0% found this document useful (0 votes)
19 views11 pages

José Neto, 2021

Uploaded by

Mikaele Silva
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
19 views11 pages

José Neto, 2021

Uploaded by

Mikaele Silva
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 11

Construction and Building Materials 299 (2021) 123935

Contents lists available at ScienceDirect

Construction and Building Materials


journal homepage: www.elsevier.com/locate/conbuildmat

Influence of phosphogypsum purification with lime on the properties


of cementitious matrices with and without plasticizer
José S. Andrade Neto a,⇑, Jéssica D. Bersch a, Thaís S.M. Silva a, Erich D. Rodríguez b, Seiiti Suzuki c,
Ana Paula Kirchheim a
a
Universidade Federal do Rio Grande do Sul, Av. Osvaldo Aranha 99, Centro Histórico, Porto Alegre/RS, CEP. 90035-190, Núcleo Orientado para a Inovação da Edificação (NORIE), Brazil
b
Universidade Federal de Santa Maria - Av Roraima 1000. Prédio 10A (Centro de Tecnologia CT), CEP 97105-900, Camobi, Santa Maria/RS, Brazil
c
InterCement, Av. das Nações Unidas, 12.495, São Paulo SP, Brazil

h i g h l i g h t s

 The replacement of NG by PG and CTPG (PG + Ca(OH)2 on Portland cement is analyzed.


-
 The addition of Ca(OH)2 on PG reduced its %F %P2O5 concentration.
 The PG treatment reduces the delay in cement hydration and setting times.
 Cements with CTPG has intermediate strengths to cements with NG and PG.
 The plasticizer delays the hydration and mechanical development of the cements.

a r t i c l e i n f o a b s t r a c t

Article history: The use of phosphogypsum as a setting controller for Portland cement is very promising. However, impu-
Received 18 February 2021 rities (such as P2O5 and F-) in its composition reduce its effectiveness and lower mechanical performance
Received in revised form 3 June 2021 at early ages within cementitious systems. This study aimed to evaluate the efficiency of a chemical
Accepted 8 June 2021
purification process based on calcium hydroxide and how this treated phosphogypsum acts as a setting
Available online 17 June 2021
controller in Portland cement systems, with and without the presence of plasticizer admixture. Three
cement types were produced with different sulfate sources: natural gypsum, phosphogypsum, and chem-
Keywords:
ically treated phosphogypsum. The hydration kinetics and reactivity of the phosphogypsum/OPC systems
Phosphogypsum
Chemical treatment
were assessed by isothermal calorimetric and the performance of the sulfate source were compared in
Setting time controller fresh state (Vicat test, mini-slump test, and rotational rheometry) and hardened state (mortar and con-
Portland cement crete compressive strength). It was observed that the presence of impurities such as P2O5 and F- in the
Hydration untreated phosphogypsum delays the hydration and lower strength is achieved in cement mortars.
Rheology The chemical treatment based on Ca(OH)2 reduced the concentration of impurities (the water-soluble
phosphorus decreased by 80.3%) and improve the performance of this alternative sulfate source as a set-
ting controller. Due to the delay in hydration, pastes with phosphogypsum showed the lowest yield stress
and viscosity. The dissolution of fluoride and phosphates through the chemical treatments based on Ca
(OH)2 might be a feasible solution from a technical and economic perspective for the potential valoriza-
tion and re-use of phosphogypsum within the cement industry.
Ó 2021 Elsevier Ltd. All rights reserved.

1. Introduction Each ton of phosphoric acid production generates between 4 and


6 tons of phosphogypsum [1]. The global generation of this by-
Phosphogypsum is a waste from phosphoric acid production, product is 280 million tons per year [2,3], being about 5.6 million
which is also considered a by-product of the fertilizer industry. tons generated per year in Brazil [4].
Most of the phosphogypsum generated (~85%) is usually dis-
⇑ Corresponding author. posed of without any treatment in large stockpiles exposed to
E-mail addresses: josedasilvaandradeneto@gmail.com (J.S. Andrade Neto), weathering processes, occupying large land areas, and inducing
jessicadbersch@gmail.com (J.D. Bersch), thaisdosocorro9@gmail.com (T.S.M. Silva), severe environmental damage (chemical and radioactive contami-
erich.rodriguez@ufsm.br (E.D. Rodríguez), seiitis@intercement.com (S. Suzuki), nation), especially in coastal regions [5]. The remaining 15% of the
anapaula.k@ufrgs.br (A.P. Kirchheim).

https://doi.org/10.1016/j.conbuildmat.2021.123935
0950-0618/Ó 2021 Elsevier Ltd. All rights reserved.
José S. Andrade Neto, Jéssica D. Bersch, Thaís S.M. Silva et al. Construction and Building Materials 299 (2021) 123935

phosphogypsum generated is recycled, used as fertilizers, soil sta- Fig. 1 presents a flowchart of phosphogypsum generation and
bilization treatments, or raw material to manufacture building reuse processes.
materials [5]. Among the treatments, washing with dissolved lime is the most
This residue/by-product consists mainly of calcium sulfate dihy- common method applied [9,11]. Water-soluble phosphates are
drate (CaSO42H2O), having a great potential to be used as a setting converted into insoluble calcium phosphates in this treatment,
controller for Portland cement, which can be a suitable substitute which do not significantly affect the cement hydration [9,11].
for natural gypsum [1,4,6]. Natural gypsum reserves are usually con- Other chemical treatments consist of using sulfuric acid [10,12],
centrated in some countries and regions. In Brazil, for instance, it is ammonium hydroxide [13] and aqueous citric acid solutions [7].
concentrated in the north/northeast region. Thus, it is necessary These treatments are effective in removing water-soluble impu-
for cement production in the southeast/south region to transport rities. However, the fraction of the phosphates and fluorides that
the gypsum over long distances, resulting in high costs and environ- replace the sulfate ions in the crystalline structure of gypsum,
mental impacts. These regions have sulfuric acid production indus- forming solid solutions, are difficult to remove [7]. For this, thermal
tries and, therefore, have a high generation of phosphogypsum, treatments followed by chemical treatments are usually needed.
which could be used as a setting controller. Furthermore, the With calcination, gypsum is converted to hemihydrate, which does
replacement of natural gypsum for phosphogypsum would decrease not form a solid solution with phosphates and fluorides, releasing
the environmental impacts of phosphogypsum disposal, such as the impurities from the crystalline structure, which be removed
atmospheric, soil, and groundwater contamination. Besides, the from the chemical treatments mentioned early [7,14].
need for large areas for phosphogypsum disposal and the amount Regarding the physical processes, the impurities can be partially
of natural gypsum extracted also would decrease. Therefore, the removed by particle size separation through sieving. According to
use of phosphogypsum in the cement industry would result in eco- Potgieter et al. [9], particles>170 mm have a high fluoride concen-
nomic and environmental benefits. tration, while particles smaller than 25 mm are rich in organic mat-
However, due to its manufacturing process, it presents impuri- ter and phosphate in the crystalline structure.
ties such as soluble phosphate (P2O5) and fluoride (F-), in addition Among all the treatments to remove the impurities of phosph-
to HPO2- 2- 2-
4 and FPO3 which substitute the SO4 ions in the crystal lat- ogypsum, mixing and grinding it with solid hydrated lime seems
tice of calcium sulfate dihydrate, resulting in solid solutions [7]. promising as it does not require water, calcination, or expensive
During the hydration of the cement with phosphogypsum, these materials, which can be highly feasible from an economic point
impurities react with the calcium ions, precipitating Ca3(PO4)2 of view within a large scale application. Mixing and grinding the
and CaF2 on the surface of the cement grains. This results in the phosphogypsum with hydrated lime reduces the harmful effects
formation of a layer that reduces the flow of water and ions, delay- of impurities on the hydration of Portland cement, decreasing the
ing the dissolution and hydration of cement [4,7,8]. As a result, initial (up to 46%) and final (up to 37%) setting times and increasing
lower mechanical performance at early ages can be identified the mechanical performance of the matrix at early ages (up to 120%
[1,4]. These effects end up restricting the use or the amount of compressive strength at 2 days, up to 83% at 7 days and up to 29%
phosphogypsum used in the cement industry. at 28 days) [9,15]. The purification treatment with hydrated lime
Removing the impurities of phosphogypsum as a purification has also been shown to effectively reduce the negative effect of
process can be carried out by chemical, physical or thermal pro- phosphogypsum, such as longer delays of setting and lower com-
cesses or a combination of these [9,10], which can be considered pressive strength at early ages (increases up to 30% in the compres-
necessary for its valorization in the Portland cement industry. sive strength at 7 days) of supersulfated cements [16,17].

Fig. 1. Flowchart of phosphogypsum generation and reuse processes. Icon made by Smashicons, Those Icons, Freepik, Pixel perfect, turkkub, Vectors Market, iconixar,
DinosoftLabs, Flat Icons, dDara, fjstudio from www.flaticon.com.

2
José S. Andrade Neto, Jéssica D. Bersch, Thaís S.M. Silva et al. Construction and Building Materials 299 (2021) 123935

However, to the best of the authors’ knowledge, no study (12.33), which is expected due to the presence of hydrated lime
regarding the effect of this treatment on the hydration and the rhe- during the purification process and agrees with the results found
ology of Portland cement has been published yet. Furthermore, few by Chen et al. [11] and by Liu et al. [18]. This is probably due to
studies [9,15] have assessed the mechanical properties of Portland the direct contribution of the hydroxide introduced by Ca(OH)2
cement with lime-purified phosphogypsum, requiring further [18], and, according to Liu et al. [17,18], purification with lime
studies with other phosphogypsum and cements to validate the removes a portion of the acidic material in the phosphogypsum,
results. Finally, there are no studies on the interaction of purified such as H3PO4, which may also have contributed to the increase
phosphogypsum with lime and plasticizer admixtures and, as in the pH.
pointed out by Holanda et al. [4], there is still a knowledge gap The content of phosphorus and fluoride - impurities responsible
about the effects of phosphogypsum impurities in cement mix- for delaying the hydration of the cement - present in the NG, PG,
tures with plasticizer admixtures. These admixtures are widely and CTPG were determined by X-ray fluorescence (PANalytical
used to adapt the rheology of cement mixtures for their applica- equipment, model Axios). Soluble phosphorus was quantified
tion. Therefore, to enable the use of phosphogypsum as a setting through Inductively Coupled Plasma Atomic Emission Spec-
regulator, it is necessary to understand its behavior in the presence troscopy (ICP-OES) using Thermo Icap 6000 with a wavelength of
of these admixtures. 213.618 nm, power of 1300 W, and flow rate of the auxiliary and
Thus, the present work aims to evaluate the purification of nebulization gas (argon) of 0.5 L/min. For this analysis, the method
phosphogypsum with lime, representing a more straightforward proposed by Holanda et al. [4] was followed, where total phospho-
and economical alternative for the cement industry when com- rus was determined after dissolving the samples in nitric acid and
pared to the purification processes reported in the literature. The soluble phosphorus after dissolving the samples in deionized
objective is to identify the influence of this purification process water.
on the hydration kinetics (calorimetry analysis) and the rheology Table 1 also shows the amounts of total fluoride and phospho-
(spread, viscosity and yield strength) of cement pastes and the rus, determined by XRF, and the amount of total and water-soluble
compressive strength of mortars and concretes. For this, a set of phosphorus, determined by ICP-OES. As expected, PG had higher
different Portland cements were produced using three sulfate amounts of fluoride (up to > 0.16%) and phosphorus (up
sources as a setting controller: natural gypsum, natural phospho- to > 1.65%) than NG.
gypsum (as received) and a chemically treated phosphogypsum The effectiveness of phosphogypsum purification is analyzed by
(mixed with hydrated lime). Complementary, the effect of these the amount of fluoride and phosphorus remaining after the chem-
sulfate sources under the performance of polyfunctional ical treatment. Regarding the fluoride, a reduction of 25.0% is
lignosulfonate-based plasticizer were also assessed. observed with the purification process with hydrated lime. The
same behavior was observed for total phosphorus with a reduction
of 22.1% (determined by XRF) and 15.1% (determined by ICP-OES).
2. Materials The amount of water-soluble phosphorus decreased by 80.3%
with a mixture of phosphogypsum and hydrated lime, showing
2.1. Calcium sulfate an excellent efficiency of the process. This behavior possibly
occurred due to reactions between calcium hydroxide and soluble
In this study, three sources of calcium sulfate were evaluated: phosphates, resulting in calcium phosphates insoluble in water
natural gypsum (NG), phosphogypsum (PG), and chemically trea- and do not significantly delay the cement hydration [4,8,11], as
ted phosphogypsum (CTPG). For the purification of phosphogyp- presented in equations (1) and (2) [8]. These reactions were possi-
sum, 12 tons of the material was spread on an industrial floor, ble due to the high humidity of phosphogypsum (approximately
and then 600 kg of hydrated lime (5%, concerning the phosphogyp- 20 wt%). Therefore, the process effectively reduced the amount of
sum mass) were placed on this material. Then, the materials were water-soluble phosphorus, but it did not show the same efficiency
mixed with the aid of a loader. The proportion of hydrated lime in reducing the amount of total phosphorus and fluoride that are
used (5%) was chosen according to Liu et al. [18] results, which also responsible for delaying the hydration of the cement.
showed that this is the optimum content to treat the phosphogyp-
sum for use as a setting controller in the supersulfated cements. CaðOHÞ2 þ H3 PO4 ¼ Ca3 ðPO4 Þ2 þ 6H2 O ð1Þ
According to the method adopted by Chen, the pH of NG, PG,
and CTPG solutions with deionized water (1:10, by mass) were
analyzed with a pH meter mPA210 from the brand MS TECHNO- CaðOHÞ2 þ 2CaHPO4 ¼ Ca3 ðPO4 Þ2 þ 2H2 O ð2Þ
PON et al. [19]. Table 1 shows the pH values of the solutions. NG The mineralogical composition of the NG, PG, and CTPG were
and PG solutions showed similar values (7.37 and 7.33). Phospho- determined by X-ray diffraction on a Bruker D8 Advance diffrac-
gypsum, when initially stored, is usually acidic (pH close to 1) due tometer with Cu K radiation, Ni filter. A step of 0.020° was used
to the presence of residual acid from the phosphoric acid produc- in a 2h interval between 5° and 70°, with a speed of 0.72°/min.
tion, but its pH increases over the time of storage due to acid evap- The X-ray diffractograms are shown in Fig. 2, in which it is possible
oration and leaching with rainwater (if is weathering to observe that all sulfate sources present gypsum (CaSO42H2O) in
exposed/stored) [20]. The CTPG solution shows a basic pH their composition. In addition, the NG also shows the presence of
calcite (CaCO3), and the CTPG presented peaks related to calcium
Table 1
carbonate (CaCO3) and calcium hydroxide (Ca(OH)2). The quantifi-
pH values of solutions in deionized water and phosphorus concentrations and fluoride cation by the Rietveld method is presented in Table 2.
in the different materials. Fig. 3 shows the TG and DTG curves of NG, PG, and CTPG,
- NG PG CTPG
obtained in a thermobalance, model TGA Q50 from TA Instruments,
using platinum vessels without a lid in an inert atmosphere (N2),
pH 7.37 7.33 12.33
Total fluoride (%F-) - XRF 0.00 0.16 0.12
with a flow rate of 40 mL/min. The test was carried out up to
Total phosphorus (%P2O5) - XRF 0.10 1.36 1.06 900 °C, with a 10 °C/min heating rate. It is possible to observe, in
Total phosphorus (%P2O5) – ICP-OES 0.07 1.72 1.46 all samples, a peak of weight loss between 85 °C and 180 °C, related
Soluble phosphorus (%P2O5) – ICP-OES 0.00 0.01 0.00 to the decomposition of calcium sulfate dihydrate (CaSO42H2O) in
Soluble phosphorus (ppm) – ICP-OES 6.00 39.00 7.70
hemihydrate (CaSO41/2H2O) and then, in anhydrite (CaSO4) [21].
3
José S. Andrade Neto, Jéssica D. Bersch, Thaís S.M. Silva et al. Construction and Building Materials 299 (2021) 123935

Fig. 3. TG and DTG curves of NG, PG, and CTPG.

Fig. 2. X-ray diffractogram of the sulfate sources analyzed (G - Gypsum, C - Calcite, Table 3
P - Portlandite). Results of the physical characterization of the cements.

Property OPC NG OPC PG OPC CTPG


Blaine surface area (cm2/g) 4790 4740 4580
Table 2
Dv90 (mm) 31.19 33.19 34.28
Quantification of crystalline phases (%) of the analyzed sulfate sources.
Dv50 (mm) 11.46 11.90 15.76
Phase ICSD* code NG PG CTPG Dv10 (mm) 1.80 1.87 2.89
DvMEAN (mm) 14.43 15.32 17.56
Gypsum 2057 98.35 100.00 94.86
Calcite 040,112 1.65 – 3.75
Portlandite 034,241 – – 1.39
GOF** 1.60 2.02 2.07 2.2. Cements
*Inorganic Crystal Structure Database; **Goodness of fit.
In an industrial process, the different sources of sulfate (7 wt%)
were mixed with clinker (73 wt%), blast furnace slag (15 wt%), and
limestone filler (5 wt%), thus resulting in three high initial strength
and sulfate resistance Portland cements (OPC Type ARI-RS in the
In the TG and DTG curves of the CTPG, it was also possible to Brazilian standard NBR 16,697 [22], similar to OPC type III in the
observe a weight loss between 400 °C and 450 °C related to the ASTM C150 / C150M – 20 [23]. The content of sulfate source
decomposition of portlandite (Ca(OH)2) and a weight loss between (7 wt%) added resulted in an SO3 content of 4.2 wt%, which was
600 °C and 750 °C referring to the decarbonation of calcium car- selected as the optimum SO3 content for the compressive strength
bonate (CaCO3) [21], possibly resulting from the hydrated lime according to the procedure described by the ASTM C563 – 19 [24].
added in the purification process. The weight loss between Table 3 shows the surface areas of the cements, obtained by the
600 °C and 750 °C was also identified in the NG curve due to the Blaine method (NBR 16,372 [25]), and also the percentile DV90,
presence of calcium carbonate as an impurity (which is coherent DV50, DV10, and DVMEAN determined by laser granulometry (using
with the XRD results). the Particle Size Analyzer CILAS 1180). The Ordinary Portland
According to Equations (3) and (4), NG present 4.10% of calcite cements with gypsum (OPC NG) and phosphogypsum (OPC PG)
and CTPG present 2.56% of portlandite and 8.48% of calcite from the show similar surface areas and particle diameters, which are
TG results. These values are higher than those determined by XRD- essential for comparison purposes. However, it is noteworthy that
Rietveld. This was expected as a part of portlandite and calcium the cement with purified phosphogypsum (OPC CTPG) presents
carbonates can be amorphous or exceedingly small, resulting in slightly lower surface area values (~4%) and values of equivalent
diffraction peaks and therefore will not be accounted in the port- diameters slightly higher than the other cements, which can affect
landite and calcite content obtained by XRD-Rietveld. its reactivity and rheology.
The chemical composition (determined by XRF) and mineralog-
ical (determined by XRD using the Rietveld method) of the
mCaðOHÞ2 cements produced were similar, as shown in Tables 4 and 5. The
CaðOHÞ2;measured ¼ WL400 C500 C  ð3Þ
mH2 O quantification of non-crystalline content from XRD data was esti-
mated by calculating the areas referring to the crystalline (sharp
XRD peaks) and non-crystalline phases (smooth background). The
mCaCO3 non-crystalline content is given by the related intensity between
CaCO3;measured ¼ WL600 C1000 C  ð4Þ the area of the crystalline and non-crystalline phases over the
mCO2
whole pattern. More details of this method can be found in Riello
Where WL400°C-500°C is the weight loss (% of initial mass) [26] and Calligaris et al. [27].
between 400 °C and 500 °C, WL600°C-1000°C is the weight loss (% of The non-crystalline contents of the cements were similar
initial mass) between 600 °C and 1000 °C, and mCaðOHÞ2 (around 17 wt%). This was expected as 15 wt% of blast furnace slag
(74 g/mol), mH2 O (18 g/mol), mCaCO3 (100 g/mol), and mCO2 was used in all cements. Additionally, as shown in Table 5,
(44 g/mol) are the molecular masses of Ca(OH)2, H2O, CaCO3, and although the sulfates do not present hemihydrate (CaSO41/2H2O)
CO2, respectively. in their mineralogical composition, this phase was identified in the

4
José S. Andrade Neto, Jéssica D. Bersch, Thaís S.M. Silva et al. Construction and Building Materials 299 (2021) 123935

Table 4 3. couple the container to the mixer in 20 s.


Chemical composition (%), in oxides, of cements. 4. turn the mixer on at 10,000 rpm and mix for 70 s.
Constituent OPC NG OPC PG OPC CTPG The hydration process of the pastes was monitored using
SiO2 20.21 20.51 20.71 isothermal calorimetry (TA Instruments 8-channel TAM AIR
Al2O3 5.74 5.77 5.83 microcalorimeter). The pastes (approximately 7 g) were placed in
Fe2O3 2.46 2.47 2.44 glass ampoules of 20 mL, and their hydration process was moni-
CaO 60.57 60.13 60.61 tored during the first 72 h. A stabilization temperature of 25 °C
MgO 2.59 2.62 2.70
SO3 4.12 4.45 3.93
was used. Also, the initial and final setting times of the cement
Na2O 0.25 0.26 0.25 pastes without admixtures were determined by the Vicat needle
K2O 0.81 0.82 0.81 method, according to the Brazilian standard NBR 16,607 [30].
Loos on ignition (950 °C) 3.66 3.88 3.98 With and without the plasticizer admixture, the workability of
the pastes was evaluated by the spread-flow-test, also known as
mini-slump [31]. In this test, the paste is inserted into a conical
trunk mold, in a single layer, and without thickening, and then
cements. This behavior probably is due to the high temperatures the mold is removed vertically, allowing the paste to flow freely.
(typically 90–130 °C) reached when grinding the cement in the ball After a few seconds, the spread is measured in two directions.
mill [28]. However, although simple and inexpensive, the mini-slump test
For the analysis of the interaction of cements with water- is a single point test, evaluating the cement paste at a single shear
reducing admixtures, a polyfunctional lignosulfonate-based plasti- rate and providing a single parameter (spreading), thus not being
cizer admixture, commonly used in the concrete industry, was possible to differentiate the viscosity and the yield stress. For a
used. The plasticizer presents, at 25 °C, pH of 5.74, density equal more accurate analysis of the rheological behavior, the rotational
to 1.2092 g/cm3, and solids content of 43.26%. Besides, deionized rheometry test was performed.
water was used to mold the pastes and drinking water from the The rheological behaviors of cement pastes, with and without
public supply network to mold the mortars and the concretes. plasticizer admixtures, were evaluated using a Brookfield R/S Plus
For the mortar molding, standard sand was used according to rotational rheometer with a conical cylinder. ASTM C1749-17 [32]
the specifications of the Brazilian standard NBR 7214 [29]. Fine recommendations were followed, using a thermal bath to maintain
sand (quartz) with a density of 2.64 g/cm3, medium sand (fine frac- the paste temperature at 25 °C. The pastes were placed on the
tion of the aggregate crushing process) with a density of 2.68 g/ rheometer 2 min after the end of the mixture (4 min after the con-
cm3, and 19.0 mm gravel density of 2.69 g/cm3 were used to cast tact with water). Fig. 4 shows the test configuration adopted, in
the concrete. which the shear rate starts at 0.129 s1 and increases on a logarith-
mic scale until reaching 50 s1 and then decreases at the same rate,
according to the studies by Vance et al. [33]. The rheological
3. Methods parameters (yield stress and plastic viscosity) were obtained by
applying the Herschel-Bulkley model in the second deceleration
3.1. Studies in pastes curve. According to the preliminary test, the deviation results
Pastes with a water/cement ratio of 0.50, without admixture within the rheometry measurement were lower than 5% for the
and with the addition of 0.75% of plasticizer admixture (concerning yield stress and viscosity, based on 20 repetitions (samples).
the cement weight), were evaluated. The content of the plasticizer
admixture was selected according to previous tests in which this
content showed the highest spread on the mini-slump test (after 3.2. Studies in mortars
this content, a plateau value was observed).
The mixture of the pastes was carried out in a vertical mixer of To assess the compressive strength of the cements, the recom-
the Makita brand, using 100 g of cement, 50 g of water, and the mendations of NBR 7215 [34] were followed, in which mortars
samples with admixture, 0.75 g of plasticizer. The mixing proce- were molded with the different cements with a proportion of 1 :
dure adopted is: 3 : 0.48 (cement : sand : water). Cylindrical specimens (/ = 50 m
1. place the water (with admixture in half of the samples) in the m, h = 100 mm) were molded, which remained in curing immersed
cement, previously disposed of in the mixing container. in water saturated with lime until the age of rupture (1, 3, 7, and
2. mix manually with the aid of a metal rod for 30 s. 28 days). Three specimens per cement were evaluated at each age.

Table 5
Percentage of phases present in the cements, determined by X-ray diffraction using the Rietveld method.

Phase ICSD* code OPC NG OPC PG OPC CTPG


Alite M1 – 31.66 30.71 30.62
Alite M3 94,742 19.02 18.54 18.89
Belite b 963 6.23 7.27 7.75
Cubic C3A 1841 4.94 5.13 5.54
Orthorhombic C3A 1880 1.29 0.89 0.66
C4AF 9197 4.44 5.06 5.34
Free lime (CaO) 75,785 0.31 0.23 0.24
Periclase 104,844 0.54 0.54 0.76
Gypsum 151,692 4.29 4.87 4.69
Hemihydrate 380,286 2.32 2.92 2.24
Calcite 79,673 7.22 6.78 7.14
Non-crystalline phases – 17.74 17.07 16.15
GOF** 1.78 2.06 2.05

*Inorganic Crystal Structure Database; **Goodness of fit.

5
José S. Andrade Neto, Jéssica D. Bersch, Thaís S.M. Silva et al. Construction and Building Materials 299 (2021) 123935

Fig. 4. Rheometry test procedure based on the shear rate as a function of time.

3.3. Studies in concretes

Concretes made with the different studied cements were evalu-


ated. A cement consumption of 330 kg/m3 was adopted in the mix
proportion to obtain concretes of moderate strength (between 20
and 40 MPa) for the concretes without admixture. Also, a slump
equal to 120 mm was fixed in the slump-test, following the proce-
dure described by NBR NM 67 [35], being the amount of water
defined from this parameter. A water/cement ratio of 0.59 was
used for concretes without admixture and a water/cement ratio
of 0.52 for concretes with 0.75% plasticizer. The reduction of
11.9% in the water to obtain the same slump due to the addition
of the plasticizer is in concordance with previous studies [36,37],
Fig. 5. Heat flow curves of cement pastes with gypsum, phosphogypsum, and
which observed reductions between 3.5 and 11% by using 0.4 to purified phosphogypsum without admixture (A) and with 0.75% of plasticizer (B).
0.8% of lignosulfonate-based plasticizers. Table 6 shows the con-
sumption of materials per cubic meter of concrete.
There are several hypotheses to explain the mechanism of retar-
Cylindrical specimens (/ = 100 mm, h = 200 mm) were molded
dation of cement hydration due to plasticizers admixtures: (i) cal-
to evaluate the compressive strength according to the Brazilian
cium complexation, (ii) slowing down of dissolution of anhydrous
Standard NBR 5739 [38]. The specimens remained immersed in
phase due to surface adsorption, (iii) nucleation poisoning of
water saturated with lime until the rupture (1, 2, 7, and 28 days).
hydrates, and (iv) presence of sugar in the plasticizers.
Two specimens per formulation were evaluated at each age.
According to hypothesis (i), there are interactions between plas-
ticizing polymers and calcium ions in the pore solution, resulting in
calcium complexation. This would remove a part of calcium ions
4. Results from the solution, delaying the build-up of calcium supersatura-
tion needed for hydrates nucleation [42]. However, according to
4.1. Study in pastes Marchon and Flatt [42], the low plasticizer contents usually used
limit the amount of calcium potentially complex, and therefore,
Fig. 5A and B show the heat flow curves of the pastes without the calcium complexations do not seem to be the main reason
admixture and with 0.75% of plasticizer, respectively, during the for the delay in cement hydration.
first 40 h of hydration. Fig. 6A and B show the cumulative heat The adsorption of the plasticizer on the surface of the cement
curves of the pastes without and with 0.75% of plasticizer, respec- particles may reduce the dissolution of anhydrous phases (ii),
tively, during the first 72 h of hydration. The induction period decreasing the rate of ion release into solution, and delaying
length and the time between the maximum of the second heat flow cement hydration [41-43]. According to Suraneni and Flatt [43],
peak and sulfate depletion, obtained by analyzing the calorimetry this occurs as the adsorption of plasticizer onto alite’s surface hin-
curves (see Figure S1), are presented in Fig. 7A and B, respectively. ders the opening of etch pits, retarding the alite dissolution.
Table S1 presents all the parameters obtained by analyzing the Plasticizers might also influence and modify the nucleation and
calorimetry curves. growth of hydration products (iii) [41,42]. The plasticizer poisons
Comparing Figs. 5, 6, and 7 and observing the data presented in the portlandite nuclei, preventing their growth. This delays C3S
Table S1, the plasticizer admixture delayed the hydration of all hydration, as the calcium saturation in the pore solution does not
pastes, prolonging the induction period, consequently delaying change. Therefore, the C-S-H precipitation is delayed, increasing
the acceleration period and the maximum peak and reducing the the induction period length [41].
total heat released, which is following several previous studies Finally, the presence of sugar in plasticizers -especially the
[36,39-41]. lignosulfonate-based plasticizer, as the one used here- also con-

Table 6
Consumption of materials, in kilograms per cubic meter of concrete.

Concrete Cement Fine sand Medium sand 19.0 mm gravel Water Plasticizer w/c ratio
Without admixture 330 409 339 1082 193 0.00 0.59
With 0.75% of plasticizer 330 417 344 1099 181 2.31 0.52

6
José S. Andrade Neto, Jéssica D. Bersch, Thaís S.M. Silva et al. Construction and Building Materials 299 (2021) 123935

tributes to the delay in cement hydration (iv). The sugars are


adsorbed into the anhydrous phases and cement hydrates, delay-
ing the onset of the acceleration period [41].
Comparing the cements, both in the absence and in the pres-
ence of plasticizer, the OPC NG showed a faster hydration reaction,
while the cements with phosphogypsum showed greater delays.
These differences are probably due to F- and P2O5 impurities, sol-
uble or present in the crystalline structure of CaSO42H2O, on phos-
phogypsum, which delays cement hydration and impairs
mechanical performance [4,7,44]. During the hydration of the
cement with phosphogypsum, the soluble phosphates and fluoride
react with the calcium ions, precipitating CaF2 and Ca3(PO4)2. Some
authors [4,7,44] state that these compounds precipitate on cement
grains, forming a layer that reduces the flow of water and ions,
delaying the dissolution and hydration of the cement. However,
the impermeability of this layer and its ability to prevent the flow
of ions and water has not yet been reported. The delay in hydration
can also be due to the process of formation of calcium phosphates
and fluorides, which consumes calcium ions from the solution,
delaying the saturation of ions Ca2+, which is known as the trigger
for the end of the induction period and the beginning of portlandite
precipitation [45]. According to some other authors [46,47], these
compounds are not responsible for the cement hydration delay,
which is retarded due to the suppression of cement grains dissolu-
tion by the adsorption of phosphate and fluoride ions onto their
preferential dissolution sites.
The OPC CTPG presented an intermediate behavior, which was
expected since the purification process reduced the amount of sol-
uble phosphorus by 80.3% and total phosphorus and total fluoride
by 22.1% and 25.0%, respectively, as shown in Table 1. Thus, the
delay caused by impurities was reduced, resulting in a higher
Fig. 6. Cumulative heat curves of cement pastes with gypsum, phosphogypsum, hydration rate compared to the cement using phosphogypsum
and purified phosphogypsum without admixture (A) and with 0.75% of plasticizer without purification.
(B). Finally, it is observed that all cements are properly sulfated
since the peaks of heat release related to the sulfate depletion after
the main hydration peak of alite are clearly defined and in accor-
dance with the literature [28,48]. Regarding the effect of the
lignosulfonate-based plasticizer admixture, the sulfate depletion
point approximated from the main peak of alite (see Fig. 7B) and
an increase in the heat released at this point, showing a greater
interaction of the admixture with the aluminates. This result
agrees with previous studies [39,42,48]. This occurs as
lignosulfonate-base plasticizers promote the ettringite formation,
leading to higher and faster consumption of sulfate ions from the
solution, anticipating the sulfate depletion [49,50]. Furthermore,
it might also be related to the slowdown of gypsum’s dissolution
due to the adsorption of the plasticizer, decreasing the rate of sul-
fate ions supply to the solution [42,48]. This lower rate of sulfate
ions supplies might not be enough to ‘‘feed” the dissolution of
C3A, creating a ‘‘false” sulfate depletion [42,48]. Finally, the better
dispersion of anhydrous particles due to the plasticizer might
enhance the dissolution of C3A, leading to a faster sulfate consump-
tion [42,48,51]. This result indicates that lignosulfonate-based
plasticizer admixture influences the sulfate optimization and,
depending on the amount of admixture added and the amount of
sulfates in the cement, the resumption of the aluminates reaction
may occur before the main peak of alite hydration, which will
result in loss of mechanical performance, especially at early ages
[28,48].
Table 7 shows the initial and final setting times, obtained by the
Vicat needle method, for cement pastes without admixture. The
results are following the calorimetry, in which the OPC NG pre-
sents the shortest initial and final setting times, and the OPC PG
the longest, presenting an initial setting time 37.5% and a final set-
Fig. 7. (A) induction period length and (B) time between the maximum of the 2nd ting time 43.2% longer than the OPC NG. The OPC CTPG, as also
peak and sulfate depletion of the cement pastes with and without plasticizer. observed in calorimetry, presented an intermediate behavior, with
7
José S. Andrade Neto, Jéssica D. Bersch, Thaís S.M. Silva et al. Construction and Building Materials 299 (2021) 123935

Table 7
Initial and final setting times of the cements, determined by the Vicat needle method.

Property OPC NG OPC PG OPC CTPG


Initial setting time [min] 150 240 220
Final setting time [min] 210 370 300

an initial setting time 9.1% and a final setting time 23.3% shorter
than the OPC PG, showing that the purification process reduced
the delay caused by phosphogypsum.
Fig. 8 shows the spread in the mini-slump test obtained for
pastes with and without admixture. It is possible to observe that
the OPC PG and OPC CTPG pastes showed the highest spread.
Besides, there is an increase in spreading when using the
lignosulfonate-based plasticizer, which is adsorbed on the cement Fig. 8. Spread of pastes with and without plasticizer in the mini-slump test.
particles and dispersed due to electrostatic forces [36].
Fig. 9A to 9F show the shear stress as a function of the shear rate
gite formation and, therefore, mechanical strength development. In
curves of the different cement pastes, with and without admixture,
addition, as also observed by Potgieter et al. [9] and Potgieter and
obtained by rotational rheometry. The addition of plasticizer
Howell-Potgieter [15], the purification process with hydrated lime
admixture reduced the torque values, showing a greater fluidity.
was effective, reducing the delay in mechanical development at
Besides, the OPC PG and OPC CTPG presented lower torque values
early ages, which corroborates the calorimetry results.
when compared to the OPC NG.
Figs. 10 and 11 show the yield stress and plastic viscosity
results at 50 rpm (the maximum rate used in this work), deter-
mined by rotational rheometry, of pastes with and without admix- 4.3. Study in concretes
ture. It is possible to observe that the plasticizer reduced the plastic
viscosity and the yield stress, which was expected due to electro- This section presents the compressive strength results in con-
static repulsion [36]. crete. As detailed in section 3.3, it was decided to fix the consis-
From the rotational rheometry and mini-slump results, it was tency in this work, adjusting for this the amount of water. Thus,
possible to observe that the OPC PG and OPC CTPG cements pre- as the amount of water required to reach 120 mm of slump was
sented a greater fluidity, with lower plastic viscosity and yield different in concretes with and without admixture, these results
stress. This result is probably due to the delay caused by phospho- should not be compared with each other. However, the differences
gypsum and purified phosphogypsum in cement hydration, as in rheology observed for pastes (section 4.1) did not significantly
observed in the calorimetry and setting time result. Due to the influence the workability of concretes, and the amount of water
presence of fluoride and phosphates in PG and CTPG, the precipita- used for concretes with different cements was the same. For each
tion of the hydrated phases (mainly ettringite in the first hours) is group (with and without admixture), the results obtained between
delayed. The precipitation of ettringite, which has a high surface the different cements can be compared, which is the focus of this
area, results in increases in viscosity and the yield stress of the section.
paste [52]. Thus, the probable greater formation of ettringite in Fig. 13A shows the results of compressive strength in concretes
the first minutes in OPC NG, justifies the lower fluidity observed without admixtures. After 1 day, OPC PG concrete had the lowest
compared to cements with phosphogypsum. strength, representing 64.1% of the highest, obtained with OPC
However, it is interesting to note that, even with 80.3%, 22.1%, NG. The concrete with OPC CTPG showed intermediate strength.
and 25.0% less, in the amounts of soluble phosphorus, total phos- This is following the results of the calorimetry and compressive
phorus, and total fluoride, the OPC CTPG showed lower yield stress strength tests in mortars, which show a shorter delay in hydration
and plastic viscosity when compared with OPC PG. This behavior and strength development at the first ages with the use of purified
may have resulted from the smaller surface area and larger particle phosphogypsum concerning the non-purified one, indicating that
diameters (as shown in Table 3). Vikan et al. [53] noted that for the the purification process was efficient. At later ages, as observed
same clinker or clinkers with a similar mineralogical composition for mortars, the strengths of the different types of concretes are
(as is the case of this work), the smaller the Blaine surface area similar, with the strength of concrete with OPC PG, at 2 days,
of the cement, the higher the fluidity of the paste. 94.7% of that obtained with OPC NG.
Fig. 13B shows the results of compressive strength in concrete
with the lignosulfonate-based plasticizer admixture. In general,
4.2. Study in mortars the samples presented higher strength than the samples without
admixture. This is due to the lower water/cement ratio (0.52 for
Fig. 12 shows the results of compressive strength in mortars concretes with plasticizer and 0.59 for concretes without
produced with different types of calcium sulfate and without plas- admixture).
ticizer admixture. After 1 day, the mortars with OPC PG and OPC Among all evaluated ages, at 1 day, the differences between the
CTPG showed strengths 24.9% and 16.5% lower than those obtained three calcium sulfate sources are more pronounced. OPC PG con-
with OPC NG, respectively. In later ages, the difference in the crete presented compressive strength of 2.9 MPa at 1 d, 75.8%
strength between the OPC PG and OPC CTPG mortars and the and 82.1% lower than the OPC CTPG (12.0 MPa) and OPC NG
OPC NG mortar decreases, being 3.5% and 5.3% at 28 days, (16.2 MPa), respectively. At the ages of 2, 7, and 28 days, this dif-
respectively. ference reduced, showing that the negative effect of phosphogyp-
The lower compressive strength in phosphogypsum mortars at sum in mixtures containing admixture occurs only at early ages,
early ages was also observed by Potgieter et al. [9], Singh et al. [13], probably due to the delay in the reaction. As shown by paste
Singh and Garg [54] and Potgieter et al. [55]. This behavior is due to calorimetry results, this result is consistent and reinforces that
the impurities of phosphogypsum, which delays C-S-H and ettrin- the use of the admixture delays the onset of hydration reactions.
8
José S. Andrade Neto, Jéssica D. Bersch, Thaís S.M. Silva et al. Construction and Building Materials 299 (2021) 123935

Fig. 9. Shear stress as a function of shear rate curves, obtained in the rotational rheometry test, from pastes (A) and (B) OPC NG without and with admixture, (C) and (D) OPC
PG without and with admixture, (E) and (F) OPC CTPG without and with admixture.

Fig. 10. Yield stress, determined by rotational rheometry, of pastes with and Fig. 11. Plastic viscosity at 50 rpm, determined by rotational rheometry, of pastes
without plasticizer. with and without plasticizer.

9
José S. Andrade Neto, Jéssica D. Bersch, Thaís S.M. Silva et al. Construction and Building Materials 299 (2021) 123935

hydration of alite. Thus, the plasticizer admixture probably alters


the optimum sulfate content, impairing mechanical performance.
Regarding rheology, pastes with phosphogypsum showed lower
yield stress and plastic viscosity, resulting in lower shear stresses
and greater spreads in the mini-slump test. This behavior is prob-
ably due to the delay in hydration, which reduces the formation of
hydration products in the first minutes (mainly ettringite). How-
ever, these differences showed in pastes did not significantly influ-
ence the workability of the concretes.
For mortars and concretes, the compressive strength at early
ages is higher with cements with gypsum than cements with phos-
phogypsum or purified phosphogypsum. The purified phosphogyp-
sum delayed the reactions less than the phosphogypsum, enabling
higher initial strengths, which shows the beneficial effect of the
Fig. 12. Compressive strength at 1, 3, 7, and 28 days. purification process of phosphogypsum with hydrated lime used
for setting control. For later ages, the use of treated or untreated
phosphogypsum did not show any adverse impact on the compres-
sive strength.

CRediT authorship contribution statement

José da Silva Andrade Neto: Conceptualization, Investigation,


Methodology, Writing - original draft, Writing - review & editing.
Jéssica Deise Bersch: Methodology, Writing - review & editing.
Thaís do Socorro Matos da Silva: Methodology, Writing - review
& editing. Erich D. Rodríguez: Supervision, Writing - review &
editing. Seiiti Suzuki: Conceptualization, Supervision, Methodol-
ogy, Funding acquisition, Writing - review & editing. Ana Paula
Kirchheim: Conceptualization, Supervision, Funding acquisition,
Writing - review & editing.

Declaration of Competing Interest

The authors declare that they have no known competing finan-


cial interests or personal relationships that could have appeared
to influence the work reported in this paper.

Acknowledgments

The authors would like to thank the CAPES (Brazilian Coordina-


tion for the Improvement of Higher Education Personnel) for the
scholarship granted to JSAN (88882.439908/2019-01) and JDB
(88882.439909/2019-01) during the period of this research and
to the Graduate Program in Civil Engineering: Construction and
Fig. 13. Compressive strength of concretes, at 1, 2, 7, and 28 days, (A) without
Infrastructure (PPGCI) of the Federal University of Rio Grande do
admixture, (B) with 0.75% of plasticizer.
Sul (UFRGS). The participation of APK and EDR was sponsored by
CNPq (Brazilian National Council for Scientific and Technological
5. Conclusions Development) through the research fellowships PQ2017 303753 /
2017-0 and 305530 / 2017-8.
From the results of calorimetry and mechanical strength, it was
observed that replacing gypsum with phosphogypsum resulted in Appendix A. Supplementary data
a delay in hydration, probably due to the presence of impurities
(F- and P2O5) in its composition. The purification of phosphogyp- Supplementary data to this article can be found online at
sum with hydrated lime reduced these impurities (mainly water- https://doi.org/10.1016/j.conbuildmat.2021.123935.
soluble phosphate), decreasing the delay effects. However, the
cement with purified phosphogypsum still showed a delay con-
References
cerning the cement with gypsum, indicating that the purification
with hydrated lime was not totally effective. Despite this, all [1] A.M. Rashad, Phosphogypsum as a construction material, J. Cleaner Prod. 166
cements showed the heat peak related to the sulfate depletion (2017) 732–743, https://doi.org/10.1016/j.jclepro.2017.08.049.
after the main hydration peak of alite, indicating that all cements [2] J. Yang, W. Liu, L. Zhang, B.o. Xiao, Preparation of load-bearing building
materials from autoclaved phosphogypsum, Constr. Build. Mater. 23 (2) (2009)
are properly sulfated. 687–693, https://doi.org/10.1016/j.conbuildmat.2008.02.011.
The addition of a lignosulfonate-based plasticizer also resulted [3] A.A. Cuadri, F.J. Navarro, M. García-Morales, J.P. Bolívar, Valorization of
in a delay in the hydration of the cements. Also, the admixture phosphogypsum waste as asphaltic bitumen modifier, J. Hazard. Mater. 279
(2014) 11–16, https://doi.org/10.1016/j.jhazmat.2014.06.058.
intensified the heat release at the peak, referring to the resumption [4] F.d.C. Holanda, H. Schmidt, V.A. Quarcioni, Influence of phosphorus from
of the aluminate reaction and brought it closer to the main peak of phosphogypsum on the initial hydration of Portland cement in the presence of

10
José S. Andrade Neto, Jéssica D. Bersch, Thaís S.M. Silva et al. Construction and Building Materials 299 (2021) 123935

superplasticizers, Cem. Concr. Compos. 83 (2017) 384–393, https://doi.org/ [30] Associação Brasileira de Normas Técnicas, NBR 16607: Portland cement -
10.1016/j.cemconcomp.2017.07.029. Determination of setting times (in portuguese), (2018).
[5] H. Tayibi, M. Choura, F.A. López, F.J. Alguacil, A. López-Delgado, Environmental [31] P.A. Wedding, D.L. Kantro, Influence of water-reducing admixtures on
impact and management of phosphogypsum, J. Environ. Manage. 90 (8) (2009) properties of cement paste - A miniature slump test, Cement, Concrete and
2377–2386, https://doi.org/10.1016/j.jenvman.2009.03.007. Aggregates. 2 (2) (1980) 95, https://doi.org/10.1520/CCA10190J.
[6] J. Rosales, S.M. Pérez, M. Cabrera, M.J. Gázquez, J.P. Bolivar, J. de Brito, F. Agrela, [32] American Society for Testing and Materials, ASTM C1749  Standard Guide for
Treated phosphogypsum as an alternative set regulator and mineral addition Measurement of the Rheological Properties of Hydraulic Cementious Paste
in cement production, J. Cleaner Prod. 244 (2020) 118752, https://doi.org/ Using a Rotational Rheometer, (2017). https://doi.org/10.1520/C1749-17A.
10.1016/j.jclepro.2019.118752. [33] K. Vance, A. Kumar, G. Sant, N. Neithalath, The rheological properties of ternary
[7] M. Singh, Treating waste phosphogypsum for cement and plaster manufacture, binders containing Portland cement, limestone, and metakaolin or fly ash,
Cem. Concr. Res. 32 (7) (2002) 1033–1038, https://doi.org/10.1016/S0008- Cem. Concr. Res. 52 (2013) 196–207, https://doi.org/10.1016/j.
8846(02)00723-8. cemconres.2013.07.007.
[8] W. Shen, G. Gan, R. Dong, H. Chen, Y. Tan, M. Zhou, Utilization of solidified [34] Associação Brasileira de Normas Técnicas, NBR 7215: Portland cement -
phosphogypsum as Portland cement retarder, J. Mater. Cycles Waste Manage. Determination of compressive strength of cylindrical test specimens (in
14 (3) (2012) 228–233, https://doi.org/10.1007/s10163-012-0065-x. portuguese), (2019).
[9] J.H. Potgieter, S.S. Potgieter, R.I. McCrindle, C.A. Strydom, An investigation into [35] A.B.D.N. Técnicas, NBR NM 67: Concrete - Slump test for determination of the
the effect of various chemical and physical treatments of a South African consistency (in portuguese), Abnt Nbr Nm 67 (1998) 1998.
phosphogypsum to render it suitable as a set retarder for cement, Cem. Concr. [36] H.Sß. Arel, E. Aydin, Effects of Ca-, Mg-, K-, and Na-lignosulfonates on the
Res. 33 (8) (2003) 1223–1227, https://doi.org/10.1016/S0008-8846(03)00036- behavior of fresh concrete, Constr. Build. Mater. 157 (2017) 1084–1091,
X. https://doi.org/10.1016/j.conbuildmat.2017.09.190.
[10] R. Moalla, M. Gargouri, F. Khmiri, L. Kamoun, M. Zairi, Phosphogypsum [37] H.Sß. Arel, The effect of lignosulfonates on concretes produced with cements of
purification for plaster production: A process optimization using full factorial variable fineness and calcium aluminate content, Constr. Build. Mater. 131
design, Environ. Eng. Res. 23 (1) (2018) 36–45, https://doi.org/10.4491/ (2017) 347–360, https://doi.org/10.1016/j.conbuildmat.2016.11.089.
eer.2017.055. [38] Associação Brasileira de Normas Técnicas, NBR 5739: Concrete — Compression
[11] X. Chen, J. Gao, Y. Zhao, Investigation on the hydration of hemihydrate test of cylindrical specimens (in portuguese), (2004).
phosphogypsum after post treatment, Constr. Build. Mater. 229 (2019) [39] S. Ng, H. Justnes, Influence of plasticizers on the rheology and early heat of
116864, https://doi.org/10.1016/j.conbuildmat.2019.116864. hydration of blended cements with high content of fly ash, Cem. Concr.
[12] E.M. van der Merwe, C.A. Strydom, Purification of South African Compos. 65 (2016) 41–54, https://doi.org/10.1016/j.
phosphogypsum for use as Portland cement retarder by a combined thermal cemconcomp.2015.10.005.
and sulphuric acid treatment method, S. Afr. J. Sci. 100 (2004) 411–414. _
[40] I.B. Topçu, Ö. Atesßin, Effect of high dosage lignosulphonate and naphthalene
[13] M. Singh, M. Garg, S.S. Rehsi, Purifying phosphogypsum for cement sulphonate based plasticizer usage on micro concrete properties, Constr. Build.
manufacture, Constr. Build. Mater. 7 (1) (1993) 3–7, https://doi.org/10.1016/ Mater. 120 (2016) 189–197, https://doi.org/10.1016/
0950-0618(93)90018-8. j.conbuildmat.2016.05.112.
[14] M.A. Taher, Influence of thermally treated phosphogypsum on the properties [41] A. Colombo, M.R. Geiker, H. Justnes, R.A. Lauten, K. de Weerdt, On the effect of
of Portland slag cement, Resour. Conserv. Recycl. 52 (1) (2007) 28–38, https:// calcium lignosulfonate on the rheology and setting time of cement paste, Cem.
doi.org/10.1016/j.resconrec.2007.01.008. Concr. Res. 100 (2017) 435–444, https://doi.org/10.1016/j.
[15] J.H. Potgieter, S.S. Howell-Potgieter, A plant investigation into the use of cemconres.2017.06.009.
treated phosphogypsum as a set-retarder in OPC and an OPC/FLY ash blend, [42] D. Marchon, R.J. Flatt, Impacts of chemical admixtures on cement hydration,
Miner. Eng. 14 (7) (2001) 791–795, https://doi.org/10.1016/S0892-6875(01) in: P.-.-C. Aïtcin, R.J. Flatt (Eds.), Science and Technology of Concrete
00074-7. Admixtures, Woodhead Publishing, Cambridge, 2016, pp. 279–304.
[16] S. Liu, L. Wang, Investigation on strength and pore structure of supersulfated [43] P. Suraneni, R.J. Flatt, Use of micro-reactors to obtain new insights into the
cement paste, Medziagotyra. 24 (2018) 319–326, https://doi.org/10.5755/j01. factors influencing tricalcium silicate dissolution, Cem. Concr. Res. 78 (2015)
ms.24.3.18300. 208–215, https://doi.org/10.1016/j.cemconres.2015.07.011.
[17] S. Liu, L. Wang, B. Yu, Effect of modified phosphogypsum on the hydration [44] A.A. Tabikh, F.M. Miller, The nature of phosphogypsum impurities and their
properties of the phosphogypsum-based supersulfated cement, Constr. Build. influence on cement hydration, Cem. Concr. Res. 1 (6) (1971) 663–678.
Mater. 214 (2019) 9–16, https://doi.org/10.1016/j.conbuildmat.2019.04.052. [45] K.L. Scrivener, P. Juilland, P.J.M. Monteiro, Advances in understanding
[18] S. Liu, P. Fang, J. Ren, S. Li, Application of lime neutralised phosphogypsum in hydration of Portland cement, Cem. Concr. Res. 78 (2015) 38–56, https://doi.
supersulfated cement, J. Cleaner Prod. 272 (2020) 122660, https://doi.org/ org/10.1016/j.cemconres.2015.05.025.
10.1016/j.jclepro.2020.122660. [46] P. Bénard, S. Garrault, A. Nonat, C. Cau-Dit-Coumes, Hydration process and
[19] X. Chen, J. Gao, C. Liu, Y. Zhao, Effect of neutralization on the setting and rheological properties of cement pastes modified by orthophosphate addition,
hardening characters of hemihydrate phosphogypsum plaster, Constr. Build. J. Eur. Ceram. Soc. 25 (11) (2005) 1877–1883, https://doi.org/10.1016/
Mater. 190 (2018) 53–64, https://doi.org/10.1016/j.conbuildmat.2018.09.095. j.jeurceramsoc.2004.06.017.
[20] O. Hentati, N. Abrantes, A.L. Caetano, S. Bouguerra, F. Gonçalves, J. Römbke, R. [47] P. Bénard, S. Garrault, A. Nonat, C. Cau-dit-Coumes, Influence of
Pereira, Phosphogypsum as a soil fertilizer: Ecotoxicity of amended soil and orthophosphate ions on the dissolution of tricalcium silicate, Cem. Concr.
elutriates to bacteria, invertebrates, algae and plants, J. Hazard. Mater. 294 Res. 38 (10) (2008) 1137–1141, https://doi.org/10.1016/j.
(2015) 80–89, https://doi.org/10.1016/j.jhazmat.2015.03.034. cemconres.2008.03.019.
[21] B. Lothenbach, P.T. Durdziński, K. de Weerdt, Thermogravimetric analysis, in: [48] J.d.S. Andrade Neto, A.G. De la Torre, A.P. Kirchheim, Effects of sulfates on the
K. Scrivener, R. Snellings, B. Lothenbach (Eds.), A Practical Guide to hydration of Portland cement – A review, Constr. Build. Mater. 279 (2021)
Microstructural Analysis of Cementitious Materials, 1st ed., CRC Press, 2016, 122428, https://doi.org/10.1016/j.conbuildmat.2021.122428.
pp. 177–212. [49] X. Wang, Y. Pang, H. Lou, Y. Deng, X. Qiu, Effect of calcium lignosulfonate on
[22] Associação Brasileira de Normas Técnicas, NBR 16697: Portland cement - the hydration of the tricalcium aluminate-anhydrite system, Cem. Concr. Res.
Requirements (in portuguese), (2018). 42 (11) (2012) 1549–1554, https://doi.org/10.1016/j.cemconres.2012.08.008.
[23] American Society for Testing and Materials, ASTM C 150/ C150M-20 Standard [50] A. Colombo, M. Geiker, H. Justnes, R.A. Lauten, K. de Weerdt, The effect of
specification for portland cement, (2020) 1–9. https://doi.org/10.1520/C0150. calcium lignosulfonate on ettringite formation in cement paste, Cem. Concr.
[24] American Society for Testing and Materials., ATSM C563-19: Standard Guide Res. 107 (2018) 188–205, https://doi.org/10.1016/j.cemconres.2018.02.021.
for Approximation of Optimum SO3 in Hydraulic Cement, (2019) 1–8. [51] D. Jansen, J. Neubauer, F. Goetz-Neunhoeffer, R. Haerzschel, W.-D. Hergeth,
[25] Associação Brasileira de Normas Técnicas, NBR 16372: Portland cement and Change in reaction kinetics of a Portland cement caused by a superplasticizer -
other powdered materials — Determination of fineness by the air permeability Calculation of heat flow curves from XRD data, Cem. Concr. Res. 42 (2) (2012)
method (Blaine method) (in portuguese), (2015). 327–332, https://doi.org/10.1016/j.cemconres.2011.10.005.
[26] F. Riello, Quantitative Analysis of Amorphous Fraction in the Study of the [52] M. García-Maté, A.G. de La Torre, L. León-Reina, E.R. Losilla, M.A.G. Aranda, I.
Microstructure of Semi-crystalline Materials, in: E.J. Mittemeijer, P. Scardi Santacruz, Effect of calcium sulfate source on the hydration of calcium
(Eds.), Diffraction Analysis of the Microstructure of Materials, Springer, 2004, sulfoaluminate eco-cement, Cem. Concr. Compos. 55 (2015) 53–61, https://
pp. 167–184, https://doi.org/10.1007/978-3-642-56680-6. doi.org/10.1016/j.cemconcomp.2014.08.003.
[27] G.A. Calligaris, T.L.T. da Silva, A.P.B. Ribeiro, A.O. dos Santos, L.P. Cardoso, On [53] H. Vikan, H. Justnes, F. Winnefeld, R. Figi, Correlating cement characteristics
the quantitative phase analysis and amorphous content of triacylglycerols with rheology of paste, Cem. Concr. Res. 37 (11) (2007) 1502–1511, https://
materials by X-ray Rietveld method, Chem. Phys. Lipids 212 (2018) 51–60, doi.org/10.1016/j.cemconres.2007.08.011.
https://doi.org/10.1016/j.chemphyslip.2018.01.003. [54] M. Singh, M. Garg, Production of beneficiated phosphogypsum for cement
[28] S.W. Dean, P.J. Sandberg, L.R. Roberts, Cement-admixture interactions related manufacture, J. Sci. Ind. Res. 61 (2002) 533–537.
to aluminate control, J. ASTM Int. 2 (6) (2005) 12296, https://doi.org/10.1520/ [55] J.H. Potgieter, S.S. Potgieter, R.I. McCrindle, A comparison of the performance of
JAI12296. various synthetic gypsums in plant trials during the manufacturing of OPC
[29] Associação Brasileira de Normas Técnicas, ABNT NBR 7214 - Standard sand for clinker, Cem. Concr. Res. 34 (12) (2004) 2245–2250, https://doi.org/10.1016/j.
cement tests - Specification (in portuguese), (2015). cemconres.2004.04.002.

11

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy