Valorization of phosphogypsum in cement-based materials: limits and potential in eco-efficient construction

1 Valorization of phosphogypsum in cement-based materials: Limits and potential in eco-efficient construction Authors: Bianca Rafaela da Silva Calderón Morales a, Antonio García-Martínez a, Paloma Pineda b, Rafael García-Tenório c a Department of Building Construction I, E. T. S. Arquitectura, Universidad de Sevilla, Avda Reina Mercedes No. 2, 41012 Seville, Spain b Department of Building Structures and Geotechnical Engineering, Research Group Transhumancias HUM-965, E. T. S. Arquitectura, Universidad de Sevilla, Avda Reina Mercedes No. 2, 41012 Seville, Spain c Departamento de Física Aplicada, E.T.S. Arquitectura, Universidad de Sevilla, Reina Mercedes s/n 41012 Seville, Spain Abstract Phosphogypsum (PG) is a type of synthetic gypsum generated during the production of phosphoric acid. Each ton of phosphoric acid generates 5 tons of phosphogypsum. This industrial process has caused significant environmental problems worldwide. After an extensive literature review, it was possible to verify that both sedimentary and igneous PG can be used as in materials building components. The use of PG up to 50% meets the limits required for index I, with a radioactivity equivalent to radio-226 and a concentration of radon-222. The data found on bricks (0,11-0,41 Bq m-2 h-1) and plate (0,16-0,41 Bq m-2 h-1) show a lower exhalation rate. It is also found that calcination contributes to the reduction of PG impurities and to the of mechanical strength increase. Mortars and concretes produced with PG-based cementitious systems cementitious systems based on PG achieve high strength (60 MPa, 70 MPa, 80 MPa) and meet the technical feasibility criteria of European standards EN 998-1, EN 998-2, EN 13813, Eurocode 1992 and the American standard ASTM C270. The setting time is influenced by the mineral phase of the PG. It is concluded that cementitious materials produced from phosphogypsum are technically viable and achieve a satisfactory performance. It is possible to advance the following lines: PG-based concrete with high performance, PG-based hydration cement, stabilized mortar, PG-based concrete with high workability and studies about chloride ingress, carbonation, sulfate attack and acid attack in PG-based concrete. Manuscript File [For Revision, Please upload clean version of Revised manuscript] Click here to view linked References 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 2 Keywords Phosphogypsum – Eco-efficient cement-based materials – Cementitious systems – Valorization - Green building materials - Eco-friendly cement Graphical abstract Abbreviations α- PG α-hemihydrate LS Limestone A-PG Anhydrite NG Natural Gypsum A-PG-II Anhydrite-II PC Portland cement A-PG-III Anhydrite-III PG Phosphogypsum 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 3 1 Introduction Sustainability is based on economic development, social equity and environmental protection. Different nations under various international agreements have committed themselves to the goal of preserving the natural resources, and the construction of a sustainable built environment [1]. Nowadays, energy efficiency and climate change are worldwide discussed [2]. In the last decades, the European Union (EU) has consistently implemented public policies on those issues. The goal is to reduce the energy consumption, the foreign energy dependence, the carbon dioxide emissions, and to foster actions aiming at reducing both the climate change, and the industrial wastes [3]. By following the public policies that guide both the energy consumption economy and the industrial processes eco-efficiency, the civil construction sector, especially the cement industry, has adhered to sustainable production processes. In this way, it seeks for methods of using industrial waste to produce cement and technologies that minimize CO2 emissions [4]. It is estimated that until 2050, there will be a 12-13% increase in cement production. The result will be an overall 4% CO2 emissions increase. However, with the use of mineral additives and industrial by-products, to be used in the clinker/cement ratio, the CO2 emission can be reduced by 37% [5]. Currently, the challenge is to find alternatives that allow for producing cement in a sustainable way. The use of alternative fuel and new technology byproducts have increased, while the consumption of conventional raw material for manufacturing cement has been reduced [1], [6]–[8]. Phosphate rock is an abundant mineral in nature which can be found predominantly in the form of calcium phosphate (apatite), and it can be used with or without industrial process [9], [10]. The physical-chemical characteristics (mineralogical and chemical composition, impurities, grain size, etc.) depend on the geographic location of the phosphate rock [11]. β- PG Β-hemihydrate PL Portland cement limestone B Binary PS Portland cement slag BA Bottom ash QA Quaternary CAC Calcium aluminate cement QI Quinary CRS Carbonate silica rock RM Red mud CSA Sulfoaluminate cement SF Silica fume Di-PG Dihydrate SS Steel Slag EAFS Electric arc furnace steel slags SSC Supersulphated cement FA Flay ash T Ternary FGD Flue-gas desulfurization gypsum GGBFS Ground granulated blast-furnace slag HCBPG Hydraulic cement-based on phosphogypsum 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 4 Thus, 85% of phosphate rock sources are of sedimentary origin, and only 15% are igneous [12]. Among the sixteen largest phosphate rock reserves in the world, Morocco (23%) and China (13%) have the largest reserves (Figure 1). In 2019, China (110Mt), Morocco (36Mt) and the United States (23Mt) were the three largest producers of phosphate rocks [13]. Figure 1: Mine production and reserves of phosphate rock in 2019. Adapted from US Geological Survey, 2020 [13] Through the chemical reaction (Equation 1) between phosphate rock (fluorapatite), sulfuric acid and water (wet chemical treatment), phosphoric acid is obtained and dihydrate phosphogypsum is generated [14]. For each 1 ton of phosphoric acid produced, 5 tons of phosphogypsum are generated [15], [16]. Approximately 85% of the generated waste is stored in areas near rivers or seas, in stacks containing millions of tons of waste, without purification treatments. Until now, only 15% of phosphogypsum is recycled, the remaining 85% are stored in areas near rivers or seas, in open air-piles, without any special treatment and causing great environmental impacts [17]. Phosphogypsum is classified as NORM waste (Naturally Occurring Radioactive Material), and its reuse can cause damage both to human beings and environment, in terms of radiation [18], [19]. Although phosphogypsum has NORM characteristics, over the years, research has been developed to make possible using it as building material component [20], [21] or as fertilizer [22]. 𝑷𝒉𝒐𝒔𝒑𝒉𝒂𝒕𝒆 𝒓𝒐𝒄𝒌 + 𝑺𝒖𝒍𝒇𝒖𝒓𝒊𝒄 𝒂𝒄𝒊𝒅 + 𝒘𝒂𝒕𝒆𝒓 → 𝑷𝒉𝒐𝒔𝒑𝒉𝒐𝒈𝒚𝒑𝒔𝒖𝒎 + 𝑷𝒉𝒐𝒔𝒑𝒉𝒐𝒓𝒊𝒄 𝒂𝒄𝒊𝒅 + 𝑯𝒚𝒅𝒓𝒐𝒈𝒆𝒏 𝒇𝒍𝒖𝒐𝒓𝒊𝒅𝒆 𝐶𝑎10(𝑃𝑂4)6𝐹2+10𝐻2𝑆𝑂4+20𝐻2𝑂 → 10𝐶𝑎𝑆𝑂4. 2𝐻2𝑂 + 6𝐻3𝑃𝑂4+2HF (1) Phosphogypsum is considered a synthetic gypsum [23], [24]. It can be used for the manufacture of coating mortars [25]. Besides, it can be also used as mineral additive in 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 5 Portland cement, and in composite cement with blast furnace slag [25], [26]. And, it is also used to regulate the cement hardening time [27]. Valorization of phosphogypsum is an active research field, and therefore, constant advances are carried out to use it as recycled material, with technologically improved levels of radioactivity [28]. García-Tenorio et al [29] assessed the environmental impact generated by NORM wastes, analysing the environmental radiological safety associated to the possible recovery. García-Díaz [30] studied the valorization and characterization of NORM industrial wastes to be used in building materials. Campos et al. [31] focused on measuring radon expiration in slabs and bricks manufactured with PG generated in Brazil. In the research carried out by Saadaoui et al. [32] was investigated the potential uses and problems caused by PG. The work by Zemni et al [33] provides a study on the generation of calcium silicate and sodium sulphate, through the chemical reactions between phosphogypsum and sodium silicate. The results showed that the solid produced during the reaction can sequester CO2. Rashad [24] assessed the effect of PG on properties such as workability, density, mechanical strength and durability within cementitious matrices. Ennaciri and Bettach [34] analysed the conversion of PG to calcite and lithium sulphate monohydrate for batteries, industrial and environmental processes. Huang et al. [35] investigated the possibility of producing Calcium belite-sulfoaluminate clinker from PG . This research is organized as follows: The first part offers a brief discussion on sustainable development and the influence of the cement industry in increasing global warming, as well as aspects of plaster generation and ways of recovery. The second part shows the methodology developed for construction of this paper, informing the types of databases used, the data classification procedure and the standards studied. The third part provides a description of the mineralogical phases of phosphogypsum. The fourth part focuses on the physical-chemical behavior of PG, in terms of on fineness and particle distribution, major elements, minor elements and radionuclides. The fifth part discusses the rheological behavior, analyzing the normal consistency of the paste, spread flow test in mortars and the slump test in concrete. Then, the sixth part shows the morphology of PG with a focus on microstructure and impurities. The seventh part presents studies on the setting time in cement, plaster and Portland cement systems. The next section discusses the types of PG- based cementitious systems, focusing in paste, mortar and concrete. The ninth part discusses the mechanical resistance of the mortars and concrete, produced with PG-based 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 6 cementitious systems. The tenth part compares the cement-based PG studies, then performance and technical feasibility are analyzed. The eleventh part discusses the durability aspects of cement-based PG, focusing on solubility, water resistance, shrinkage by drying and expandability. In the twelfth and thirteenth parts, other ways of enhancing phosphogypsum are shown, such as geotechnical materials. The study of sustainable building materials contributes to the promotion of a circular economy. In this context, this paper was developed, with the objective of investigating and evaluating the performance and technical feasibility of different cementitious systems based on phosphogypsum. Based on the data provided, it is expected to contribute to its enhancement, offering a basis for taking the PG out of the classification of industrial byproduct and transforming it into an input potentially capable of being used in cement-based without causing risks to human health. 2 Methodology This work provides a literature review from the most relevant advances on cement-based phosphogypsum. The data are analysed following these steps: (i) collection of the data, (ii) the classification of data according to the specific topic, (iii) the organization of data in the form of tables and graphs, (iv) the description of the topic, and (v) conclusions per item. 2.1 Data Selection The data were selected from a comprehensive literature review, focusing on research on the feasibility of formulating HCBPG (Hydraulic Cement Based on Phosphogypsum) and the application in building materials. The selected articles are obtained from scientific literature, including indexed journals related to cement, mortar, concrete and plaster. The following sources were reviewed: Science Direct, Scopus, Mendeley, Google Scholar, Springer Link, Taylor and Francis, International Journal of Innovative Science Engineering and Technology (IJISET), International Journal on Emerging Researches in Engineering Science and Technology, MATEC Web of Conferences, International Meeting of Laboratories and Matters Experiments, Construction and Listening Systems (RILEM), American Society for Testing and Materials (ASTM), and European Standard and American Standard. 2.2 Classification of the literature A review was performed based on the keywords phosphogypsum, mortar, cement-based, binder, concrete, durability, industrial by-product, raw material, supplementary cementitious materials, setting time, curing, filler, gypsum and synthetic gypsum. After that stage, the data were classified as follows: 1) Search; 2) Classification; 3) Creation of tables; 4) New 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 7 classification; 5) New tables; 6) Definitive tables. Table 1 shows the methodology used to classify the literature used in this work. Table 1 summarizes the classification of the literature. Table 1: Classification of the literature Stage Name Result obtained 1 Search Data search and grouping. Use of Mendeley and JabRef software to organize and create tables of the references found. The result was 192 publications related to phosphogypsum. 2 Classification Classification of the data in fourteen study subjects based on the titles 3 Creation of tables Creation of tables on concrete studies, binders, phosphogypsum and lime-based cements, pastes and mortars and sulphated cement. 4 New classification Review of the tables and changes in the classification of themes based on step 1. 5 New tables Creation of new tables based on cement systems, mortars, concrete, self-levelling mortars and chemical composition of phosphogypsum. 6 Definitive tables Definition of tables on the subjects concerning chemical composition, paste, mortar, concrete and geotechnical construction materials and building components. 2.3 Standards used To verify the potential ad usefulness of the reviewed data, European codes and American codes were taken as a basis for comparison. It was possible to compare the following properties: setting time of cementitious systems, mechanical strength and expandability. The codes and scientific works are provided in Table 2. Properties Used in Type of binder Specification or References used as parameter Ref. Consistency Paste Dihydrate Singh (2002) [25] Plaster Lin (2019) [36] HCBPG and Portland cement replacement by PG Neville and Brooks (2010) [37] Setting time Paste Plaster EN 13279-1 and ASTM C28 [38], [39] HCBPG EN 15743, EN 14216, EN 14647, ASTM C91, ASTM C1600 and ASTM C 595 [40]–[45] Portland cement EN 197-1 and ASTM C 150 [46], [47] Compressive strength Mortar HCBPG and Portland cement replacement by PG EN 998-1, EN 998-2, EN 13813, and ASTM C270 [48]–[51] Concrete HCBPG and Portland cement replacement by PG Eurocode 1992 and EN 206 [52], [53] Expansibility (Soundness) Paste HCBPG EN 15743 [42] Portland cement EN 197-1 [46] Shrinkage Paste Plaster, HCBPG and SAC/PC Hewlett and Liska (2019) Lamond and Pielert, EN 1367-4, EN 12620 [54]–[57] Spread flow Mortar HCBPG ASTM C270 EN 1015-2 and EN 13454-1 [51], [58], [59] 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 8 Table 2: Reviewed Standards and scientific works 3 Phosphogypsum: mineralogical phases description Gypsum is the mineralogical term used for calcium sulphate [63]. It is defined as a nonhydraulic and air binder [64], [65]. Natural calcium sulphate is found in three different mineral phases, which are distinguished by the degree of hydration: Dihydrate (CaSO4.2H2O) hemihydrate (CaSO4.0,5H2O) and anhydrite (CaSO4) [66]. However, when the dihydrate (Di-PG) mineral phase is heated (calcination), dehydration process occurs changing the crystalline structure. The first mineral phase after dehydration is known as hemihydrate (α-PG and β-PG). Anhydrite III (A-PG III) is the second mineral phase, anhydrite II (A-PG II) is the third mineral phase, and anhydrite I ( A-PG I) is the fourth mineral phase [67]. Figure 2 shows the dehydration process of calcium sulphate. Figure 2: Dehydration process of calcium system. Adapted from Kuntze [63] and Odler [67] Dihydrate mineral phase is gypsum in the raw state [68] and remains stable until the temperature of less than 45ºC [67]. The Hemihydrate phase can occur in two ways: αhemihydrate due to the autoclaving process or β-hemihydrate through calcination. It has metastable thermodynamic behaviour and both processes occur at temperatures between 45ºC-200ºC and is popularly known as Plaster of Paris [69], [70]. The process of mineral phase changes to arise the anhydrite mineral phase, consuming a large quantity of energy. Anhydrite-III (CaSO4.III-α and CaSO4.III-β) occurs through the calcination process of the gypsum hemihydrate between 110-220ºC and has metastable thermodynamic behaviour. Whereas, anhydrite-II (CaSO4.II) comes from the calcination of anhydrite III at 300ºC. It has rhombic crystalline structure and thermodynamic stability. Anhydrite I is the last mineralogical phase that were the dihydrate gypsum arrives. The Slump Concrete HCBPG and Portland cement replacement by PG EN 206 and Neville and Brooks (2010) [37], [53] Water-resistant paste Plaster and HCBPG Kovler (2001), Selim et al (2020) and Piasta and Zarzycki (2017) [60]–[62] Dihydrate (Di-PG) •(CaSO4.2H2O) •Raw material •stability < 45ºC •Stable Hemihydrate (α-PG and β-PG) •(CaSO4.0,5H2O) •Phase I •>45ºC-200ºC •Metastable • α and β Anhydrite III (A-PG III) •CaSO4 •Phase II •110ºC-220ºC •Metastable •Soluble • α and β Anhydrite II (A-PG II) •CaSO4 •Phase III •300ºC •Unsoluble • α and β Anhydrite I (A-PG I) •CaSO4 •Phase IV •1180ºC •unsoluble •High temperature anhydrite • α and β 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 9 process occurs through the calcination of anhydrite II with a temperature higher than 1180ºC [70]. It has a defined structure and exhibits thermodynamic stability. It can be concluded that the dihydrate or hemihydrate phase consume low amount of energy than anhydrite II, II and III phases, during the heat treatment. The use of PG in phases A- PG III, A-PG II e A-PG I, in spite of contributing to the mechanical strength of cementbased materials, has a high production cost and impacts on the environment, due to the high energy required for calcination. 4 Physical-chemical properties Based on the literature review, the fineness and particle distribution and chemical composition of the phosphogypsum were analysed. 4.1 Particle fineness and distribution The fineness of the binders is a crucial property for cementitious materials. It is determined through the specific surface area and is directly related to cement hydration and the development of mechanical strength. The analysis of the particle size distribution is also important, as it helps to determine the total surface area of the particles [37], [71]. Raw phosphogypsum (Di-PG) when extracted from the piles is like wet sand (Figure 3). However, in the dry condition at room temperature, it is a material like silty sand (fine sand). The grains are soft, with uniform granulometry and are considered fine (less than 250 μm). It presents excellent compaction that, consequently, influences the compressive strength [72], [73]. It is worth mentioning that the granulometric distribution and the morphology of the grains, depends on the type of phosphate rock, beneficiation form, distribution of impurities [74], [75] and the time of the attack of sulfuric acid during the production of phosphoric acid [72], [73].Furthermore, has an influence on the hardening time, water demand, fluidity, consistency, microstructure and resistance [76]. Di-PG particles can have a maximum size of 0.5mm (No. 40 sieve) to 1.0 mm (No. 20 sieve). The material passing through the 0.075 mm sieve (200 sieve) can range between 50-75% [72].The silt present in the Di-PG can be classified as soil A-4 in the soil classification system AASHTO20. Generally, the phosphogypsum hemihydrate particle is larger in size than Di- PG [72]. PG particles can be almost 10 times thinner than GGBFS particles [77]. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 16 radioactive material [113], such as, polonium (Po) and radium (Ra) being consequently a source of gaseous radon (Rn). These radionuclides can be leached into underground aquifers and surface water systems, being potentially incorporated in the food chain, and suspended particles can be inhaled by workers and population. Consequently, they can be an external and internal source of radiation [127]. During the production of phosphoric acid, most of U and Th are dissolved, however, most of Ra is deposited in PG. Therefore, the radionuclide 226Ra is the most worrying element in PG [113]. Heavy metals in phosphogypsum appear in different amounts and may contain rare earth elements (REE) [128]. The most common toxic heavy metals are: chromium (Cr), lead (Pb), cadmium (Cd), zinc (Zn), copper (Cu), arsenic (As) [129], manganese (Mn) [113]. The amount and type of minor elements in the composition of PG can range widely, and also depends on the origin of the phosphate rock [115]. PG of sedimentary origin may contain a high concentration of heavy metals and REE, Cd, As, Sb, V, Cr, Zn, Cu, Ni. PG of igneous origin has low concentrations of Cd, As and Hg. These heavy metals are potentially harmful to human health, and through water and absorption by plants, they can be consumed directly or indirectly by humans [128]. In Table 4, the concentration of heavy metals is provided. In the literature, PG of sedimentary origin is considered to have a moderate-high level of danger, whereas PG of igneous origin is considered to have a low level of danger [129]. Among the sixteen chemical elements with toxic characteristics, the PG of Morocco presented the largest amount (8) and the highest concentrations of As, Cd, Cr, Zn, Th, La, Sc and Y. The PG of Brazil presented high level of Ba concentration, while PG from USA (Idaho) showed high concentration of Pb. PG from Canada showed a high level of concentration of the elements Cu, Ni, Zr and U. It is noteworthy that data on the Po element were not found in the review. It is also observed that PG from Russia contains low levels of heavy metal concentration. Table 5 shows the concentration of the most common radionuclides in phosphogypsum and the geological origin. Most of the data on radionuclides present in phosphogypsum were of sedimentary geological origin. And it is worth mentioning that, although the PG used in China is of sedimentary origin, in general, the concentration of radionuclides is low in comparison to other data of the same geological origin and in comparison, to the PG data of igneous origin. Table 4: Comparison of heavy metals between Morocco, EUA, Canada, Brazil and Russia phosphogypsum. Adapted from [130], [131]. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 17 Table 5: Comparison of radionuclides between Australia, Brazil, China, Egypt, EUA, India, Indonesia, Spain and Sweden phosphogypsum. Adapted from [130], [131]. Use in PG rock origin Radionuclides (Bq/Kg) Geological origin Ref. 238U 226Ra 210Pb 210Po 230Th Sedimentary Igneous Australia Numerous 10 500 - - - x [115] Brazil Brazil 42 695 894 677 560 x [133] Brazil Brazil 49 744 1061 - 782 x [134] China Keiyan 15 85 82 82 - x [135] USA Central Florida 130 1140 1370 1030 113 x [115] Egypt Nile Valley rock - 100 - 445 - x [135] India Vadorado 60 510 490 420 - x [135] Indonesia PT Petrokimia Gresik 43 473 480 450 - x [135] Spain Morocco 140 620 - 82 280 x [16] Sweden Kola (Russia) - 15 - - - x [115] It is possible to verify the variety of heavy metals present in PG and the variation in concentration between PG of sedimentary origin and that of igneous origin. In general terms, the table shows that the Brazilian PG has a lower concentration of heavy metals, while the PG of Morocco presented the highest concentration of heavy metals. In relation to radionuclides, it appears that the PG generated in Russia has a lower concentration of 226Ra, while the PG generated in USA (Florida) contains the highest concentration of radionuclides. Although the PG of China is of sedimentary origin, the concentration levels of radionuclides are low compared to the other countries mentioned in this study. Geological Origin Heavy metals (ppm) REE (ppm) Sedimentary Igneous Ref. As Ba Cd Cr Cu Ni Pb Sr Zn Zr Th U Po La Sc Y x Morroco - 98 6 20 21 <1 6.2 709 8 6.2 4.0 8.3 - 86 4.7 144 x [132] Morroco 11 - 30 291 22 - 7 - 345 - - - - - - - x [129] USA (Florida) - 43 <2 5 3.4 5 10 750 6.4 54 1.5 4.5 - 39.8 <0,05 71 x [132] USA (Idaho) - 47 10.7 48 11.4 5 13 660 31 48 < 0.5 7.3 - 73,1 <0.05 125 x [132] Canada (Alberta) - 140 28 70 41.7 15 7 670 112 110 1 13.3 - - - - x [113] Brazil (Imbituba) - 836 0.8 7 6 6 3 12090 15 - - - - - - - x [20] Russia 1 - 0.1 23.3 30 - 3 - 19 - - - - - - - x [129] 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 18 5 Rheological behaviour Study of the rheological behaviour of cementitious materials in the fresh state is directly related to their efficiency [136]. Workability is a property that is associated with cement-based consistency [137], and therefore with the fluidity of concrete and mortars [138]. 5.1 Normal consistency of paste The normal consistency of paste is the term used to define a standardized mixture between the binder (cement, plaster or lime) and water. This properties has direct influence on the rheological behavior of cementitious systems.In addition, the normal consistency of paste, interferes with the workability of cement-based materials [139]. It determines the appropriate degree of plasticity of the paste [140] as well as the amount of water that is required. In addition, it is related to the initial and final setting times, and to the expansibility of the cement paste [37]. Figures 4-6, provide the relation between the compressive strength and the normal consistency of the plaster, HCBGP and Portland cement, in the mineral phases dihydrate (Di-PG), hemihydrate (α-PG and β -PG) and insoluble anhydrite (A-PG II and A-PG I). In this review, the values of normal consistency of phosphogypsum dihydrate, plaster, HCBGP and Portland cement with replacement PG, were compared to the data from the literature. It is worth mentioning that it was not possible to find standardized data. The normal consistency of natural dihydrate plaster is 60% consistent [25]; while the normal consistency of the α-PG plaster has a value of 30-45% and the β-PG plaster 65-85% [36]. The normal consistency of 26-33% [37] was used as reference in cementitiuous systems (HCBGP) and Portland cement with partial replacement PG. Figure 4 shows the percentage of the normal plaster consistency in relation to the compressive strength. The analysed works include the following phases: mineral dihydrate (Di-PG), hemihydrate (α-PG and β-PG), and insoluble anhydrite (A-PG II and A-PG I). In general, it is observed that there is a relation between the compressive strength, the mineral phase of the phosphogypsum and the normal consistency: if the value of normal consistency increases, the value of compressive strength decreases. It is noted that the consistency of Di-PG is directly related to the purity of the material, and the use of retardant does not cause interference. Normal Di-PG consistency can range from 66% to 80% % [25], [96]. When the normal consistency of natural calcium sulphate dihydrate is compared, it is observed that both types (selenite and phosphogypsum) have similarities in the values of compressive strength and normal consistency [25]. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 19 The consistency of α-PG without setting retardant is 58% consistent and its compressive strength is 15 MPa. However, the use of setting retarder reduces consistency and increases setting time. The use of 0.20% sodium succinate reduces the consistency to 35% and increases the strength to 25.58 MPa [141]. The consistency of β-PG without the use of setting retardants can vary between 60% to 66% [141]– [143]. Washed purified β-PG has a consistency of 64%, and the compressive strength is 13.5 MPa. A-PG I has a very low percentage of normal consistency (22.14%-23.53%) compared to other mineral phases. The compressive strength is high (35.1 MPa -37.2 MPa), regardless of the amount of setting retarder [144]. A-PG II calcined at 800ºC, presented a low consistency of 26.14% and resistance to compression of 56 MPa. Figure 4: Normal plaster consistency Figure 5 shows the relation between normal consistency of HCBPG and compressive strength. No results of normal HCBPG consistency were found in the literature for Di-PG, A-PG I and % A- PG III. The normal consistency of the α-PG + FA + lime system is 44%, and the compressive strength is 26.6MPa when the HCBPG curing temperature is 50ºC. The normal consistency of the α-PG + GGBFS + marble dust + chemical activator system was 41% and the compressive strength 66 66 70 70 80 80 58 35 38 40 60 60 62 66 64 22.14 23.53 23 26.14 10.68 8.9 10.6 9.6 5.9 5 15 28.58 23.7 20.6 13.2 13.3 9.8 10.5 13.5 36 37.2 35.1 56 0 5 10 15 20 25 30 35 40 45 50 55 60 0 10 20 30 40 50 60 70 80 90 100 Compressive Strength (MPa) Normal Consistency (%) Normal consistency (%) Natural gypsum Di-PG (60%), Singh 2002 Natural gypsum α-PG (30%), Lin 2019 Natural gypsum β-PG (65%), Lin 2019 Compressive Strength (MPa) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 20 was 32.2 MPa, due to the curing temperature of 50ºC [141]. Studies of the α-PG + FA + lime + PC system (15% + 60 + 5% + 20%) showed that the normal consistency of 22% for the compressive strength of 35MPa [145]. Normal consistency of the β-PG + GGBFS + PC system is 36% and the compressive strength is 35MPa [143].The β-PG + FA + PC system showed a 34% consistency and a compressive strength of 8.78MPa [146]. The β-PG + FA + 10% lime system has a normal consistency of 34% and a curing temperature of 50ºC, and a compressive strength of 22.41MPa [147]. Standard consistency of the β-PG + FA + lime + clay system was 37% and the compressive strength was 27.2MPa [148]. The normal consistency of the 50% A-PG II + 50% GGBFS system was 27.42% and the compressive strength is 35 MPa [149]. Figure 5: Normal HCBPG consistency 44 44 41 41 35 36 34 34 34 37 27.42 7.2 26.4 20.6 32.2 27.5 35 8.78 8.78 22.41 27.2 35 0 5 10 15 20 25 30 35 40 0 10 20 30 40 50 60 Compressive Strength (MPa) Normal Consistency (%) Nornal consistency (%) Consistency (26%), Neville and Brooks (2010) Compressive Strength (MPa) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 21 The Figure 6 shows that studies on normal consistency with partial replacement of Portland cement were performed only with Di-PG. The binary Di-PG + PC system showed a normal consistency between 26% -26.5% and the compressive strength of 30MPa-36.8MPa respectively [27], [150]. The Di-PG-PS binary system had a normal consistency of 23.7% and a compressive strength of 50MPa [150]. The normal consistency of both systems 40% Di-PG + 60% NG + PC and 60% Di-PG + 40% NG + PC is 26.3% and the compressive strength is 50 MPa, despite the percentage of natural plaster used and phosphogypsum, are different in both systems [150].The normal consistency for both systems 40% Di-PG + 60% NG + PS and 60% PG-dih + 40% NG + PS is 23%. However, the compressive strength of the 60% PG-dih + 40% NG + PS system is 45 MPa [150]. Figure 6: normal consistency in Portland cement with partial PG replacement In view of the normal consistency data with PG, it can be concluded that the normal consistency value of natural gypsum [25] is close to the value of the Di-PG phosphogypsum paste and the β- PG plaster. The normal consistency in Di-PG with the presence of impurity, increases the normal consistency. The treated α-PG plaster has a consistency value close to the value of the untreated β-PG plaster and the reference value [36]. The use of the sodium succinate, potassium citrate and sodium sulphate retardants has an influence on reducing the normal consistency of the α-PG plaster and on increasing the resistance to compression, in addition, the values found for normal consistency are close to the normal 26 26.5 23.7 26.3 26.4 23 23 30 36.8 50 50 50 40 45 0 10 20 30 40 50 60 0 10 20 30 40 50 60 Compressive Strength (MPa) Normal consistency (%) Nornal consistency (%) Consistency (26%) , Neville and Brooks (2010) Compressive Strength (MPa) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 22 consistency value of the reference [36]. References to the normal consistency of phosphogypsum A-PG II and A-PGI was not found, however, it is concluded that the normal consistency is lower when compared to Di-PG, α-PG and β-PG. It is also concluded that the consistency, curing temperature and mineral phase can influence the mechanical strength of the paste. The normal consistency of the HCBPG can vary between 37-44% and for all the analysed results, it can be verified that they are above the reference value (26%) [37]. HCBPG with α -PG have higher values of normal consistency compared to HCBPG with β-PG and A-PG II. It is worth mentioning that the PG in the mineral phases α -PG and A-PG II and A-PG I has low porosity [67], [143], [149] and therefore the normal consistency will be lower in comparison to the porosity of the mineral phases Di-PG and β-PG [67]. The normal consistency of studies with partial replacement of Portland cement by phosphogypsum varied between 23-26.5%, and the compressive strength of the paste varied between 30-50 MPa. All the studies have values close to the reference [37]. The compressive strength values are directly related to the normal consistency. It is also concluded that the temperature during the curing period and the mineral phase of PG influences the mechanical strength of the paste. It should be noted that the normal consistency with low water requirement, influences the reduction of porosity and the production of cement-based with high performance [151]. 5.2 Spread flow and slump test The Spread flow is a test that assess the ability to deformation a mortar under the action of loads [152]. In the laboratory, spread flow is evaluated with the aid of the flow table and the cone trunk mold (Spread flow test) [153] . The slump test is performed on concrete to assess its fluidity [154]. There are few studies that explain the relation between the components of mortars, their fluidity and slump [155]. It is important to note that the current state of the research has scattered information regarding the flow spread and the slump test. Thus, the found results were of mortars produced with Di-PG, α-PG, β-PG and A-PGII. Spread flow data were compared with the American standard ASTM C270 [51] wich requires the mortar spread to be between 110 ± 5 mm. The European standards EN 1015-2 [50] relates the density of the mortar in the fresh state with the spreading and requires the spread flow value between 120-175mm. The standard EN 13454-1 [59] requires the following spread flow: fluid (greater than or equal to 220mm), highly plastic (150-220mm) and thick (110-140mm). 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 23 The test slump data were compared with the standard EN 206 [53] and the Neville and Brooks reference [37]. Table 6 and Table 7 shows the data summary referring to the spread flow and the Slump test, respectively. Table 6: Summary of the spread flow data Mineral phase Composite binder Spread Flow (mm) Ref. Di-PG 10%Di-PG + PC 300 [156] 50%Di-PG + PC 175 [156] Di-PG + PC+ 1%Lime 301 [157] Di-PG +PC + 4% Lime 247 [157] Di-PG + FA + lime + PC 342-375 [158] α- PG 33%α-PG + FA + lime+PC 342-375 [158] 52%α-PG + FA + lime+PC 190-270 [159] β-PG β-PG+GGBFS+PC 105 [160] β-PG+FA+PC 105 [160] β-PG+FA+Clay+lime 100-105 [148] A-PG III unkown unkown unkown A-PG II 10%A-PGII + PC 300 [156] 50%A-PGII + PC 250 [156] A-PGII + FA + lime + PC 342-375 [158] A-PG I unkown unkown unkown Table 7: Summary of the Slump test Mineral phase Composite binder Slump (mm) Ref. Di-PG 15%Di-PG+PC 80 [161] 5%Di-PG+PC 1 85.45 [162] 10%Di-PG+PC 1 80 [162] 2,5%Di-PG+PC 2 72.30 [162] 5%Di-PG+PC 2 78.5 [162] 5%Di-PG+PC 3 68 [162] 10%Di-PG+PC 3 62 [162] Di-PG + GGBFS + Lime 155 [163] α- PG unkown unkown unkown β-PG unkown unkown unkown A-PG III unkown unkown unkown A-PG II A-PG II + GGBFS + Lime 55 [163] A-PG I unkown unkown unkown 1 Concrete produced with Class of strength M20 2 Concrete produced with Class of strength M25 3 Concrete produced with Class of strength M30 The studies show that mortars produced with the β-PG+GGBFS+PC, β-PG+FA+PC and β- PG+FA+Clay+lime systems can achieve a flow between 100-105mm. It was not possible to find the percentage of each component in the literature [148], [160]. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 24 Mortars with the formulations Di-PG + FA + lime + PC, α-PG + FA + lime+PC and A-PGII + FA + lime + PC had a flow variation between 342-375mm. Each mixture was composed of 33% PG, 42% FA, 20% PC and 3% lime [158]. Mortar made with the formulation α-PG + FA + lime + PC with 52% PG, 33% FA, 10% PC and 3% lime showed that the flow can vary between 190- 270mm [159]. Mortars with cement replacement by PG, were carried out with Di-PG and A-PGII + PC. The variation used for PG, for both studies, was between 10-50%. Mortars made with 10% Di-PG had an approximate flow of 300 mm, while mortars with 50% Di-PG had an approximate flow of 175 mm. Mortars made with 10% A-PG II had a flow of 300 mm, while mortars made with 50% A- PGII had an approximate flow of 250 mm [156]. The study with Di-PG + PC + lime showed that the flow can vary between 274-320 mm. The study consisted of varying the percentage of lime by 0-4%, with 1% substitution by 1%. Mortars with 1% lime showed a flow of 301 mm, while mortars with 4% lime showed a flow of 274 mm [157].The partial replacement of Portland cement by Di-PG at 5% provides a slump of 100 mm and with 15% of Di-PG the slump can reach 80 mm [161]. The concrete grades can also be tested for partial replacement of PC with PG. Class M20 concrete with 5% Di-PG replacement reaches a slump of 85.45 mm, while 10% Di-PG reaches a slump of 80 mm. Class M25 concretes with 2.5% Di-PG can have a slump of 72.3 mm and the use of 5% Di-PG influences a slump of 78.5 mm. M30 grade concretes with 5% Di-PG can have a 68 mm slump and with 10% Di-PG the slump can reach 62 mm [162]. The concrete produced with the Di-PG + GGBFS + Lime system with 17% PG, 82% GGBFS and 1% lime has a 155 mm slump and 2.5% incorporated air. The A-PGII + GGBFS + Lime system with 11% PG, 81% GGBFS and 8% lime has a 55 mm slump and the incorporated air content is less than 2% [163]. In view of the spread flow data, it can be concluded that: most of the data found on spread flow comply the standard EN 13454-2 requirements. Mortars can be classified as fluid (great than 220 mm) and used as self-leveling mortars. It was verified that only mortars produced with β-PG + GGBFS + PC, β-PG + FA + PC and β-PG + FA + Clay + lime cementitious systems, accomplish the ASTM C270 requirements. Whereas from the EN 1015-2 standard requirements, only, mortar produced with 50% Di-PG + PC accomplishes the spread flow requirements. It is also verified that this mortar can exhibit light density (great than 1200kg/m³), in the fresh state [152]. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 25 All the data about slump test of concrete produced with partial replacement of Portland cement by PG and cementitious systems based on PG, are in accordance with the EN 206 standard. Most studies on concretes showed a slump between 55-85.45 mm. These data, according to the consistency class S2 (50-90 mm), are related to the standardized consistency and can be applied in foundation, slabs and general structures. Only the concrete produced with the Di-PG + GGBFS + Lime cementitious system reaches the consistency class S4 (160-210 mm), due to its 155 mm slump. This type of concrete can be used in foundations, slabs, pumping and piling concretes. Furthermore, it is considered a concrete with high workability, due to its fluid consistency, released and compacted with little or no effort and with sufficient cohesion to be handled without segregation or exudation [164]. 6 Phosphogypsum morphology Scanning electron microscopy (SEM) has been used to study the morphology of phosphogypsum. By using this technique, it is possible to identify the shape and growth of the crystals. In addition, the formation of the PG microstructure can be monitored [146]. SEM can also evaluate the mineralogical formation of PG [165]. In Figure 7, the phosphogypsum in the dihydrate, hemihydrate, anhydrite III, anhydrite II and anhydrite I can be observed. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 32 mim). And the final setting time of all the cementitious systems meet the requirements of the standard ASTM C150 (375 mim). Figure 9: Fast and medium setting time of hydraulic cement-based on phosphogypsum It can be seen in Figure 10 that the A-PG II + GGBFS system (calcined at 800ºC) has an initial time of 210 minutes and a final setting time of 360 minutes [142]. It should be noted that the initial and final setting times of A-PG II + GGBFS system (calcined at 750ºC) may vary depending on the proportion of PG and GGBFS used. The longest setting times occur with 70% A-PG II and 24% GGBFS. The shortest initial and final setting times occur with 50% A-PG II and 50% GGBFS [149]. Besides, it possible verify that the A-PG II + Slag + lime system [175] present an initial setting time of 259 min and a final setting time of 550 min. The Di-PG + slag + Lime system [175] presents the initial setting time of 365 min and the final setting time of 580 min. The 10% A-PG II + 2% Di-PG + slag + lime system [175] has an initial setting time of 330 min and a final setting time of 45 24 17 25 23 26 55 70 70 95 60 90 45 40 42 106 150 145 150 120 0 50 100 150 200 250 300 350 400 Setting time (mim) Initial setting time Final setting time ASTM C595 (<7min) EN 15743, ASTM C1600 and ASTM C595, ASTM C150 (≥ 45 min) EN 15743 (≥60 min) EN 15743 and EN C14216 (≥ 75 min) ASTM C1600 (10min) Final ASTM C150 (375min) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 33 555 min. While the Di-PG + Steel Slag + limestone system [176] has an initial setting time of 485 min and a final setting time of 717 min. The use of alkaline activators in the composition of cementitious systems and its the influence on the initial and final setting time are also depicted in Figure 10. The Di-PG + SS + LS + NaOH system [177] has an initial setting time of 129 min and a final setting time of 252 min. The Di-PG + GGBFS + Alkali activator + sodium silicate + sodium hydroxide system [178] has an initial setting time of 48 min and a final setting time of 222 min. The Di-PG + GGBFS + Alkali activator + Sodium hydroxide system (27.8) [178] has an initial setting time of 90 min and a final setting time of 318 min, while the Di-PG + GGBFS + Alkali activator + Sodium hydroxide (18.5) [178] has an initial setting time of 108 min and a final setting time of 300 min. The initial setting time with HCBPG was compared to the requirements of the European standard EN 15743 [179] as the Supersulphated cement is composed of calcium sulphate, Portland cement and blast furnace slag. In addition, the American standard ASTM C 595 [44] was also used in the comparison, due to its specifications about ternary systems. It can seen that all the data found on the HCBPG initial setting time are in accordance with the ASTM C 595 standard [44] requirements. It can be concluded that the systems α-PG + FA + lime, α-PG + GGBFS + marble dust + chemical activation, 8% α-PG + FA + lime + PC, 15% α- PG + FA + lime + PC, and 10% α-PG + FA + lime + PC do not accomplish the requirements of the initial setting time prescribed by the standard EN 15743 [179]. Most cement systems accomplish the final setting time of ASTM C 595, overtaking the established limit. it is important to mention that all cementitious systems can be classified as a binder with an low setting time. Besides, only the initial setting time of cementitious systems A-PG II + slag + lime, Di-PG + slag + lime, A-PG II + GGBFS 800ºC, 10% A-PG II + 2% Di-PG + slag + lime, Di- PG + steel slag + slag + limestone meet the requirements of the standard ASTM C91 (great equal to 120 mim). In addition, can seen that the initial setting time of cementitious systems 70% A-PG II + 24% GGBFS 750ºC, 65% A-PG II + 30% GGBFS 750ºC, 60% A-PG II + 30% GGBFS 750ºC, 50 % A-PG II + 50% GGBFS 750ºC and Di-PG + GGBFS + alkali activated + sodium silicate + sodium hydroxide (20) (Supersulphated cement) do not meet the requirements of EN14647 and ASTM C91 (great equal to 90 mim). Only the final setting time of cementitious systems A-PG II + slag + lime, Di-PG + slag + lime + 10% A-PG II + 2% Di-PG + slag + lime, Di-PG + steel slag + slag + limestone ASTM C150 requirements. It should be noted that all cementitious systems can be classified as agglomerates 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 34 with a slow setting time and meet the final setting time of the ASTM C91 standard (less than 1000 mim). Figure 10: Slow setting time of hydraulic cement-based on phosphogypsum 7.3 Setting time of Portland cement Figure 11 shows the initial setting time and the final setting time of Portland cement with partial replacement of PG. It is observed that the mineral phase found in most studies was Di-PG, and the percentage of substitution used by researchers was 5% PG [25], [150], with the exception of the Di-PG system + PC that used only 1% replacement of Portland cement with PG [27]. 295 365 210 90 75 65 54 330 485 129 48 90 108 550 580 360 300 240 210 186 555 717 252 222 318 300 0 100 200 300 400 500 600 700 800 900 1000 1100 1200 Setting time (mim) Initial setting time Final setting time final ASTM C 150 (≥45min) EN 14647 ( ≥90 min) Inicial ASTM C 91 S and M (≥ (90 mim) Inicial ASTM C 91 N (≥ 120 mim) Final ASTM C595 (≤ 375min) Final ASTM C91 (≤ 1000min) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 35 Figure 11: Setting time of cementitious systems with Portland cement It is noted that the initial setting time can vary between 110 and 325 minutes [25], [27], [150]. While the final setting time can vary between 215 and 500 minutes Most Portland cements were produced with five percent phosphogypsum. Only the PC with an initial setting time of 110 m and a final setting time of 215 mim used one percentage of phosphogypsum in the composition. Di-PG treated with 4% citric acid has an initial setting time of 130 minutes and a final setting time of 270 minutes [25]. In the Di-PG + 40% GGBFS + 60% Clinker system (Di-PG + PS), the initial setting time was 255 minutes and the final setting time was 380 minutes [150].The Di-PG + 50% GGBFS + 50% Clinker system (Di-PG + PS) presented the initial setting time of 275 minutes and the final setting time of 390 minutes. It appears that the Di-PG + OS system has a longer setting time and a final setting time, if compared to the Di-PG + PC system Purification of Di-PG + 40% + GGBFS + 60% Clinker (Di-PG + PS) and Di-PG + 50% GGBFS + 50% Clinker (Di-PG + PS) systems with citric acid causes a reduction in the inicial and final setting time. The Di-PG + 40% + GGBFS + 60% Clinker system had an initial setting time of 185 minutes and a final setting time of 260 minutes. Meanwhile, the Di-PG + 50% GGBFS + 50% Clinker system presented an initial setting time of 245 minutes and a final setting time of 330 minutes [25]. It is noted that the purification of the systems and Di-PG + 40% + GGBFS + 60% 325 110 250 130 270 255 275 185 245 188 191 201 222 200 100 409 215 500 270 380 380 390 260 330 270 286 275 299 320 260 0 200 400 600 800 Setting time (mim) Initial setting time Final setting time Inicial EN 197-1 and ASTM C150 (45min) Final ASTM C 150 (375 min) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 36 Clinker and Di-PG + 50% GGBFS + 50% Clinker contributed to the reduction of the initial setting time and the final setting time. Data of the mixture of PG + NG were also found in the literature. The initial setting time of the 40% Di-PG + 60% NG + PC system was 188 minutes, and the final setting time was 270 minutes. The 60% Di-PG + 40% NG + PC system has an initial setting time of 191 minutes and a final setting time of 286 minutes. It is suggested that the increase in the percentage of PG in the system may influence the increase in the initial and final setting times [150]. The 40% Di-PG + 60% NG + PS system has an initial setting time of 201 minutes and a final setting time of 275 minutes. While the 60% Di-PG + 40% NG + PS system presented an initial setting time of 222 minutes and a final setting time of 299 minutes [150]. The increase in the initial and final setting times may occur due to the increase in the percentage of PG. In addition, the addition of GGBFS in the cement can also influence the increase in the initial and final setting time [169]. The β-PG + PC system presented the initial setting time of 200 minutes and the final setting time of 320 minutes [25]. Although the β-PG plaster has a initial setting time between 5-8 minutes and the final setting time is less than 30 minutes [36] and being considered a material of fast hardening, it appears that there was no influence on the reduction of the initial and final setting time of the β- PG + PC system. However, it can be observed that the treatment with 3% citric acid, can contribute to the reduction of the initial and final setting time [25]. Regarding the initial setting time of Portland cement with PG replacement, the obtained data are in accordance with the American standard ASTM C150 [47] and the European standard EN 197- 1[46]. Di-PG + PC, Di-PG + PS, Di-PG + 50% GGBFS + 50% Clinker, Di-PG + 40% GGBFS + 60% Clinker systems do not accomplish the ASTM C150 standard final setting time requirements [47]. 8 Phosphogypsum-based cementitious systems Cementitious systems can be composed of binary, ternary or quaternary mixtures [180]. Binary, ternary or quaternary designations indicate the number of materials or fillers that are included in the cement mixture, without indicating its nature or content [181]. During cement production, the clinker may be partially replaced by materials such as fly ash, blast furnace slag and limestone to produce composite cement. Construction industry constantly seeks to develop cement mixtures and ways to standardize mixtures. In addition to seeking material availability by region [182], [183]. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 37 International standards for the development of non-cement Portland have been studied by the RILEM Technical Committee 224-AAM, in partnership with business and society to achieve satisfactory results [184]. Investigations on calcium sulphate-based cement systems have become a promising alternative to produce cement and concrete at room temperature, due to the availability of raw materials. In addition, combining hemihydrate or anhydrite calcium sulphate with moderate amounts of calcium aluminate or silica contributes to stabilize the cementitious system. High calcium sulphate cements are better suited for work on fiberglass reinforcement than steel reinforcement [185]. Thus, it is observed that the study of calcium sulphate-based cement systems is a promising field and contributes to the production of cements with greater eco-efficiency. 8.1 Studies on phosphogypsum pastes Most studies focused on phosphogypsum are related to the control of the setting time and the microstructure properties [77], [117], [186]–[188]. In view of this fact, a comprehensive literature search was conducted to list the different possibilities of producing Hydraulic Cement Based on Phosphogypsum-based (HCBPG) applied in paste application. The results obtained can be seen in Table 8- Appendix A. As far as cementitious formulations are concerned, it is observed that almost all the studies were developed by using ternary (T), quaternary (QA) and Supersulphated cement (SSC) systems, although Belz et al. [189] and Hua et al. [190] developed binary systems. Regarding the use of phosphogypsum, dihydrate (Di-PG) and βhemihydrate (β-PG) phases are found in almost all the works. The anhydrite phase (A-PG) appears only in the works of Singh and Garg [142], Erdem and Olmez [191], Singh and Garg [149], [173] and Michel et al. [192]. However, PG αhemihydrate (α-PG) was not found in any paste study. As aforementioned, there are impurities in the chemical composition of PG dihydrate, and because of this, the mechanical performance of HCBPG might be compromised. The quality of materials produced with PG dihydrate can also be influenced by their porosity, as PG dihydrate is composed of interconnected crystals, with porosity up to 70% and its mechanical properties are influenced by porosity [193]. The works with PG β-hemihydrate show better performance compared to PG dihydrate, due to modifications of its crystal structure [66]. Furthermore, its brittleness after heat treatment can be attenuated by reducing its porosity, increasing the specific area of the particles, and increasing the speed of mechanical development [194]. Meanwhile, PG α-hemihydrate, with autoclaving, will present a higher degree of inter-crystalline bonding and lower porosity, bearing a higher degree of 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 38 resistance [195]. Anhydrite phase can react with water and convert to plaster, and therefore to be used in building materials. Although the setting time is slow, its reactivity can be increased considerably after grinding, by increasing the specific area and adding activators [196]. The fly ash (FA) and blast furnace slag (GGBFS) were the most by-products used in PG-based cementitious systems [197], [198].The silica fume (SF) [199]–[201], bottom ash (BA) [202] and carbonate rock (CSR) [200] were in some studies. The use of ordinary Portland cement (OPC) has been found in almost all the works. However, the work by Liu and Wang [203] uses the PC with blast furnace slag, while Gaiducis et al. [200] used the PC with limestone filler. Replacing OPC with industrial by-products during cement production contributes considerably to reducing CO2 emissions, because less clinker will be consumed [204]. Between 2017-2019, most publications are related to the development of Supersulphated cements (SSC). This fact reflects that researchers have shown interest in defining the appropriate content of phosphogypsum in cementitious systems, such as SSC [179], [205]–[208]. Sulfoaluminate cement (CSA) and Calcium aluminate cement (CAC) have been used also in recent research [209], [210].These two types of cements are also considered more eco-efficient materials due to their low energy consumption and low carbon dioxide emissions [185], [211]. Research on HCBPG shows that it is possible to develop ternary cement systems without the presence of Portland cement [73]. Different authors have developed formulations based on phosphogypsum (PG), fly ash (FA) and/or blast furnace slag (GGBFS), hydrated lime and/or alkaline activating solutions ([103], [142], [148], [212]–[215]) . Due to the use of alkaline activators, the cementitious systems also acquire pozzolanic characteristics [216], [217]. It is concluded that most of the studies were with ternary, quaternary and Supersulphated cement systems. PG dihydrate and PG β-hemihydrate were the most used mineral phases. The use of PG dihydrate influences material quality due to impurities and porosity. PG α-hemihydrate has lower porosity compared to PG β-hemihydrate and, therefore, HCBPG may have higher mechanical resistance. HCBPG with anhydrous PG have higher setting time. The use of Portland cement predominated, however, studies were also performed with Calcium aluminate cement (CAC), Calcium sulfoaluminate cement (CSA). Fly ash (FA) and blast furnace slag (GGBFS) were the supplementary cementitious materials used in all the studies. Also, it can be noted that lime was the most used alkaline activator. 8.2 Studies on phosphogypsum mortars Cement-based mortars are cementitious materials characterized by the mixture of cement paste and sand [218]. They can be used in different applications such as outer coating, inner coating or 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 39 consolidation injections. Each type of application has different requirements despite of using the same building materials [219]. Research on the application of phosphogypsum in mortars aimed at identifying the study parameters (water/binder, binder/sand and spread) used for the preparation of the product, as well as the most used mineral phases, the type of cement, the presence of cementitious systems and the mixing ratio. The results obtained are presented in Table 9-Appendix B (from 1992 to 2020). Based on the information found in the literature, various industrial by-products with phosphogypsum can be combined to produce mortars. However, it is observed that the percentage of used materials can vary considerably, especially the percentage related to PG, OPC, FA and GGBFS. PG ranges from 10% to 90%, GGBFS ranges from 1% to 92%, FA 1.5% to 60%, PC 3% to 90%, Lime 1% to 50%, and SS 5% -13, 6%. Moreover, the works by Singh and Garg [160] and Wang et al. [220] did not provide data on formulations. And Singh and Garg [174] reported only the percentage of PG that was used. The review has also shown that several types of phosphogypsum mineral phases have been studied. Dihydrate phosphogypsum (Di-PG) was used in almost all the works, followed by the phases βhemihydrate (calcined) and α-hemihydrate (autoclaved). Ren [159] and Hyung [221] employed phase II anhydrite. Mun et al [175] used PG anhydrite II. The study by Mun et al.[222] evaluated the effect of PG on dihydrate, anhydrite (II and III) and β-hemihydrate phases for all the manufactured cement systems. No studies have been found with anhydrite I. The fact that PG dihydrate appears in most mortar studies suggests an attempt to value PG through more economical techniques, since dehydration (calcination, autoclaving and semi-drying) is an expensive process [223]. Most cementitious systems were composed of ternary and quaternary systems, including the use of blast furnace slag (GGBFS) and/or fly ash (FA), hydrated lime and/or Portland cement. The works by Mun et al. [175], Huang and Lin [176], Hyung et al. [221], Li et al [224] and Wang et al. [220] developed ternary systems for phosphogypsum-based hydraulic binders, without clinker. Gong et al. [156] produced mortars using binary systems (PG-PC). Around 330Mt/year of GGBFS and 900Mt/year of FA are generated [211]. Due to the high global availability, the low cost and the high-quality control FA and GGBFS are considered basic raw materials for the formulation of eco-efficient cement and cement systems [225], [226]. Portland cement was the only hydraulic binder used in cement-based mixtures. Despite the diversity of cement types on the market [204], due to the standardization of research on Portland cement (PC), other types of cement are no longer used. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 40 Therefore, it can be concluded that the formulations used in mortars have very different formulations. PG dihydrate is present in most studies, however studies have also been performed with anhydrite (II and II) and hemihydrate (α-PG and β-PG). The analysed cementitious systems contain the following compositions: (I) without Portland cement: PG-FA-lime; PG-GGBFS-lime; PG-GGBFS-FA-lime, (II) with Portland cement: PC-PG-FA-lime; PC-PG-GGBFS-lime; PC-PG- GGBFS-FA-lime. The other systems found were binary (PC-PG) or systems developed for Supersulphated cements (GGBFS-PG-PC). Only Portland cement was used, while fly ash (FA) and blast furnace slag (GGBFS) were the supplementary cementitious materials used in all studies. 8.3 Studies on phosphogypsum concretes Concrete is a building material produced from the mixture of fine aggregate, coarse aggregate and cement. Mineral additions and chemical additives are also used to improve concrete performance [227]. Constant research is carried out in order to develop more eco-efficient concretes, and for this, industrial waste and agro-industrial waste are used [228]–[230]. Table 10-Appendix C provides relevant concrete studies that used PG. Between 1999-2019, 14 articles were published and most of them used Portland cement as raw material. Almost all the studies are related to the partial replacement of Portland cement by phosphogypsum [231], [232]. The results demonstrated that the properties of synthetic gypsum cement are similar to those of obtained by using natural gypsum [233]. Although PC prevails in the concrete production, the study by Tian et al. [234] highlights from an eco-efficient framework, as it manufactures concrete with less environmental impact. By using phosphogypsum-based Sulfoaluminate cement (CSA), they produced a material that consumes less energy, emits less CO2 and has a lower cost. That is due to a greater incorporation of gypsum (15- 25%) [46], [107], [235] compared to Portland cement which incorporates only 5% [107], [236]. The works by Yang et al. [237], Ding et al. [238], Buhari e Raju [161] and Sukaman et al. [239] show that the production of concrete with ternary cement systems are possible, while Yang et al. [237] proposed a quaternary system. The concretes developed by these authors used phosphogypsumbased hydraulic binders and cementitious supplements such as blast furnace slag and fly ash. Ding et al. [238] proposed a quinary system with PG, PC, GGBFS and SS to develop a Supersulphated cement. The use of binary systems is more common in concrete production than ternary systems [229]. However, the possibility of creating cementitious systems with 3 or more materials may contribute to improve the concrete properties [240]. PG dihydrate is present in almost all the works. However, 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 41 in the work by Smaddi et al. [241] was observed the use of β-PG and A-PG, while the study developed by Yoo et al. [163] formulated HCBPG with the simultaneous presence of Di-PG and A-PG, and did not use PC in the formulation of the cement system Based on the literature review, the use of PG in PC, CSA and HCBPG is observed. Furthermore, different cementitious systems, with different types of raw materials and study parameters were used. It is important to highlight that only Reddy et al. [242] reported the whole mix preparation data (%W, water/binder ratio, optimal water/binder ratio, fineness modulus) as well as the effect on compressive strength with the use of phosphogypsum. It can be concluded that many studies use PG to partially replace cement, however, there are also studies that used PG in concrete, with and without Portland cement. PG in the dihydrate phase was the most used, although some studies have also used PG hemihydrate (α-PG and β-PG). Portland cement is the most used cement, but the use of calcium sulfoaluminate cement is also observed. The ternary system with GGBFS is present in most studies. The studies with the application of PG in concrete have different values in the material production parameters. The water/cement ratio varied between 0.4-0.65, the optimal water/cement ratio varied between 0.4- 0.6. The percentage of phosphogypsum considered as excellent varied between 4.5%-55%, the fineness module of the fine aggregate varied between 2.73-3.71, while the fineness module of the coarse aggregate showed a variation of 3.87- 7.2. The effect of compressive strength in relation to the optimal percentage of PG varied between a 20-25% increase. 9 Mechanical strength of mortar and concrete with phosphogypsum Studies related to the mechanical strength of cementitious matrices are directly linked to safety and structural stability. The compressive strength is one of the properties that could indicate possible quality variations. Based on the literature review, it is evident that a main objective of the studies on the application of PG in mortars and concretes is the investigation on the mechanical properties. The obtained results were aimed at analyzing compressive and flexural strengths. The studies also provide the microstructure of mortars and concretes through tests such as: Scanning Electron Microscopy (SEM), Energy Dispersive Spectroscopy (EDS), X-ray Fluorescence (XRF) and X-ray Diffraction (XRD). These tests provide the chemical characterization, including the pozzolanic reaction, as well as the formation of ettringite and C-S- H. and consequently, the resistance increase. Thermo-gravimetric and differential thermal analysis (TG-DTA) tests were used to verify at which temperature the PG modified the mineral phase (dihydrate to α-hemi, β-hemi, Anhydrite-III, Anhydrite-II and Anhydrite). The phosphogypsum in the mineral phase β-hemi is present in almost 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 48 studies were performed to compare the compressive strength of concretes with natural additive cement and synthetic gypsum. Figure 15 – Compressive strength in concrete with replacement Portland cement by Analysing the results, most partial substitutions of Portland cement (PC) by phosphogypsum occurred with PG dihydrate, and in most studies the percentage of PG used is above 5%. However, concrete produced with 4.5% of PG achieves a value of 32.5MPa. [244]. Concrete produced with partial replacement of cement by 10% PG reaches 49.3MPa [242]. However, studies show that the production of concrete with partial replacement of cement in 10% PG, can achieve a compressive strength of 30.5 MPa [245]. The use of 7.5% PG as partial replacement of cement can achieve 38.48 MPa [162]. However, concretes produced with 10% β-hemihydrate (β-PG) and dihydrate (Di-PG) can rise 36.3 MPa and 32.2 MPa, respectively [246]. Nevertheless, the studies showed that concretes produced with 8% PG dihydrate reach only 27.93 MPa. These studies do not inform the cement strength class [247] . From the literature review it can be concluded that the concretes made with partial replacement of Portland cement by phosphogypsum, accomplish the Eurocode EM 1992-1-1 requirements in terms of compressive strength. The resistance class refers to a minimum of 25 MPa for cylindrical specimens and 30 MPa for cubic specimens [52]. 32.5 49.3 30.15 27.93 32.2 38.48 36.3 0 10 20 30 40 50 60 Compreswsive strength (MPa) Compressive strength Strength class EN 1992-1-1 (C25MPa) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 49 10 Comparison of cement-based studies with and without Portland cement In order to make easier the analysis of phosphogypsum application in cementitious matrices, Tables 11 and 12 summarize the most significant results on compressive strength of mortars and concrete, respectively, produced with HCBPG, as well as the production parameters. Table 11-Appendix D shows that various types of PG mineral phases were used. The Anhydrite II, β-hemihydrate and dihydrate phases are present in studies without PC mortars. Whereas studies with PC mortars have most mineral phases (α-hemihydrate, anhydrite II, anhydrite III, βhemihydrate, dihydrate). PG dihydrate is present in most results without PC mortars, while most PC mortars have PG anhydrite II. In addition, it can be observed that mortars with PC had studies with greater variation of the PG mined phases compared to mortars without PC. In both types of mortar (without and with PC) results can be observed with PG dihydrate above 44 MPa. Among the materials used in the formulation of the mortars, can seen several cementitious systems with blast furnace slag (GGBFS) and without Portland cement. It is important to mention that fly ash (FA) was the material most used in mortars with PC. The mortars without PC used only GGBFS as supplementary cementing materials. Only one study reported the flow between mortars without PC, while most studies with PC mortars reported the flow, water/cement ratio and binder/sand ratio. Table 12-Appendix E provides a comparison of different concretes produced with PG. PG dihydrate was used as partial replacement in most of the studies, although PG β-hemihydrate was also used. The compressive strength results accomplish the strength requirements, ranging from 27.93 MPa to 49.3 MPa. In addition, studies on setting times were reported in most studies, as well as the water/cement ratio and fineness modulus. Only two studies showed the effect of compressive strength on the ideal PG amount. Studies on concrete with cementitious systems have also achieved satisfactory results, being in accordance with the minimum requirements of compressive strength (25MPa) [53]. Portland cement, slag (GGBFS) and hydrated lime were the most used materials. The 70 MPa result can be observed in the formulation PG dihydrate + anhydrite II + GGBFS + lime. Chemical activator and aluminium powder were also used as raw material. Regarding the analysed parameters, it was verified that the water/cement ratio was not informed, providing only the fineness module of the coarse aggregate. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 50 11 Durability of phosphogypsum-based cement The durability analysis is the study of the agents and mechanisms that have an effect on the material degradation. Through this study, it is possible to guarantee the performance and integrity of the material, from its production to the time of use (useful life). Phosphogypsum-based cement is prone to undergo moisture, due to the ease of solubilization in the presence of water. Therefore, its use is restricted to indoor environments. Some studies have been developed to improve resistance [142], [143], [146], [248]–[250]. As it will be detailed in the next paragraphs, the solubility of phosphogypsum has been analysed, as well as the water resistance of cement-based PG, shrinkage by drying, and the expandability of the paste containing PG. 11.1 Solubility The solubility of calcium sulphate, in all mineral phases, changes if temperature increases. Figure 16 shows the solubility of calcium sulphate in water as a function of temperature Calcium sulphate dihydrate solubilizes when the water reaches 20°C. When the water temperature is above 40 ºC the solubilization decreases significantly, resulting in the deposition of the material. The hemihydrate phase solubilizes from 20ºC, and it has greater solubility than both the dihydrate and the anhydrous phase, due to its lower thermodynamic stability [251]. The calcium sulphate βhemihydrate solubilization coefficient (0.92g/100 ml) is lower than that of calcium sulphate αhemihydrate (1.2g/100 ml) [252]. Anhydrite mineral phase, anhydrous I and II forms are insoluble. However, anhydrite III (soluble anhydrite) has good solubility and the crystalline structure is an anhydrous state of the hemihydrate. Furthermore, it can return to the hemihydrate phase very quickly, when in contact with water (vapor or liquid) at room temperature. For also presenting thermodynamic instability, greater care is required during preparation [253]. Summarizing, in cement-based materials, the calcium sulphate solubility differs depending on the mineral phase, and the importance of this phenomenon is verified. In this way, it is possible to regulate the setting time during the hydration of the elemental composites. In addition, calcium sulphate can also improve both mechanical strength and water resistance by adding pozzolan, cement and hydrated lime. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 51 Figure 16: Solubility of calcium sulphate in water as a function of temperature. Reprinted from Hoang, Ang and Rohl [251] with permission from Elsevier 11.2 Water-resistant phosphogypsum Calcium sulphate is a product that has low water resistance, as it has high solubility [126]. In humid environments, crystals tend to lose their adhesion strength. As a result, dissolution occurs under water contact. The environments that are prone to freezing, can also yield to different cracking or collapse patterns, due to the expansion of water that is in the pores [36]. Cement-based materials can be improved through calcination. The phosphogypsum hemihydrate crystals gain water resistance [67].The water resistance increase in cement-base calcium sulphate can be achieved by the addition of hydrated lime, cement and pozzolana (fly ash, smoke silica, blast furnace slag) [254], [255]. From the performed literature review, the data on the water absorption of phosphogypsum Di- PG, α-PG, β-PG, A-PG II, A-PG I system are summarized in Table 13-Appendix F. However, it is worth noting that it was not possible to find data on the A-PG III. In addition, either data on the compressive strength values or data on water resistance coefficient were not found. Due to the correlation between the compressive strength and the percentage of water absorption, the performance of cement-based PG and plaster is evaluated in this paper. The Di-PG + PC water absorption was 2% and the compressive strength 32 MPa [156]. The Di- PG + PC + GGBFS system showed a water absorption of 3% and a compressive strength 22 MPa [160]. The water absorption for the Di-PG + PC + FA system was 10% with a compressive strength of 19.6 MPa [160]. The Di-PG + FA + Lime system had a water absorption of 28% and a compressive strength of 12 MPa [256]. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 52 The water absorption of α-PG plaster is 32% [257]. While α-PG + FA + PC + Lime shows water absorption and 14% for compressive strength of 27.5MPa [145]. When the α-PG + FA + Lime system is subjected to a 50ºC curing, the water absorption increases 19.6% and the compressive strength decreases to 26.4 MPa [141]. However, the PG + GGBFS + Marble dust + chemical activate system with a curing temperature of 50ºC can reach 32.2 MPa, with a water absorption of only 4% [141]. From the reviewed data, β-PG plaster can be more soluble compared to α-PG plaster, with a water absorption of 30% to leaching. The compression strength values range from 13.20 to 14 MPa, and the water resistance coefficient is 0.35 [142], [143], [171], [257]–[259]. When analysing the β-PG + PC + GGBFS system, it is noted that the compressive strength can reach 35 MPa, with a variation in water absorption between 3-7% % [160], [259]. The use of organic retardant in the β-PG + PC + GGBFS system maintains the compressive strength at 35 MPa, however the water absorption can vary between 2.89-6% [143], [160], [257]. The study of the β-PG + FA + PC system showed that water absorption can vary between 6.5- 10% and the compressive strength can vary between 25.6-21.3 MPa, respectively. When this system is cured at 50ºC, the water absorption remains at 10% and the compressive strength reduces to 20 MPa, and the use of a retarder causes a compression strength increase of 25 MPa with 8 % water absorption [146], [160], [171], [250], [260], [261]. The composition 70% β-PG + 30% SAC showed a resistance coefficient from 0.70 to 22% water absorption [258]. The water absorption of the A-PGII + GGBFS system can vary between 9.55 and 10.43% and the compressive strength follows the variation of 35-56 MPa, respectively [142], [149]. When blast furnace slag is replaced by Portland cement to form the A-PGII + PC system, water absorption reduces to 4%, but the compressive strength is slightly reduced to 34.5MPa [156]. The study of the PG + FA + Red Mud system showed that it is possible to achieve a compressive strength of 40 MPa and a water absorption of 4% [21]. In view of the obtained data, it can be concluded that there is a relation between the water content absorbed by the paste and the compressive strength. That is, the greater the water absorption is, the lower the resistance to compression is. It appears that the plasters Di-PG, α-PG and β-PG had the highest water absorption rates. The Di-PG + FA + Lime, β-PG + Fa + Lime and 70% β-PG + 30% SAC systems also showed high water absorption and therefore, it can be concluded that the systems with lime, exhibit high absorption of water compared to systems composed of PG + GGBFS + PC, PG + FA + PC, PG + PC. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 53 It can be seen that most of the data on water absorption of cement-based PG are in agreement with the result of water absorption of cement paste (16%), presented in the study by Selim et al [61]. According to Kovler [255] cementitious systems based on calcium sulphate with water absorption up to 15%, are considered materials with high resistance to water. In the early ages, the absorption water varies between 35-40% on cementitious systems produced with calcium sulphate + silica fume + Portland cement. At 28 days of age the water absorption decreases (30-35%), due to the formation of C-S-H. 11.3 Drying shrinkage Drying shrinkage is a phenomenon that causes the reduction of the volume of cementitious materials, through the decrease of the internal relative humidity of the pores [262]. When there are internal or external restrictions and the volume reduction is impaired, residual tensile stresses may accumulate, and strains and/or micro cracks may occur [262]. The drying shrinkage depends on the following parameters: relative humidity, type of material used (binder, fineness of the binder, addition of pozzolana, volume of cement paste), water / cement ratio, type and size of the aggregate, porosity and fineness and size of the part that will undergo retraction [263]. The drying shrinkage data obtained from the literature review are depicted in Table 14. It is worth noting that it was possible to find information only for the mineral phases Di-PG and α-PG. The performance and technical feasibility of cement systems were compared with the European standards EN 12620 [57], EN 1367-4 [56] and the bibliographic references Hewlett and Liska [54] and; Lamond and Pielert [55]. Table 14: Drying shrinkage Mineral phases Drying Shrinkage (%) PG (%) Binder Ref. Di-PG 3 10 HCBPG [264] Di-PG 0.06 40 SAC/PC [265] Di-PG cured at 80ºC 0.06 50 HCBPG [190] Di-PG cured at 50ºC 0.16 50 HCBPG [190] α-PG system +60%FA 0,11 15 HCBPG [145] α-PG system +50%FA 0,11 15 HCBPG [145] α-PG system +40%FA 0,13 15 HCBPG [145] β-PG unknown unknown unknown unknown A-PG III unknown unknown unknown unknown A-PG II unknown unknown unknown unknown A-PG I unknown unknown unknown unknown 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 54 Mortars produced with Di-PG + FA + lime with 10% PG resulted in 3% shrinkage when drying (when curing was carried out in water). For outdoors curing, there was no change in volume [264]. The study of self-levelling mortars with the Di-PG + SAC / PC system, if SAC/PC is in the 6: 4 ratios, showed that the use of 40% PG results in 0.06% shrinkage by drying. The study of mortars produced with PGS-optimal (CM: PG: SG: Lime: SF: USZ: BXF200-L: Defoaming Agent) with 50% PG resulted in a drying retraction of 0.06% for the curing temperature 80ºC, and in a 0.16% retraction by drying, for a curing temperature 50ºC. It suggests that the higher the curing temperature is, the lower the drying shrinkage for mortars produced with PGS-optimal HCBPG is [190]. The α-PG + FA + PC + Lime system, containing 15% PG, was studied in mortars. The α-PG system + 60% FA and α-PG system + 50% FA showed a shrinkage of 0.11%. While the α-PG system + 40% FA showed a 0.13% shrinkage on drying. In the early ages, the drying shrinkage of Portland cement pastes, with 47% relative humidity varies between 0.35-0.70%. At the age of 6.5 months, mainly the pastes with the largest amount of pores, reaches 0.65% retraction, while pastes at the age of 30 months show a 0.45% drying retraction, when the relative humidity is by 47% [54]. In concrete, the relative humidity of 50% can influence the shrinkage due to drying between 0.04- 0.08%. In reinforced concrete with normal amount of steel, the retraction can vary between 0.02- 0.03%. In structural lightweight concrete the retraction varies between 0.04-0.15%. It is considered a great shrinkage effect on drying, in cellular concrete, when the shrinkage varies between 0.3-0.6% [55]. According to standard EN 1367-4, when the effect of the aggregate on the concrete shrinkage exceeds 0.006 mm and 12% the test must be considered unsatisfactory, and another test must be performed. The EN 12620 standard determines that the shrinkage due to drying in the concrete, occurs because of the aggregate, and it cannot exceed 0.075%. The shrinkage by drying can vary between 0.06-3% for systems with HCBPG. While drying shrinkage is very low in sulfoaluminate cement-based materials. In view of the data from the literature review, it can be concluded that the data presented in Table 12 are in accordance with the standards EN 1367-4, EN 12620, and with the literature on shrinkage by drying [54], [55]. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 55 11.4 Expansibility (soundness) In cementitious materials, in order to guarantee both durability and an adequate useful life, it is essential to control and minimize the changes in volume [37]. The expandability in cementitious matrices is directly related with the reaction of tricalcium aluminate (free lime), magnesium sulphate and the calcium sulphate of the phosphogypsum. During the chemical reaction of these elements, the formation of ettringite crystals and the subsequent expansion of the cement matrix occur [266]. Therefore, the cement matrix breaks in the hardened state [37]. In the literature, the data on expandability was not found only for the mining phase A-PG III. In Table 15-Appendix G, the expansion data for Di-PG, α-PG, β-PG, A-PG II, A-PG I can be observed. It is noteworthy that it was not possible to identify, most of the phosphogypsum percentage used in expandability studies. The expandability study with the Di-PG plaster showed the result of 0.6mm [174]. The expandability of Portland cement with partial replacement by Di-PG with 12.5-25% PG, can vary between 6-8 mm [27], [245], [267]. The Di-PG + PC + GGBFS system presented the result of expandability of 3.6 mm [160], [174]. The expandability of the Di-PG + PC + FA system was 1.6mm [160], [174], while the Di-PG + PC + Red Mud system showed an expandability of 1.6mm [174]. Among the results on expandability, the Supersulphated cement composed of Di-PG + PC + GGBFS + SS + LS + NaOH and 45% PG, gave 13.5mm [177]. Only the expansion value of α-PG + PC + FA + Lime system with 10% PG was found. The expandability value was 2.3mm [145]. The value for the β-PG plaster was 0.6mm [171], [174]. The study of partial replacement of Portland cement with 70% PG (β-PG + PC) showed an expansibility of 0.10 mm, while the partial replacement of pozzolanic cement with 70% PG (β-PG + PCC) presented 0.106 mm expandability [241]. The study of the β-PG + PC + GGBFS system with retarder and without organic retarder showed an expandability of 1.60 mm [160], [174], [249], [257], [259]. The β-PG + GGBFS + tartaric acid system with 15% PG showed an expandability of 1.2 mm [173]. The β-PG + PC + FA system showed an expansion range between 0.88-1.00 mm [160], [174]. When a retarding additive is used for the same system, the expandability value is 0.88 mm [250]. And when the β-PG + PC + FA system is cured at 50ºC, the expandability is 1.8mm [146]. The β-PG + FA + PC + Lime + chemical additives system with PG between 70-75% showed an expandability of 2.0mm. The β-PG + PC + Red Mud system showed an expandability of 1.1 mm. The highest expansion value found in the literature with the use of β-PG was 12 mm for Supersulphated cement composed of β-PG + GGBFS + SS + SL with 45% PG [176]. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 56 The expandability of A-PG II plaster is 0.5 mm [142]. The A-PG II + GGBFS system presents a range of expandability between 0.81-0.90mm, for 70% PG and 15% PG, respectively [149], [173]. While the A-PGII + GGBFS + system chemical activator showed an expandability of 1.5 mm [142]. The highest expandability value with A-PGII was 3.0 mm for the PG + PC + GGBFS system, with 10% of PG relative to Supersulphated cement [191]. Only a result of 0.06 mm for A- PG I + GGBFS + 2% citric acid system is obtained from the literature [21]. It can be concluded that the plaster has practically no expandability, because the values are very low. However, when there is a mixture of PG in any mineral phase (with fly ash, blast furnace slag, Portland cement, setting retarder or chemical activator) the expandability increases. The highest expansibility values will occur when Portland cement is partially replaced by PG. It should be noted that only the PG + GGBFS + SS + SL (Supersulphated cement) system did not agree with the expandability requirements of the European standards EN15743 [179] and EN 197-1 [268]. 12 Application of phosphogypsum in geotechnical materials The production of Portland cement-based geotechnical materials require combustion, furthermore, they consume a lot of energy and pollute the environment [269]. To minimize the environmental impacts, the construction industry has adopted the use of industrial or agro-industrial by-products over the years as alternative raw materials, as they provide low cost and lower energy consumption [270]. Based on the literature review of works that were done between 1992-2019, Table 16-Appendix H shows the studies on phosphogypsum-based geotechnical and building components materials. In total were found 8 studies on materials to be used as road basement; 5 studies on tile and ceramic; 1 study on lime; 5 studies on self-levelling mortars; 7 studies on non-hydraulic binders; 3 studies on plasterboard; 13 on blocks and bricks and 7 studies on plaster. Recent studies (2017-2019) have not focused on a single theme. Contreras et al. [271] applied PG as an additive in the production of ceramic bricks. The works by Silva et al. [272] and Dutta [273] investigated paving and geotechnical materials by using phosphogypsum. In addition, the works developed by Qiang et al. [274] and Schaefer et al [71] show that it is possible to incorporate PG in Sulfoaluminate cement and Calcium aluminate cement (CAC) for the production of self-levelling mortars. In the work by Ding et al. [275], it was found that PG can be used for soil stabilization. The work by Zhou et al. [276] was focused on the manufacture of ceramics without the combustion process, and Romero-Hermida [277] investigated the possibility of producing lime from phosphogypsum as a substitute for commercial lime. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 57 Regarding the use of PG, it is noted that the most used mineral phases correspond to phosphogypsum dihydrate (Di-PG) and β-hemihydrate (β-PG). The works by Schaefer et al [71] and Singh e Garg [21], [165] were performed with anhydrous phase phosphogypsum (A-PG). The studies of Garg et. al [96]; Garg e Pundir [145] used the phosphogypsum in α-hemihydrate (α-PG) phase. Portland cement (PC) is the most used hydraulic binder in all the works, and fly ash (FA) compared to blast furnace slag was more prominent in the manufacture of building material components. In addition hydrated lime is present in most of the works due to studies on alternative phosphogypsum-based hydraulic binders in ternary, quaternary and quaternary systems [160], [174], [249], [259], [260], [264], [278]. In view of the discussed issues, it can be concluded that phosphogypsum can also be applied in geotechnical materials, self-levelling mortars, blocks, tiles, slabs, lime, binders, with the use of PG β-hemihydrate and dihydrate. 13 Study of radiation in building materials with phosphogypsum Natural radionuclides are present in the Earth's geological formation. They are found in rocks, soils, building materials and industrial by-products. In construction materials, the following radionuclides appear more frequently: uranium series (U-238), thorium series (Th-232) and potassium isotopes (K-40) [279]. Building materials manufactured with inputs of terrestrial origin (sand, granite, clay, marble, gravel, wood) contain less radionuclides than materials produced with industrial by-products (phosphogypsum, blast furnace slag and fly ash) [280], [281]. Radiation in building materials occurs in two ways: (i) external radiation - caused by radio isotopes (Ra) with the emission of gamma radiation (Ra-226 generated by the U238 decay); (ii) internal radiation - caused by radon isotopes (Rn) with gas exhalation from Rn-222 (generated by the decay of U238) and exhalation of gas Rn-220 (generated by the decay of Th232) [282]. Radon and its daughters emit short-lived alpha radiation. Large levels of radiation can be transmitted indoors. Contact with high concentration can cause respiratory changes and greater chances of developing cancer when exposed to gas [283], [284]. The control of radiation in building materials produced with phosphogypsum is essential to understand the limit amount of incorporation of PG. Depending on the uranium source, the activity of Ra-226 in PG occurs one or two times stronger, compared to natural gypsum [285]. Because of this, constant research seeks to establish basic safety standards for protection against the dangerous ones arising from radiation exposure [286]. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 64 compressive strength of mortars with PC varied between 10-70 MPa. Besides, most mortars produced with cementitious systems without and with Portland cement accomplish the compressive strength requirements the standards ASTM C 270 standard, EN 998-1, EN 998-2 and EN 13813. The data on phosphogypsum mortars point out the feasibility of using them in plastering mortars for indoor and outdoor environments, structural mortar (masonry foundation and masonry below grade), in addition to selflevelling mortars. 12) As far as the concrete compressive strength is concerned, the compounds that are produced with cementitious systems ranged between 38.6-80 MPa, accomplishing the Eurocode 1992-1-1 requirements. The cementitious systems Di-PG + CSA + GGBFS and Di-PG + FA + Lime do not meet the minimum requirements because they have a compressive strength below 25 MPa. All the analysed mixtures, with partial replacement of cement by PG, accomplish the Euro Code 1992-1-1 prescriptions. 13) Regarding durability, the data show that the plaster, regardless of the mineral phase, has high water absorption and low resistance to compression. However, cementitious systems and partial substitutions of PC for PG, have less water absorption and greater compressive strength. The drying shrinkage of the SAC/PC binary cement system is lower compared to HCBPG. The expandability of the Plaster is low compared to the binary system PG + PC and HCBPG, in all mineral phases. 14) Phosphogypsum can also be applied in geotechnical materials, self-levelling mortars, blocks, tiles, slabs, lime, binders. Most of the studies were developed with Di-PG and β- PG. 15) Other forms of PG valorization occur with the production of portlandite and sodium sulphate. CO2 capture contributes to the mitigation of PG, as well as the treatment of bauxite and use in agriculture. 16) The health risks, due to radiation exposure of phosphogypsum building materials, depends on the origin of the PG, as well as the density and thickness of the material. The literature showed that up to 50% phosphogypsum can be used without causing contamination by radon-222 and up to 30% of PG can be used without causing contamination by radio-226. The data mentioned in this paper show an overview on the feasibility of using phosphogypsum in cementitious materials accomplishing relevant requirements of the current Standards. The use of PG of sedimentary origin is still a challenge, and therefore ways of controlling or minimizing the effect of radiation should be studied, as well as the application of raw state PG (igneous and sedimentary). In addition, it is also necessary to research on methods of removing impurities in an 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 65 economical way. In future research the studies on high performance concrete, hydration of cementitious systems, stabilized mortar, high workability concrete, (by using PG-based cementitious systems), could contribute to foster both the research and the use of more to ecoefficient building materials. Moreover, circular economy would be promoted within the construction industry by valorizating the phosphogypsum: now it is waste but a it could be a raw material in the future. This could contribute in reducing the negative environmental impacts construction sector produces. 17 Appendix: supplementary data 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 Table 8 - Appendix A: Studies on phosphogypsum pastes 1 Table 9-Appendix B: Studies on phosphogypsum mortars 2 Author Year Phosphogypsum (PG) mineral phase Supplementary cementitious materials Cement types Alkali-activated materials Cementitious systems % Weight the raw materials Di- PG A- PG α- PG β- PG GGBFS FA SS EAFS BA CSR SF PC PS PL CAC CSA Lime Solution SSC B T QA QI PG GGBFS FA SS EAFS BA CSR SF PC PS PL CAC CSA Lime Bijen and Niel [206] 1981 x x x x x 14-15 83-79 - - - - - - 2-3 - - - - - Valenti et al. [212] 1988 x x x x 20 - 60 - - - - - - - - - - 40 Singh and Garg [142] 1990 x x x x - - - - - - - - - - - - - - 1990 x x x x - - - - - - - - - - - - - - Singh and Garg [160] 1992 x x x x x - - - - - - - - - - - - - - 1992 x x x x - - - - - - - - - - - - - - Erdem and Ölmez [191] 1993 x x x 10-18 77-85 - - - - - - 2-5 - - - - - Singh and Garg 1993 x x x x x - - - - - - - - - - - - - - Belz et al. [189] 1994 x x x x x x x 18,5- 28,5 61,5-69 43- 50 - - - - - 10- 20 - - - - 1-28,5 Singh and Garg [149] 1995 x x x x - - - - - - - - - - - - - - Sherman et al. [307] 1995 x x x x x - - - - - - - - - - - - - - Singh, Garg, kumar [213] 1996 x x x x - - - - - - - - - - - - - - Singh and Garg [143] 1996 x x x x x 40 - 40 - - - - - - - - - - 20 Singh and Garg [146] 1997 x x x x x - - - - - - - - - - - - - - Singh and Garg [250] 2000 x x x x - - - - - - - - - - - - - Singh and Garg [150] 2000 x x x x - - - - - - - - - - - - - Singh and Garg [173] 2002 x x x x 15 75 - - - - - - 10 - - - - - 2002 x x x x x 15 75 - - - - - - 10 - - - - - 2002 x x x x 15 75 - - - - - - 10 - - - - - Min et al. [214] 2008 x x x x 15 - 68- 85 - - - - - - - - - - 15-35 Michel et al. [192] 2011 x x x x x 15,5 85 - - - - - - - - - - 73,5 - Gaiducis et al. [200] 2011 x x x x x 60-85 - - - - - 10- 20 5- 10 - 10- 20 - - - Gao et al. [308] 2012 x x x x x - - - - - - - - - - - - - - Mei et al. [201] 2012 x x x x x x - - - - - - - - - - - - - - Dvorkin et al. [309] 2013 x x x x x x 5-10 80-85 - - - - - - 5 -10 - - - - - Ding et. al. [310] 2014 x x x x x x 45 48 - 2 - - - - 5 - - - - - Ukrainczyk et al. [202] 2013 x x x x x 12,5-30 - - - 19- 38,85 13,4- 22 - - - - - - - - Rashad [215] 2015 x x x x 5-15 - 85- 100 - - - - - - - - - - - Hua, Wang, Yao [190] 2016 x x x x x x 50 30 - - - - - 3 20 - - - - 2 2016 x x x x x 50 30 - - - - - - 20 - - - - 2-6 2016 x x x x 50 30 - - - - - - 10- 50 - - - - - Gracioli et al. [311] 2017 x x x x x 10-20 89-90 - - - - - - - - - - - - Bumanis et al. [312] 2018 x x x x - - - - - - - - - - - - - - Liu and Wang [203] 2018 x x x x - - - - - - - - - - - - - - Lam [313] 2018 x x x x x x 30-60 30-60 - - - - - - 5-10 - - - - 5-10 Kramar et al. [210] 2019 x x x 20 5-10 - - - - - - - - - - - - Liu, Wang, and Yu [209] 2019 x x x x x x 15 81 - - - - - - 3 - - 1 - - Gijbels et al. [178] 2019 x x x x 10 90 - - - - - - - - - - - - Abbreviations: α- PG: α-hemihydrate; A-PG: Anhydrite; A-PG-II: Anhydrite-II; A-PG-III: Anhydrite-III; β- PG: Β-hemihydrate; B: Binary; BA: Bottom ash; CAC: Calcium aluminate cement; CRS: Carbonate silica rock; CSA:Sulfoaluminate cement; Di-PG: Dihydrate; EAFS: Eletric arc furnace fteel slags; FA: Fly ash; FGD: Flue-gas desulfurization gypsum GGBFS: Ground granulated blast-furnance slag; HCABPG: Hydraulic Cement Based on Phosphogypsum; LS: Limestone: PC: Portland Cement: PG: Phosphogypsum; PL: Portland cement limestone; PS: Portland cement slag; QA: Quaternary; QI: Quinary; RM: Red mud; SF: Silica fume; SS: Steel slag; SSC: Supersulphaated cement; T: Ternary 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 67 3 Author Year Phosphogypsum mineral phase Supplementary cementitious and Alkaliactivated Cement types Cementito us systems % Weigth the raw materials Studies parameter Di- P G A- P G A IIPG A IIIPG α- PG β- PG GGBF S S S FA R M LS Cla y Lim e P C P S P L CA C CS A B T QA PG GGBF S FA PC Lime clay SS Water/ binder Binder/Sa nd Flow (mm) Singh and Garg [160] 1992 x x x x - - - - - - - - 1:2-1:5 105 1992 x x x x - - - - - - - - 1:2-1:5 105 Singh and Garg [148] 1999 x x x x 10 - 45 - 20 25 - 0,5 1:3-1:5 100-105 Singh and Garg [174] 2001 x x x x 65- 70 - - - - - - - 1:2-1:5 - 2001 x x x x 65- 70 - - - - - - - 1:2-1:5 - 2001 x x x x 65- 70 - - - - - - - 1:2-1:5 - Mun et al. [175] 2007 x x x x 8-14 91-85 - - 1 - - 0,5 2,45 - 2007 x x x x x 2-10 87 - - 1 - - 0,5 2,45 - 2007 x x x x 12- 19,5 87- 79,5 - - 1 - - 0,5 2,45 - 2007 x x x x 11 87-79 - - 2-10 - - 0,5 2,45 - Mun and So [222] 2008 x x x x - x x 5-10 - - 87- 95 - - - 0,45 2 - 2008 x x x x x - x x 7,5 - 2,5 90 - - - 0,45 2 - 2008 x x x x x - x x 7,5 2,5 90 - - -- 0,45 2 - 2008 x x x x x x - x x 7,5 1 1,5 90 - - 0,45 2 - Huang and Li A [176] 2010 x x x x - x 25- 65 22,2- 47,8 - - - - 5-13,6 - 1:3 - Huang and Li B [177] 2010 x x x x 45 40 - - - - 5 - - Garg and Pundir [145] 2012 x x x x 8-10 - 56- 60 15- 20 5- 20 - - 1:6 1:3-1:6 - Ren et al. [158] 2012 x x x x x 33 - 42 20 3 - 2:1 1:3 342 –375 2012 x x x x x 33 - 42 20 3 - 2:1 1:3 342 –375 2012 x x x x x 33 - 42 20 3 - 2:1 1:3 342 –375 Hyung et al. [221] 2015 x x x x 7-15 92-84 - - 1-1,5 - - - - - 2015 x x x x 7-15 92-84 - - 1-1,5 - - - - - Huang et al. [314] 2016 x x x x 45 - - 3- 15 30-50 - - 1:2 1:3 - Ren et al. [159] 2017 x x x x x 52 - 33 10 3 - - 2:1 1:3 190-270 Li et al. [315] 2018 x x x x 90- 60 - 7,5- 30 - 2,5- 10 - - 0,2-1,0 1:1 -1:4 - Wang et al. [220] 2019 x x x x - - - - - - - 0,5 0,25 - Gong et al. A [156] 2020 x - x x 10- 50 - - 50- 90 - - - 0,5 0,5 175-325 2020 x x x x 10- 50 - - 50- 90 - - - 0,5 0,5 175-325 2020 x x 10- 50 - - 50- 90 - - - 0,5 0,5 175-325 Gong et al. B [157] 2020 x x x x - - - - - - - - - 274-320 Abbreviations: α- PG: α-hemihydrate; A-PG: Anhydrite; A-PG-II: Anhydrite-II; A-PG-III: Anhydrite-III; β- PG: Β-hemihydrate; B: Binary; BA: Bottom ash; CAC: Calcium aluminate cement; CRS: Carbonate silica rock; CSA:Sulfoaluminate cement; Di-PG: Dihydrate; EAFS: Eletric arc furnace fteel slags; FA: Fly ash; FGD: Flue-gas desulfurization gypsum; GGBFS: Ground granulated blast-furnance slag; HABPG:Hydraulic alternative binder phosphogypsum-based; LS: Limestone: PC: Portland Cement: PG: Phosphogypsum; PL: Portland cement limestone; PS: Portland cement slag; QA: Quaternary; QI: Quinary; RM: Red mud; SF: Silica fume; SS: Steel slag; SSC: Supersulphated cement; T: Ternary 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 68 4 5 Ref. Year Phosphogypsum mineral phases Cement types Supplementary Cementitious and Alkali-activated Materials Cementitious systems Proportion of raw materials (W%) Parameter studied PG- Di α- PG β- PG PG- A PC CSA GGBFS FA SF SS Lime Hydrated Alkaline Activating Solutions T QA QIA Replacement partial PG PC CSA GGB FS FA Lime SS SF Water/ binder Optimus water/binder Optimus % PG Fineness modulus (mm) Effect on compressive strengh in respect to the optimum replacement of phosphogypsum Fine aggregate Coarse aggregate Smaddi et al. [241] 1999 x x X 10- 90 10-90 - - - - - - - 0,6 10 - - - 1999 x x X 10- 90 10-90 - - - - - - - 0,6 10 - - - Diop et al. [244] 2005 x x x 1-10 20-80 - - - - - - - - 4,5 - - - Reddy et al. [242] 2010 x x x 10-40 60-90 - - - - - - 0,4-0,65 0,4 10 2,9 6,9 25 % increased Bagade and Satone [245] 2012 x x x 10-20 80-90 - - - - - - 0,4 0,4 10 2,9 6,9 - Yang et al.[237] 2013 x x x x x 0-55 15 - 0-35 - 2-16 - - 0,43- 0,45 0,45 55 - - - Ding et al. [238] 2014 x x x x x 45 48 - 5 - - 2 - - - 45 2,73 - - Dhinakaran and Shanthi [247] 2015 x x x 2-10 90-98 - - - - - - - 0,5 8 - - 20 % increased Nigade and Bagade[267] 2015 x x x 5-25 75-95 - - - - - - 0,45 - 0,50 10 - - - Yoo et al. [163] 2015 x x x x 12 - - 87 - 1 - - - - 12 - 6,68 - 2015 x x x x 6 - - 87 - 1 - - - - 6 - 6,68 - 2015 x x x x x 17 - - 82 - 1 - - - - 17 - 6,68 - 2015 x x x x 11 - - 85 - 4 - - - - 11 - - - 2015 x x x x 11 - - 81 - 8 - - - - 11 - - - Tian et al. [234] 2016 x x x x 40-70 - 25 - 20- 50 0-12- - - - - 45-55 - - - 2016 x x x x 40-70 - 25 20- 50 - 0-12 - - - - 45-55 - - - Buhari and Raju [161] 2016 x x x x 5-15 75-90 - - - - - 10 - 0,45 10 3,71 3,87 - Islam et al. [246] 2017 x x x 5-15 85-95 - - - - - - - 0,45 10 2,7 - - Srinivasulu and Raghava [162] 2017 x x x 2,5-10 90-97,5 - - - - - - - - 7,5 2,8 7,2 - Sukmana et al. [239] 2019 x x x x x x 35-45 26-34 - - - 8 -12 - - - - 35 - - - Abbreviations: α- PG: α-hemihydrate; A-PG: Anhydrite; A-PG-II: Anhydrite-II; A-PG-III: Anhydrite-III; β- PG: Β-hemihydrate; B: Binary; BA: Bottom ash; CAC: Calcium aluminate cement; CRS: Carbonate silica rock; CSA:Sulfoaluminate cement; Di-PG: Dihydrate; EAFS: Eletric arc furnace fteel slags; FA: Fly ash; FGD: Flue-gas desulfurization gypsum; GGBFS: Ground granulated blast-furnance slag; HABPG:Hydraulic alternative binder phosphogypsum-based; LS: Limestone: PC: Portland Cement: PG: Phosphogypsum; PL: Portland cement limestone; PS: Portland cement slag; QA: Quaternary; QI: Quinary; RM: Red mud; SF: Silica fume; SS: Steel slag; SSC: Supersulphated cement; T: Ternary Table 10-Appendix C: Studies on phosphogypsum concretes 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 69 Table 11- Appendix D: Comparison of cement-based studies with and without Portland cement - mortars Table 12- Appendix E: Comparison of cement-based studies with and without Portland cement - Concrete Mortar without cement Portland - EN 998-1, EN 998-2, EN 1015-2, EN 13454-2 and ASTM C270 PG phases Ref. Compressive strength (MPa) Flow (mm) Water/ binder Binder/ sand Setting time (min) %Formulation Initial Final PG FA GGBFS Lime PC SS LS Chemical activator Anhydrite II [175] 70 - 0,5 2,45 05:15 09:30 12 91 1 - - [221] 42 - - - - - 7 92 1 - - β-hemihydrite [148] 6,3 100-105 0,5 1:3 02:21 05:10 - - - - - - Dihydrate [175] 60 - 0,5 2,45 06:05 09:40 17 82 1 - [176] 47 - - 1:3 08:05 11:57 33 41,4 - 11,8 11,8 [177] 44,7 - - - 02:19 04:12 45 40 - - 14 1 [221] 40 - - - - - 12 87 1,5 - - [220] 20.5 - 0,5 0,25 - - - - - - - - - - Dihydrate + AnhydriteII [175] 65 - 0,5 2,45 05:30 09:15 12 (6+6) 87 1 - Mortar with cement Portland EN 998-1, EN 998-2, EN 1015-2, EN 13454-2 and ASTM C270 PG phases Ref. Compressive strength (MPa) Flow mm Water/ binder Binder/ sand Setting time (min) %Formulation Initial Final PG FA GGBFS Lime PC SS LS Chemical activator α-hemihydrite [158] 20 342-375 2:1 1:3 1404 2020 33 42 - 3 20 - - [145] 10 - 1:6 1:3 22 38 15 60 - 5 20 - - Anhydrite II [159] al. 2017 31,8 190-270 2:1 1:3 - - 52 33 3 10 - - 2 [222] 72 - 0,4 2 - - 7,5 2,5 90 - - [158] 30 342-375 2:1 1:3 517 705 33 42 3 20 - - [156] 40 275 0,5 0,5 - - 30 70 - - Anhydrite III [222] 70 - 0,45 2 - - 7,5 2,5 90 - - [156] 30 325 0,5 0,5 - - - - β-hemihydrite [160] 13,31 105 - - 70 145 - - - - - - - - [222] 60 - 0,45 2 - - 7,5 1,5 1 9 - - [157] 36,71 283 - 311 493 - - - - - - - β-hemi (PC+PG) [222] 70 - 0,45 2 - - 7,5 92,5 - - Dihydrate [158] 15 342-375 2:1 1:3 2934 3625 33 42 - 3 20 - - [243] 45 - 1:2 1:3 - - 45 48 7 - - [156] 34,5 225 0,5 0,5 - - 30 70 - - 1 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 70 Table 13- Appendix F: Water-resistant phosphogypsum- Kovler (2001); Piasta and Zarzycki (2017); Selim et al (2020) Replacement Portland cement by phosphogypsum – EN 206:2013 and Eurocode 1992 Phosphogypsum phases Ref. Compressive strength (MPa) Strength class (MPa) Setting time %Formulation Water/binder Fineness modulus Effect on compressive strength in respect to the optimum replacement of PG Initial Final PC PG GGBFS Lime SS Chemical activator aluminium powder Fine aggregate coarse aggregate β-hemihydrite [246] 36,33 175 285 90 10 - - - - - 0,45 2,7 - - Dihydrate [244] 32,5 32,5 - - 90 4,5 - - - - - - - - - [242] 49,3 53 35 190 90 10 - - - - - 0,4 2,9 6,9 25 % increased [245] 30,15 53 185 870 90 10 - - - - - 0,4 2,9 6,9 - [247] 27,93 - - - 92 8 - - - - - 0,5 - - 20 % increased [246] 32,2 - 72 159 90 10 - - - - - 0,45 2,7 - - [162] 38,78 - - - 92,5 7,5 - - - - - - 2,8 7,2 - Concrete with cementitious systems - EN 206:2013 and Eurocode 1992 Phosphogypsum phases Ref. Compressive strength (MPa) Strength class (MPa) Setting time %Formulation Water/binder Fineness modulus Effect on compressive strength in respect to the optimum replacement of PG Initial Final PC PG GGBFS Lime SS Chemical activator aluminium powder Fine aggregate coarse aggregate Dihydrate +Anhydrite II [163] 70 - - - - 12 (6+6) 87 1 - - 6,8 - Anhydrite II [163] 80 - - - - 12 87 1 - - 6,8 - Dihydrate [237] 48 - - - 15 55 30 7 1,6 0,074 - - - - [238] 38,6 - - - 5 45 48 - 2 - 2,73 - - [163] 75 - - - - 17 82 1 - - 6,8 - [316] 2,92 - - - 25 50 25 - - - - [239] 20,93 - - - 34 35 10 - - - - 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 71 Table 15- Appendix G: Expansibility cold (soundness) – EN 15743/ EN 197-1 Mineral phase Water Absorption (%) Compressive Strength (MPa) Composite binder Water resistance coeficient Ref. Di-PG 2 32 PG+PC Unknown [156] 3 22 PG+PC+GGBFS Unknown [160] 10 19.6 PG+PC+FA Unknown [160] 28 12 PG+FA+Lime Unknown [256] α-PG 32 Unknown PG Unknown [257] 14 27.5 PG+FA+PC+Lime Unknown [145] 19,6 26.4 PG+FA+Lime 50ºC Unknown [141] 4 32.2 PG+ GGBFS+ Marble dust+ chemical activator 50ºC Unknown [141] β-PG leaching 13.20 PG Unknown [142] Leaching 14 PG Unknown [171] 30 Unknown PG Unknown [257] 30.5 Unknown PG 0.35 [258] 32.09 Unknown PG Unknown [143] 42 Unknown PG Unknown [259] 7 35 PG+PC+GGBFS Unknown [259] 3 35 PG+PC+GGBFS Unknown [160] 2,89 Unknown PG+PC+GGBFS+ organic retarder Unknown [143] 6 35 PG+PC+GGBFS+ organic retarder Unknown [249], [257] 10 21.3 PG+FA+PC Unknown [160] 6.5 25.6 PG+FA+PC Unknown [171] 10 20 PG+FA+PC 50ºC Unknown [146] 8 25 PG+FA+PC+ retarder Unknown [250] 21.6 4.11 PG + FA + Lime Unknown [260] 10 12 PG+FA+Lime Unknown [256] 22 Unknown 70%PG+30%SAC 0.70 [258] A-PG III Unknown Unknown Unknown Unknown Unknown A-PGII 10.43 56 PG+GGBFS Unknown [142] 9.55 35 PG+GGBFS Unknown [149] 4 34.5 PG+PC Unknown [156] A-PG I 4 40 PG+FA+Red Mud Unknown [21] 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 72 Mineral phases Soundness Cold expansion (mm) %PG System binder Ref. Di-PG 0.6 100 PG Plaster [174] 6 12.5 PG +PC Parcial replacement [27] 8 20 PG +PC Parcial replacement [245] 9 25 PG +PC Parcial replacement [267] 3.6 Unknown PG+PC+GGBFS HABPG [160], [174] 1.10 Unknown PG+PC+FA HABPG [160], [174] 1.6 Unknown PG+PC+Red Mud HABPG [174] 13.5 45 PG+PC+GGBFS+SS+LS+NaOH Supersulphated [177] α-PG 2.3 10 PG+PC+FA+Lime HABPG [145] β-PG 0.6 100 PG Plaster [174], [171] 0.1 70 PG+PC Parcial replacement [241] 0.106 70 PG+PPC POZOLANIC Parcial replacement [241] 1.60 Unknown PG+PC+GGBFS HABPG [160], [174], [259] 1.60 Unknown PG+PC+GGBFS+ organic retarder HABPG [249], [257] 1.2 15 PG + GGBFS + tartaric acid HABPG [173] 0.88 Unknown PG+PC+FA HABPG [160] 0.88 Unknown PG+FA+PC+ retarder HABPG [250] 1.00 Unknown PG+PC+FA HABPG [174] 1.8 Unknown PG+PC+FA 50ºC HABPG [146] 2.0 70-75 PG+FA+PC+Lime+chemical additives HABPG [171] 1.1 Unknown PG+PC+Red Mud HABPG [174] 12 45 PG+GGBFS+SS+SL Supersulphated [176] A-PG III Unknown Unknown Unknown Unknown Unknown A-PG II 0.5 100 PG Plaster [142] 0.81 70 PG+GGBFS HABPG [149] 0.90 15 PG+GGBFS HABPG [173] 1.5 Unknown PG+ GGBFS+ chemical activator HABPG [142] 3.0 10 PG+PC+GGBFS Supersulphated [191] A-PG I 0.06 PG+ GGBFS+2% citric acid HABPG [21] 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 73 Ref Year Phosphogypsum mineral phase Raw materials Cement types Cement based Phosphogypsum Application in other materials building studies Studies parameter Di- PG A- PG α- PG β- PG GGBFS FA SS Clay Adobe Fibre RM Fluoro LH PC CSA CAC B T QA QI Road base Brick and Blocks Tile and Ceramic Binder Plaster Selfleveling Board %PG Optimum % replacement of phosphogypsum Study objective Sing and Garg [249] 1992 x x x - - Binder Singh and Garg [160] 1992 x x x x x x x - - Binder x x x x x x x - - Binder Sing and Garg [259] 1993 x x x x x x - - Binder Singh and Garg [257] 1994 x x x x - - Plaster board Gutti et al. [317] 1996 x x x x 8 - Soil stabilization Singh and Garg [165] 2000 x x - - Anhydrite cement Kumar [318] 2000 x x x x x 10 -40 - Binder Sing and Garg [174] 2001 x x - - Plastering Kumar [260] 2003 x x x x x 10 -40 - Binder Péra and Ambroise [319] 2004 x x x x x 25 - Topping mortar Singh and Garg [144] 2005 x x - - Plastering x x x x x x - - Tile Degirmenci et al. [278] 2007 x x x x x 2,5-5 - Soil stabilization She et al. [320] 2007 x x x x x 0-92 18-23 Binder Degirmenci et al A [261] 2008 x x x x x 10 -50 - Binder x x x x x 10-50 - Binder Degirmenci et al B [321] 2008 x x x - 25 adobe stabilization Ajam et al. [322] 2009 x x x 5-40 30 Fired clay W. Shen et al. [323] 2009 x x x x x 0-5 2,5 solidified material Yang et al. [324] 2009 x x x x x 35 - Wall x x x x x 30-50 30 Wall Garg, Minocha and Jain [96] 2011 x x - 100 plaster x x x x - - Multiphase x x x x x x x x - - Binder x x x - - boards x x x x x - - flooring Garg and Pudir [145] 2012 x x - 100 Plaster x x x x x x 8-15 15 Binder x x x x x x 8-15 15 Binder Zhou et al. [325] 2014 x x x x 75 65-85 Non-fired Yang, Zhang and Yan [265] 2016 x x x x x 40-55 45 mortar Zhou et al. [326] 2016 x x - 100 Green tiles Dutta et al. [273] 2017 x x x x x 2-10 2 geotechnical Schaefer et al. [116] 2017 x x x x x - 19,8 underlayments 2017 x x x x x - 19,8 underlayments Contreras et al. [327] 2018 x x x 5-10 7,5 Additive Ding, Jianwen et al. [275] 2019 x x x x x - 26 Stabilized soil Qiang et al [274] 2019 x x x x x x - - mortar Romero-Hermida et al. [328] 2019 x x - - Rheological characteriza lime Silva et al. [329] 2019 x x x x - 11 Pavements 2019 x x x x - 41 Pavements Zhou et al. [276] 2019 x x - 100 non-fired ceramic Abbreviations: α- PG: α-hemihydrate; A-PG: Anhydrite; A-PG-II: Anhydrite-II; A-PG-III: Anhydrite-III; β- PG: Β-hemihydrate; B: Binary; BA: Bottom ash; CAC: Calcium aluminate cement; CRS: Carbonate silica rock; CSA:Sulfoaluminate cement; Di-PG: Dihydrate; EAFS: Eletric arc furnace fteel slags; FA: Fly ash; FGD: Flue-gas desulfurization gypsum; GGBFS: Ground granulated blast-furnance slag; HABPG:Hydraulic alternative binder phosphogypsum-based; LS: Limestone: PC: Portland Cement: PG: Phosphogypsum; PL: Portland cement limestone; PS: Portland cement slag; QA: Quaternary; QI: Quinary; RM: Red mud; SF: Silica fume; SS: Steel slag; SSC: Supersulfated cement; T: Ternary Table 16-Appendix H: Application of phosphogypsum in geotechnical materials in construction and components 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 80 Formation and exploration: An invited paper,” Econ. Geol., vol. 112, no. 3, pp. 483–516, 2017, doi: 10.2113/econgeo.112.3.483. [94] P. BECKER, Phosphates and phosphoric acid. raw materials, technology, and economics of the wet process. 1989. [95] S. Çoruh and O. N. Ergun, “Use of fly ash, phosphogypsum and red mud as a liner material for the disposal of hazardous zinc leach residue waste,” J. Hazard. Mater., vol. 173, no. 1–3, pp. 468–473, Jan. 2010, doi: 10.1016/J.JHAZMAT.2009.08.108. [96] M. Garg, A. K. Minocha, and N. Jain, “Environment hazard mitigation of waste gypsum and chalk: Use in construction materials,” Constr. Build. Mater., vol. 25, no. 2, pp. 944–949, 2011, doi: 10.1016/j.conbuildmat.2010.06.088. [97] N. F. Mihelja, U. N, S. Leakoviæc, and J. Šipušiæb, “Waste Phosphogypsum–Toward Sustainable Reuse in Calcium Sulfoaluminate Cement Based BuildingMaterials,” Chem. Biochem. Eng. Quaterly, vol. 27, no. 2, pp. 219–226, 2013, [Online]. Available: http://hrcak.srce.hr/index.php?show=clanak&id_clanak_jezik=154035. [98] Y. Shen, J. Qian, J. Chai, and Y. Fan, “Calcium sulphoaluminate cements made with phosphogypsum: Production issues and material properties,” Cem. Concr. Compos., vol. 48, pp. 67–74, 2014, doi: 10.1016/j.cemconcomp.2014.01.009. [99] H. El-Didamony, H. S. Gado, N. S. Awwad, M. M. Fawzy, and M. F. Attallah, “Treatment of phosphogypsum waste produced from phosphate ore processing,” J. Hazard. Mater., vol. 244–245, pp. 596–602, Jan. 2013, doi: 10.1016/J.JHAZMAT.2012.10.053. [100] T. Mashifana, F. N. Okonta, and F. Ntuli, “Geotechnical properties and application of lime modified phosphogypsum waste,” Medziagotyra, vol. 24, no. 3, pp. 312–318, 2018, doi: 10.5755/j01.ms.24.3.18232. [101] K. Gijbels et al., “Radiological and leaching assessment of an ettringite-based mortar from ladle slag and phosphogypsum,” Cem. Concr. Res., vol. 128, p. 105954, Feb. 2020, doi: 10.1016/J.CEMCONRES.2019.105954. [102] P. M. Rutherford, M. J. Dudas, and J. M. Arocena, “Trace elements and fluoride in phosphogypsum leachates,” Environ. Technol. (United Kingdom), vol. 16, no. 4, pp. 343–354, 1995, doi: 10.1080/09593331608616276. [103] G. Bumanis, J. Zorica, D. Bajare, and A. Korjakins, “Technological properties of phosphogypsum binder obtained from fertilizer production waste,” Energy Procedia, vol. 147, pp. 301–308, Aug. 2018, doi: 10.1016/J.EGYPRO.2018.07.096. [104] F. do C. Holanda, H. Schmidt, and V. A. Quarcioni, “Influence of phosphorus from phosphogypsum on the initial hydration of Portland cement in the presence of superplasticizers,” Cem. Concr. Compos., vol. 83, pp. 384–393, Oct. 2017, doi: 10.1016/j.cemconcomp.2017.07.029. [105] A. Arman and R. K. Seals, “A preliminary assessment of utilization alternatives for phosphogypsum,” in Proceedings of the Third International Symposium on Phosphogypsum, 1990, pp. 562–575. [106] P. S. Bardella and G. Camarini, “Recycled Plaster: Physical and Mechanical Properties,” Adv. Mater. Res., vol. 374–377, pp. 1307–1310, 2011, doi: 10.4028/www.scientific.net/AMR.374-377.1307. [107] S. Suárez, X. Roca, and S. Gasso, “Product-specific life cycle assessment of recycled gypsum as a replacement for natural gypsum in ordinary Portland cement: Application to the Spanish context,” J. Clean. Prod., vol. 117, pp. 150–159, 2016, doi: 10.1016/j.jclepro.2016.01.044. [108] R. H. Moreira, F. S. Queiroga, H. A. Paiva, N. H. Medina, G. Fontana, and M. A. Guazzelli, “Extraction of natural radionuclides in TENORM waste phosphogypsum,” J. Environ. Chem. Eng., vol. 6, no. 5, pp. 6664–6668, Oct. 2018, doi: 10.1016/J.JECE.2018.10.019. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 81 [109] UNSCEAR, Source and effects of ionizing radiation, vol. 36, no. 3. 2008. [110] M. Singh, “Effect of phosphatic and fluoride impurities of phosphogypsum on the properties of selenite plaster,” Cem. Concr. Res., vol. 33, no. 9, pp. 1363–1369, 2003, doi: 10.1016/S0008-8846(03)00068-1. [111] A. A. Tabikh and F. M. Miller, “The nature of phosphogypsum impurities and their influence on cement hydration,” Cem. Concr. Res., vol. I, pp. 663–678, 1971. [112] Y. Huang et al., “Influence of phosphorus impurities on the performances of calcium sulfoaluminate cement,” Constr. Build. Mater., vol. 149, pp. 37–44, Sep. 2017, doi: 10.1016/j.conbuildmat.2017.05.028. [113] S. M. Luther, M. J. Dudas, and P. M. Rutherford, “Radioactivity and chemical characteristics of Alberta phosphpgypsum,” Water, Air, Soil Pollut., vol. 69, pp. 277–290, 1992. [114] A. Pundir, M. Garg, and R. Singh, “Evaluation of properties of gypsum plastersuperplasticizer blends of improved performance,” J. Build. Eng., vol. 4, pp. 223–230, Dec. 2015, doi: 10.1016/J.JOBE.2015.09.012. [115] P. M. Rutherford, M. J. Dudas, and R. A. Samek, “Environmental impacts of phosphogypsum,” Sci. Total Environ., vol. 149, no. 1–2, pp. 1–38, Jun. 1994, doi: 10.1016/0048-9697(94)90002-7. [116] C. O. Schaefer, M. Cheriaf, and J. C. Rocha, “Production of synthetic phosphoanhydrite and its use as a binder in self-leveling underlayments (SLU),” Materials (Basel)., vol. 10, no. 8, 2017, doi: 10.3390/ma10080958. [117] H. Ölmez and E. Erdem, “The effects of phosphogypsum on the setting and mechanical properties of Portland cement and trass cement,” Cem. Concr. Res., vol. 19, no. 3, pp. 377–384, May 1989, doi: 10.1016/0008-8846(89)90027-6. [118] Y. Huang, J. Qian, J. Liang, N. Liu, F. Li, and Y. Shen, “Characterization and calorimetric study of early-age hydration behaviors of synthetic ye’elimite doped with the impurities in phosphogypsum,” J. Therm. Anal. Calorim., vol. 123, no. 2, pp. 1545–1553, 2016, doi: 10.1007/s10973-015-5009-y. [119] C. Koopman, “Purification of gypsum from the phophoric acid production by recrystallization with simultaneous extraction,” Delft, 2001. [120] M. Singh, M. Garg, and S. S. Rehsi, “Purifying phosphogypsum for cement manufacture,” Constr. Build. Mater., vol. 7, no. 1, pp. 3–7, 1992, doi: 10.1016/0950- 0618(93)90018-8. [121] H. Ölmez and V. T. Yilmaz, “Infrared study on the refinement of phosphogypsum for cements,” Cem. Concr. Res., vol. 18, no. 3, pp. 449–454, May 1988, doi: 10.1016/0008- 8846(88)90079-8. [122] J. H. Potgieter, S. S. Potgieter, R. I. McCrindle, and C. A. Strydom, “An investigation into the effect of various chemical and physical treatments of a South African phosphogypsum to render it suitable as a set retarder for cement,” Cem. Concr. Res., vol. 33, no. 8, pp. 1223–1227, Aug. 2003, doi: 10.1016/S0008-8846(03)00036-X. [123] S. Liu, J. Ouyang, and J. Ren, “Mechanism of calcination modification of phosphogypsum and its effect on the hydration properties of phosphogypsum-based supersulfated cement,” Constr. Build. Mater., vol. 243, p. 118226, May 2020, doi: 10.1016/J.CONBUILDMAT.2020.118226. [124] G. Aouad, A. Laboudigue, N. Gineys, and N. E. Abriak, “Dredged sediments used as novel supply of raw material to produce Portland cement clinker,” Cem. Concr. Compos., vol. 34, no. 6, pp. 788–793, 2012, doi: 10.1016/j.cemconcomp.2012.02.008. [125] C. Zhu, J. Zhang, W. Yi, W. Cao, J. Peng, and J. Liu, “Research on degradation mechanisms of recycled building gypsum,” Constr. Build. Mater., vol. 173, pp. 540–549, Jun. 2018, doi: 10.1016/J.CONBUILDMAT.2018.04.060. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 82 [126] N. Lushnikova and L. Dvorkin, “25 - Sustainability of gypsum products as a construction material,” in Sustainability of Construction Materials (Second Edition), Second Edi., J. M. Khatib, Ed. Woodhead Publishing, 2016, pp. 643–681. [127] C. W. Berish, “Potential environmental harzads of phosphogypsum storage in central Florida,” Miami, Florida, USA, 1990. [128] I. Aydin, F. Aydin, A. Saydut, E. G. Bakirdere, and C. Hamamci, “Hazardous metal geochemistry of sedimentary phosphate rock used for fertilizer (Mazıdag, SE Anatolia, Turkey),” Microchem. J., vol. 96, no. 2, pp. 247–251, Nov. 2010, doi: 10.1016/J.MICROC.2010.03.006. [129] H. Ohtake and S. Tsuneda, Phosphorus Recovery and Recyclings, 1 first. Tokyo: Springer, 2019. [130] H. Tayibi, M. Choura, F. A. López, F. J. Alguacil, and A. López-Delgado, “Environmental impact and management of phosphogypsum,” J. Environ. Manage., vol. 90, no. 8, pp. 2377–2386, Jun. 2009, doi: 10.1016/J.JENVMAN.2009.03.007. [131] M. Rentería-Villalobos, I. Vioque, J. Mantero, and G. Manjón, “Radiological, chemical and morphological characterizations of phosphate rock and phosphogypsum from phosphoric acid factories in SW Spain,” J. Hazard. Mater., vol. 181, no. 1–3, pp. 193–203, Sep. 2010, doi: 10.1016/J.JHAZMAT.2010.04.116. [132] M. Rentería-Villalobos, I. Vioque, J. Mantero, and G. Manjón, “Radiological, chemical and morphological characterizations of phosphate rock and phosphogypsum from phosphoric acid factories in SW Spain,” J. Hazard. Mater., vol. 181, no. 1–3, pp. 193–203, Sep. 2010, doi: 10.1016/J.JHAZMAT.2010.04.116. [133] B. Mazzilli, V. Palmiro, C. Saueia, and M. . Nisti, “Radiochemical characterization of Brazilian phosphogypsum,” J. Environ. Radioact., vol. 49, no. 1, pp. 113–122, May 2000, doi: 10.1016/S0265-931X(99)00097-1. [134] C. H. R. Saueia and B. P. Mazzilli, “Distribution of natural radionuclides in the production and use of phosphate fertilizers in Brazil,” J. Environ. Radioact., vol. 89, no. 3, pp. 229–239, Jan. 2006, doi: 10.1016/J.JENVRAD.2006.05.009. [135] W. C. Burnett, M. K. Schultz, and C. D. Hull, “Radionuclide flow during the conversion of phosphogypsum to ammonium sulfate,” J. Environ. Radioact., vol. 32, no. 1–2, pp. 33–51, Jul. 1996, doi: 10.1016/0265-931X(95)00078-O. [136] R. N. González-Kunz, P. Pineda, A. Bras, and L. Morillas, “Plant biomass ashes in cement-based building materials. Feasibility as eco-efficient structural mortars and grouts,” Sustain. Cities Soc., vol. 31, pp. 151–172, 2017, doi: 10.1016/j.scs.2017.03.001. [137] RILEM13-M, 13-MR committee on mortars and Renderings: Determination of mortar consistence, vol. 11. 1982, pp. 207–216. [138] M. L. Kosmatka, Steven H.; Wilson, Design and Control of Concrete Mixtures -The Guide to Applications, Methods and Materials. 2011. [139] M. Kismi, J. C. Saint-Arroman, and P. Mounanga, “Minimizing water dosage of superplasticized mortars and concretes for a given consistency,” Constr. Build. Mater., vol. 28, no. 1, pp. 747–758, 2012, doi: 10.1016/j.conbuildmat.2011.10.056. [140] ASTM C219, “Standard terminology relating to Hydraulic Cement,” ASTM (American Soc. Test. Mater., pp. 1–4, 2019, doi: 10.1520/C0219-19. [141] M. Garg, N. Jain, and M. Singh, “Development of alpha plaster from phosphogypsum for cementitious binders,” Constr. Build. Mater., vol. 23, no. 10, pp. 3138– 3143, 2009, doi: 10.1016/j.conbuildmat.2009.06.024. [142] M. Singh, M. Garg, and S. S. Rehsi, “Durability of phosphogypsum based waterresistant anhydrite binder,” Cem. Concr. Compos., vol. 20, pp. 271–276, 1990. [143] M. Singh and M. Garg, “Relationship between mechanical properties and porosity of water-resistant gypsum binder,” Cem. Concr. Res., vol. 26, no. 3, pp. 449–456, 1996, doi: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 83 10.1016/S0008-8846(96)85032-0. [144] M. Singh and M. Garg, “Study on anhydrite plaster from waste phosphogypsum for use in polymerised flooring composition,” Constr. Build. Mater., vol. 19, no. 1, pp. 25–29, Feb. 2005, doi: 10.1016/J.CONBUILDMAT.2004.04.038. [145] M. Garg and A. Pundir, “Comprehensive study of fly ash binder developed with fly ash - Alpha gypsum plaster - Portland cement,” Constr. Build. Mater., vol. 37, pp. 758–765, 2012, doi: 10.1016/j.conbuildmat.2012.08.018. [146] M. Singh and M. Garg, “Durability of cementitious binder derived from industrial wastes,” Mater. Struct., vol. 30, no. 10, pp. 607–612, 1997, doi: 10.1007/bf02486902. [147] M. Singh and M. Garg, “Phosphogypsum — Fly ash cementitious binder — Its hydration and strength development,” Cem. Concr. Res., vol. 25, no. 4, pp. 752–758, May 1995, doi: 10.1016/0008-8846(95)00065-K. [148] M. Singh and M. Garg, “Cementitious binder from fly ash and other industrial wastes,” Cem. Concr. Res., vol. 29, no. 3, pp. 309–314, 1999, doi: 10.1016/S0008- 8846(98)00210-5. [149] M. Singh and M. Garg, “Activation of Gypsum Anhydrite-Slag Mixtures,” Cem. Concr. Res., vol. 25, no. 2, pp. 332–338, 1995. [150] M. Singh, “Influence of blended gypsum on the properties of Portland cement and Portland slag cement,” Cem. Concr. Res., vol. 30, no. 8, pp. 1185–1188, 2000, doi: 10.1016/S0008-8846(00)00306-9. [151] P. Aïtcin, High-Performance Concrete. . [152] H. Carasek, “Argamassas,” Mater. construçao Civ. e Princípios Ciência dos Mater., pp. 863– 904, 2007. [153] ASTM C230, “Standard Specification for Flow table for use in tests of hydraulic cement,” ASTM, pp. 4–9, 2020, doi: 10.1520/C0230. [154] ASTM C143, “Standard Test Method for Slump of Hydraulic-Cement Concrete,” ASTM (American Soc. Test. Mater., vol. 04, no. 1, pp. 1–4, 2015, doi: 10.1520/C0143. [155] Y. Ghasemi, M. Emborg, and A. Cwirzen, “Exploring the relation between the flow of mortar and specific surface area of its constituents,” Constr. Build. Mater., vol. 211, pp. 492– 501, 2019, doi: 10.1016/j.conbuildmat.2019.03.260. [156] X. Gong, J. Liu, Z. Sun, and F. Li, “Effects of Phosphogypsum and Calcined Phosphogypsum Content on the Basic Physical and Mechanical Properties of Portland Cement Mortar,” J. Test. Eval., vol. 48, no. 5, p. 20180380, 2020, doi: 10.1520/jte20180380. [157] X. Gong, J. Liu, T. Zhang, and Z. Jiao, “Effect of Modified Phosphogypsum on Properties of Cement Mortar,” J. Test. Eval., vol. 48, no. 4, p. 20180702, 2020, doi: 10.1520/jte20180702. [158] T. Ren, J. Zhu, W. Liu, X. Zhu, and J. Yang, “Water resistance of composite binders containing phosphogpysum with different pretreatment processes,” Adv. Cem. Res., vol. 24, no. 2, pp. 111–120, 2012, doi: 10.1680/adcr.11.00014. [159] T. Ren, R. Xi, and Jiankuan Yang, “Influence of autoclave pretreatment on the properties of phosphogypsum-based composite binders,” IOP Conf. Ser. Earth Environ. Sci., vol. 81, 2017, doi: doi :10.1088/1755-1315/81/1/012015. [160] M. Singh and M. Garg, “Investigation of a durable gypsum binder for building materials,” Constr. Build. Mater., vol. 6, no. 1, pp. 52–56, 1992, doi: 10.1016/0950- 0618(92)90030-3. [161] B. M. Buhari and R. Raju, “Flexural Behavior of Ternary Blended Concrete Containing Silica Fume and Phosphogypsum,” Int. J. Sci. Technol. Eng., vol. 3, no. 03, pp. 9– 14, 2016. [162] K. Srinivasulu and P. Raghava, “EFFECTS OF PARTIAL REPLACEMENT OF CEMENT WITH PHOSPHOGYPSUM ON STRENGTH CHARACTERISTICS OF 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 84 CONCRETE,” Int. J. Eng. Res., vol. 5, no. 2, 2017. [163] S. Yoon, K. Mun, and W. Hyung, “Physical Properties of Activated Slag Concrete Using Phosphogypsum and Waste Lime as an Activator,” J. Asian Archit. Build. Eng., vol. 14, no. 1, pp. 189–195, 2015, doi: 10.3130/jaabe.14.189. [164] P. K. Mehta and P. J. M. Monteiro, Concrete: Microstruture, properties and Materials. 2006. [165] M. Singh and M. Garg, “Making of anhydrite cement from waste gypsum,” Cem. Concr. Res., vol. 30, no. 4, pp. 571–577, 2000, doi: 10.1016/S0008-8846(00)00209-X. [166] B. Ma, W. Lu, Y. Su, Y. Li, C. Gao, and X. He, “Synthesis of α-hemihydrate gypsum from cleaner phosphogypsum,” J. Clean. Prod., vol. 195, pp. 396–405, 2018, doi: 10.1016/j.jclepro.2018.05.228. [167] M. Singh and M. Garg, “Making of anhydrite cement from waste gypsum,” Cem. Concr. Res., vol. 30, no. 4, pp. 571–577, Apr. 2000, doi: 10.1016/S0008-8846(00)00209-X. [168] R. H. Geraldo et al., “Calcination parameters on phosphogypsum waste recycling,” Constr. Build. Mater., vol. 256, p. 119406, 2020, doi: 10.1016/j.conbuildmat.2020.119406. [169] J. Newman and B. Seng Choo, Advanced Concrete Technology Constituent Materials, First. Oxford: Elsevier, 2003. [170] EUROPEAN COMMITTEE FOR STANDARDIZATION, “EN 13279-2:2014. Gypsum binders and gypsum plasters - Part 2: Test Methods.,” Eur. Stand. En 13279-2, pp. 3–23, 2014. [171] M. Garg, A. K. Minocha, and N. Jain, “Environment hazard mitigation of waste gypsum and chalk: Use in construction materials,” Constr. Build. Mater., vol. 25, no. 2, pp. 944–949, Feb. 2011, doi: 10.1016/J.CONBUILDMAT.2010.06.088. [172] M. Singh, M. Garg, C. L. Verma, S. K. Handa, and R. Kumar, “An improved process for the purification of phosphogypsum,” Constr. Build. Mater., vol. 10, no. 8, pp. 597–600, 1996, doi: 10.1016/S0950-0618(96)00019-0. [173] M. Singh and M. Garg, “Calcium sulfate hemihydrate activated low heat sulfate resistant cement,” Constr. Build. Mater., vol. 16, no. 3, pp. 181–186, 2002, doi: 10.1016/S0950- 0618(01)00026-5. [174] M. Singh and M. G. And, “Phosphogypsum Based Composite Binders,” J. Sci. Ind. Res. (India)., vol. 60, no. 10, pp. 812–817, 2001. [175] K. J. Mun, W. K. Hyoung, C. W. Lee, S. Y. So, and Y. S. Soh, “Basic properties of non-sintering cement using phosphogypsum and waste lime as activator,” Constr. Build. Mater., vol. 21, no. 6, pp. 1342–1350, 2007, doi: 10.1016/j.conbuildmat.2005.12.022. [176] Y. Huang and Z. Lin, “Investigation on phosphogypsum–steel slag–granulated blastfurnace slag–limestone cement,” Constr. Build. Mater., vol. 24, no. 7, pp. 1296–1301, Jul. 2010, doi: 10.1016/J.CONBUILDMAT.2009.12.006. [177] Y. Huang and Z. Lin, “Effect of sodium hydroxide on the properties of phosphogypsum based cement,” J. Wuhan Univ. Technol. Mater. Sci. Ed., vol. 25, no. 2, pp. 342–345, 2010, doi: 10.1007/s11595-010-3342-7. [178] K. Gijbels, R. I. Iacobescu, Y. Pontikes, S. Schreurs, and W. Schroeyers, “Alkaliactivated binders based on ground granulated blast furnace slag and phosphogypsum,” Constr. Build. Mater., vol. 215, pp. 371–380, 2019, doi: 10.1016/j.conbuildmat.2019.04.194. [179] EN15743, “Supersulfated cement - Composition, specifications and conformity criteria,” CEN (Comité Eur. Norm., p. 28, 2010. [180] P. C. Nkinamubanzi and P.-C. Aïtcin, “The Use of Slag in Cement and Concrete in a Sustainable Development Perspective,” in WABE International Symposium on Cement and Concrete, 1999, pp. 85–110. [181] P.-C. Aitcin and S. Mindess, Sustainability of concrete, vol. 17 H1-B. 2011. [182] M. Glavind, “Sustainability of cement, concrete and cement replacement materials in construction,” in Sustainability of construction materials, J. M. Khatib, Ed. Oxford: Woodhead 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 85 Publishing Limited, 2009, p. 313. [183] P. C. Aitcin, “Binder and concrete of yesterday,” in Binder for Durable and Sustainable Concrete, vol. 91, London: Taylor & Francis, 2008. [184] M. C. G. Juenger, F. Winnefeld, J. L. Provis, and J. H. Ideker, “Advances in alternative cementitious binders,” Cem. Concr. Res., vol. 41, no. 12, pp. 1232–1243, 2011, doi: 10.1016/j.cemconres.2010.11.012. [185] E. Gartner, “Industrially interesting approaches to ‘low-CO2’ cements,” Cem. Concr. Res., vol. 34, no. 9, pp. 1489–1498, 2004, doi: 10.1016/j.cemconres.2004.01.021. [186] J. Rosales et al., “Treated phosphogypsum as an alternative set regulator and mineral addition in cement production,” J. Clean. Prod., vol. 244, p. 118752, Jan. 2020, doi: 10.1016/J.JCLEPRO.2019.118752. [187] X. Chen, J. Gao, C. Liu, and Y. Zhao, “Effect of neutralization on the setting and hardening characters of hemihydrate phosphogypsum plaster,” Constr. Build. Mater., vol. 190, pp. 53–64, 2018, doi: 10.1016/j.conbuildmat.2018.09.095. [188] X. Chen, J. Gao, and Y. Zhao, “Investigation on the hydration of hemihydrate phosphogypsum after post treatment,” Constr. Build. Mater., vol. 229, p. 116864, Dec. 2019, doi: 10.1016/J.CONBUILDMAT.2019.116864. [189] G. Belz, J. Beretka, R. Cioffi, L. Santoro, N. Sherman, and G. L. Valenti, “Development of Cementitious Products Using Industrial Process Wastes as Sources of Reactive Sulfate and Alumina,” Stud. Environ. Sci., vol. 60, pp. 579–588, Jan. 1994, doi: 10.1016/S0166-1116(08)71491-2. [190] S. Hua, K. Wang, and X. Yao, “Developing high performance phosphogypsumbased cementitious materials for oil-well cementing through a step-by-step optimization method,” Cem. Concr. Compos., vol. 72, pp. 299–308, Sep. 2016, doi: 10.1016/j.cemconcomp.2016.05.017. [191] E. Erdem and H. Ölmez, “The mechanical properties of supersulphated cement containing phosphogypsum,” Cem. Concr. Res., vol. 23, no. 1, pp. 115–121, Jan. 1993, doi: 10.1016/0008-8846(93)90141-U. [192] M. Michel, J.-F. Georgin, J. Ambroise, and J. Péra, “The influence of gypsum ratio on the mechanical performance of slag cement accelerated by calcium sulfoaluminate cement,” Constr. Build. Mater., vol. 25, no. 3, pp. 1298–1304, Mar. 2011, doi: 10.1016/J.CONBUILDMAT.2010.09.015. [193] Z. C. S. S. K. T. Faber, “A hierarchical study of the mechanical properties of gypsum,” no. January, pp. 4444–4453, 2010, doi: 10.1007/s10853-010-4527-z. [194] M. Sophia and N. Sakthieswaran, “Gypsum as a Construction Material- A Review of Recent Developments,” Int. J. Innov. Res. Sci. Technol., vol. 2, no. 12, pp. 315–323, 2016. [195] A. J. Lewry and J. Williamson, “The setting of gypsum plaster - Part II The development of microstructure and strength,” J. Mater. Sci., vol. 29, no. 21, pp. 5524–5528, 1994, doi: 10.1007/BF00349943. [196] T. Sievert, A. Wolter, and N. B. Singh, “Hydration of anhydrite of gypsum (CaSO4.II) in a ball mill,” Cem. Concr. Res., vol. 35, no. 4, pp. 623–630, Apr. 2005, doi: 10.1016/J.CEMCONRES.2004.02.010. [197] S. M. A. El-Gamal and F. A. Selim, “Utilization of some industrial wastes for ecofriendly cement production,” Sustain. Mater. Technol., vol. 12, pp. 9–17, 2017, doi: 10.1016/j.susmat.2017.03.001. [198] S. Samad and A. Shah, “Role of binary cement including Supplementary Cementitious Material (SCM), in production of environmentally sustainable concrete: A critical review,” Int. J. Sustain. Built Environ., vol. 6, no. 2, pp. 663–674, Dec. 2017, doi: 10.1016/J.IJSBE.2017.07.003. [199] S. Hua, K. Wang, and X. Yao, “Developing high performance phosphogypsum- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 86 based cementitious materials for oil-well cementing through a step-by-step optimization method,” Cem. Concr. Compos., vol. 72, pp. 299–308, 2016, doi: 10.1016/j.cemconcomp.2016.05.017. [200] S. Gaidučis, J. Žvironaite, R. Mačiulaitis, and G. Jakovlev, “Resistance of phosphogypsum cement pozzolanic compositions against the influence of water,” Mater. Sci., vol. 17, no. 3, pp. 308–313, 2011, doi: 10.5755/j01.ms.17.3.599. [201] F. Mei, J. Hou, and Z. Liu, “Research on activity characteristics on composite cementitious materials based on phosphogypsum,” Procedia Eng., vol. 43, pp. 9–15, 2012, doi: 10.1016/j.proeng.2012.08.003. [202] N. Ukrainczyk, N. F. Mihelj, and J. Šipušiæa, “Calcium Sulfoaluminate Eco-Cement from Industrial Waste,” Chem. Biochem. Eng. Q, vol. 27, no. 1, pp. 83–93, 2013. [203] S. Liu and L. Wang, “Investigation on strength and pore structure of supersulfated cement paste,” Medziagotyra, vol. 24, no. 3, pp. 319–326, 2018, doi: 10.5755/j01.ms.24.3.18300. [204] M. Schneider, M. Romer, M. Tschudin, and H. Bolio, “Sustainable cement production—present and future,” Cem. Concr. Res., vol. 41, no. 7, pp. 642–650, Jul. 2011, doi: 10.1016/J.CEMCONRES.2011.03.019. [205] C. Angulski da Luz and H. R. D., “Influence of curing temperature on the process of hydration of supersulfated cements at early age,” Cem. Concr. Res., vol. 77, pp. 69–75, Nov. 2015, doi: 10.1016/J.CEMCONRES.2015.07.002. [206] J. Bijen and E. Niël, “Supersulphated cement from blastfurnace slag and chemical gypsum available in the Netherlands and neighbouring countries,” Cem. Concr. Res., vol. 11, no. 3, pp. 307–322, May 1981, doi: 10.1016/0008-8846(81)90104-6. [207] H. G. Midgley and K. Pettifer, “The micro structure of hydrated super sulphated cement,” Cem. Concr. Res., vol. 1, no. 1, pp. 101–104, Jan. 1971, doi: 10.1016/0008- 8846(71)90086-X. [208] A. Gruskovnjak et al., “Hydration mechanisms of super sulphated slag cement,” Cem. Concr. Res., vol. 38, no. 7, pp. 983–992, Jul. 2008, doi: 10.1016/J.CEMCONRES.2008.03.004. [209] S. Liu, L. Wang, and B. Yu, “Effect of modified phosphogypsum on the hydration properties of the phosphogypsum-based supersulfated cement,” Constr. Build. Mater., vol. 214, pp. 9–16, Jul. 2019, doi: 10.1016/j.conbuildmat.2019.04.052. [210] S. Kramar, L. Žibret, E. Fidanchevska, V. Jovanov, B. Angjusheva, and V. Ducman, “Use of fly ash and phosphogypsum for the synthesis of belite-sulfoaluminate clinker,” Mater. Construcción, vol. 69, no. 333, p. 176, 2019, doi: 10.3989/mc.2019.11617. [211] K. L. Scrivener, V. M. John, and E. M. Gartner, “Eco-efficient cements: Potential economically viable solutions for a low-CO 2 cement-based materials industry,” Cem. Concr. Res., vol. 114, no. March, pp. 2–26, 2018, doi: 10.1016/j.cemconres.2018.03.015. [212] G. L. Valenti, R. Cioffi, L. Santoro, and S. Ranchetti, “Influence of chemical and physical properties of Italian fly ashes on reactivity towards lime, phosphogypsum and water,” Cem. Concr. Res., vol. 18, no. 1, pp. 91–102, Jan. 1988, doi: 10.1016/0008- 8846(88)90125-1. [213] M. Garg, M. Singh, and R. Kumar, “Some aspects of the durability of a phosphogypsum-lime-fly ash binder,” Constr. Build. Mater., vol. 10, no. 4, pp. 273–279, 1996, doi: 10.1016/0950-0618(95)00085-2. [214] Y. Min, Q. Jueshi, and P. Ying, “Activation of fly ash-lime systems using calcined phosphogypsum,” Constr. Build. Mater., vol. 22, no. 5, pp. 1004–1008, 2008, doi: 10.1016/j.conbuildmat.2006.12.005. [215] A. M. Rashad, “Potential use of phosphogypsum in alkali-activated fly ash under the effects of elevated temperatures and thermal shock cycles,” J. Clean. Prod., vol. 87, no. 1, pp. 717–725, 2015, doi: 10.1016/j.jclepro.2014.09.080. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 87 [216] J. L. Provis and J. S. J. van Deventer, RILEM State of the Art Reports: Alkali Activated Materials. 2014. [217] H. F. W. Taylor, Cement Chemistry. London: ACADEMIC PRESS, 1990. [218] E. Herve, S. Care, and J. P. Seguin, “Influence of the porosity gradient in cement paste matrix on the mechanical behavior of mortar,” Cem. Concr. Res., vol. 40, no. 7, pp. 1060– 1071, 2010, doi: 10.1016/j.cemconres.2010.02.010. [219] F. M. A. Henriques and P. Faria, “O papel da investigação no estudo das argamassas de substituição na conservação do património,” Construction, pp. 29–37, 2008. [220] Q. Wang, S. Zhuang, and R. Jia, “An investigation on the anti-water properties of phosphorus building gypsum (PBG)-based mortar,” J. Therm. Anal. Calorim., vol. 136, no. 4, pp. 1575–1585, 2019, doi: 10.1007/s10973-018-7825-3. [221] W. Hyung, D. Han, and D. Seo, “Basic Properties of Non-Sintered Cement Mortar Using Industrial Byproducts as Activators,” J. Asian Archit. Build. Eng., vol. 14, no. 3, pp. 649–656, 2015, doi: 10.3130/jaabe.14.649. [222] K. Mun and S. So, “Properties of cement mortar with phosphogpysum under steam curing condition,” Adv. Mater. Sci. Eng., vol. 2008, pp. 1–6, 2008, doi: 10.1155/2008/382490. [223] G. Jiang, A. Wu, Y. Wang, and W. Lan, “Low cost and high efficiency utilization of hemihydrate phosphogypsum: Used as binder to prepare filling material,” Constr. Build. Mater., vol. 167, pp. 263–270, 2018, doi: 10.1016/j.conbuildmat.2018.02.022. [224] L. G. Li, Y. M. Wang, Y. P. Tan, A. K. H. Kwan, and L. J. Li, “Adding granite dust as paste replacement to improve durability and dimensional stability of mortar,” Powder Technol., vol. 333, pp. 269–276, 2018, doi: 10.1016/j.powtec.2018.04.055. [225] M. H. Samarakoon, P. G. Ranjith, T. D. Rathnaweera, and M. S. A. Perera, “Recent advances in alkaline cement binders: A review,” J. Clean. Prod., vol. 227, pp. 70–87, Aug. 2019, doi: 10.1016/J.JCLEPRO.2019.04.103. [226] S. K. Antiohos, V. G. Papadakis, E. Chaniotakis, and S. Tsimas, “Improving the performance of ternary blended cements by mixing different types of fly ashes,” Cem. Concr. Res., vol. 37, no. 6, pp. 877–885, Jun. 2007, doi: 10.1016/J.CEMCONRES.2007.02.017. [227] P. Vashistha, V. Kumar, S. K. Singh, D. Dutt, G. Tomar, and P. Yadav, “Valorization of paper mill lime sludge via application in building construction materials: A review,” Constr. Build. Mater., vol. 211, pp. 371–382, Jun. 2019, doi: 10.1016/J.CONBUILDMAT.2019.03.085. [228] D. X. Xuan, L. J. M. Houben, A. A. A. Molenaar, and Z. H. Shui, “Mixture optimization of cement treated demolition waste with recycled masonry and concrete,” Mater. Struct. Constr., vol. 45, no. 1–2, pp. 143–151, 2012, doi: 10.1617/s11527-011-9756-3. [229] S. Samad and A. Shah, “Role of binary cement including Supplementary Cementitious Material (SCM), in production of environmentally sustainable concrete: A critical review,” Int. J. Sustain. Built Environ., vol. 6, no. 2, pp. 663–674, Dec. 2017, doi: 10.1016/J.IJSBE.2017.07.003. [230] P. Shafigh, H. Bin Mahmud, M. Z. Jumaat, and M. Zargar, “Agricultural wastes as aggregate in concrete mixtures – A review,” Constr. Build. Mater., vol. 53, pp. 110–117, Feb. 2014, doi: 10.1016/J.CONBUILDMAT.2013.11.074. [231] I. O. Toma, D. Covatariu, A. M. Toma, G. Taranu, and M. Budescu, “Strength and elastic properties of mortars with various percentages of environmentally sustainable mineral binder,” Constr. Build. Mater., vol. 43, pp. 348–361, 2013, doi: 10.1016/j.conbuildmat.2013.02.061. [232] H. El Nouhy, E. Khattab, and S. Zeedan, “Behavior of Cement Pastes and Mortar Containing Phosphogypsum,” Key Eng. Mater., vol. 668, pp. 181–188, Oct. 2015, doi: 10.4028/www.scientific.net/kem.668.181. [233] S. Suárez Silgado, L. Calderón Valdiviezo, S. Gassó Domingo, and X. Roca, “Multi- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 88 criteria decision analysis to assess the environmental and economic performance of using recycled gypsum cement and recycled aggregate to produce concrete: The case of Catalonia (Spain),” Resour. Conserv. Recycl., vol. 133, no. June 2017, pp. 120–131, 2018, doi: 10.1016/j.resconrec.2017.11.023. [234] T. Tian, Y. Yan, Z. Hu, Y. Xu, Y. Chen, and J. Shi, “Utilization of original phosphogypsum for the preparation of foam concrete,” Constr. Build. Mater., vol. 115, pp. 143–152, Jul. 2016, doi: 10.1016/j.conbuildmat.2016.04.028. [235] F. P. Glasser and L. Zhang, “High-performance cement matrices based on calcium sulfoaluminate–belite compositions,” Cem. Concr. Res., vol. 31, no. 12, pp. 1881–1886, Dec. 2001, doi: 10.1016/S0008-8846(01)00649-4. [236] EN197-1:, “Cement Part 1: Composition, Specifications and Conformity Criteria for Common Cements,” CEN (Comité Eur. Norm., no. November, p. 50, 2011. [237] L. Yang, Y. Yan, and Z. Hu, “Utilization of phosphogypsum for the preparation of non-autoclaved aerated concrete,” Constr. Build. Mater., vol. 44, pp. 600–606, 2013, doi: 10.1016/j.conbuildmat.2013.03.070. [238] S. Ding, Z. Shui, W. Chen, J. Lu, and S. Tian, “Properties of supersulphated phosphogysumslag cement (SSC) concrete,” J. Wuhan Univ. Technol. Mater. Sci. Ed., vol. 29, no. 1, pp. 109–113, 2014, doi: 10.1007/s11595-014-0876-9. [239] N. C. Sukmana, M. I. Khifdillah, A. S. Nurkholil, and U. Anggarini, “Optimization of non-autoclaved aerated concrete using phosphogypsum of industrial waste based on the taguchi method,” IOP Conf. Ser. Mater. Sci. Eng., vol. 509, no. 1, 2019, doi: 10.1088/1757- 899X/509/1/012095. [240] A. Mardani-Aghabaglou, C. Yüksel, A. Beglarigale, and K. Ramyar, “Improving the mechanical and durability performance of recycled concrete aggregate-bearing mortar mixtures by using binary and ternary cementitious systems,” Constr. Build. Mater., vol. 196, pp. 295–306, Jan. 2019, doi: 10.1016/J.CONBUILDMAT.2018.11.124. [241] M. M. Smadi, R. H. Haddad, and A. M. Akour, “Potential use of phosphogypsum in concrete,” Cem. Concr. Res., vol. 29, no. 9, pp. 1419–1425, 1999, doi: 10.1016/S0008- 8846(99)00107-6. [242] T. S. S. Reddy, D. R. Kumar, and H. S. Rao, “A study on strength characteristics of phosphogypsum concrete,” Asian J. Civ. Eng., vol. 11, no. 4, pp. 411–420, 2010. [243] X. Huang, X. Zhao, S. Bie, and C. Yang, “Hardening Performance of Phosphogypsum-Slag-Based Material,” Procedia Environ. Sci., vol. 31, pp. 970–976, Apr. 2016, doi: 10.1016/j.proenv.2016.03.002. [244] M. B. Diop et al., “Phosphogypsum Based Cement Formulation in Agreement With Standard Portland Cement in the Framework of a Concrete Code Policy for Thermal and Energetic Efficiency of the Building,” African Concr. Code Symp., pp. 306–322, 2005, [Online]. Available: http://www.acmc-ngo.org/pdf/paper 19.pdf. [245] M. A. Bagade and S. R. Satone, “An experimental investigation of partial replacement of cement by various percentage of Phosphogypsum in cement concrete,” Int. J. Eng. Res. Appl., vol. 2, no. 4, pp. 785–787, 2012. [246] G. M. S. Islam, F. H. Chowdhury, M. T. Raihan, S. K. S. Amit, and M. R. Islam, “Effect of Phosphogypsum on the Properties of Portland Cement,” Procedia Eng., vol. 171, pp. 744–751, 2017, doi: 10.1016/j.proeng.2017.01.440. [247] S. Dhinakaran and R. M. Shanthi, “Experimental Investigation on Concrete with Phosphogypsum,” Int. J. Emerg. Res. Eng. Sci. Technol., vol. 2, no. 3, pp. 2–5, 2015. [248] M. Singh and M. Garg, “Durability of cementing binders based on fly ash and other wastes,” Constr. Build. Mater., vol. 21, no. 11, pp. 2012–2016, 2007, doi: 10.1016/j.conbuildmat.2006.05.032. [249] M. Singh and M. Garg, “Glass fibre reinforced water-resistant gypsum-based 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 89 composites,” Cem. Concr. Compos., vol. 14, no. 1, pp. 23–32, 1992, doi: 10.1016/0958- 9465(92)90036-U. [250] M. Singh and M. Garg, “Fibre reinforced gypsum binder composite, its microstructure and durability,” Mater. Struct., vol. 33, no. 8, pp. 525–528, 2000, doi: 10.1007/bf02480530. [251] T. A. Hoang, H. M. Ang, and A. L. Rohl, “Effects of temperature on the scaling of calcium sulphate in pipes,” Powder Technol., vol. 179, no. 1–2, pp. 31–37, Oct. 2007, doi: 10.1016/J.POWTEC.2006.11.013. [252] K. Ishikawa, “Bone substitute fabrication based on dissolution-precipitation reactions,” Materials (Basel)., vol. 3, no. 2, pp. 1138–1155, 2010, doi: 10.3390/ma3021138. [253] J. D. Bapat, Mineral admixtures in cement and concrete. 2012. [254] A. Bentur, K. Kovler, and A. Goldman, “Gypsum of improved performance using blends with Portland cement and silica fume,” Adv. Cem. Res., vol. 6, no. 23, pp. 109–116, 1994, doi: 10.1680/adcr.1994.6.23.109. [255] K. Kovler, “Strength and water absorption for gypsum-cement-silica fume blends of improved performance,” Adv. Cem. Res., vol. 10, no. 2, pp. 81–92, 1998, doi: 10.1680/adcr.1998.10.2.81. [256] N. Değirmenci, “Utilization of phosphogypsum as raw and calcined material in manufacturing of building products,” Constr. Build. Mater., vol. 22, no. 8, pp. 1857–1862, Aug. 2008, doi: 10.1016/J.CONBUILDMAT.2007.04.024. [257] M. Singh and M. Garg, “Gypsum-based fibre-reinforced composites: an alternative to timber,” Constr. Build. Mater., vol. 8, no. 3, pp. 155–160, 1994, doi: 10.1016/S0950- 0618(09)90028-9. [258] Z. Jin, B. Ma, Y. Su, W. Lu, H. Qi, and P. Hu, “Effect of calcium sulphoaluminate cement on mechanical strength and waterproof properties of beta-hemihydrate phosphogypsum,” Constr. Build. Mater., vol. 242, p. 118198, May 2020, doi: 10.1016/J.CONBUILDMAT.2020.118198. [259] M. Singh and M. Garg, “Microstructure of glass fibre reinforced water resistant gypsum binder composites,” Cem. Concr. Res., vol. 23, no. 1, pp. 213–220, 1993, doi: 10.1016/0008-8846(93)90152-Y. [260] S. Kumar, “Fly ash-lime-phosphogympsum hollow blocks for walls and partitions,” Build. Environ., vol. 38, no. 2, pp. 291–295, 2003, doi: 10.1016/S0360-1323(02)00068-9. [261] N. Degirmenci, “Utilization of phosphogypsum as raw and calcined material in manufacturing of building products,” Constr. Build. Mater., vol. 22, no. 8, pp. 1857–1862, Aug. 2008, doi: 10.1016/J.CONBUILDMAT.2007.04.024. [262] C. Di Bella, M. Wyrzykowski, and P. Lura, “Evaluation of the ultimate drying shrinkage of cement-based mortars with poroelastic models,” Mater. Struct. Constr., vol. 50, no. 1, pp. 1–13, 2017, doi: 10.1617/s11527-016-0870-0. [263] P. Acker and F. J. Ulm, “Creep and shrinkage of concrete: Physical origins and practical measurements,” Nucl. Eng. Des., vol. 203, no. 2–3, pp. 143–158, 2001, doi: 10.1016/S0029-5493(00)00304-6. [264] N. Deǧirmenci, “Utilization of phosphogypsum as raw and calcined material in manufacturing of building products,” Constr. Build. Mater., vol. 22, no. 8, pp. 1857–1862, 2008, doi: 10.1016/j.conbuildmat.2007.04.024. [265] L. Yang, Y. Zhang, and Y. Yan, “Utilization of original phosphogypsum as raw material for the preparation of self-leveling mortar,” J. Clean. Prod., vol. 127, pp. 204–213, Jul. 2016, doi: 10.1016/J.JCLEPRO.2016.04.054. [266] A. R. K. Kobayashi, “Avaliação da influência do tipo de cimento na expansibilidade das misturas de fosfogesso e cimento,” São Carlos, São Carlos, Brasil, 2000. [267] S. Nigade and M. Bagade, “An Experimental Investigation of Partial Replacement of 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 No conflict of interest Conflict of Interest