Crystal chemistry and temperature behavior of the natural hydrous borate colemanite, a mineral commodity of boron

Abstract

Colemanite, CaB₃O₄(OH)₃⋅H₂O, is the most common hydrous Ca-borate, as well as a major mineral commodity of boron. In this study, we report a thorough chemical analysis and the low-temperature behavior of a natural sample of colemanite by means of a multi-methodological approach. From the chemical point of view, the investigated sample resulted to be relatively pure, its composition being very close to the ideal one, with only a minor substitution of Sr²⁺ for Ca²⁺. At about 270.5 K, a displacive phase transition from the centrosymmetric P2₁/a to the acentric P2₁ space group occurs. On the basis of in situ single-crystal synchrotron X-ray (down to 104 K) and neutron diffraction (at 20 K) data, the hydrogen-bonding configuration of both the polymorphs and the structural modifications at the atomic scale at varying temperatures are described. The asymmetric distribution of ionic charges along the [010] axis, allowed by the loss of the inversion center, is likely responsible for the reported ferroelectric behavior of colemanite below the phase transition temperature.

Appendices

Appendix

A thorough chemical analysis of the studied colemanite sample has been performed by adopting a multi-methodological approach. In the following, a detailed description of the experimental techniques, reported in the section “Chemical analysis”, is given.

Titrimetric determination of boron

120–200 mg of the sample of colemanite were placed in a 50-ml plastic test tube. 5 ml of water and 3 ml of HCl 1M were added. The plastic test tube was then covered and transferred in an ultrasound bath for 1–2 h. The resulting clear solution was transferred in a 200–300 ml beaker with water up to about 100 ml of total solution.

A combined glass electrode (InLab^® Routine Pro – Mettler Toledo) was immersed in the solution and the pH was adjusted to 5.5–6.5, by the addition of HCl 0.1–1 M and NaOH 0.1–1 M. 5–6 g of mannitol were added and stirred until complete dissolution. The resulting solution was titrated with NaOH 0.1 M up to pH 8.3–8.7.

Due to the absence in the sample of hydrolyzing elements, as well as of elements able to influence the acidity of the solution, it can be inferred that the titrated acid content is entirely due to the presence of boric acid in solution. The results are reported in Table S1 (supplementary materials).

Ethylenediaminetetraacetic acid (EDTA) titrimetric method of calcium

40–80 mg of colemanite were placed in a 50-ml plastic test tube along with 5 ml of water and 1 ml of HCl 1M. The plastic tube was subsequently covered with lid and transferred in an ultrasound bath for 1–2 h.

The resulting clear solution was transferred in a 300–400-ml beaker and diluted to 200 ml with water. 10 ml of buffer solution (pH = 10 mixture of ammonium chloride/ammonia) and 3–4 drops of Eriochrome black T solution (2g/l in ethanol) were added. The solution was titrated with a standard solution of EDTA 0.01 M [solution of EDTA with 5 × 10⁻⁴ mol/l of magnesium chloride hexahydrate (MgCl₂⋅6H₂O)], until its purple color was altered to blue. The total volume of EDTA added to the solution is proportional to the average content of Ca and Sr in the colemanite sample (Table S2, supplementary material).

Determination of carbon and hydrogen by thermal decomposition and detection by infrared absorption (C, H)

100–300 mg of the natural sample of colemanite were decomposed at 950 °C in an elementary analyzer LECO Truspec CHN, in excess of oxygen for 90 s. The products of decomposition were passed through a second furnace (Afterburner) at 850 °C for a further oxidation and particulate removal. The gases, collected and homogenized in a container of 4.5 liters at 50 °C, were sent to the detectors for infrared absorption for the measurement of CO₂ and H₂O (i.e., carbon is measured in the form of CO₂ and hydrogen in the form of H₂O).

EDTA, sodium tetraborate decahydrate (borax, Na₂B₄O₇·10H₂O), boric acid (H₃BO₃), calcium carbonate (CaCO₃), sodium nitrate (NaNO₃) and oxalic acid (dehydrated) (HO₂CCO₂H·2H₂O) were used as calibration standards. The results are reported in Table S3 (supplementary material).

Determination of fluorine by ion selective electrode

5–20 mg of the colemanite sample were placed in a 50-ml plastic test tube along with 5 ml of water and 3 ml of hydrochloric acid 1M. The plastic test tube was covered and transferred in an ultrasound bath for 1–2 h. Later, 2–3 ml of total ionic strength adjustment buffer (commercial solution TISAB III) were added to the solution and diluted to 20 ml with water. The content of fluorine was determined using a perfectION^™ combination fluoride ion selective electrode (from Mettler Toledo) adopting the standard addition method of certified reference material solution of F from 0.1 to 2 mg/l (Table S4, supplementary material).

Determination of minor elements by ICP-AES

All the measurements were performed in radial view mode with an ICP/AES Perkin Elmer Optima 7000DV spectrometer. For the rare earth elements, 5–20 mg of colemanite sample were placed in a 50-ml plastic test tube along with 5 ml of water and 3 ml of hydrochloric acid 1M. The plastic test tube was covered and transferred in an ultrasound bath for 1–2 h. The solution was transferred and diluted with water in a 25-ml volumetric flask. A blank solution and series solution for calibrations were prepared carrying out the same procedure without the sample. A certified reference material (CRM) multi-elemental standard mix for ICP (50 mg/l) of each element was used for preparing the solution series for calibration (five solutions from 0.05 to 1 mg/l for each element).

For the analysis of the other elements investigated by ICP-AES (Table S5, supplementary material), the decomposition of the colemanite sample was obtained by alkaline fusion of 5–20 mg in a platinum crucible with 100 mg of Na₂CO₃ or K₂CO₃ in a muffle furnace at 1000 °C for 5 min, followed by dissolution in 10 ml of water and 1 ml of H₂SO₄ 1M or 1 ml of HCl 1M. The resulting clear solution was transferred and diluted with water in a 25-ml volumetric flask containing 2.5 ml of Sc solution (100 mg/l) as internal standard.

Determination of chloride, bromide, iodide by ion chromatography

25 mg of the colemanite sample were placed in a 50-ml plastic test tube along with 20 ml of water and one drop of nitric acid. The plastic test tube was covered and transferred in an ultrasound bath for 1–2 h. The solution was transferred and diluted with water in a 25ml volumetric flask. A blank solution and series solutions for calibrations were made carrying out the same procedure without the sample. CRM containing 100 mg/l of each element was used for preparing the solution series for calibration (five solutions from 0.1 to 2 mg/l for each element). The analysis was performed using a Dionex ICS-1600 Standard Integrated IC system equipped with Columns ION PACK AG23 Guard 4 × 50 mm + AS23 4 × 250 mm. A solution of KOH 10 mM was used as eluent. The results are reported in Table S6 (supplementary material).

Determination of water content by heating

500–600 mg of the colemanite sample were placed in a quartz crucible with lid and gradually heated in a muffle furnace from ambient temperature up to 800 °C. Between 300 and 400 °C, the sample was partially lost during heating, the weight loss being due to H₂O and partially to inorganic matter. B, Ca and Sr were analyzed in the residual sample to provide suitable corrections. The results are reported in Table S7 (supplementary material).