Using bio-based polymers for curing cement-based materials
J. Zlopasa
1*, E.A.B. Koenders
1,2, S.J. Picken
1(1) Delft University of Technology, Delft, The Netherlands
(2) COPPE-UFRJ, Rio de Janeiro, Brazil
Abstract: Curing is the process of controlling the rate and extent of moisture loss from the surface of
cement based materials. It is the final stage in the production of cement-based materials and it is the essential part for achieving continuous hydration of cement, while avoiding cracking due to drying shrinkage. Continuous cement hydration also guarantees a strong bond between aggregate, fewer voids, and depercoliation of capillary pores. Thus, a properly cured cement-based material is prepared for a long service life. Using environmentally friendly, water based bio-polymers could help to achieve more durable cement-based materials, and, therefore preventing a premature end of service life of building materials. Rapid Chloride Migration tests and Environmental Scanning Microscope are employed to investigate the functional properties, e.g. transport property, and microstructure properties, respectively. Mortar samples were cured in air and applied by water-based curing compound, made of sodium alginate. We observed strong beneficial effects of applying sodium alginate as a curing compound in terms of microstructure and hydration development. Based on these results, a less porous microstructure and an improved durable cement-based material was achieved that was prepared for longer service life.
Keywords: curing, bio-based polymer, transport property, microstructure
1 Introduction
Cement-based materials are composite materials that consist of a continuous phase of cement paste and a discontinuous phase of aggregates. The aggregates phase is viewed as an inert filler that contributes to volume stability and higher durability, and are bonded by hydration products of cement clinker in cement paste [1]. Cement clinker is composed of four mineral phases Ca3SiO5 Ca2SiO4, Ca3Al2O6 and Ca4Al2Fe2O10. Two major phases (around 80wt.% for most
cements) of cement clinker are tri- and di-calcium silicate. Tri-calcium silicate reacts faster and is responsible for strength development during the first weeks, whereas di-calcium silicate reacts slower and contributes to the long-term strength of cement-based materials. In general, the reactions of both silicate phases are presented as follows:
(CaO)b(SiO2) + (b-Ca/Si+y)(H2O) (CaO)(Ca/Si)(SiO2)×y(H2O) + (b-Ca/Si)Ca(OH)2 (1)
where b=2 or 3 (di- or tri calcium silicate, respectively), Ca/Si=1.7-1.8 [2], and y=4. Since water is an essential component of hydration, and cement hydration will only proceed in a water-filled space, sustaining the hydration process requires that inter-particle voids remain filled with water [3]. Powers et al. showed that hydration of cement is greatly slowed down when the relative humidity in cement capillary pores goes below 80% [4], which most likely happens at the surfaces of e.g. concrete elements. Curing is a name given to procedures that aim to avoid this. They are used for promoting the hydration of cement, and consists of a control of moisture movement (and temperature) from and into the cement-based materials [1] through the exposed surfaces. By preventing the loss of water from cement-based materials, continuous hydration could be achieved and drying shrinkage be avoided, leading to a minimum of surface cracks, a stronger bond between aggregates, fewer voids, and lower connectivity of pores. Such a microstructure is denser and can prevent slower penetration of aggressive fluids that can be
*
harmful, e.g. for corrosion of steel reinforcement [5]. Therefore, a properly cured cement-based material is better prepared for a long service life. Generally, curing can be performed by adding water or by hindering water escape from the cement-based material’s surface. Continuing adding water by means of water ponding, water spraying, and/or by the use of wet burlaps usually gives the best end results. However, this technique requires workers on site that keep the concrete moist, which can be costly. In addition, this method can be especially costly in places where there is a scarcity of water. As said, the second way of curing cement-based materials is by preventing it from water evaporation. This can be accomplished by covering the surface with a plastic sheets or by spraying it with a curing compound (polymer solutions/emulsions) that creates a film that hinders water evaporation. Curing compounds can be water based or solvent based [6]. In general, curing compounds based on organic-solvents show better performance when compared with water-based ones [7]. As a drawback, there can be an environmental impact, especially when using it in low ventilated environments. In this study, we propose the use of a water-soluble bio-based polymer, sodium alginate, as potential curing compound for cement-based materials. Alginates are linear water-soluble polysaccharides comprising of (1— 4) linked units of α-ᴅ-mannuronate (M) and β-L-guluronate
(G) at different proportions and different distributions within the chains. Functional properties are strongly correlated with composition (M/G ratio) and with the sequence of the uronic acids. They are present in brown algae and can also be found in metabolic products of bacteria, e.g. pseudomonas and azotobacter [8-11]. Alginates are commonly used as food additives, gelling agents, wound dressings and for drug delivery[12,13]. The gelling property of this polymer gives a unique opportunity to be applied in various fields. Alginates gel either by lowering the pH below the pKa value of the uronic residue or in the presence of polyvalent ions [14,15].
Polyvalent ions act as bridges between different G units of chains, as shown in Figure 1.
Figure 1. Structure of alginic acid (left) and schematic representation of crosslinking of alginate by
polyvalent ions (right), from [15]
Following Equation 1, a large amount of Ca2+ is produced by cement hydration, which is freely
available (usually for carbonation). Since sodium alginate reacts rapidly with Ca2+ by
cross-linking, forming non-water soluble calcium alginate, we have investigated the possibility to use sodium alginate as external curing compound. For this purpose, we compared air-cured and alginate-cured mortar samples. We have examined functional and microstructure properties of mortar samples with and without application of sodium alginate solution, by the use of Environmental Scanning Microscope (ESEM) and Rapid Chloride Migration (RCM) Tests.
2 Experimental plan
2.1 Materials and mortar preparation
Sodium alginate with an average molecular weight of 150,000 and M/G ratio of 1.56 was obtained from Sigma Aldrich Co. Water solution of 3wt.% sodium alginate was prepared using demineralised water under vigorous stirring.
Mortar samples were prepared using commercial Portland cement Cem I 52.5R (produced by ENCI Heidelberg Cement group) with w/c of 0,5 and aggregate particles that follow Fuller distribution. Table 1 gives an overview of the chemical composition of the cement used. The mineral content of the cement calculated, using the Bogue method [16], is 68,3% Ca3SiO5, 6,2%
Ca2SiO4, 6,3% Ca3Al2O6 and 10,0% Ca4Al2Fe2O10. Table 2 gives an exact mix design of the
mortar.
Table 1 Chemical composition (X-ray Flouresence Analysis) of Cem I 52,5R Chemical composition Weight fraction
- % CaO 64,9 SiO2 20,1 Al2O3 4,5 Fe2O3 3,3 K2O 0,46 SO3 3,3 MgO 1,4 P2O5 0.4 TiO2 0.2 Loss on ignition (900°C) 1,1
Samples were cast in cylindrical moulds with 100mm diameter which was followed by compaction using a vibrational table. After compaction the sodium alginate solution was poured on top of the three mortar samples while other three sample were air-cured (surface of mortar was left uncovered). Furthermore, samples were placed in a lab with constant environmental conditions of 50% RH and 20°C for 28 days, after which they were de-moulded analysed.
Table 2 Mortar mix design Cem I 52,5 R [kg/m3] 506.96
Water [kg/m3] 253.48 Aggregates [kg/m3] 1520.88
2.2 Characterization methods
In order to evaluate differences in functional properties of the mortar samples, the chloride migration coefficient is determined by use of the Rapid Chloride Migration, RCM, test. The method is described in NT Build 492 [17]. The principle behind this method is in the application of external electrical current axially across the mortar specimen which forces the chloride ions
to migrate into the specimen at higher rate. Since curing of cement-based materials mostly affects the near surface area, we have modified the standard RCM test slightly by not removing the first 10-20 mm of the mortar surface from the samples. A sketch of the RCM test is shown in Figure 2. After the samples were de-moulded they were preconditioned by placing them in vacuum for 3 hours and then, with the vacuum pump still running, a saturated Ca(OH)2 solution
was introduced and samples were completely submerged in it. After 1 hour the vacuum pump was turned off and air was allowed to enter the container and the specimens were further keep in the saturated Ca(OH)2 solution for 18 hours. Once the preconditioning was completed the
samples were placed in the RCM setup, which consists of a rubber sleeve in which the samples are placed in. Electrodes are immersed in the anolyte (0,3 M NaOH) and catholyte (2 M NaCl) solutions and connected to the power supply unit. Initial current through the sample at 30 V is recorded and the voltage was adjusted according to the standard which also states the duration of the test. The voltage is adjusted in order for the chlorides to penetrate through about half of the sample. Initial and final temperatures of both anolyte and chatolyte were also recorded. After the described test duration specimens were removed from the RCM setup and split axially into two pieces. On the freshly split surface 0,1 M AgNO3 was sprayed and after few minutes
white silver chloride started to precipitate (see Figure 3) and change colour. The precipitated silver chloride represents the chloride penetration depth, from which the migration coefficient is calculated using Equation 2:
Dnssm=(0.0239×(273+T) ×L)/((U-2) ×t) ×(xd-0.0238×√((273+T) ×L×xd)/(U-2)) (2)
where T is the temperature, L thickness of the sample, t test duration, U applied voltage and xd is
the chloride penetration depth.
Figure 2 Schematic representation of RCM test
Difference in developed microstructure of mortar samples were investigated using environmental scanning microscope (ESEM) in backscattered electron mode (BSE), in conjunction with energy-dispersive spectroscopy for qualitative analysis (Ca, Si, C, Al, Mg, Fe, Na, K). Polished sections for ESEM and energy-dispersive X-ray spectroscopy (EDS) is done as follows. The sampling was done 30 mm from the surface, cement hydration was stopped by submerging the samples in liquid nitrogen and subsequently moved into a freeze-dryer (sublimation) with temperature of -24°C and under vacuum for 21 day. The dried samples were
impregnated under a vacuum with epoxy resin. After the resin hardened, the surface of specimens were grounded and polished.
3 Results and discussion
From the penetration depths, measured after silver chloride precipitation, the chloride non-steady state migration coefficient, Equation 2, was calculated. It can be seen in Figure 3 (right) that the migration coefficient decreased substantially when applying Sodium Alginate as an external curing compound. Chloride migration coefficient for alginate-cured mortar was 1,6 times lower than for the air-cure mortar sample.
Figure 3 Chloride penetration (left) and chloride migration coefficient obtained from RCM test results (right)
It has also to be noted that, as mentioned, in this study we used Cem I, which has a low amount of supplementary cementitious materials that go through slower secondary hydration reactions. This allows more time for the water, that is intended for hydration, to evaporate. And since in marine environments, e.g. The Netherlands, cements with supplementary cementitious materials are frequently used.
Figure 4 ESEM-BSE micrograph of near surface area of alginate-cured mortar sample
In Figure 4 and 5 visual observation of near surface area by means of electron microscopy was done. Microstructure of the mortar sample that was covered with alginate exhibited a clearly denser microstructure when compared to the microstructure that was air-cured only.
Also, the microstructure that was cured with alginate as an external curing compound exhibited fewer cracks which is a usual phenomena when there is excessive drying of young cement-based material at the surface, e.g. drying or plastic shrinkage.
Figure 5 ESEM-BSE micrograph of near surface area of air-cured mortar sample
Figure 6 shows a near surface area analysed by element mapping, where the elements of interest are calcium and carbon. Calcium is the most abundant polyvalent cation. We analysed calcium leaching due to carbonation or due to the reaction with alginate. In contrast to what was expected, due to the calcium reaction with alginate, a lower calcium content in near surface area was observed for the air-cured sample. The calcium is leached due to bleed water that comes on the surface of the sample from the interior, subsequently Ca2+ reacts with CO2 and forms CaCO3,
which leads to a concentration gradient and more calcium will be drawn to surface. In [18] sodium alginate was mixed with gypsum in order to reduce calcium leaching. Authors observed chelation of Mg2+ and Ca2+, which depends on the initial concentration of alginate mixed in with
gypsum. Since the samples were impregnated with epoxy resin, analysis of carbon content will indicate the porous and cracked regions in the sample. Air-cured mortar sample showed a higher intensity for carbon, which is in agreement with chloride migration results and observed microstructure, e.g. more porous regions.
Figure 6 ESEM-BSE image with EDS mapping of Ca (yellow) and C (red) of near surface area of A) alginate cured mortar sample and B) air cured mortar sample.
4 Conclusions
In this work the influence of alginate as external curing compound for cement-based materials was studied and compared with air-cured samples. Objective was to see whether a water-soluble bio-based polymer, sodium alginate, can be used as an external curing compound for
cement-based materials. Rapid cross-linking was observed once the sodium alginate solution arrives at the surface of mortar sample. This indicated that sodium alginate was cross-linking to calcium alginate by means of rapid ion exchange with excess of calcium produced by cement hydration. After 28 days RCM test were performed and significant differences were observed between sodium alginate and air-cured mortar samples. Sodium alginate cured samples had a 1,6 times lower migration coefficient, which gives an indication that the mortar samples cured with sodium alginate could have a potentially longer service life than air-cured. This was confirmed by the use of ESEM, where a more porous and cracked microstructure was observed. Element mapping of near surface are showed more porous regions for air-cured samples compared to sodium alginate cured mortar samples. Future work will focus on the use of sodium alginate as an external curing compound for Cem III/B, which consist of a high mass replacement of cement by supplementary cementitious materials and is therefore more sensitive to curing conditions. Also, an improvement of alginate barrier properties will be investigated by addition of natural clays.
Acknowledgements
The authors like to acknowledge the Dutch National Science foundation STW. The research conducted within this project is financed by STW as a part of the IS2C program (www.is2c.nl), number 10962. Arjan Thijsen assistance in ESEM measurement is acknowledged
5 References
[1]Neville AM (1995), Properties of concrete 4th edn. Wiley, New York
[2] Richardson IG (2000), The nature of the hydration products in hardened cement pastes, Cem. Concr. Compos. 22(2) 97–113.
[3]Hover KC (2011), The influence of water on the performance of concrete, Constr. Build. Mater 25(7):3003-3013.
[4]Powers TC, Brownyard TL (1946-1947), Studies of the physical properties of hardened Portland cement paste, J. Am. Concrete I. 18:1-9.
[5]Meeks KW and Carino NJ (1999), Curing of high-performance concrete: Report of the state-of the art. NISTIR 6298, US Department of Commerce.
[6]Wang J, Dhir RK, Levitt M (1994), Membrane curing of concrete: moisture loss, Cement Concrete Res. 24(8):1463:1474
[7]Al-Gahtani AS (2010), Effect of curing methods on the properties of plain and blended cement concretes, Constr. Build. Mater. 24:308-314.
[8]Grasdalen H, Larsen B, Smidsrød (1981), 13C-n.m.r. studies of monomeric composition and sequence in alginate, Carbohyd. Res. 89(2):179-191.
[9]Draget KI, Skjåk-Bræk G, Smidsrød O (1997), Alginate based new materials, Int. J. Biol. Macromol. 21(1):47-55.
[10]Linker JB, Jones RS (1966), A new polysaccharide resembling alginic acid isolated from Pseudomonads, J. Bio Chem 241:3845-3851
[11]Gorin PAJ, Spencer JFT (1966), Exocellular alginic acid from Azotobacter vinelandii, Can. J. Chem 44:993-998.
[12]Laurienzo P (2010), Marine polysaccharides in pharmaceutical application: an overview, Marine drugs 8(9):2435-2465.
[13]Matthew IR, Browne RM, Frame JW, Millar BG (1995), Subperiosteal behaviour of alginate and cellulose wound dressing materials, Biomaterials 16(4):275-278.
[14]Russo R, Malinconico M, Santagata G (2007), Effect of cross-linking with calcium ions on the physical properties of alginate films, Biomacromolecules 8:3193-3197.
[15]Narayanan RP, Melman G, Letourneau NJ, Mendelson NL, Melman A (2012), Photodegradable iron (III) cross-linked alginate gels, Biomacromolecules 13(8):2465-2471.
[16]Taylor HFW (1997), Cement Chemistry 2nd edn.Thomas Telford publishing, London
[17]NT Build 492 (1999), Concrete, mortar and cement-based repair materials: chloride migration coefficient from non-steady-state migration experiments, Nordtest Finland.
[18]Belcarz A, Janczarek M, Kolacz K, Urbanik-Sypniewska T, Ginalska G, (2013), Do Ca2+-chelating polysaccharides reduce calcium ion release from gypsum-based materials?, Cent. Eur. J. Biol. 8(8):735-746.
