiol 1990;69(6):86470. [35] Longworth TI. Contribution of construction activity to aggressive ground conditions causing the thaumasite form of sulfate attack to concrete in pyritic ground. Cem Concr Compos 2003;25(8):100513. [36] Maeda T, Negishi A, Komoto H, Oshima Y, Kamimura K, Sugio T. Isolation of iron-oxidizing bacteria from corroded concretes of sewage treatment plants. J Biosci Bioeng 1999;88(3):3005. [37] Parker CD. The function of Thiobacillus concretivorus (nov. spec.) in the corrosion of concrete exposed to atmospheres containing hydrogen sulde. Aust J Exp Biol Med Sci 1945;23:918. [38] Nica D, Davis JL, Kirby L, Zuo G, Roberts DJ. Isolation and characterization of microorganisms involved in the biodeterioration of concrete in sewers. Int Biodeter Biodegr 2000;46(1):618. [39] Waksman SA, Joffe JS. Microorganisms concerned in the oxidation of sulfur in the soil II. Thiobacillus thiooxidans, a new sulfur-oxidizing organism isolated from the soil. J Bacteriol 1922;7:23956. [40] Jin SM, Yan WM, Wang ZN. Transfer of IncP plasmids to extremely acidophilic Thiobacillus thiooxidans. Appl Environ Microbiol 1992;58(1):42930. [41] Gu JD, Ford TE, Berke NS, Mitchell R. Biodeterioration of concrete by the fungus Fusarium. Int Biodeter Biodegr 1998;41(2):1019. [42] Tazawa EI, Morinaga T, Kawai K. Deterioration of concrete derived from metabolites of microorganisms. In: Proceedings 3rd CANMET/ACI international conference durability of concrete. American Concrete Institute. Nice, France; 1994. p. 108797. [43] Monteny J, De Belie N, Vincke E, Verstraete W, Taerwe L. Chemical and microbiological tests to simulate sulfuric acid corrosion of polymer-modied concrete. Cem Concr Res 2001;31(9):135965.
Studies of the biological processes behind the corrosion of wastewater infrastructure, with particular reference to the role of sulfate-reducing and sulfur-oxidising bacteria. Studies of the chemical effects of sulfates and sulfuric acid on concrete mixes. Laboratory-based research methodologies, especially those incorporating the biological effect on concrete. Chemical tests alone do not fully represent the microbial effects on concrete, although they may help in assessing the types of damage that can occur. Some researchers have carried out fullscale laboratory analysis, but it is worth noting that the equipment necessary to adequately mimic in situ conditions is invariably complicated, cumbersome and custom built [4,63]. The realisation of resources required to undertake such research continues to be an obstacle to addressing this topic. The use of such complex research apparatus in routine performance-based specication is impractical. Although there exists signicant quantities of data on the topics of sulfate, sulfuric acid and biogenic corrosion of concrete, little has been achieved in the way of formulating an accepted mathematical model of deterioration that incorporates agreed parameters of signicance. This represents a signicant knowledge gap and acts as a technical barrier towards using material design as a means of controlling corrosion due to biochemical attack. This continues to inhibit the design of durable concrete wastewater infrastructure and has signicant implications for public expenditure in this area. The need to consider the interaction of biological and chemical processes may hold the key to achieving greater progress and allow practitioners to use concrete mix design as a means of delivering intended service lives. Acknowledgements The authors gratefully acknowledge the nancial support provided by Enterprise Ireland Innovation Partnership Project IP/ 2008/540 and Ecocem Ireland. References
[1] Congressional budget ofce. future investment in drinking and wastewater infrastructure. ISBN 0160512433, Congressional Budget Ofce; 2002. [2] Attiogbe EK, Rizkalla SH. Response of concrete to sulfuric acid attack. ACI Mater J 1988;84(6):4818. [3] Beeldens A, Monteny J, Vincke E, De Belie N, Van Gemert D, Taerwe L, et al. Resistance to biogenic sulfuric acid corrosion of polymer-modied mortars. Cem Concr Compos 2001;23(1):4756. [4] Monteny J, Vincke E, Beeldens A, De Belie N, Taerwe L, Van Gemert D, et al. Chemical, microbiological, and in situ test methods for biogenic sulfuric acid corrosion of concrete. Cem Concr Res 2000;30(4):62334. [5] Mori T, Nonaka T, Tazaki K, Koga M, Hikosaka Y, Noda S. Interactions of nutrients, moisture and pH on microbial corrosion of concrete sewer pipes. Water Res 1992;26(1):2937. [6] Neville A. The confused world of sulfate attack on concrete. Cem Concr Res 2004;34(8):127596. [7] Yamanaka T, Aso I, Togashi S, Tanigawa M, Shoji K, Watanabe T, et al. Corrosion by bacteria of concrete in sewerage systems and inhibitory effects of formates on their growth. Water Res 2002;36(10):263642. [8] Ayoub GM, Azar N, El Fadel M, Hamad B. Assessment of hydrogen sulde corrosion of cementitious sewer pipes: a case study. Urban Water J 2004;1(1):3953. [9] De Belie N, Monteny J, Beeldens A, Vincke E, Van Gemert D, Verstraete W. Experimental research and prediction of the effect of chemical and biogenic sulfuric acid on different types of commercially produced concrete sewer pipes. Cem Concr Res 2004;34(12):222336. [10] Kawamura S. Integrated design and operation of water treatment facilities. Wiley; 2000. [11] Parker CD. The isolation of a species of bacterium associated with the corrosion of concrete exposed to atmospheres containing hydrogen sulde. Aust J Exp Biol Med Sci 1945;23:8190. [12] Fan CY, Field R, Pisano WC, Barsanti J, Joyce JJ, Sorenson H. Sewer and tank ushing for sediment, corrosion, and pollution control. J Water Res Pl 2001;127(3):1946.
M. OConnell et al. / Cement & Concrete Composites 32 (2010) 479485 [44] Thomson G. Corrosion and rehabilitation of concrete access/inspection chambers. In: Proceedings of 63rd annual water industry engineers and operators conference. Brauer College, Warrnambool, Australia; 2000. p. 2835. [45] Bassuoni MT, Nehdi ML. Resistance of self-consolidating concrete to sulfuric acid attack with consecutive pH reduction. Cem Concr Res 2007;37(7): 107084. [46] Building Research Establishment. BRE special digest 1, concrete in aggressive ground. Part 2: specifying the concrete and additional protective measures, 2nd ed.; 2003. [47] Hill J, Byars EA, Sharp JH, Lynsdale CJ, Cripps JC, Zhou Q. An experimental study of combined acid and sulfate attack of concrete. Cem Concr Compos 2003;25(8):9971003. [48] Vincke E, Wanseele EV, Monteny J, Beeldens A, Belie ND, Taerwe L, et al. Inuence of polymer addition on biogenic sulfuric acid attack of concrete. Int Biodeter Biodegr 2002;49(4):28392. [49] Allahverdi A, Skvara F. Acidic corrosion of hydrated cement based materials. Part 2 kinetics of the phenomenon and mathematical models. Ceram Silikty 2000;44(4):15260. [50] De Belie N. Evaluation of methods for testing concrete degradation in aggressive solutions. In: Proceedings of workshop on performance of cement-based materials in aggressive aqueous environments characterisation, modelling, test methods and engineering aspects. RILEM Publications SARL; 2008. p. 7990. [51] Gollop RS, Taylor HFW. Microstructural and microanalytical studies of sulfate attack. IV. Reactions of a slag cement paste with sodium and magnesium sulfate solutions. Cem Concr Res 1996;26(7):101328. [52] Tulliani JM, Montanaro L, Negro A, Collepardi M. Sulfate attack of concrete building foundations induced by sewage waters. Cem Concr Res 2002;32(6): 8439.
485
[53] Rendell F, Jauberthie R. The deterioration of mortar in sulphate environments. Constr Build Mater 1999;13(6):3217. [54] Building Research Establishment. BRE digest 363, sulfate and acid resistance of concrete in the ground, 2nd ed.; 1991. [55] Collepardi MA. State-of-the-art review on delayed ettringite attack on concrete. Cem Concr Compos 2003;25(4):4017. [56] Hekal EE, Kishar E, Mostafa H. Magnesium sulfate attack on hardened blended cement pastes under different circumstances. Cem Concr Res 2002;32(9): 14217. [57] Santhanam M, Cohen MD, Olek J. Sulfate attack research whither now? Cem Concr Res 2001;31(6):84551. [58] Gollop RS, Taylor HFW. Microstructural and microanalytical studies of sulfate attack. V. Comparison of different slag blends. Cem Concr Res 1996;26(7): 102944. [59] Irassar EF, Bonavetti VL, Gonzlez M. Microstructural study of sulfate attack on ordinary and limestone Portland cements at ambient temperature. Cem Concr Res 2003;33(1):3141. [60] Pava S, Condren E. Study of the durability of OPC versus GGBS concrete on exposure to silage efuent. J Mater Civil Eng 2008;20(4):31320. [61] Al-Amoudi OSB. Performance of 15 reinforced concrete mixtures in magnesiumsodium sulphate environments. Constr Build Mater 1995;9(3): 14958. [62] Clark LA. Thaumasite expert group one-year review. London: Department of the Environment, Transport and the Regions; 2000. [63] De Muynck W, De Belie N, Verstraete W. Effectiveness of admixtures, surface treatments and antimicrobial compounds against biogenic sulfuric acid corrosion of concrete. Cem Concr Compos 2009;31(3):16370.
Assessment of ultrasonic signals to determine the early age properties of concretes incorporating secondary cementitious materials OConnell, M., McNally, C., Donohue, S., Bonal, J. and Richardson, M.G. Summary Secondary cementitious materials (SCMs) such as ground granulated blast-furnace slag (GGBS) are used in increasing quantities in concrete practice internationally. While these materials offer benefits such as reduced CO2 and a more dense microstructure, they also have drawbacks in terms of slower initial gain of strength. There are significant financial implications associated with this, as it can lead to delays in the construction process. Key to overcoming this challenge is the development of a methodology to assess the early-age stiffness development in concretes manufactured using GGBS. This paper presents the results of a study into the application of ultrasonic sensors to assess the early age concrete stiffness. A novel wavelet-based approach is used to overcome the difficulties associated with wave reflections and classical wave theory is used to determine the concrete small-strain stiffness based on P and S wave velocities. It was found that the results are largely in agreement with those obtained using standard strength testing, suggesting potential practical applications of this method.
Near Surface 2009 15th European Meeting of Environmental and Engineering Geophysics Dublin, Ireland, 7 - 9 September 2009
Introduction Secondary cementitious materials (SCMs) such as ground granulated blast-furnace slag (GGBS) and pulverized fuel ash (PFA) are used in increasing quantities in concrete practice internationally. These materials are derived from industrial by-products and offer the significant advantages of abating the high CO2 emissions associated with cement manufacture, while also making the concrete more dense and durable in the long term (McNally et al, 2005). However a significant drawback is that SCMs can lead to the concrete being relatively weak in the initial few days after being cast, leading to costly delays in the construction process. Decisions on construction sites on when such a concrete is sufficiently strong to resist applied loading are often based on empirical site practices, which can be quite punitive to SCMs. In this context, the importance of being able to assess non-destructively the early-age strength and stiffness of concrete with or without SCMs becomes clear. Ultrasonic sensors have the potential to allow such an assessment, but to date there is no clear approach available on the characterisation of concrete at early ages. Materials The experimental procedure required three concrete mixes to be prepared with a varying amount of GGBS to be used as a cement replacement. The cement used was a CEM II A-L Portland-limestone-cement and GGBS was introduced at replacement levels of 0%, 50% and 70%. Details of the mix design are outlined in Table 1. A series of eight 100mm cubes and eight 100mm x 200mm concrete cylinders were prepared for each mix with a water/binder ratio of 0.45. These were left to cure at room temperature for 24 hours prior to demoulding after which they were stored in a water curing tank at 20C for up to 28 days.
Quantities (kg/m ) Mix ID MA MB MC Cement 360 180 108 GGBS 0 180 252 20mm. 810 805 805 10mm. 405 400 400 Sand 685 680 162 Water 162 162 162
3
Table 1: Concrete mix designs Wavelet Analysis of Ultrasonic Signals It is well known that the Youngs modulus of a material can be determined from the velocity of the P and S waves across a sample. This approach may be applied using ultrasonic sensors on concrete samples to allow quantification of the increase in concrete stiffness due to further hydration of the cementitious materials present. However for this to be done accurately, it is key that the arrival time of the P and S waves can be determined. For the former this is a relatively straightforward task as the P-wave arrival can be visually detected from the 1st discontinuity in the signal. However for the S-wave arrival the situation is less clear, as there exists a significant amount of interference from the presence of reflected waves off the side boundaries of the sample. It is in this context that a wavelet-based algorithm was developed to allow detection of the discontinuity in the signal associated with the S-wave arrival. A signal may be observed through two main domains: time domain and frequency domain. The Fourier Transform and its inverse connect these domains and are the main mathematical tools for signal analysis. The Fourier Transform is perfectly adequate for stationary and periodic signals and provides a global description of frequency distribution, energy and
Near Surface 2009 15th European Meeting of Environmental and Engineering Geophysics Dublin, Ireland, 7 - 9 September 2009
overall regularity. However, it involves the complete loss of local time information such as the location of singularities. A wavelet is a mathematical function used to divide a given function or continuous-time signal into different frequency components and study each component with a resolution that matches its scale. A wavelet transform is the representation of a function by wavelets. The wavelets are scaled and translated copies (known as "daughter wavelets") of a finite-length or fast-decaying oscillating waveform (known as the "mother wavelet"). Wavelet transforms have advantages over traditional Fourier transforms for representing functions that have discontinuities and sharp peaks, and for accurately deconstructing and reconstructing finite, non-periodic and/or non-stationary signals. The Wavelet Transform is motivated by the possibility of finding a singularity as it decomposes the signal into elementary building blocks that are well localized both in time and frequency (Mallat and Hwang, 1992). The local detail is matched to the scale of the wavelet, so it can characterise coarse (low frequency) features on large scales and fine (high frequency) features on small scales. By assuming the shear wave to be both plane and homogenous, the shear wave arrival in a bender element test is characterised by the arrival of a broad-band energy. The shear wave is however preceded by the near-field effect, waves reflecting off the boundaries of the sample which create a first singular point in the output signal, as illustrated in Fig. 2. It induces an opposite phase wave preceding the S-wave. The main S-wave arrival creates a second singular point; however its location in the time domain is usually scrambled by the near-field effect and the noise. Nevertheless this singularity exists and is defined as a discontinuity in the first derivative at this time. Bonal et al. (2008) have presented a novel method of analysing signals using wavelets based on the Lipschitz exponent (Mallat and Hwang, 1992). The Lipschitz exponent is a well known tool used to estimate function differentiability. A key feature of the Lipschitz exponent is its ability to distinguish between singularities due to noise, and those due to events such as the arrival of the S-wave. In order to accurately determine the first arrival of a shear wave using wavelets, Bonal et al (2008) followed several maxima lines of the wavelet transform modulus across a range of scales. The local Lipschitz exponents are determined and the discontinuities sorted based on this result. They observed the first arrival to be among these singular points with a local Lipschitz exponent of almost 1; this value is variable but must be compatible with the characteristics of the input function. This algorithm was then implemented using a Python script; this approach for wavelet based singularity detection does not give automated results, but nonetheless directs us to points of interest within the signal. Methodology The investigation required S-wave and P-wave ultrasonic readings to be taken on cubes at one, three, eight, fourteen and twenty-eight days following casting of the specimens. Two sets of transducers were used for this purpose in conjunction with a square wave pulser-receiver in through-transmission mode. The S-wave transducers had a frequency of 0.1 MHz, the Pwave transducers 50 kHz. For ultrasonic S-wave measurements each transducer was coupled directly opposite the other using a shear wave couplant gel. A 100V amplitude pulse was emitted into the specimen for a duration of two minutes and readings were recorded using a digital oscilloscope connected to a laptop computer. A similar procedure was used for Pwaves with a P-wave couplant gel. The previously described Python script was now employed to locate the first S-wave arrival; the application of this script is shown in Fig. 2. To calculate static Youngs modulus for each mix at eight, fourteen and twenty-eight days tests were carried out in accordance with BS 1881-121 (1983). Extensometers were used to determine the mean micro-strain of the cylinder at the upper and basic applied stresses.
Near Surface 2009 15th European Meeting of Environmental and Engineering Geophysics Dublin, Ireland, 7 - 9 September 2009
Fig 1: Ultrasonic measurement experimental set-up: (L-R: laptop, oscilloscope, pulserreceiver, transducers, 100mm cube and couplant gels) With the stresses known and the micro-strain readings given by the extensometers, a static Youngs modulus value to the nearest 100 N/mm2 was calculated for each sample tested. From ultrasonic testing, the velocity of the P-wave (VP) and the S-wave (VS) can be calculated. Knowing these, the small strain shear modulus and the Poissons ratio can be determined from Eqn 1 and 2 respectively. Using these values the small strain Youngs modulus can then be ascertained (Eqn 3).
Eqn 1
Eqn 2
Eqn 3
Fig. 2: S-wave signal showing singularity detection using a Python script Results The results for the experimental testing programme are shown below in Fig. 3. The trends observed in small strain Youngs modulus, static Youngs modulus and compressive strength are broadly in line with what is expected for each of the three concrete mixes. The 70% GGBS mix is clearly defined as being the slowest to develop strength and stiffness at early ages, although it can be seen that this does gradually improve over time. The shape of the curves for strength and small-strain stiffness are also quite similar, giving confidence in the results produced using the wavelet singularity detection algorithm. The approach also appears more sensitive than the standard static modulus of elasticity, which suggests very little difference in the concrete properties, regardless of GGBS content.
Near Surface 2009 15th European Meeting of Environmental and Engineering Geophysics Dublin, Ireland, 7 - 9 September 2009
Fig. 3: Results of experimental testing programmes Conclusions Based on the presented results a number of conclusions can be drawn:
A wavelet-based approach is a valid tool for determining the shear wave velocities in concrete samples, and is capable of overcoming the issues associated with reflected waves. However care is needed in determining appropriate Lipschitz exponents. The small strain Youngs modulus and compressive strengths for early age concrete follow similar trends for all GGBS contents; the static modulus of elasticity appears less sensitive to variations in the GGBS. Ultrasonic sensors have the potential to be used in this application, with potentially significant savings to companies seeking to use secondary cementitious materials in concrete practice.
Acknowledgements The authors would like to acknowledge the financial support of Ecocem Ireland Ltd. and the Enterprise Ireland Innovation Partnership programme (grant no IP/2008/0540). The assistance of Mr. Eoin Dunne of Trinity College Dublin is also gratefully acknowledged. References Bonal, J., Donohue, S. & McNally, C., (2008) Examination of a novel wavelet-based approach for bender element testing. In: Cannon, E., West, R. & Fanning, P. eds. Bridge and Infrastructure Research in Ireland, BRI 2008, Galway, Ireland, pp.435442 . Mallat, S. and Hwang, W.L., (1992) Singularity detection and processing with wavelets. Information Theory, IEEE Transactions, Vol. 38 (2) pp.617643, 1992. McNally, C., Richardson, M.G., Evans, C., & Callanan, T., (2005) Determination of chloride diffusion coefficients for use with performance-based specifications. Proceedings of the 6th International Congress on Global Construction: Ultimate Concrete Opportunities; Application of Codes, Designs and Regulations Dundee, pp.321327.
Near Surface 2009 15th European Meeting of Environmental and Engineering Geophysics Dublin, Ireland, 7 - 9 September 2009
Days Exposure
Days 0 28 56 84 112 140 175 196 224 252 280 308 336 364 392 420 448 476 504
L (%) P1 0 0 0.006 0.012 0.025 0.027 0.029 0.035 0.043 0.045 0.051 0.06 0.066 0.073 0.081 0.101 0.105 0.119 0.141
L (%)P2 0.000 0.000 0.003 0.007 0.018 0.031 0.029 0.034 0.046 0.053 0.062 0.077 0.090 0.109 0.128 0.154 0.182 0.222 0.268
L (%)P3 0 0.01 0.004 0.01 0.019 0.024 0.026 0.031 0.041 0.046 0.045 0.053 0.058 0.069 0.077 0.093 0.102 0.119 0.138
L (%)P4 0 0.01 0.01 0.016 0.027 0.029 0.031 0.037 0.046 0.047 0.05 0.058 0.062 0.072 0.078 0.092 0.097 0.114 0.131
Days Exposure
Days 0 28 56 84 112 140 175 196 224 252 280 308 336 364 392 420 448 476 504
L (%)P1 0 0.004 0.006 0.012 0.019 0.015 0.013 0.015 0.017 0.018 0.017 0.018 0.018 0.018 0.021 0.021 0.022 0.026 0.027
L (%)P2 0.000 0.006 0.007 0.012 0.018 0.018 0.016 0.017 0.017 0.020 0.022 0.022 0.024 0.024 0.026 0.028 0.030 0.030 0.032
L (%)P3 0 0.004 0.008 0.009 0.013 0.016 0.015 0.016 0.017 0.017 0.020 0.021 0.022 0.022 0.026 0.025 0.027 0.030 0.030
L (%)P4 0 0.006 0.005 0.011 0.012 0.015 0.019 0.018 0.020 0.021 0.024 0.026 0.026 0.027 0.03 0.031 0.032 0.033 0.035
Days Exposure
Days 0 28 56 84 112 140 175 196 224 252 280 308 336 364 392 420 448 476 504
L (%)P1 0 0.005 0.004 0.008 0.010 0.013 0.017 0.02 0.019 0.022 0.024 0.026 0.026 0.026 0.029 0.028 0.034 0.033 0.033
L (%)P2 0.000 0.002 0.004 0.010 0.010 0.015 0.012 0.014 0.014 0.017 0.018 0.021 0.020 0.021 0.024 0.026 0.030 0.030 0.031
L (%)P3 0 0.008 0.002 0.012 0.014 0.016 0.015 0.017 0.017 0.020 0.022 0.023 0.023 0.024 0.027 0.027 0.029 0.031 0.035
L (%)P4 0 0.005 0.004 0.012 0.010 0.014 0.014 0.016 0.017 0.019 0.02 0.025 0.023 0.024 0.029 0.027 0.03 0.032 0.036
CEM I 100%
0.8 0.7 0.6 Expansion % 0.5 0.4 0.3 0.2 0.1 0 0 56 112 168 224 280 336 392 448 504 Prism 1 Prism 2 Prism 3 Prism 4
Days Exposure
Days 0 28 56 84 119 140 168 196 224 252 280 308 336 364 392 420 448
L (%)P1 0 0.013 0.025 0.029 0.037 0.048 0.059 0.072 0.088 0.109 0.123 0.153 0.185 0.249 0.309 0.407 0.546
L (%)P2 0.000 0.003 0.015 0.024 0.031 0.043 0.055 0.068 0.084 0.104 0.122 0.153 0.184 0.239 0.297 0.377 0.502
L (%)P3 0 0.008 0.022 0.031 0.041 0.056 0.071 0.086 0.107 0.134 0.162 0.209 0.258 0.341 0.445 0.571 0.725
L (%)P4 0 0.014 0.026 0.034 0.046 0.058 0.072 0.089 0.109 0.132 0.155 0.196 0.236 0.315 0.404 0.52 0.664
Days Exposure
Days 0 28 56 84 119 140 168 196 224 252 280 308 336 364 392 420 448
L (%)P1 0 0.003 0.006 0.015 0.013 0.018 0.016 0.019 0.02 0.021 0.023 0.023 0.024 0.028 0.03 0.031 0.033
L (%)P2 0.000 0.008 0.012 0.016 0.017 0.021 0.024 0.026 0.028 0.033 0.034 0.034 0.039 0.041 0.042 0.046 0.049
L (%)P3 0 0 0.006 0.009 0.014 0.017 0.018 0.021 0.023 0.025 0.026 0.028 0.029 0.034 0.034 0.035 0.037
L (%)P4 0 0.002 0.008 0.013 0.021 0.024 0.026 0.028 0.031 0.034 0.036 0.038 0.039 0.042 0.044 0.046 0.047
SRPC 100%
0.08 0.07 0.06 Expansion % 0.05 0.04 0.03 0.02 0.01 0 0 56 112 168 224 280 336 392 448 Prism 1 Prism 2 Prism 3 Prism 4
Days Exposure
Days 0 28 56 84 112 140 168 196 224 252 280 308 336 364 392 420
L (%)P1 0 0 0 0.003 0.008 0.012 0.015 0.022 0.024 0.024 0.032 0.041 0.048 0.051 0.064 0.071
L (%)P2 0 0 0.003 0.007 0.012 0.016 0.02 0.024 0.029 0.031 0.036 0.044 0.053 0.059 0.068 0.075
L (%)P3 0 0 0.002 0.004 0.01 0.014 0.018 0.023 0.027 0.031 0.036 0.044 0.051 0.058 0.067 0.073
L (%)P4 0 0.001 0.006 0.009 0.012 0.014 0.019 0.023 0.026 0.028 0.028 0.037 0.042 0.045 0.054 0.06
CEM II 100%
0.18 0.16 0.14 Expansion (%) 0.12 0.1 0.08 0.06 0.04 0.02 0 0 5 10 15 Time ^ 0.5(Days) 20 25
Time ^ 0.5(Days)
CEM I 100%
0.4 0.35 0.3 Expansion (%) 0.25 0.2 0.15 0.1 0.05 0 0 5 10 15 Time ^ 0.5(Days) 20 25
SRPC 100%
0.08 0.07 0.06 Expansion (%) 0.05 0.04 0.03 0.02 0.01 0 0 5 10 15 Time ^ 0.5(Days) 20 25
MA53
MA16
y = -0.0283x + 6.1883
10 Time (mins)
15
20
6 5.9 Ln(Pressure)
MB15
Ln(Pressure)
MB16
y = -0.013x + 6.200
10 Time (mins)
15
20
6 5.9 Ln(Pressure)
MC15
Ln(Pressure)
MC16
5.8 5.7 5.6 5.5 5.4 5.3 0 5 10 Time (mins) 15 20 y = -0.0555x + 6.2162
y = -0.018x + 6.221
10 Time (mins)
15
20
MD15
Ln(Pressure)
MD16
y = -0.015x + 6.224
10 Time (mins)
15
20
6.05 6 5.95 5.9 5.85 5.8 5.75 y = -0.0345x + 6.1994 5.7 5.65 0 5 10 Time (mins) 15 20
ME15
ME16
Ln(Pressure)
Ln(Pressure)
y = -0.014x + 6.235
10 Time (mins)
15
20
SR15
Ln(Pressure)
SR16
y = -0.082x + 6.1198
y = -0.006x + 6.223
10 Time (mins)
15
20
10 Time (mins)
15
20
MA53
MA16
y = 1.78E-08x - 9.05E-09
2.5
3 Time ^0.5
3.5
MB15
MB16
y = 1.30E-08x + 1.40E-08
MC15
Water (m3)
MC16
1.00E-07 8.00E-08 6.00E-08 4.00E-08 2.00E-08 0.00E+00 2 2.5 3 Time ^0.5 3.5 4 y = 3.12E-08x - 2.85E-09
y = 1.96E-08x + 3.96E-08
2.5
3 Time ^0.5
3.5
MD15
1.40E-07 1.20E-07 Water (m3) 1.00E-07 8.00E-08 6.00E-08 4.00E-08 2.00E-08 0.00E+00 2
MD16
2.50E-07 2.00E-07 1.50E-07 1.00E-07 5.00E-08 0.00E+00 2 2.5 3 Time ^0.5 3.5 4 y = 7.57E-08x + 1.96E-08
y = 2.68E-08x + 1.26E-08
2.5
3 Time ^0.5
3.5
ME15
ME16
SR15
Water (m3)
SR16
y = 6.44E-08x + 2.96E-08
2.5
3 Time ^0.5
3.5
2.5
3 Time ^0.5
3.5
Water sorptivity
6E-08 5E-08 Water (m3) Water (m3) 4E-08 3E-08 2E-08 1E-08 0 2 2.5 3 Time ^0.5 3.5 4 y = 1.56E-08x - 6.25E-09 2.00E-08 0.00E+00 2 2.5 3 Time ^0.5 3.5 4
MA53
MA16
y = 2.27E-08x + 1.59E-08
MB15
Water (m3)
y = 5.07E-09x + 6.55E-09
MB16
y = 4.04E-09x + 1.76E-08
2.5
3 Time ^0.5
3.5
2.5
3 Time ^0.5
3.5
MC15
Water (m3)
MC16
8.00E-08 6.00E-08 4.00E-08 y = 2.59E-08x - 4.23E-09 2.00E-08 0.00E+00 2 2.5 3 Time ^0.5 3.5 4
y = 1.59E-08x - 3.35E-08
2.5
3 Time ^0.5
3.5
MD15
Water (m3)
MD16
1.00E-07 5.00E-08 0.00E+00 2 2.5 3 Time ^0.5 8E-08 7E-08 6E-08 3.5 4 y = 3.97E-08x + 2.36E-08
ME15
ME16
Water (m3)
5E-08 4E-08 3E-08 2E-08 1E-08 0 2 2.5 3 Time ^0.5 2.50E-07 2.00E-07 3.5 4 y = 2.44E-08x - 2.35E-08
y = 2.66E-08x + 5.75E-09
2.5
3 Time ^0.5
3.5
SR15
SR16
Water (m3)
1.50E-07 1.00E-07 5.00E-08 0.00E+00 2 2.5 3 Time ^0.5 3.5 4 y = 5.35E-08x + 2.76E-08
2.5
3 Time ^0.5
3.5
1 2449 2437 2445 2438 2451 2451 2461 2464 2462 2511 2487 2500 3550 3527 3582 3611
2 2448 2436 2444 2438 2451 2451 2468 2473 2472 2517 2493 2502 3548 3526 3593 3623
3 2449 2435 2443 2439 2451 2451 2446 2449 2448 2498 2475 2486 3550 3528 3569 3588
4 2453 2435 2445 2439 2455 2452 2442 2446 2444 2497 2473 2485 3544 3528 3570 3582
8 2436 2446 2453 2453 2357 2344 2406 2415 3550 3527 3395 3422
12
16
2446
2447
2446
2446
2454
2454
2454
2454
2238
2147
2071
1999
WEEK 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
2 17 33 17 33 33 50 250 283 317 545 564 527 -12 -104 243 313
4 100 17 33 50 100 67 -183 -167 -150 182 200 218 -58 -81 -23 -162
12
16
20
50 0 67 83
50
67
50
50
100
100
100
100
-1650 -1817 -3583 -5100 -6367 -7567 -1018 -1055 -2818 -4200 -5509 -6709 12 -93 -2052 -2017 23 -81 -4000 -3942 23 -81 -5461 -5426 12 -70 -6678 -6701 35 -70 -7896 -7849
MASS (g) CUBE No. 3 4 5 6 7 8 9 10 11 12 13 14 PRISM No. 1 2 3 4 MASS (g/m2) CUBE No. 3 4 5 6 7 8 9 10 11 12 13 14 PRISM No. 1 2 3 4
WEEK 0 2423 2434 2421 2424 2423 2436 2445 2429 2442 2485 2485 2456 3485 3468 3494 3498
1 2426 2436 2422 2426 2426 2437 2441 2425 2440 2484 2485 2452 3486 3470 3483 3496
2 2426 2436 2422 2427 2424 2438 2438 2426 2434 2488 2487 2455 3487 3470 3488 3504
3 2426 2437 2423 2426 2425 2436 2422 2405 2417 2467 2472 2435 3484 3468 3466 3480
4 2424 2436 2422 2425 2425 2437 2411 2394 2406 2458 2463 2427 3485 3470 3448 3463
12
16
MIX 20
MB 24
2422 2425 2425 2437 2273 2281 2361 2325 3485 3470 3281 3303
2423
2423
2423
2424
2438
2438
2438
2438
2175
2099
2016
1956
WEEK 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
12
16
20
17 17 33 17
33
33
33
50
33
33
33
33
-2600 -2683 -4450 -5717 -7100 -8100 -2255 -2382 -3964 -5364 -6800 -7945 0 23 -2470 -2261 12 23 -4128 -3965 23 23 -5472 -5264 12 35 -6852 -6574 23 58 -7884 -7571
MASS (g) CUBE No. 3 4 5 6 7 8 9 10 11 12 13 14 PRISM No. 1 2 3 4 MASS (g/m2) CUBE No. 3 4 5 6 7 8 9 10 11 12 13 14 PRISM No. 1 2 3 4
WEEK 0 2437 2434 2439 2421 2439 2438 2407 2435 2447 2478 2494 2472 3449 3472 3515 3475
1 2438 2437 2441 2424 2440 2440 2404 2433 2443 2478 2498 2475 3451 3477 3512 3476
2 2439 2436 2441 2425 2441 2440 2410 2439 2451 2485 2504 2481 3452 3476 3520 3486
3 2438 2436 2441 2423 2441 2439 2394 2427 2438 2471 2491 2470 3452 3475 3507 3471
4 2437 2435 2440 2423 2440 2439 2387 2420 2424 2461 2482 2457 3450 3476 3494 3452
12
16
MIX MC 20 24
2441 2442 2442 2440 2442 2424 2441 2439 2440 2440 2438 2441 2311 2311 2205 2122 2051 1986 2379 2353 2272 2198 2135 2078 3450 3475 3370 3334 3452 3476 3228 3193 3451 3475 3107 3064 3450 3475 2998 2957 MIX 3453 3478 2912 2868 MC 24
WEEK 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
12
16
20
33 50 33 17
50
50
17
50
33
33
50
-2067 -2267 -4033 -5417 -6600 -7683 -2091 -2164 -3636 -4982 -6127 -7164 12 35 -1681 -1635 35 46 -3328 -3270 23 35 -4730 -4765 12 35 -5994 -6006 46 70 -6991 -7038
MASS (g) CUBE No. 3 4 5 6 7 8 9 10 11 12 13 14 PRISM No. 1 2 3 4 MASS (g/m2) CUBE No. 3 4 5 6 7 8 9 10 11 12 13 14 Prism No. 1 2 3 4
WEEK 0 2519 2516 2516 2527 2515 2500 2450 2446 2446 2502 2496 2481 3527 3499 3512 3503
1 2522 2518 2519 2529 2517 2504 2435 2433 2432 2496 2496 2480 3528 3504 3497 3484
2 2522 2518 2518 2529 2517 2502 2436 2434 2432 2497 2496 2481 3530 3505 3497 3489
3 2521 2518 2518 2529 2516 2503 2419 2419 2414 2482 2481 2468 3530 3503 3477 3461
4 2520 2517 2518 2528 2517 2502 2409 2411 2404 2475 2474 2457 3528 3502 3464 3444
12
16
MIX MD 20 24
2518 2519 2519 2519 2518 2529 2517 2503 2504 2503 2503 2502 2298 2297 2191 2104 2020 1957 2377 2364 2270 2199 2124 2063 3529 3503 3325 3297 3531 3503 3174 3136 3530 3504 3061 3024 3532 3504 2950 2916 MIX 3531 3504 2867 2838 MD 24
WEEK 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
12
16
20
33 33 33 50
50
50
50
33
67
50
50
33
-2467 -2483 -4250 -5700 -7100 -8150 -2164 -2127 -3836 -5127 -6491 -7600 23 46 -2168 -2388 46 46 -3919 -4255 35 58 -5229 -5554 58 58 -6516 -6806 46 58 -7478 -7710
MASS (g) CUBE No. 3 4 5 6 7 8 9 10 11 12 13 14 PRISM No. 1 2 3 4 MASS (g/m2) CUBE No. 3 4 5 6 7 8 9 10 11 12 13 14 PRISM No. 1 2 3 4
WEEK 0 2447 2437 2423 2427 2411 2446 2432 2420 2433 2471 2486 2470 3532 3471 3494 3469
1 2450 2437 2424 2427 2412 2449 2432 2422 2435 2483 2493 2480 3536 3476 3468 3489
2 2448 2440 2424 2427 2413 2448 2437 2425 2440 2486 2498 2484 3537 3476 3492 3478
3 2448 2438 2425 2428 2413 2448 2416 2407 2418 2460 2481 2464 3536 3475 3473 3451
4 2448 2438 2425 2427 2413 2448 2408 2396 2409 2458 2472 2455 3536 3473 3460 3440
12
16
MIX ME 20 24
2425 2426 2425 2425 2426 2428 2413 2448 2449 2448 2449 2449 2303 2313 2210 2127 2042 1982 2386 2372 2284 2215 2144 2080 3536 3474 3331 3329 3536 3475 3198 3199 3535 3475 3084 3076 3535 3476 2974 2952 MIX 3535 3476 2896 2870 ME 24
WEEK 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
12
16
20
33 17 33 33
50
33
33
50
50
33
50
50
-1950 -2000 -3717 -5100 -6517 -7517 -1818 -1782 -3382 -4636 -5927 -7091 46 35 -1890 -1623 46 46 -3432 -3130 35 46 -4754 -4557 35 58 -6029 -5994 35 58 -6933 -6945
MASS (g) CUBE No. 3 4 5 6 7 8 9 10 11 12 13 14 PRISM No. 1 2 3 4 MASS (g/m2) CUBE No. 3 4 5 6 7 8 9 10 11 12 13 14 PRISM No. 1 2 3 4
WEEK 0 2481 2459 2470 2478 2482 2458 2461 2478 2464 2520 2509 2522 3543 3550 3542 3555
1 2482 2461 2471 2480 2483 2461 2451 2472 2452 2526 2514 2530 3546 3551 3541 3556
2 2482 2462 2472 2479 2483 2460 2448 2469 2450 2524 2512 2529 3546 3549 3537 3552
3 2482 2460 2471 2480 2483 2460 2431 2447 2434 2508 2498 2512 3546 3550 3512 3524
4 2482 2460 2471 2480 2483 2460 2423 2437 2425 2500 2490 2507 3545 3550 3504 3514
12
16
MIX SR 20 24
2461 2472 2472 2472 2472 2471 2484 2461 2462 2461 2462 2460 2337 2326 2190 2105 2028 1962 2415 2421 2309 2232 2159 2092 3545 3550 3395 3395 3546 3552 3211 3205 3546 3553 3075 3077 3547 3553 2965 2968 MIX 3548 3553 2875 2881 SR 24
WEEK 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
8 33 33 33 50
12
16
20
33
33
33
17
67
50
67
33
-2350 -2300 -4567 -5983 -7267 -8367 -1709 -1836 -3873 -5273 -6600 -7818 23 0 -1704 -1855 35 23 -3838 -4058 35 35 -5414 -5542 46 35 -6690 -6806 58 35 -7733 -7814
Week MA 0 1 2 3 4 8 12 16 20 24 0 261 380 29 11 1601 3586 5047 6314 7505 MB 0 51 25 335 513 2442 4127 5454 6832 7875
Mass Loss (g/m2) MC 0 6 105 107 285 1984 3567 4974 6182 7219 MD 0 153 137 408 567 2300 4065 5402 6728 7735 ME 0 66 139 198 337 1844 3415 4762 6117 7121 SR 0 15 55 353 483 1959 4084 5553 6841 7933
MIX
WEEK 0 1 2 3 4 8 12 16 20 24 0 91 218 255 182 -1127 -2982 -3818 -4527 -5091 0 91 255 291 273 -836 -2800 -3582 -4473 -4927
WEEK 0 1 2 3 4 8 12 16 2469 2484 2498 2516 2512 2390 2294 2258 2494 2511 2526 2542 2548 2409 2310 2272
MIX WEEK 0 1 2 3 4 8 12 16 20 0 273 527 855 782 -1436 -3182 -3836 -4527 0 309 582 873 982 -1545 -3345 -4036 -4636
WEEK 0 1 2 3 4 8 12 16 2459 2482 2496 2516 2529 2394 2299 2255 2484 2477 2517 2534 2549 2426 2329 2291
MC 24 2231 2257 MC
WEEK 0 1 2 3 4 8 12 16 20 24 0 418 673 1036 1273 -1182 -2909 -3709 -4273 -4145 0 -127 600 909 1182 -1055 -2818 -3509 -4036 -4127
WEEK 0 1 2 3 4 8 12 16 2553 2563 2575 2580 2581 2471 2378 2331 2544 2557 2569 2580 2578 2453 2366 2325
MD 24 2266 2267 MD
WEEK 0 1 2 3 4 8 12 16 20 24 0 182 400 491 509 -1491 -3182 -4036 -4836 -5218 0 236 455 655 618 -1655 -3236 -3982 -4691 -5036
WEEK 0 1 2 3 4 8 12 16 2478 2498 2512 2525 2534 2424 2326 2281 2483 2503 2513 2531 2541 2424 2338 2288
MIX WEEK 0 1 2 3 4 8 12 16 20 0 364 618 855 1018 -982 -2764 -3582 -4000 0 364 545 873 1055 -1073 -2636 -3545 -4055
WEEK 0 1 2 3 4 8 12 16 2510 2526 2540 2549 2540 2435 2337 2291 2529 2549 2556 2557 2552 2459 2351 2303
SR 24 2220 2229 SR
MASS (g/m2) WEEK CUBE No. 0 1 2 3 4 8 12 16 0 291 545 709 545 -1364 -3145 -3982 0 364 491 509 418 -1273 -3236 -4109
Week MA 0 1 2 3 4 8 12 16 20 24 0 61 158 183 153 982 2891 3700 4500 5009 MB 0 194 370 577 589 1491 3264 3936 4582 4727
Mass Loss (g/m2) MC 0 97 425 649 820 1118 2864 3609 4155 4136 MD 0 140 286 377 383 1573 3209 4009 4764 5127 ME 0 243 389 577 692 1027 2700 3564 4027 3991 SR 0 219 346 407 323 1318 3191 4045 4882 5364
Compressive Strength: Mix MA Days 0 28 56 168 Water 57 57 58 60 Acid 57 50 35 21 % Loss 0% 12% 40% 65%
Cube strength (MPa) 80 70 60 50 40 30 20 10 0 0
28
56
84 Days
112
140
168
Compressive Strength: Mix MB Days 0 28 56 168 Water 61 65 65 70 Acid 61 53 35 19 % Loss 0% 18% 46% 74%
Cube strength (MPa) 80 70 60 50 40 30 20 10 0 0
28
56
84 Days
112
140
168
Days 0 28 56 168
Water 58 58 63 65
Acid 58 51 37 22
80 70 60 50 40 30 20 10 0 0
28
56
84 Days
112
140
168
Days 0 28 56 168
Water 65 71 75 77
Acid 65 56 43 19
80 70 60 50 40 30 20 10 0 0 28
MD - 100% CEM I
56
84 Days
112
140
168
Days 0 28 56 168
Water 54 57 59 69
Acid 54 49 34 18
80 70 60 50 40 30 20 10 0 0
28
56
84 Days
112
140
168
Days 0 28 56 168
Water 60 58 68 75
Acid 60 59 41 21
80 70 60 50 40 30 20 10 0 0
SR - 100% SRPC
28
56
84 Days
112
140
168
Ref Bar (mm) 5.981 5.989 6.618 6.851 6.189 6.192 6.436 6.424 6.229 6.249 6.596 6.621 6.433 6.439 Ref Bar (mm) 5.975 5.997 6.802 6.882 6.191 6.192 6.258 6.47 6.243 6.252 6.61 6.621 6.438 6.438
Reading (mm) 2.621 2.625 3.345 3.476 2.807 2.823 3.138 3.047 2.887 2.861 3.213 3.23 3.046 3.047 Reading (mm) 2.884 2.874 3.685 3.725 3.031 3.039 3.22 3.328 3.056 3.082 3.427 3.424 3.311 3.222
L (%) 0.036 -0.004 -0.007 -0.002 0.026 -0.005 0.009 -0.010 -0.008 -0.011 -0.009 -0.011 L (%) 0.002 -0.014 -0.015 -0.012 0.034 -0.008 -0.026 -0.019 -0.024 -0.030 -0.002 -0.037
6.608 Ref Bar (mm) 5.956 5.994 6.664 6.869 6.19 6.192 6.234 6.448 6.232 6.251 6.605 6.62 6.437 6.439 Ref Bar (mm) 5.985 6 6.848 6.883 6.192 6.192 6.396 6.484 6.245 6.254 6.624 6.618 6.438 6.431
3.347 Reading (mm) 2.75 2.761 3.422 3.631 2.942 2.937 3.014 3.209 2.99 3.005 3.352 3.357 3.176 3.178 Reading (mm) 2.851 2.822 3.625 3.665 2.981 2.976 3.188 3.246 3.013 3.019 3.354 3.372 3.177 3.171
-0.016 L (%) -0.004 -0.002 -0.006 -0.009 0.005 -0.002 -0.004 -0.005 -0.008 -0.012 -0.011 -0.011 L (%) -0.018 -0.016 -0.013 -0.015 -0.012 -0.024 -0.022 -0.023 -0.037 -0.027 -0.033 -0.033
Ref Bar (mm) 6.135 6.132 6.034 6.038 6.214 6.244 6.05 6.058 6.25 6.248 6.612 6.62 6.434 6.434 Ref Bar (mm) 6.133 6.133 6.035 6.039 6.233 6.247 6.058 6.056 6.251 6.253 6.614 6.619 6.437 6.438
Reading (mm) 3.182 3.172 3.07 3.064 3.224 3.277 3.075 3.075 3.276 3.273 3.635 3.635 3.451 3.451 Reading (mm) 2.949 2.951 2.833 2.83 3.07 3.037 2.829 2.831 3.015 3.016 3.385 3.379 3.196 3.194
L (%) -0.002 -0.006 -0.012 -0.003 -0.006 -0.009 -0.006 -0.006 -0.007 -0.010 -0.009 -0.009 L (%) -0.007 -0.010 0.008 -0.010 -0.018 -0.016 -0.021 -0.021 -0.018 -0.022 -0.023 -0.024
Ref Bar (mm) 6.134 6.133 6.035 6.039 6.212 6.247 6.057 6.059 6.25 6.252 6.612 6.619 6.435 6.436 Ref Bar (mm) 6.132 6.132 6.033 6.039 6.241 6.247 6.054 6.059 6.244 6.251 6.616 6.62 6.433 6.437
Reading (mm) 3.085 3.076 2.885 2.893 3.157 3.142 2.895 2.833 3.027 3.045 3.386 3.381 3.203 3.202 Reading (mm) 3.017 2.881 2.721 2.724 2.959 2.93 2.74 2.73 2.917 2.918 3.386 3.29 3.088 3.091
L (%) -0.037 -0.036 0.001 -0.019 -0.042 -0.068 -0.066 -0.060 -0.068 -0.072 -0.070 -0.071 L (%) -0.024 -0.026 -0.012 -0.026 -0.025 -0.031 -0.030 -0.033 0.008 -0.032 -0.038 -0.038
Ref Bar (mm) 6.147 6.15 6.463 6.468 6.242 6.249 6.054 6.056 6.249 6.253 6.61 6.612 6.444 6.446 Ref Bar (mm) 6.149 6.15 6.466 6.464 6.248 6.25 6.056 6.059 6.252 6.255 6.611 6.612 6.443 6.446
Reading (mm) 3.134 3.1 3.374 3.396 3.141 3.145 2.951 2.953 3.145 3.148 3.502 3.495 3.32 3.323 Reading (mm) 3.042 3.013 3.235 3.226 2.995 3.01 2.8 2.803 2.986 2.982 3.338 3.331 3.151 3.153
L (%) -0.016 -0.009 -0.020 -0.022 -0.021 -0.021 -0.022 -0.022 -0.023 -0.027 -0.030 -0.029 L (%) -0.038 -0.040 -0.046 -0.041 -0.048 -0.048 -0.052 -0.054 -0.054 -0.058 -0.062 -0.062
Ref Bar (mm) 6.152 6.15 6.466 6.466 6.246 6.25 6.055 6.057 6.25 6.255 6.611 6.612 6.443 6.446 Ref Bar (mm) 6.151 6.15 6.464 6.464 6.246 6.25 6.054 6.056 6.253 6.257 6.611 6.611 6.444 6.447
Reading (mm) 2.98 2.966 3.298 3.274 3.219 3.108 2.98 2.884 3.114 3.07 3.432 3.427 2.253 3.254 Reading (mm) 3.132 3.162 3.402 3.399 3.172 3.163 2.984 2.969 3.169 3.162 3.514 3.517 3.334 3.334
L (%) 0.006 -0.003 0.063 0.017 0.044 0.004 0.019 0.000 0.002 0.000 -0.402 -0.003 L (%) -0.017 -0.018 -0.022 -0.027 -0.020 -0.027 -0.026 -0.030 -0.031 -0.030 -0.036 -0.038
Ref Bar (mm) 5.962 5.819 6.464 6.444 6.148 6.153 6.049 6.053 6.253 6.254 6.614 6.616 6.44 6.443 Ref Bar (mm) 5.964 5.716 6.467 6.46 6.153 6.153 6.052 6.054 6.253 6.254 6.615 6.616 6.443 6.443
Reading (mm) 2.717 2.559 3.215 3.187 2.891 2.883 2.794 2.794 2.989 2.985 3.348 3.347 3.174 3.174 Reading (mm) 2.662 2.574 3.209 3.259 2.988 2.981 2.882 2.87 3.075 3.062 3.405 3.402 3.222 3.213
L (%) 0.004 0.001 0.001 -0.004 0.002 0.000 -0.002 -0.004 -0.002 -0.004 -0.002 -0.004 L (%) 0.018 0.040 0.055 0.052 0.053 0.047 0.050 0.044 0.037 0.035 0.032 0.029
Ref Bar (mm) 5.964 5.825 6.466 6.434 6.152 6.153 6.051 6.055 6.253 6.254 6.615 6.616 6.441 6.442 Ref Bar (mm) 5.797 5.732 6.466 6.466 6.151 6.153 6.054 6.055 6.25 6.257 6.615 6.616 6.442 6.443
Reading (mm) 2.876 2.732 3.372 3.337 3.047 3.043 2.995 2.959 3.148 3.147 3.508 3.506 3.332 3.33 Reading (mm) 2.582 2.51 3.254 3.257 2.893 2.907 2.787 2.789 2.982 2.981 3.34 3.339 3.158 3.156
L (%) 0.000 -0.002 -0.005 -0.007 0.015 -0.001 -0.005 -0.006 -0.006 -0.007 -0.006 -0.008 L (%) 0.004 0.005 -0.014 -0.010 -0.018 -0.018 -0.018 -0.022 -0.021 -0.022 -0.025 -0.026
Ref Bar (mm) 5.683 5.955 6.458 6.464 6.146 6.15 6.054 6.054 6.255 6.229 6.609 6.61 6.442 6.445 Ref Bar (mm) 5.937 5.956 6.463 6.465 6.15 6.149 6.058 6.056 6.255 6.239 6.611 6.597 6.444 6.445
Reading (mm) 2.672 2.889 3.224 3.185 2.869 2.853 2.748 2.743 2.942 2.905 3.297 3.297 3.123 3.119 Reading (mm) 2.681 2.689 3.129 3.119 2.806 2.804 2.696 2.69 2.893 2.87 3.226 3.201 3.053 3.047
L (%) -0.067 -0.085 -0.084 -0.092 -0.096 -0.098 -0.099 -0.103 -0.098 -0.099 -0.101 -0.104 L (%) -0.027 -0.032 -0.031 -0.031 -0.038 -0.040 -0.038 -0.041 -0.047 -0.052 -0.050 -0.052
Ref Bar (mm) 5.927 5.955 6.461 6.46 6.15 6.152 6.055 6.058 6.255 6.236 6.61 6.608 6.444 6.445 Ref Bar (mm) 5.95 5.957 6.462 6.461 6.151 6.154 6.059 6.058 6.255 6.245 6.61 6.6 6.444 6.446
Reading (mm) 2.592 2.631 3.131 3.083 2.817 2.758 2.711 2.678 2.875 2.841 3.222 3.213 3.045 3.044 Reading (mm) 2.715 2.72 3.162 3.156 2.827 2.825 2.734 2.718 2.88 2.923 3.246 3.228 3.067 3.061
L (%) 0.002 -0.017 0.001 -0.024 -0.004 -0.018 -0.018 -0.024 -0.021 -0.024 -0.026 -0.026 L (%) -0.025 -0.027 -0.035 -0.037 -0.035 -0.041 -0.055 -0.034 -0.051 -0.054 -0.056 -0.059