Activity concentration (Bq/kg) of radionuclides present in different phosphate rock samples.
The recycling of waste generated by industrial production processes is a topic of considerable environmental and economic interest. The minimization of waste disposal, avoiding its direct release into the environment, generates environmental benefits for industries in addition to the manufacture of the main product. Some of these wastes, measured by their radioactive element content, may be considered as naturally occurring radioactive material (NORM). Two of these NORMs are phosphogypsum (PG) and ilmenite mud (IM) come from the fertilizer industry and TiO2 pigment industry, respectively. This chapter discusses the viability of valorization and/or recycling of PG and IM in the manufacture of sulfur polymer cement/concrete, Portland cement, and ceramic materials.
- llmenite mud
- sulfur polymer concrete/cement
- portland cement
Some minerals or raw materials used in industrial process can contain radionuclides at concentrations which cannot be disregarded. The main natural radionuclides derive from the 238U and 232Th decay series. In some cases, industrial processing can lead to further enhancement of the concentrations in the product, by‐product or in the waste materials.
These materials are known as naturally occurring radioactive material (NORM) used to describe materials that contain radioactive elements, radionuclides, found in the natural environment. Long‐lived radioactive elements of interest include uranium, thorium, and potassium, and any of their radioactive decay products such as radium and radon. However, the term is used more specifically for all naturally occurring radioactive materials where human activities have increased the potential for exposure compared with the unaltered situation. Sometimes artificially concentrated NORM is called technologically enhanced naturally occurring radioactive material (TENORM).
Environmental management of NORMs has taken on a new urgency in the last few decades. Increasingly, the governments are adopting more stringent regulations on radiation and environmental protection in general. Nowadays, the recycling and/or valorization of wastes generated by different industries are increasing exponentially [1–6]. Reduction in the disposal of these wastes has environmental and economic benefits for the industries and for the population.
This chapter is focused mainly in the valorization for construction materials of two NORMs coming from the fertilizer industry—phosphogypsum (PG) and the TiO2 pigment industry—ilmenite mud (IM).
1.1. Fertilizer industry: phosphgypsum
The continuous rise in world population increases the demand for food production which involves a constant increase in the use of phosphate fertilizer  and the development of chemical industries related to their production. Phosphate rocks are used for manufacturing phosphoric acid and chemical fertilizers. These are mainly composed of carbonate fluorapatite (Ca10F2(PO4)6CaCO3), quartz, goethite; minor mounts of Al‐phosphates, anatase, monazite, barite, magnetite; and heavy metals and trace elements (Cd, Ni, As, and Sr) [8, 9]. The phosphate rock also contains certain levels of natural radioactivity being a major NORM due to its uranium (238U) and thorium (232Th) content , Table 1. So the fertilizer industry has an impact in conventional and radioactive environmental contamination [11, 12].
Currently, over 90% of phosphoric acid production is by the wet acid method, involving the chemical attack of phosphate rock ore with sulfuric acid, generating a waste called phosphogypsum (CaSO4·2H2O), Eq. (1). This process is economic, however, large amounts of PG are generated—around 5 tons per ton of phosphate rock.
CaSO4·H2O is the main component of phosphogypsum, making up over 90% of it. Other impurities such as residual acids, fluorides (NaF, Na3AlF6, Na3FeF6 Na2SiF6, CaF2), fluoride sulfate ions, H3PO4, Ca3(PO4)2, CaHPO4·2H2O, Ca(H2PO4)2H2O, residual acids, some trace metals (e.g., Cr, Cu, Zn and Cd), and organic matter can be attached to the surface of the gypsum crystals [13–16]. Due to the residual acid compounds, PG has a pH 3 and is thus classed as an acidic by‐product.
The wet process causes the selective separation and concentration of naturally occurring radionuclides present in phosphate rock. Depending on the source rock, PG can contain as much as 60 times the level of radionuclides normally found prior to processing, with around 80% of 226Ra, 86% of U, and 70% of Th ending up in the phosphoric acid, Table 2 . In particular, a great variation of 226Ra activity is observed depending on the nature of the phosphate rock. For example, this ranges between 15 Bq/kg in Sweden and 114 Bq/kg in Florida. It is important to be aware of this radionuclide content due to the fact that 226Ra produces radon gas (222Rn) which has a short half‐life of 3.8 days and is therefore highly active, that is, has a high radiation output and can cause significant damage to internal organs. High levels of 238U and 210Po activity have also been reported by various authors.
|Country||Refs.||PG rock origin||238U||236Ra||210Pb||210Po||230Th|
|Egypt||||Nile Valley Rock||–||100||–||445||–|
PG is stored without any prior treatment, requires a large land area, and can cause serious environmental contamination of soils, water, and the atmosphere. So PG management is one of the most serious problems currently faced by the phosphate industry.
1.2. TiO2 industry: ilmenite mud
The titanium product most widely used is not in the form of titanium metal or alloy. It is a white powder—titanium dioxide (TiO2) pigment. Because it provides whiteness and opacity, this is widely used in a huge range of products and industries from coatings and plastics, printing inks, as flux in glass manufacture, filler material for paper industries and rubber. It is also an important substance, although not used in such high volumes, in important sectors such as pharmaceuticals, foodstuffs, and cosmetics [19, 20]. The annual TiO2 worldwide production is around 4.5 million t .
Two different commercial processes are used in TiO2 pigment production—the chloride process using chloride gas and the sulfate process using concentrated sulfuric acid. Around 60% of total TiO2 production is by the chloride process and 40% by the sulfate process [20–22].
The main feedstocks for the TiO2 industry are rutile, synthetic rutile, and chloride slag, all used in the chloride route, and ilmenite and sulfate slag, used for the sulfate process. These ores can contain levels of naturally occurring radioactive material so this is a NORM industry, Table 3.
In the sulfate route, the main raw material used to produce TiO2 is ilmenite ore (FeTiO3). This contains approximately 43–65% TiO2. A titanium slag (a co‐product of the ilmenite ore coming from smelting process) which contains 70–80% titanium dioxide [23–25] is also used. The ilmenite contains around 300 Bq/kg of the radionuclides from the thorium series and around 100 Bq/kg for the radionuclides of the uranium series. This requires it to be classed as a NORM industry .
TiO2 production begins with a highly exothermic reaction involving the digestion of a mix of ilmenite and titanium slag by highly concentrated (98%) sulfuric acid and water. The general steps for the digestion reaction are dissolution of the raw material, given in Eq. (2), followed by TiO2 precipitation, Eq. (3).
|Raw materials||TiO2 (%)||238U||232Th (Bq/g)|
After the sulfuric acid dissolution, the liquid contains titanylsulfate (TiOSO4) and iron sulfate (FeSO4).The liquor is then passed to a clarification tank, where the un‐dissolved ore (solid ilmenite mud) is separated by flocculation and filtration. It is then neutralized and stored in a safe area (Figure 2).
After ilmenite mud separation, the TiO2 is precipitated from the clarified liquor by boiling it for several hours followed by cooling. Then, the TiO2(S) is separated by vacuum filters from the liquor, which is a “strong” acid, with a content 20–25% H2SO4. After filtration, the hydrated titanium dioxide slurry is placed in rotatory kilns where the TiO2 crystals grow to their final size and the water and sulfur traces are removed . The production of half a ton of TiO2 pigment requires one ton of raw materials.
During the feedstock digestion with sulfuric acid, the radionuclides are distributed; U and Th believe will pass forward in liquor; Ra has low solubility and suggests 75–95% is not digested; and Pb has low solubility and expects majority to remain in the solid phase. Considering that after the digestion, the solid volume is reduced by a factor of 5–6 and the overall impact of the radionuclides activity in the residue, ilmenite mud, is increased. The activity concentration for the ilmenite mud is typically given as, 238U series = 600–1000 Bq/kg (in a feedstock 100–200 Bq/kg) and 232Th series = 1200–2500 Bq/kg (in a feedstock 400–500 Bq/kg) [24, 26].
The ilmenite mud remaining at the end is enriched in radium isotopes, 226Ra (860 Bq/kg) and 228Ra (2590 Bq/kg) as well as 40K (280 Bq/kg). The increased radium concentration is due to the insolubility of radium as a sulfate . The relatively high concentration of 40K in ilmenite mud is because the main form of 40K (the raw material had 18 Bq/kg) in the feedstock, and this remains unaltered by the process .
So the ilmenite mud is a highly acid waste that contains a large amount of natural radionuclides and also certain metals. All of these can adversely affect the environment and must be managed adequately.
2. Norm wastes application for construction materials
2.1. Physico‐chemical and radiological characterization of raw material
Table 4 shows the chemical composition for the major elements in PG and IM and in the raw materials [gravel, sand, red stoneware (RSM), and clinker] used in the various valorization processes. The PG and IM were collected from a PG piles and TiO2 factory located at the Southwest Spain in the city of Huelva, respectively.
PG is composed of 40% CaO and 54% SO3. The main mineralogical phases by X‐ray diffraction (XRD) are gypsum (CaSO4·2H2O) and basanite (CaSO4·1/2H2O). Also, a 0.95% of P2O5 is founded in PG sample, which remains later the washing in the industrial process .
The IM sample had a high concentration of titanium oxides (53%) and appreciable amounts of iron and silicon oxides (12.5 and 11.9%, respectively). Similar compositions have been observed elsewhere . In the ilmenite mud, ilmenite (22%) and rutile (34%) were the main mineral phases observed by XRD. Other species such as Fe and Ti oxides (Fe3Ti3O10), zircon (ZrSiO4), and quartz (SiO2) at 18.2, 13, and 12.1%, respectively, were also present.
In relation to the valorization studies for these industrial wastes, it is important take into account the particle sizes. The d (0.5) value of the particles of phosphogypsum and mud measured was ≤53 and 30 µm, respectively.
To evaluate the potential radiological impacts of these wastes when used in the production of building materials, the EU has proposed reference values for the natural radionuclide concentrations , defining an external risk index (I), also called the activity concentration index, according to the following equation:
where C226Ra, C232Th, and C40K are the activity concentrations for 226Ra, 232Th, and 40K, respectively, expressed in Bq/kg.
This index should not exceed one (I ≤ 1) for any material used in bulk quantities, for example, cement or concrete. It should not exceed six (I ≤ 6) for superficial materials, for example, tiles or boards. This is to ensure that the external dose received by occupants does not exceed the reference value of 1 mSv/year [28–29]. Other countries use the equivalent radium concentration parameter, Ra(eq), given in Eq. (5):
where C(226Ra), C(228Ra), and C(40K) are the activity concentrations for 226Ra, 228Ra, and 40K in Bq/kg, respectively. Three hundred and seventy Bq/kg is taken as the reference.
The concentration of natural radionuclides from the 238U decay series in phosphate rock (PR) used in Huelva was around 1650 Bq/kg (Table 5). This is over 50 times higher than the concentrations in unperturbed soils which are around 25 Bq/kg of 238U and 232Th . Ilmenite ore (ILM) is also a NORM mineral since it is high in natural radionuclides from both the Th and U series with a total concentration of over 400 Bq/kg for 238U and 232Th nuclides .
|PR||1600 ± 90||1650 ± 70||25 ± 2||1580 ± 80||22 ± 2||<18||5.4||1612|
|PG||624 ± 37||97 ± 6||8.2 ± 1.0||589 ± 34||8 ± 1||<18||3.9||621|
|ILM||82 ± 4||105 ± 10||315 ± 20||110 ± 10||300 ± 20||30 ± 5||1.9||541|
|IM||247 ± 11||184 ± 12||250 ± 15||521 ± 30||1919 ± 112||334 ± 21||11.4||3291|
The activity concentration of 238U and 226Ra in the pure PG used in this work is around 100 and 600 Bq/kg, respectively, which are significantly higher than the average world‐wide values from unperturbed soils. The ilmenite mud (IM) also has a total radionuclide concentration higher than 1000 Bq/kg. 226Ra and 228Ra are the radionuclides with the highest activity concentrations in IM at around 500 and 2000 Bq/kg, respectively. This is a considerable fraction of the radioactivity content originally present in the raw material.
2.2. Use of phosphogypsum and ilmenite mud in sulfur polymer cement/concrete manufacturing
The manufacture of Portland cement is associated with a high energy consumption (850 kcal/kg of clinker) due to the high temperature required to produce Portland cement clinker (around 4500°C). This has negative environmental effects as around 850 kg of CO2is emitted into the atmosphere per ton of clinker produced, as well as other greenhouse gases and NOXand SOx. It also involves massive quarrying for raw materials (1.7 ton to produce 1 ton of clinker) [3, 4].
In contrast, the manufacture of traditional binders such as sulfur polymer concretes (SPCs) does not require a heavy energy input. SPCs are based basically in the sulfur properties. To manufacture, a thermoplastic material is melted and mixed with an aggregate after which the mixture poured, molded and allowed to harden. These materials have been studied extensively over the past few decades.
Various trials have been carried out in which sulfur was used as the basis of construction materials [31–34]. In these studies, SPCs were prepared by heating of a mix of elemental sulfur (un‐modified sulfur) and mineral aggregates. However, the SPCs manufactured using elemental sulfur exhibited durability problems. They failed under repetitive cycles of freezing and thawing, humid conditions, or immersion in water [31, 35].
Extensive research has established the reason for the failure of these SPCs and determined ways of preventing it. When the hot mixture of elemental sulfur and aggregate is cooled to cast SPCs the liquid sulfur binder crystallizes at 114°C as monoclinic sulfur (Sβ) with a decrease in volume of around a 7%. When the temperature goes below 96°C, the Sβ begins to transform to orthorhombic sulfur Sα which is the stable form of sulfur at ambient temperature . This transformation is rapid (<24 hours) and causes the material failure. The Sα form is denser than Sβ. In the material, high strength is induced by the change in the material. Thus, the sulfur binder can become highly stressed and cause durability problems, mainly when it is exposed to water conditions.
It is necessary modify the sulfur‐based concrete to reduce the expansion and contraction during thermal cycling and achieved a product with good durability. Several chemical substances to inhibit the transformation of monoclinic to orthorhombic sulfur and to improve SPC products have been studied. Such modifiers are now available. Among the modifiers, commonly used are dicyclopentadiene (DCPD) or a combination of dicyclopentadiene, cyclopentadiene, and dipentene [31–36].
Sulfur concrete has properties, which for certain applications are superior to Portland cement concrete (PCC). Compared to traditional PCC, the main advantage of sulfur concrete is its high durability. SPC is also resistant to aggressive chemical environments and has good mechanical properties, high frost resistance, low absorbability, antiseptic qualities, water resistance, and good thermal insulating and dielectric properties . SPCs have a fast setting time reaching nearly full mechanical strength in <24 hours. These features make SPCs an ideal material for a wide range of applications in different fields, such as in road and railway infrastructure, hydrotechnical constructions, waste water treatment plants, and landfills . Due to their generally good properties and higher adhesive strength, SPCs are can also be used for stabilization/solidification in aggregate recycling and radioactive waste storage [35, 38–40].
Valorization of PG and IM by incorporating them into sulfur polymer concretes was studied by the authors of this chapter.
Different mixtures were prepared with percentages of PG and IM between 10 and 30 wt%. Each sample was denominated as SPC‐XPG‐Y and SPC‐XIM‐Y, where “X” is the percentage wt% of elemental sulfur and “Y” is the percentage wt% of phosphogypsum or ilmenite mud of the mixture. The reference sample was SPC.
The SPCs were prepared by first heating the aggregates (gravel/sand and NORM waste) at 130–135°C for 4 h. Then, the sulfur, gravel, sand, and NORM waste are heated together in a preheated mixing bowl, where the temperature is controlled at 135–140°C for 10 minutes. After this time, the modified sulfur STXTM is added with continuous mixing. Finally, the mixture is stirred at 140–145°C for 4–5 minutes. The molds (4 × 40 × 160 mm) are preheated to approximately 120°C before adding the sulfur polymer concrete. The material is compacted using a vibration table at 3000 rpm for 30 seconds. The storage of the moulds is done at a room temperature. The specimens are de‐molded 24 hours after placement in the steel moulds.
The compressive Cs and flexural strength Fs in SPC cements are shown in Table 6. The compressive strength of the SPCs has values between 55–62 MPa for PG and 58–64 MPa for IM. These values are in line with those reported by other authors. For example, López et al.  present values of 54 and 58 MPa for metacinnabar SPC with sulfur/HgS ratios of 0.4 and 2.6, respectively. Abdel‐Mohsen et al.  obtained a compressive strength of 54 MPa for SPC with fly ash, where the sulfur/fly ash proportion was 0.9. SPCs with lead wastes, have been studied by Lin et al. , and the final compressive strength of the SPC was 48.5 MPa. Between 65 and 73 MPa was the SPC strength obtained using a recycled aggregate coming from the ceramic industry .
|Fs (Mpa)||Cs (Mpa)||Ratios sulfur/waste|
|SPC‐21||7.07 ± 0.66||57.70 ± 1.98||–|
|SPC‐17PG‐10||9.36 ± 0.24||55.41 ± 1.37||1.70|
|SPC‐19PG20||11.20 ± 0.14||62.11 ± 0.86||0.95|
|SPC‐21PG30||10.72 ± 0.60||56.76 ± 5.20||0.70|
|SPC‐17IM10||9.83 ± 1.82||58.45 ± 1.62||1.70|
|SPC‐21IM20||13.25 ± 0.77||64.38 ± 1.62||1.05|
|SPC‐21IM30||9.40 ± 0.52||36.77 ± 2.58||0.70|
These results can be compared with Portland cement, which has a Cs > 52.5 MPa. At 28 days of curing Portland cement has a flexural strength of 10.1 ± 1.2 MPa and compressive strength of 61.3 ± 1.0 MPa . These values of strength are similar or slightly lower than the values obtained for the SPC with a percentage of between 20 and 30% of PG or between 10 and 20% of IM at 1 day old.
The strength obtained tends to increase as the sulfur/waste ratio increases up to between 0.95 and 1.05 (Table 6). A thin layer of sulfur‐coated PG particles is a good binder for aggregates increasing the mechanical strength. However, a decrease in the compressive strength is observed with a larger proportions of sulfur. This is due to the fact that this level of sulfur increases the thickness of the sulfur layer around the particles and leads to brittleness.
Table 7 shows the water activity coefficient (WAC) results recorded for SPC‐21PG‐30 and SPC‐21IM‐20. Water absorption by capillarity was studied gravimetrically in prismatic concrete specimens (50 × 29 × 9 mm) in water at room temperature. The water level was sufficient to wet only the lower surface of the test specimens. After 28 days, the WAC coefficient values of the samples are very low and certainly less than that of the SPC. However, all WAC values for SPC‐PG and SPC‐IM samples are smaller than those for concrete made from Portland cement as reported by several authors [42, 43]. WAC coefficients of 5.0 and 6.2 kg/m2 after 28 days of immersion in water are measured for an ordinary Portland cement and for a concrete made of Brazilian Portland cement (CPII E‐02) (slag‐modified Portland cement), respectively. The use of a sulfur/modified sulfur mix reduces the WAC coefficient producing a very impermeable concrete.
|Mixture||Water absorption (g cm-2)|
|3 hours||72 hours||7 days||28 days|
|210Pb||70 ± 5.0||143 ± 9.0||219 ± 13.0||28 ± 4||40 ± 4||76 ± 7|
|238U(234Th)||21 ± 2.0||12 ± 2.0||38 ± 3.0||25 ± 4||61 ± 5||84 ± 7|
|232Th(212Pb)||9.4 ± 0.7||8.6 ± 0.6||8.1 ± 0.5||54 ± 3||115 ± 7||182 ± 11|
|226Ra||63 ± 4.0||115 ± 7.0||179 ± 11.0||51 ± 3||123 ± 7||194 ± 11|
|228Th||8.7 ± 0.7||<6||6.8 ± 0.7||50 ± 2||100 ± 6||168 ± 11|
|228Ra(228Ac)||8.6 ± 0.8||8.8 ± 0.8||6.9 ± 0.7||212 ± 13||426 ± 26||674 ± 39|
|40K||528 ± 32.0||394 ± 24.0||347 ± 21||512 ± 31||493 ± 32||413 ± 24|
Taking into account that the wastes come from a NORM industry, it is necessary to carry out a radioactivity characterization of the SPCs obtained (Table 8). The radionuclides with the highest activity concentration in the PG‐SPCs are 226Ra (and its daughters of short half life) and 40K. As expected, the activity concentration of 226Ra shows a linear dependence (y = 5.49x + 8.95; R2 = 0.9988) on the percentage (x) of PG added. However, 40K concentration decreases linearly (y = -5.82x + 529; R2 = 0.9041). This is due to the fact that the potassium content in PG is practically negligible (<18 Bq/kg for 40K, or <0.06% in natural potassium).
In SPC‐IM samples, the radionuclide with the highest activity concentration is 228Ra. The activity concentration of 228Ra also shows a linear dependence (y = 18.7x + 59.6; R2 = 0.9976) on the percentage (x) of IM added to the SPC‐IM samples. A similar behavior was showed for 210Pb, 234Th, and 232Th. In relation to the 40K, the activity concentration decreases when we increase the IM content in the sample.
The activity concentration index (I) in the SPC‐PG samples is below the EU reference values for bulk building materials (Table 8). The Ra(eq) is also below the threshold of 370 Bq/kg considered in USA and some other countries. However, in the SPC‐IM samples, the index I is higher than the EU reference values for materials used in bulk amounts. Nevertheless, these could have other civil construction applications, for example, in marine platforms or bridges. The values of Ra(eq) are also higher than the reference level of 370 Bq/kg.
Due to the high radionuclide content to assess the risk of the use is important study the environmental mobility of these. In relation to the toxicity characteristic leaching procedure leaching test (TCLP, USEPA) , it can be confirmed that none of the radionuclides 234Th, 226Ra, 228Ra 40K, and 210Pb were detected in the leachates as their concentrations were below 1 Bq/L (leaching coefficients LC < 0.1%).
|Sample||pH||Ac (wt%)||238U(Bq/L)||210Pb(Bq/L)||210Pb Lc (%)||238U Lc (%)|
|SPC‐21||2||1.63||0.078 ± 0.003||<0.01||<0.01||0.27|
|4||1.01||0.143 ± 0.012||0.039 ± 0.019||0.19||0.48|
|6||1.25||0.078 ± 0.005||0.017 ± 0.011||0.09||0.26|
|8||0.95||0.0106 ± 0.0003||<0.01||<0.01||0.03|
|SPC‐21PG‐30||2||1.50||1.09 ± 0.05||0.97 ± 0.07||1.17||0.18|
|4||1.89||0.73 ± 0.05||0.15 ± 0.03||0.78||0.03|
|6||1.99||1.14 ± 0.04||0.067 ± 0.017||1.22||0.01|
|8||1.46||0.09 ± 0.03||<0.01||0.09||<0.01|
|10||0.87||0.05 ± 0.01||<0.01||0.05||<0.01|
|SPC‐21IM‐20||2||0.37||0.463 ± 0.025||0.871 ± 0.020||0.86||0.26|
|4||0.36||0.363 ± 0.078||0.891 ± 0.027||0.88||0.30|
|6||0.34||0.340 ± 0.020||0.862 ± 0.020||0.84||0.20|
|8||0.38||0.111 ± 0.008||0.906 ± 0.023||0.90||0.18|
|10||0.46||0.009 ± 0.004||0.807 ± 0.026||0.77||0.05|
In acidic media, pH 2 and 4, the concentration of 238U in SPC‐21 was between 0.078 and 0.143 Bq/L and for pH 8 0.011 Bq/L. These values are in agreement with those recorded for inland water bodies which are between 0.005 and 0.5 Bq/L . The SPC‐21PG‐30 leachate has a mean 238U concentration of around 1 Bq/L in acidic media higher than SPC‐21, falling to 0.07 Bq/L for basic media (close to the upper limits allowed for groundwater affected by mining residues [1 Bq/L]) . This, plus the results outlined in Table 9, shows that the contamination of water by uranium isotopes by SPCs containing 30% waste would be negligible.
The activity concentration of 210Po in SPC‐21 is between 0.039 and 0.017 Bq/L for acid media, below the mean 0.01 Bq/L recorded for basic media. In acid media, the 210Po concentration is higher in SPC–21PG–30 with values around the levels for inland water bodies but lower at pH of 8 and 10, that is, 0.01 Bq/L. The 238U concentration is around twice that of 210Po in the leaching so Po tends to be more strongly fixed to a particular material than U.
The activity concentration of 238U in the acidic media in SPC‐21IM‐30 is around 0.38 Bq/L, this value falling between 0.1 and 0.009 Bq/L for pH = 8 and 10, respectively. The contamination of water by uranium isotopes due to the SPC‐IM would be negligible. At all pH values, the 210Po concentration is very similar, around 0.8 Bq/L, with the leaching coefficient for 210Po being in general higher than for 238U by a factor of 3. For pH = 10, it can be higher by a factor of 15.
This result confirms that phosphogypsum and ilmenite mud can be successfully immobilized and valorized in the manufacture of SPC construction materials, which also allow for the safe disposal of waste sulfur.
2.3. Use of ilmenite mud in commercial cement manufacturing
Industrial wastes are widely used within the cement industry [3, 4, 47–49] either as partial or total substitution of raw material in clinker production or as an addition to clinker for cement production. In addition, it is important to note that the use in cement production of waste generated in several industrial activities should be undertaken with caution. Certain requirements should be met: (a) the physical and chemical properties of the cement generated should be at least comparable to the physical and chemical properties of the “normal” cement; and (b) the “new” cement should not cause any additional environmental problems (in other words, one environmental problem should not be replaced by another).
It is known that the moderate addition (5% or less) of pure TiO2 in the manufacture of Portland clinker increases the hydraulic activity (the capacity to react with water in order to gain high resistance) of the cement, acts as a mineralizer, reduces the porosity and increases the resistance [50, 51].
Taking into account that IM contains around 50% TiO2, we decided to explore the effects of adding the ilmenite mud (IM) directly to cement. The mixtures were IMA (97.5% Cem–2.5% IM), IMB (95% Cem–5% IM) and IMC (90% Cem–10% IM). Below we analyze and evaluate the main mechanical, elastic and thermodynamic properties of these cements. Their properties were compared with those of ordinary Portland cement (OPC)—Type I (clinker 97% and natural gypsum 3%) with compressive strength around 52.5 N/mm2).
The first step after the formation of the three dried mixtures of clinker with ilmenite mud was to determine the appropriate water/cement (W/C) relation in weight terms for obtaining a “normal” consistency of the paste formed . The results of experimentation show that the optimum water–cement ratio for each of the samples was 0.27 for OPC, IMA, and IMB and 0.28 for IMC. The setting times of the three cements formed using different proportions of IM as an additive were obtained with a Vicat apparatus . The setting time for OPC is 139 min. This time is extended to 158, 171, and 203 min for IMA, IMB, and IMC, respectively. On the other hand, the final setting time was 224 for OPC, and 201, 206, and 263 min for IMA, IMB, and IMC. The setting times for the three IM cements are comparable to those obtained for commercial cement and fulfill the requirements of national legislation. Note that the initial and final setting times are prolonged by increasing the proportion of IM used.
Mechanical resistance tests (compressive and flexural strength) were carried out for the new cements with ilmenite mud added and for the commercial cement (CEM). The results obtained are given in Table 10
|Sample||Flexural strength||Compressive strength|
|2 days||28 days||2 days||28 days|
|Commercial cement (CEM)||6.8 ± 0.3||10.1 ± 1.2||34.4 ± 0.4||61.3 ± 1.1|
|IMA (2.5% IM)||6.8 ± 0.1||9.8 ± 0.6||38.1 ± 1.4||55.3 ± 1.9|
|IMB (5% IM)||6.7 ± 0.3||9.0 ± 0.3||34.8 ± 0.9||55.9 ± 1.3|
|IMC (10% IM)||4.8 ± 0.3||8.0 ± 0.2||24.2 ± 0.8||45.5 ± 1.0|
These results show that the addition of a little ilmenite mud to the cement (2.5 and 5%) improves the mechanical strength. In particular, after to 2 days, the bending strength of the new cement that had 2.5% IM added is not affected but its compressive strength increased by approximately a 10%. After 28 days, the values decrease slightly, but is still within acceptable limits (>52.5 MPa) .
In order to analyze the possible risks of short‐time expansion, which can affect the ilmenite mud cements formed, the Le Chatelier test was performed . The results obtained for the new cements show that the presence of IM in the cements formed does not cause any significant modification in their expansion behavior in comparison with the commercial cement. This applies irrespective of the proportion of IM used as an additive, within the range tested.
Calorimetric analyses of the three IM cements were performed (Table 11) in order to determine the total heat liberated in their hydration process. These values clearly indicate that the addition of IM diminishes the rate of heat release in the acceleration–deceleration peak, down to 12.4 kJ/kg h in the IMC sample. This is 28% lower than that obtained for the commercial cement. The addition of IM also retards the appearance of this peak until 7.23 h in the IMB sample or 8.61 h in the IMC sample. The total heat liberated in the cements formed with IM is slightly lower in all the samples, at around 245 kJ/kg.
|Sample||Acceleration–deceleration peak||Total heat after 60 hours (kJ/kg)|
|V(kJ/kg h)||tV (h)||QV (kJ/kg)|
Finally, a thorough analysis from the environmental point of view was carried out. The external risk index (I) of samples was measured. For IMA, the value is 0.68, that is, below unity. Therefore, this sample can be used without restrictions. However, sample IMB presents an index slightly above unity at 1.09. IMC has a value close to 2 which is a problem from a radiological point of view. In relation with radium‐equivalent Ra(eq), the values obtained for IMA, IMB, and IMC were 190, 308, and 544 Bq/kg, respectively, showing that IMA and IMB are below the 370 Bq/kg level, but IMC is above it.
To assess the risk of the use of IM to the environment, a TCLP leaching test was carried out . The pollutant concentrations (metals and radionuclides) in the leaching solutions obtained from the mobility tests were measured. The leaching test shows that the concentrations of radioisotopes are of the same order of magnitude as typical ones in continental waters , and it can be ensured that their potential radiological impact is negligible.
The content of Cr(VI) in the samples must also be analyzed, because European directive  prohibits the marketing and use of cement which, when hydrated, contains more than 2 mg/kg of water‐soluble hexavalent chromium, determined as percentage by mass of dry cement. In Table 12, we can see that the concentration of Cr (total) in the IMC sample is 70 µg/L and taking into account that the dilution factor used was 20, the final concentration of Cr by mass of dry cement is 1.4 mg/kg, lower than 2 mg/kg. Also, values for other heavy metals are similar or even lesser than the cement taken as reference.
In order to gain a better understanding of how the ilmenite mud impurities are incorporated in the final cement product, a detailed study by SEM was made of the IMC cement sample. This sample is the one that contains the biggest proportion of ilmenite mud (10%). In Figure 3, one can see a main dark matrix (point 1) corresponding to the cement formed, with a composition typical of a calcium silicate hydrate (C–S–H) gel and a Ca–Si ratio = 2.2 ± 0.1. The bright particles embedded in the matrix are formed by iron plus titanium (point 2). This can be deduced from the corresponding X‐ray spectra obtained by XRMA at these points, which are also shown in Figure 3.
All the SEM‐XRMA results indicate that the particles trapped in the matrix are high in iron and titanium. This is important because it reduces the potential leaching of the iron and titanium contaminants in the cement formed and, consequently, limits their potential harm to the environment.
2.4. Ilmenite mud as additive in the manufacture of ceramic
In line with the research to date, and noting the composition of the IM , another possible valorization method for IM is its use in the production of ceramic tiles. Three fundamental factors need to be borne in mind when considering the possibility of using ceramic tiles for the recycling of IM. Firstly, it is essential to study whether the presence of this residue modifies the mechanical properties of the tile compared with commercial tile. Secondly, these studies require prior physico‐chemical analysis (elemental, mineralogical, and morphology) of the waste and the components used in the generation of tile tested. Finally, it is essential to check its environmental impact mainly associated to the potential problem of leaching of metals and radionuclides included in the tile matrix [55–58].
Ilmenite mud exhibits a high content of iron oxide (Fe2O3 ≈ 10%). This is a similar value to that of red stoneware (RSM) which is made from natural clays with high iron oxide content (Fe2O3 > 7%), (Table 4). For this reason, its valorization in ceramic tiles has been trialed . Mixtures of a commercial red stoneware (RSM) with different concentrations of ilmenite mud (3, 5, 7, 10, 30, and 50%) were shaped by uniaxial pressing (Nannetti S hydraulic press) at 40 MPa in a steel die to produce tiles measuring 50 × 50 × 5 mm, which were fired in an electric furnace at 1150°C following a fast‐firing process for 8 minutes.
The sintering behavior of the ceramic was evaluated on the basis of water absorption, apparent porosity, and bulk density. The water absorption E (wt%), was measured according to EN ISO 10545‐3  for ten representative specimens. The apparent porosity and the bulk density were measured according to ASTMC373‐88 . Also, the linear shrinkage, LS (%), and the bending strength, BS (MPa), were measured according to EN 843‐1 .
The results shown in Table 13 indicate that the linear shrinkage increases with the concentration of ilmenite mud showing values up to 7%. Lower values are advantageous because they reduce cracking and volume changes during firing. Also, apparent porosity generally increases with the concentration of ilmenite mud. This physical property is very important, and it is directly related with the water absorption and open porosity . Moreover, the water absorption decreases with the addition of 3 and 5% of sludge (2.82 and 2.48%, respectively). In accordance with the European Standard EN 14411 , tiles are categorized according to the water absorption coefficient (E). So tiles with low water absorption belong to group I, which in turn, is divided in two sub‐groups (a) E ≤ 0.5% (called BIa group) and (b) 0.5% < E ≤ 3% (called BIb group). Tiles with 3% < E ≤ 6% and 6% < E ≤ 10% belong to the BIIa and BIIb groups, respectively. This fact is important because of low values of both E and P provides a high resistance to freeze‐thaw cycles, requiring less drying time.
|Linear shrinkage LS (%)||Apparent porosity|
|100/0||3.8 ± 0.1||12.5 ± 0.7||5.30 ± 0.06||2.36 ± 0.02||35.1 ± 0.9|
|97/3||5.6 ± 0.1||6.71 ± 0.3||2.82 ± 0.05||2.38 ± 0.04||41.1 ± 0.8|
|95/5||6.0 ± 0.1||5.98 ± 0.2||2.48 ± 0.03||2.42 ± 0.05||40.2 ± 0.8|
|93/7||6.0 ± 0.1||10.0 ± 0.5||4.18 ± 0.04||2.40 ± 0.03||37.5 ± 0.7|
|90/10||6.4 ± 0.2||12.0 ± 0.7||4.72 ± 0.06||2.55 ± 0.03||36.5 ± 0.7|
|70/30||6.3 ± 0.1||19.8 ± 0.8||9.02 ± 0.10||2.22 ± 0.02||33.2 ± 0.8|
|50/50||6.6 ± 0.1||20.9 ± 0.9||9.18 ± 0.12||2.27 ± 0.05||30.8 ± 0.9|
The bulk density increases with the concentration of sludge up to 10% due to the high density of sludge (3.7 g cm-3). The bending strength increases with the addition of IM up to 10% due to the beneficial effect of ilmenite on sintering during firing as denoted by the decrease in porosity (Table 13).
The 97/3 and 95/5 samples show the highest values of bending strength, that is, 41 and 40 MPa, respectively. These are above 30 MPa, the minimum value required for group BIb . On the other hand, values corresponding to 100/0, 93/7, and 90/10 samples are greater than for the BIIa group (22 MPa). Finally, in the 70/30 and 50/50 samples, bending strength decreases due to the increase of interconnecting open pores.
In order to analyze how the ilmenite mud is incorporated to the final product, a detailed study by FESEM was made. The images (Figure 4) show good sintering and a homogeneous microstructure free of internal defects such as cracks.
It is important to note the existence of both open (<5 µm) and closed (<10 µm) pores in the samples. Once the allotropic transformation is carried out at 573°C, quartz particles produce shrinkage giving rise to cracks when the piece cools . The addition of the moderate percentages of ilmenite mud has a benefit, decreasing the open porosity as we can see in 97/3 and 95/5 samples.
Recall that the original mud had a total radionuclide concentration of 2000–3000 Bq/kg, with the highest activity concentrations for 226Ra and 228Ra being around 500 and 2000 Bq/kg, respectively, (Table 5). In the tiles, the activity concentration of these radionuclides increased with increasing ilmenite mud content, as did those of 228Ra, 228Th, and 238U. The activity concentration of 40K was constant in all samples. The index (I) for ceramic samples is lower than six for all of the analyzed materials. This means that ilmenite mud suitable material for use in the ceramic industry in comparison with other additives .
A TCLP leaching test was also applied to evaluate the potential environmental impacts generated by hazardous metals and radionuclides contained in tiles. The results indicate that increasing the proportion of ilmenite mud in mixture decreases the leached metals. Obviously, the firing process makes metal less leachable. These values are lower than the limits imposed by the US EPA and will not have a significant impact when this material is released into the environment. The only exception is 210Po (Table 14).
|RSM||29 ± 3||1.5||36 ± 4||1.8||50 ± 17||1.7||129 ± 28||7.6||7.3 ± 1.9||0.4|
|97/3||10.1 ± 2.1||0.6||10.1 ± 2.1||0.5||18 ± 6||0.6||54 ± 10||2.5||12 ± 2||0.7|
|95/5||4.8 ± 1.4||0.2||6.3 ± 1.7||0.3||33 ± 11||1.0||75 ± 17||3.0||9.6 ± 1.9||0.6|
|93/7||6.3 ± 1.4||0.3||5.1 ± 1.2||0.3||10.5 ± 3.6||0.4||24 ± 5||1.3||10.8 ± 2.9||0.3|
|90/10||6.4 ± 1.4||0.4||6.4 ± 1.4||0.4||5.1 ± 1.6||0.2||8.0 ± 2.0||0.4||5.5 ± 0.8||0.3|
|IM||2.1 ± 0.7||0.1||5.3 ± 1.2||0.3||37 ± 9||1.0||99 ± 18||4.3||9.7 ± 2.8||0.1|
The motilities of U‐ and Th‐isotopes are of the same order of magnitude for the various samples studied, that is, 10–100 mBq/L, apart from Po which is one order of magnitude lower (<10 mBq/L). This fact is explained by the tendency of Po to bind onto the particulate material .
Therefore, that the impact of the use of ilmenite mud as an additive in ceramics is negligible from the environmental point of view.