Degradation processes of reinforced concretes by combined sulfate–phosphate attack
Introduction
The alteration of concrete structural elements due to the action of external chemical degrading agents is one of the main causes for the reduction of the service life in modern constructions [1]. These dynamics are particularly aggressive in environments characterized by heavy soil and atmospheric pollution, leading to a progressive decalcification of the cementitious matrix and oxidation of the rebars [2], [3]. Among the various forms of concrete chemical degradation, the external sulfate attack has been widely studied over the last two decades. Several theories on the actual contribution of this alteration dynamic on the physical and mechanical decay of concrete structures have been formulated [4], [5], [6]. Furthermore, the leaching and decalcification processes due to the exposure of concrete to acidic solutions have been extensively investigated in several scientific studies [7], [8], [9], [10]. Such chemical alteration dynamics are extremely detrimental to the durability of cement-based materials, leading to the dissolution of cement hydration products – mainly portlandite and C–S–H – and consequent loss of concrete strength.
Yet, there are few literature references on phosphate attack in cementitious materials, despite the interactions between phosphate ions and calcium-saturated systems that are widely demonstrated [11], [12], [13]. Some authors refer to limited decalcification on hydrated cement pastes under the action of phosphoric acid [14]. Moreover, an experimental and modeling study on concrete samples in contact with phosphate-rich solutions has been carried out to assess potential phosphate ion–concrete interactions [15]. The results of this integrated study showed that, while the formation of apatite on the expense of portlandite is thermodynamically favored in such systems, this mineral phase is not found experimentally, most likely due to kinetic factors. Nevertheless, the interaction between phosphates and concrete has never been evaluated on systems far from equilibrium characterized by local pH and chemical concentration gradients, corresponding to actual situations in attacked concrete structures materials. Moreover, the interaction between phosphate ions and other chemical species (e.g. sulfates) in case of concrete degradation by several types of chemical pollutants has never been taken into consideration before, as well as the possible evolution of the altered cementitious systems under variable temperature regimes, typical for temperate climates.
As for the influence of pH in the evolution of a hypothetical phosphate-related alteration of concrete, several studies confirmed the strong correlation between this parameter and the type of calcium phosphate phase formed [16], [17], [18]. The most common and least soluble mineral, hydroxyapatite (Ca5(PO4)3(OH), commonly abbreviated HAP), is stable under neutral or basic conditions [17]. On the other hand, phases such as brushite (Ca(HPO4)·2(H2O), commonly abbreviated as DCPD) and octacalcium phosphate (Ca8(HPO4)2(PO4)4·5(H2O), commonly abbreviated as OCP) are generally crystallized from more acidic solutions [16], [18]. Both brushite and octacalcium phosphate are reported to be precursors of apatite at acidic and neutral pH, respectively [17]. The higher solubility and lower stability of the two phases with respect to apatite cause their progressive transformation into apatite, both by dissolution and re-precipitation processes and by epitaxial growth [17].
Apart from the alteration dynamics related to the sole calcium–phosphate ion interaction in concrete, it is necessary to take into account other mutual interactions between phosphate compounds, concrete constituents and other external chemical species. Regarding this topic, the possible synergic action of phosphate and sulfate-based pollutants toward the alteration of cement-based materials needs to be considered, due to the extensive literature devoted to the study of calcium phosphate and sulfate phases in solid solution [19], [20], [21], [22]. Gypsum and brushite both belong to the monoclinic system and have similar structures, despite the fact that they are not isomorphous (a = 5.182 Å, b = 15.180 Å, c = 6.239 Å, β = 116.43°, sp.gr. I2/a for brushite; a = 5.670 Å, b = 15.201 Å, c = 6.533 Å, β = 118.60°, sp.gr. I2/c for gypsum [21]). Although they have different sizes (the phosphate group is larger than the sulfate one), the substitution of sulfate and phosphate groups can occur within certain limits [20], and a solid solution between the two phases, which also comprehends the stoichiometric mineral ardealite (Ca2(SO4)(HPO4)·4(H2O), monoclinic system, Cc space group), can form. Furthermore, it has been demonstrated [22] that the relative proportions between the two end members of the solid solution have relevant influences on the dehydration behavior of the resulting compounds, being significantly different from the ones of the pure end members within certain composition intervals. Supposedly, such behavior influences the evolution of the secondary phases in case of alteration of the concrete elements exposed to significant seasonal temperature variations [23], as it happens in temperate climates.
Furthermore, the interaction between phosphate ions, nitrogen-based pollutants and magnesium ions deriving both from concrete cementitious matrix and aggregate fraction, needs to be considered, both due to the extensive scientific literature regarding the detrimental action of ammonium salts on concrete [24], [25], [26] and the occurrence of ammonium phosphate phases like struvite under certain chemical–physical conditions. Struvite ((NH4)MgPO4·6(H2O)) is a hydrated phase consisting of magnesium, ammonium and phosphate in equal molar concentrations. Ammonium can be substituted to a certain extent by alkaline ions, in particular potassium [27]. Struvite precipitation is controlled by pH, degree of supersaturation, temperature and the presence of other ions in solution such as calcium, and can occur when the concentrations of magnesium, ammonium and phosphate ions exceed the solubility product for struvite. Among the various factors, pH is crucial for the formation of this phase, being sparingly soluble in neutral and alkaline, but readily soluble in acid conditions [28].
Finally, the possible interaction between phosphate ions and concrete rebars due to hypothetic phosphate attack of reinforced concrete shall be considered, taking into account the reported occurrence of iron phosphate phases like vivianite (Fe3(PO4)2·8(H2O)) and mitridatite (Ca2Fe3(PO4)3O2·3(H2O)) under specific chemical–physical conditions [29].
The cementitious materials studied in this work were sampled from one of the dismissed structures of the Ex-Agrimont industrial facility, building C3, built in the 1950s near the coastline of Porto Marghera (Venice area, north-eastern Italy) (Fig. 1).
The Porto Marghera industrial area suffered from heavy pollution during the 20th century, due to uncontrolled emissions from chemical industries and oil refineries [30]. All production plants of the area were responsible for severe forms of atmosphere, water and soil pollution.
As for atmospheric emissions, a detailed monitoring during the years of maximum activity of the area is not available. Yet, an intensive monitoring campaign performed by the Regional Association for Environmental Protection (ARPA) during the years 1998–2004 [31], when most of the factories had already closed, still showed particularly severe atmospheric pollution from gaseous sulfur- and nitrogen-based pollutants, like hydrogen sulfide, sulfur dioxide and trioxide and ammonia.
Furthermore, during the years of maximum activity in the area, unauthorized dumps for the storage of toxic waste were frequently excavated, and by-product slurries and non-depurated process waters were dumped and dispersed with no control. All these activities contributed to the progressive pollution of the groundwater and soils of Porto Marghera with abnormal amounts of chlorinated and aromatic solvents, aromatic hydrocarbons, phenols, ammonia, aromatic amines, sulfates, phosphates, lead, cadmium, zinc, mercury and arsenic. The Ex-Agrimont area was particularly affected by soil and water pollution due to the leakage of process water from the industrial plant, which produced phosphate- and ammonia-based fertilizers by chemical extraction from insoluble minerals from 1927 until its closure in 1997.
The main sources of phosphate minerals are naturally-occurring sedimentary deposits of phosphorites, sedimentary rocks containing significant amounts (about 30%) of cryptocrystalline apatite. After the mining, phosphates are separated from accessory minerals (mainly clay minerals), and the concentrated material is then chemically processed to obtain phosphate-based fertilizers, mainly through what is known as the wet process method [32]. In this process, the phosphate raw materials react with sulfuric acid to produce phosphoric acid. More precisely, the reaction combines calcium from the phosphate with sulfate to form phosphoric acid and calcium sulfate, mainly gypsum (called phosphogypsum), according to the following simplified reaction:
During the reaction, gypsum crystals precipitate and are separated from the acid by filtration and washing. Wet processed phosphoric acid normally contains 26 wt.% to 30 wt.% of P2O5, and depending on the types of fertilizer to be produced, it is usually concentrated from 40 wt.% to 55 wt.% of P2O5 through a series of vacuum evaporators, and subsequently, after reaction with ammonia, granulated to produce diammonium phosphate (DAP) and monoammonium phosphate (MAP). Granular triple superphosphate (GTSP), a non-nitrogen fertilizer, is produced by adding phosphoric acid to ground phosphate rock.
The production of fertilizers requires and produces a large volume of process water, a dilute mixture of phosphoric, sulfuric, and fluosilic acids with very low pH (about 1 to 2) [32]. Furthermore, it contains several other ions found in phosphate rock, as well as ammonia from the solid fertilizer-manufacturing process [32]. In order to reduce its environmental impact, process water commonly undergoes a series of chemical treatments, by sequential additions of calcium hydroxide and acids, to obtain a final solution with a pH of about 6.5 from which all solids and ammonia have been removed [32].
The studied structure (Fig. 1a) is a 20 m tall building, constituted by 22 couples of load-bearing reinforced concrete semi-arches. The structure is enclosed by a reinforced concrete wall up to 2 m high. Above this height the semi-arches are connected by prefabricated non-structural reinforced concrete beams which constituted the backing for the original fiber-cement roofing. A binder classifiable as CEM I-type cement (European Standard EN 197-1) was used for the manufacturing of all concrete elements, employing dosages between 300 and 350 kg/m3 and with w/c ratios varying from 0.6 to 0.7. No chemical admixtures were added to the fresh mix. The aggregate fraction is characterized by a 20 mm maximum diameter and a granulometric distribution close to the theoretical Bolomey curve. The coarse aggregates consist mainly of carbonate rocks. Among the various carbonate species, sparry and micritic limestones and dolostones are dominant. The fine aggregate fraction consists of carbonate rock fragments of the same nature as the coarse aggregates and quartzite grains.
An intensive inspection survey highlighted the poor condition of the concrete structural frame of the building, despite the overall quality of the concrete and its good mechanical properties [33]. All structural elements are covered by superficial black crusts, which have caused dusting and superficial scaling of the cement matrix. The pervasive alteration caused, in some cases, complete consumption and detachment of concrete covers, triggering extensive oxidation on the underlying rebars. The weathering also affects the higher sections of the structure, indicating that atmospheric pollutants were responsible for the alteration of the concrete.
The degradation of reinforced concrete elements is further accentuated in the south-western section of the structure, near the platform originally used for the direct load of final products and by-products on trains. In this area, up until 2 m of height, the reinforced concrete of the arches and other structural elements is heavily scaled. The concrete cover is extensively detached showing pervasive oxidation of the rebars, and globular superficial concretions are present (Fig. 1b, c). Furthermore, the cementitious matrix is characterized by an anomalous white–yellowish color. It shows a relevant loss of cohesion, leading to pulverization in the worst cases.
Given the poor conditions of the C3 building and the heavy pollution affecting this area over the last few decades, the building has been thoroughly studied to verify the possible occurrence of a form of phosphate attack contributing to the decay of the concrete structural elements. More in detail, the objectives of the study are the following:
- −
To identify any possible interaction between saturating solutions rich in phosphate pollutants and reinforced concrete constituents (hardened cement paste, aggregate and steel rebars);
- −
To recognize the sequences of secondary phase crystallization;
- −
To verify the possible influence of pH gradients in the type and stability of secondary calcium phosphates crystallized within the altered materials;
- −
To verify the possible formation of gypsum–brushite solid solution phases within the altered materials due to the simultaneous occurrence of phosphate- and sulfate-based degrading agents, also considering a possible crystallization heterogeneity due to the presence of chemical gradients;
- −
To investigate the possible occurrence of secondary ammonium phosphates due to the interaction between phosphate- and nitrogen-based pollutants and concrete;
- −
To study the interaction between phosphate pollutants and steel rebars;
- −
To improve the knowledge on the dehydration behavior of the gypsum–brushite solid solution under temperature regimes compatible with the ones acting on concrete structures in temperate climates, also taking into account for possible crystal size and kinetic factors on the thermal stability of the compounds;
- −
To elaborate a qualitative model for the evolution of the degradation, verifying it through simple thermodynamic calculations.
Section snippets
Materials
Due to the considerable lack of coherence of the materials, samples were mechanically separated from the substrate with a metal chisel and superficial concretions were collected with a scalpel. Nine samples of altered concrete were taken from different structural elements of the building, below 2 m of height, where the superficial scaling is more pervasive (Fig. 2): 3 from adjacent arches (B2A1, B2A2, B2A3), 3 from the outer walls of the structure (B2M1, B2M2, B2M3) and 3 from the railway
Characterization of the altered materials
The diffraction patterns of the finer fraction of the altered concrete samples revealed a complex mineral assemblage. Besides the occurrence of aggregate-related phases (quartz, calcite, dolomite, plagioclase, biotite and chlorite), peaks of secondary sulfate and phosphate phases were detected (Fig. 4). More specifically, apatite, associated with gypsum and brushite in solid solution, octacalcium phosphate, a mineral phase with crystal structure similar to ardealite and struvite were observed
Discussion
On the basis of the analytical results, a qualitative model, based on four main phases, is hereby proposed to describe the alteration processes occurring in the studied concrete.
The results of the previous analyses performed on unaltered portions of the concrete structure [33] suggest that the cementitious matrix of the degraded conglomerate was constituted mainly by C –S–H and CH, and the alkaline environment of the conglomerate ensured the permanence of a passivation layer around the rebars.
Conclusions
This study demonstrated that the action of pollutants in aggressive environments caused a severe degradation in the concrete of the C3 building of the Ex-Agrimont area. A significant phosphate–concrete interaction was observed and verified, also thermodynamically, highlighting a form of chemical degradation not widely considered yet. Phosphate and sulfate attack resulted to be synergic, with the formation of solid solution phases. Moreover, heterogeneous sequences of crystallization and diffuse
Acknowledgments
This research project was financially supported by Mapei S.p.A. in the framework of its research agreement with the University of Padova. Mapei S.p.A. is also acknowledged for providing materials and facilities during several experimental sessions of the project. Prof. Gilberto Artioli and Prof. Claudio Mazzoli of the Department of Geosciences, University of Padova, are deeply acknowledged for their theoretical and analytical contributions on the research topic and for their useful reviews of
References (52)
- et al.
Durability of concrete — Degradation phenomena involving detrimental chemical reactions
Cem. Concr. Res.
(2008) - et al.
Sulfate attack of concrete building foundations induced by sewage waters
Cem. Concr. Res.
(2002) - et al.
The effects of atmospheric multipollutants on modern concrete
Atmos. Environ.
(2003) A state-of-the art review on delayed ettringite attack on concrete
Cem. Concr. Compos.
(2003)The confused world of sulfate attack on concrete
Cem. Concr. Res.
(2004)- et al.
Acidic attack of cement based materials a review. Part 1. Principle of acidic attack
Constr. Build. Mater.
(2001) - et al.
Acidic attack of cement-based materials—A review Part 2. Factors of rate of acidic attack and protective measures
Constr. Build. Mater.
(2002) Acidic attack of cement based materials — A review Part 3: research and test methods
Constr. Build. Mater.
(2004)- et al.
Durability of concrete exposed to leaching and external sulphate attacks
Cem. Concr. Res.
(2009) - et al.
Observations upon calcium oxalate crystallization kinetics of simulated urine
J. Urol.
(1977)
The system CaHPO4·2H2O–CaSO4·2H2O: crystallization from calcium phosphate solutions in the presence of SO42−
J. Cryst. Growth
Physicochemical study of the alteration surface of concrete exposed to ammonium salts
Cem. Concr. Res.
Deterioration of high-performance concrete subjected to attack by the combination of ammonium nitrate solution and flexure stress
Cem. Concr. Res.
Struvite formation, control and recovery
Water Res.
Scanning electron microscopy imaging of hydraulic cement microstructure
Cem. Concr. Compos.
The thermal dehydration of synthetic gypsum
Thermochim. Acta
Calcium phosphate solid phase induction by dioleoylphosphatidate liposomes
J. Colloid Interface Sci.
Crystal growth of calcium phosphates. Epitaxial considerations
J. Cryst. Growth
Conversion of amorphous calcium phosphate into hydroxyapatite
J. Cryst. Growth
Sulfate Attack on Concrete
Calcium Phosphates in Biological and Industrial Systems
Phosphates: Geochemical, Geobiological, and Materials Importance
Calcium Phosphate: Structure, Synthesis, Properties, and Applications
Concrete Technology
Laboratory investigation on effect of phosphate ions on concrete
The role of brushite and octacalcium phosphate in apatite formation
Crit. Rev. Oral Biol. Med.
Cited by (22)
Investigation of diffusion behavior, stability of testing ions, interconversion feasibility and microstructures of iodide and chloride in concrete
2024, Journal of Building EngineeringInvestigation of natural diffusion behavior in concrete using iodide replacing chloride ions: The impact of mineral admixtures types and dosages
2024, Journal of Materials Research and TechnologyDecay products of historical cements from the Palace of Knossos, Crete, Greece
2023, Journal of Cultural HeritagePhosphogypsum circular economy considerations: A critical review from more than 65 storage sites worldwide
2023, Journal of Cleaner ProductionThe sponge effect of phosphogypsum-based cemented paste backfill in the atmospheric carbon capture: Roles of fluorides, phosphates, and alkalinity
2023, Separation and Purification TechnologyDegradation resistance of different cementitious materials to phosphoric acid attack at early stage
2022, Cement and Concrete ResearchCitation Excerpt :Phosphates (H2PO4−, HPO42− or PO43−), are one of the main components in silage effluent discharged by agricultural practices with a resultant acidic condition (pH ≈ 3.5–5) [20], posing severe challenges for cementitious materials used in sewer networks [8,21,22]. According to related research [23], fertilizer production could produce a dilute mixture of phosphoric, sulphuric and even fluosilic acids with pH level as low as 1 to 2. Previous studies have shown that the exposure of OPC concretes to phosphoric acid (pH = 2, 3 or 4) or phosphate-rich acidic environments resulted in significant degradation [1,8,24].