Valorisation of Waste Ilmenite Mud in The Manufacture of Sulphur Polymer Cement
Valorisation of Waste Ilmenite Mud in The Manufacture of Sulphur Polymer Cement
Valorisation of Waste Ilmenite Mud in The Manufacture of Sulphur Polymer Cement
1
Departamento de Física Aplicada, Universidad de Huelva 21071, Huelva, Spain
2
Centro Nacional de Investigaciones Metalúrgicas (CENIM), CSIC, Avda. Gregorio del Amo 8,
Corresponding author: Manuel Jesús Gázquez *Phone: +34 959 21 97 98; Fax: +34
ABSTRACT
This paper reports the preparation of sulphur polymer cements (SPCs) incorporating
waste ilmenite mud for use in concrete construction works. The ilmenite mud raw
chemically and radiologically. The optimal IMC-SPC mixture had a sulphur/mud ratio
(w/w) of 1.05 (mud dose 20 wt %); this cement showed the greatest compressive
strength (64 MPa) and the lowest water absorption coefficient (0.4 g.cm-2 at 28 days).
containing 10% and 20% ilmenite mud met the demands of international standards for
1. INTRODUCTION
industrial wastes is an area of great environmental and economic interest (Chen et al.,
2010; Cruz-Yusta et al., 2011; Puertas et al., 2008;Yan et al., 2011). This paper
describes the stabilisation and valorisation of waste ilmenite mud, generated in the
titanium dioxide industry, as a component of sulphur polymer cement (SPC) that could
TiO2production begins with the mixing of ilmenite and highly concentrated sulphuric
acid (80-95%). The liquor generated is passed to a clarification tank where the un-
attacked solid - ilmenite mud - is allowed to settle. This mud is finally separated from
the liquor by decantation and filtration (Gázquez et al., 2011). It is then neutralised and
Sulphur polymer cements have advantages over regular Portland cement in that they
harden in under 24 h, are of high compressive strength, show resistance to fatigue, are
little permeable to water, show exceptional resistance to acid and salt (allowing their use
in aggressive environments such as sea water, and under all weather conditions),and are
recyclable (ACI Committee 548, 1993; Amo and El Gamal, 2009). In addition, SPC
manufacture could make use of large amounts of sulphur waste, e.g., from oil refineries
and the metallurgical industry. SPCs can also be used as stabilising agents for other
kinds of waste (López et al., 2009; López et al., 2011; Mohamed and Gamal, 2007;
Sandrolini et al., 2006), including – potentially - ilmenite mud. Unfortunately, this type
2011); its immobilisation in SPCs used in the construction of occupied buildings might
112, 1999) that propose reference values for natural radionuclide concentrations in
building materials; Eq. (1) shows how the exposure risk index (I) is calculated:
226 232 40
Eq. (1)
300 / 200 / 3000 /
where C(226Ra), C(228Ra) and C(40K) are the respective activity concentrations for 226Ra,
228
Ra and 40K in the building material considered (expressed in Bq kg−1). It is
recommended that building material-induced indoor gamma doses do not exceed 1 mSv
per year.
In addition, ilmenite mud also contains a number of metals. If these were to leach out
of ilmenite mud-containing SPCs (IMC-SPCs), they too could cause environmental and
The aims of the present work were: (1) to study the stabilisation of ilmenite mud as an
additive in SPCs, allowing its valorisation as a component of building materials, and (2)
2.1 Materials
The raw materials used to make the SPCs examined in this work were granular
elemental sulphur (99.4 wt%, size <60 µm, type Rubber Sul 10) (Repsol-YPF, Madrid,
Spain), gravel (<6.3 mm) and a siliceous sand (<4 mm). A modified sulphur-containing
thermoplastic material in all the SPCs made. STXTM prevents crystalline changes inside
SPCs when these are exposed to changing temperatures over time(STARTcreteTM). The
ilmenite mud sample was supplied by Huntsman-Tioxide (Huelva, Spain). It was dried
Table 1 shows the SPCs produced. These included three IMC-SPCs made with
different doses of ilmenite mud (10, 20 and 30% w/w [SPC17-10, SPC21-20 and
SPC21-30 respectively]), and a control SPC (SPC21-0) with no ilmenite mud but
containing 99.5% pure, inert calcium carbonate (7.69% w/w) (Panreac(R), Barcelona,
Spain). (Note: the SPC17-10 was prepared with 17% instead of 21% sulphur to
optimise the workability of the product). All these SPCs were prepared using a
sulphur/STXTM ratio of 10 and a gravel/sand ratio of 0.5. Previous studies (Lopez et al.,
2009; McBee, 1979;Sandrolini et al., 2006) have shown these ratios to produce final
To prepare the SPCs, the aggregates (gravel, sand and mud) were heated in an oven to
135–140ºC for 4 h. The sulphur was liquefied in a mixing bowl within the same
temperature range for 10 min. These materials were then mixed into a homogeneous
viscous paste. At this point, STXTM was added, stirring for 4–5 min at 140–145ºC.
induce unwanted reactions between the STXTM and the sulphur, leading to a poorly
workable, highly viscous final material. Steel moulds (40×40×160 mm) were preheated
to approximately 120ºC and the forming SPCs poured into them. A vibration table set
at 3000 rpm for 30 s was employed to compact the cements in the moulds before storing
them at room temperature for 24 h. The hardened SPCs casts were then de-moulded.
Major elements were determined in the ilmenite mud, sand and gravel by X-ray
fluorescence (XRF) using a Bruker S4 Pioneer system (4 kW, Rh front window and
anode, five analysing crystals [LIF200, Ge, PET, OVO55 and OVOC] and two X-ray
possible. Thus 1 g samples of each dry SPC or original ilmenite mud were ground using
a pestle and mortar. The ground samples were then mixed with 10 g of lithium
tetraborate and 5 drops of 20% lithium iodide to form a homogenous glass ready for
examination.
Granulometric analyses of the ilmenite mud, gravel and sand were performed using a
Mastersize 2000 APA granulometer (Malvern Instruments Ltd.). For this, some 20-30 g
of each of these raw materials were placed in deionised water for 24 h. They were then
placed in a flask and mixed using a magnetic stirrer at a constant speed to ensure the
granulometric analysis.
The compressive (Cs) and flexural (Fs) strengths of the SPCs were measured according
Only one day of curing was necessary before performing this test with the SPC casts
since 80% of the final values reached are achieved within this time (López et al., 2009;
López et al., 2011; Sandrolini et al., 2006). Values for the OPC casts were measured at
two of the IMC-SPCs made with ilmenite mud (SPC17-10 and SPC21-20), and the
control SPC (SPC21-0) (Table 1), were dried at 50°C in an oven until a constant mass
was reached. They were then placed on a grating in a dish of water (at room
temperature), the liquid being allowed to wet only their lower surfaces. The lower parts
of the sides adjoining the inflow face were sealed with a polyethylene sheet to prevent
any water being absorbed into their surface pores. All experiments were performed in
triplicate. The WAC values were then determined using Eq. (2):
∆ 0
16 Eq. (2)
where Mj is the sample mass after 28 days, Mo the initial mass, and ‘16’ the wetted
The coefficients of absorption (AC) with respect to the pH were determined by placing
the different casts in deionised water for 24 h. They were then weighed, immersed for
42 days in buffers at pH 2, 4, 6, 8 or 10, and reweighed. The AC value for each cement
3 1
% 100 Eq. (3)
1
where P3i is the mass after 42 days of immersion in buffer, and P1i the initial weight of
2.4 Radioactive characterisationof the starting ilmenite mud, SPCs and SPC
leachates
Samples of the SPCs and mud were ground with a pestle and mortar. The activity
concentration of 226Ra was then determined at 352 keV (via the emission of 214Pb), that
of 228Radetermined at 911 keV (via the emission of 228Ac), and that of 228Th at 583 keV
(via the emission of 208Tl) (always taking into account the branching ratios in the decay
of these nuclides), using a gamma spectrometer (38% relative efficiency, full width at
half maximum FWHM 0.95 keV at the 122 keV line of 57Co and 1.9 keV at the 1333
210
Po and the Th and U isotopes were dissolved out of the SPCs and isolated using a
poor solubility of the ilmenite mud required the use of a total dissolution method based
on KHSO4 fusion previously validated by our group (Mantero et al., 2010). After
isolation, the U and Th isotopes were electrodeposited independently onto stainless steel
discs. The 210Po was obtained by self-deposition on silver discs. The activity
concentration of all three nuclides was then measured using an EG&G Ortec alpha-
The 210Po and the Th and U isotopes in the leachates from the SPCs (see below) were
electrodeposited in the same way before analysis. The isotopes of Ra were measured
directly.
The mobilities of the radionuclides and metals in the manufactured SPCs were
determined by leaching tests performed at different pHs. The procedures followed were
the same as those described for the determination of Ac. At 42 days, the radionuclide
radionuclide concentrations of the original buffers (i.e., before being used in the
The leaching coefficient (Lc) for 238U and 210Po at each pH was determined according to
1
% 1
100 Eq. (4)
1
(expressed in Bq L-1), Vt is the total volume of the solution after 42 days (0.250 L in all
cases), Asis the concentration of the corresponding radionuclide in the cement sample
(expressed as Bq g-1), and P1i the mass of the sample after 24 h of immersion in water
(expressed in g).
ULTIMA 2 apparatus. The system was calibrated using appropriate standards. Leachate
samples were diluted to 2% in a nitric acid solution for introduction into the apparatus.
Table 2 shows the concentrations (% dry weight) of the major elements in the raw
materials and ilmenite mud used to make the SPCs. The mud had a TiO2 content of 53
wt%, and contained appreciable amounts of iron oxide (12.5 wt%) and silicon oxide
(11.9 wt%). These figures are similar to those reported by other authors (Gázquez et al.,
concentration (expressed as SO3) was approximately 7.8 wt%; this is not surprising
given that the digestion process in which the mud is generated involves the use of
Figure 1a shows the distribution of raw gravel and sand particle size against the Spanish
Standard for structural concretes (Real Decreto 2661/1998); both materials met the
requirements for use. Figure 1b shows the percentage (by volume) of particles in the
ilmenite mud with a diameter between 1 and 2000 µm. The median d0.5 of the particle
Fig. 1. Particle size distribution for: a) sand and gravel, and b) ilmenite mud.
XRD. The minerals detected included ilmenite, rutile, zircon (ZrSiO4), quartz (SiO2)
and Fe3Ti3O10. The presence of rutile, zircon and quartz is not surprising since all are
Figure 3 shows the diffractograms for sulphur and STXTM. The diffraction lines
appearing at 22.8º and 27.8° (2θ) for the modified sulphur STXTM correspond to the Sβ
crystalline phase. These lines are better observed in the magnifications provided in the
3.2 Compressive and flexural strengths of the SPCs, and their WAC and Ac values
Table 3 shows the Fs and Cs values recorded for the different SPCs. SPC21-20
returned the highest Cs (64 MPa). López et al. (2009), reported values of 54 and 58
MPa for metacinnabar-containing SPCs with sulphur/HgS ratios (w/w) of 0.4 and 2.6
sulphur/phosphogypsum ratio of 0.9 (López et al., 2011). Mohamed and Gamal (2007)
reported a Cs of 54 MPa for an SPC manufactured with fly ash and with a sulphur/fly
ash ratio (w/w) of 0.9. Sandrolini et al. (2006) reported a Cs of 65-73 MPa for an SPC
made with an aggregate recycled from the ceramics industry. Ordinary Portland cement
(OPC) contains clinker (97%) and natural gypsum (3%), and has a Cs of >52.5 MPa
the present work, the OPC had an Fs of 10.1±1.2 MPa after 28 days of curing, and a Cs
of 61.3±1.0 MPa, values similar to or slightly lower than those returned by SPC17-10
and SPC21-20 after just one day of curing (Table 3). The Cs increased as the
which point all the particles would have been coated with a thin layer of sulphur. This
would act as a good binder for the aggregate particles, increasing the final Cs. However,
the addition of more sulphur (i.e., in the SPC21-30 mixture) would have produced a
thick layer of sulphur around the aggregates, increasing the brittleness of the final
composite. Given the poor Cs value recorded for SPC21-30, this cement was left out of
After 28 days, SPC21-20 returned a WAC value of 0.6 kg m-2, slightly lower than the
0.9 kg m-2 of the SPC21-0. However, both values are smaller than those reported for
Portland cements. For example, Khatib and Clay (2004) report a WAC of 5.0 kg m-2 for
OPC after 28 days of immersion in water, while Medeiros and Helene (2009) report a
value of 6.2 kg m-2 for Brazilian Portland cement (CPII E-02 or slag-modified Portland
cement) after the same immersion time. SPC21-20 would therefore appear to be a very
impermeable SPC.
Table 5 shows the Ac values for the different SPCs. The SPC21-0 control had a higher
Ac than SPC17-10 and SPC21-20 over the full pH range tested (1.63% at pH 2 to 0.75%
3.3.1 Activity concentrations and activity concentration indices of the ilmenite mud and
SPCs
Table 4 shows the results of the radioactive characterisation of the ilmenite mud and
SPCs. The original mud had a total radionuclide concentration (2.5–3.0 Bq g-1) that
surpassed the 1 Bq g-1 EU safety threshold for naturally occurring radioactive material
(NORM) wastes (IAEA, 2004). The radionuclides with the highest activity
concentrations were 226Ra and 228Ra at around 520 and 1900 Bq kg-1 respectively
(Gázquez et al., 2009b; Gázquez et al., 2011). These concentrations are significantly
higher than the average worldwide values for soils (25 Bq·kg-1 of 238U and 232Th in
secular equilibrium with their daughters) (UNSCEAR, 1993). In the IMC-SPCs, the
did those of 210Pb, 228Th, and 238U. The activity concentration of 40K decreased as the
The activity concentration index (I) of all the SPCs was higher than the EU reference
value for occupied buildings (Radiation Protection, 112, 1999). This index should not
exceed 1 for any material used in bulk quantities, e.g., concrete. Nevertheless, the
present IMC-SPCs could have other civil construction applications, e.g., in marine
shown below.
Neither 226Ra, 234Th, 228Ra, 40K nor 210Pb were detected in the leachates of any SPC,
238
leading to inappreciable Lc values. U and 210Po were detected among the alpha
particle emitters. Th isotope levels were below the detection limit in most of the
samples tested (0.5-1 mBq). This is explained in that Th tends to bind strongly to
particulate matter.
The highest activity concentration of 238U in the SPC21-0 leachate was 0.14 Bq L-1 at
pH 4, and 0.01 Bq L-1 at pH 8, similar to the values recorded for inland water bodies
(typical range 0.005–0.5 Bq L-1) (Más et al., 2006). The SPC17-10 leachate had a 238U
activity concentration of around 0.12 Bq L-1 in the acidic buffers (pH 2-6), falling to
0.01 Bq L-1 at pH 8, and 0.002 Bq L-1 at pH 10. The SPC21-20 leachate returned similar
therefore be negligible.
Table 5 shows the activity concentrations recorded for 210Po to be one third those
recorded for 238U (Más et al., 2006) at pH 2, 4 and 6 in all the SPC leachates examined.
The 210Po activity concentrations for SPC21-0 were 0.039 Bq L-1 at pH 4 and 0.017 Bq
L-1 at pH 6, values higher than those obtained at pH 8 and pH 10. The SPC17-10
leachate returned 210Po activity concentrations over the pH 2-6 range similar to those of
SPC21-0, while at pH 8 and pH 10 the values were higher, although of the same order
of magnitude as reported for continental waters (Más et al., 2006). The 210Po in the
SPC21-20 leachate clearly decreased with pH but levels were below the Spanish
Table 6 shows that Pb, Cd, As and other elements in the leachates were below their
detection limits. No significant differences were seen between SPC17-10 and SPC21-
20, indicating these pollutants to be very well bound to the cement matrix. Mg, Al, Ti,
Mn and Fe were more mobile at low pHs (especially pH2). The K concentrations of the
4. Economic viability
In Spain, the Huntsman-Tioxide company produces 64,000 t of TiO2 every year, and
about 30,000 t/year of ilmenite mud waste. According to Decision 2000/532/EC of the
European Union Landfill legislation (2003/33/EC) establishes that this waste must be
safely stored. The costs of such storage differ by country. In Spain, which has great
storage capacity but little industrial activity at the present time, these costs are very low
at about €100/t. In contrast, in the UK, they are much greater. The cost of making
SPC21-20 (including the raw materials and energy required) is some €125/t according
mud).
SPC21-20 could be used in exactly the same way as the type of concrete used in marine
environments. In Spain, such concrete costs about €75/m3 or €32.6/t. The economic
reflected by the equation: storage cost + sale price – manufacturing costs. In Spain,
production may not be economically viable, but, for example, in the UK, it may well be.
4. CONCLUSIONS
The present results show that ilmenite mud from the TiO2 industry can be successfully
was seen between those of SPC21-20 and SPC21-0. The WAC values recorded for the
IMC-SPCs were slightly lower than that obtained for the control SPC,and much lower
The leaching tests showed the mobility of radionuclides and metals in the IMC-SPCs to
be negligible. The leaching coefficients obtained in acidic media were always higher
than those obtained in alkaline media, but were always lower than international
recommendations.
Although not suitable for the construction of occupied buildings, SPC21-20 in particular
could be safely used in concrete construction works, e.g., in the construction of pump
foundations, walls for storage areas, underwater joint protections for steel pipes, and
sewer systems.
ACKNOWLEDGEMENTS
This work was supported by the Regional Government of Andalusia through the
Huelva for their environmental management and control” (grant RNM‐6300). Dr. Irene
García expresses her gratitude to the Spanish National Research Council (CSIC) for her
Social Fund.
REFERENCES
under document number C(2000) 1147). Official Journal L 226, 06/09/2000 pp. 0003 –
0024.
procedures for the acceptance of waste at landfills pursuant to Article 16 of and Annex
ACI Committee 548, 1993. Guide for mixing and placing sulphur concrete in
Chen, Y., Chang, J., Shih ,P., Ko, M., Chang, Y., Chiang, L., 2010. Reusing pretreated
Cruz-Yusta, M., Marmol, I., Morales, J., Sanchez, L., 2011. Use of olive biomass fly
ash in the preparation of environmentally friendly mortars. Environ. Sci. Technol. 45,
6991–6996.
Gázquez, M.J., Bolívar, J.P, Garcia-Tenorio, R., Vaca, F., 2009a. Physicochemical
characterization of raw materials and co-products from the titanium dioxide industry, J.
Gázquez, M.J., Mantero, J., Bolívar, J.P, Garcia-Tenorio, R., Galán, F., 2009b.
industry, 1st Spanish National Conference on Advances in Materials Recycling and Eco
Gázquez, M.J., Mantero, J., Bolívar, J.P., García-Tenorio, R., Vaca, F., Lozano, R.L.,
of the Concepts of Exclusion Exemption and Clearance, Safety guide No RS-G 17.
STI/PUB/1202.
Khatib, J.M., Roger, M., 2004. Absorption characteristics of metakaolin concrete. Cem.
López, F.A., Román, C.P., Padilla, I., López-Delgado, A., Alguacil, F.J., 2009. The
of the 1st Spanish National Conference on Advances in Materials Recycling and Eco-
Energy (RECIMAT'09) Madrid, Spain, November 12-13, 2009. Editors: F.A. López, F.
López, F.A., Gázquez, M.J., Alguacil, F.J., Bolivar, J.P., García-Díaz, I., López Coto, I.,
Mantero, J., Lehritane, M., Hurtado, S., García-Tenorio, R., 2010. Radioanalytical
Mas, J.L., García-León, M., García-Tenorio, R., Bolívar, J.P., 2006. Radionuclide
Bureau of mines report no. RI 8346. US Bureau of Mines, Washington, DC, p 21.
Medeiros, M.H.F, Helene, P., 2009. Surface treatment of reinforced concrete in marine
Mohamed O, A.M, Gamal, M., 2007. Sulfur based hazardous waste solidification.
Oliveira, J.M., Carvalho, F.P., 2006. Sequential extraction procedure for determination
Perez Moreno, J.P., San Miguel, E.G., Bolívar, J.P., Aguado, J.L., 2002 . A
Puertas, F., García-Díaz, I., Barba, A., Gazulla, M.F., Palacios, M., 2008. Ceramic
wastes as alternative raw materials for Portland cement clinker production. Cement.
Sandrolini, F., Manzi, S., Andrucci, A., 2006. Sulfur-polymer matrix composites from
37, 695-702.
Sandrolini, F., Manzi, S., Andrucci, A., 2006. Sulfur-polymer matrix composites from
resistencias mecánicas.AENOR.
UNE-EN 480-5: 2006. Admixtures for concrete, mortar and grout. Test methods. Part 5:
Committee on the effects of Atomic Radiation. Report to the General Assembly with
Yan, S., Sagoe-Crentsil, K., Shapiro, G., 2011.Reuse of de-inking sludge from waste