Magnesium silicate processing

Abstract

Methods of processing magnesium silicate materials are described to produce a number of products including magnesium hydroxide. Related methods of use of processed magnesium silicate and other reaction products are described for energy production, cement manufacture and carbon sequestration. In one embodiment the method comprises subjecting a magnesium silicate source to an acid digestion; increasing the digested liquid pH to produce a magnesium salt solution; subjecting the magnesium salt solution to electrolysis; and recovering magnesium hydroxide produced from electrolysis. By-products such as silica, iron oxy(oxides) and others are also described along with further reaction products such as magnesium oxide and magnesium carbonate.

Claims

1. A method of processing a magnesium silicate source by the steps of: selecting a magnesium silicate source selected from: olivine, serpentine, pyroxene, amphiboles, phyllosilicates, clays, and combinations thereof, subjecting the selected magnesium silicate source to water digestion at a temperature of less than 120? C. and wherein the ratio, by mass, of magnesium silicate source to water in the water digestion is 1 part magnesium silicate to 1 to 20 parts water to form a washed magnesium silicate; subjecting the washed magnesium silicate to acid digestion, to produce a digested solution, acid digestion comprising mixing the washed magnesium silicate at a temperature of less than 120? C. with hydrochloric acid or sulfuric acid sufficient to decrease the pH of the mixture to ?1-6 and, recovering evolved hydrogen gas during acid digestion; completing a base wash by increasing the digested solution pH by addition of an alkali solution at a temperature of less than 120? C., to produce a magnesium salt solution consisting of magnesium chloride, magnesium sulphate, or both magnesium chloride and magnesium sulphate, wherein the base wash comprises two pH increasing steps including: a first increase in digested solution pH by at least 1 to 3 pH greater than the digested solution pH and removing precipitated silica from the magnesium salt solution; and a subsequent pH increase in digested solution pH to a pH of 6.0 or higher and removing precipitated iron oxide from the magnesium salt solution; subjecting the magnesium salt solution to electrolysis at a temperature of less than 120? C.; and recovering magnesium hydroxide produced from electrolysis from an electrolysis cathode and oxygen and chlorine from an electrolysis anode.

2. The method as claimed in claim 1 wherein the magnesium silicate source is processed to a reduced particle size prior to or during acid digestion.

3. The method as claimed in claim 2 wherein the mean particle size is less than approximately 2 mm.

4. The method as claimed in claim 1 wherein a further water digestion step occurs after acid digestion and before the base wash, the further water digestion step occurring at a temperature less than 120? C. and wherein the ratio, by mass, of digested solution to water in the further water digestion step is 1 part digested solution to 1 to 20 parts water to form a washed digested solution.

5. The method as claimed in claim 1 wherein the alkali is selected from the group consisting of: magnesium hydroxide, calcium hydroxide, potassium hydroxide, sodium hydroxide, and combinations thereof.

6. The method as claimed in claim 1 wherein the electrolysis anode is selected from: carbon, graphite, activated titanium, mixed metal oxides (MMO), and combinations thereof.

7. The method as claimed in claim 1 wherein the electrolysis cathode is selected from: platinum, activated titanium, mixed metal oxides (MMO), nickel based alloy, and combinations thereof.

8. The method as claimed in claim 1 wherein an additional step is completed after recovery of the magnesium hydroxide of: dehydrating the magnesium hydroxide to produce magnesium oxide (MgO).

9. The method as claimed in claim 8 wherein dehydration occurs at a temperature of less than 1200? C.

10. The method as claimed in claim 1 wherein the method comprises an additional step of: reacting the recovered magnesium hydroxide with carbon dioxide to form a magnesium carbonate containing compound.

11. The method as claimed in claim 1 wherein the magnesium silicate source is processed to a reduced particle size prior to or during acid digestion and wherein the reduced particle size is of variable size and shape and not crystalline.

12. The method as claimed in claim 1 wherein acid digestion occurs at a temperature of approximately 60? C.

13. The method as claimed in claim 1 wherein the acid used during acid digestion is sulfuric acid only.

14. The method as claimed in claim 1 wherein sufficient acid is added during acid digestion to reduce the pH to 2-5.

15. The method as claimed in claim 1 wherein the electrolyzer comprises a porous membrane for separation of the electrolysis anode and the electrolysis cathode.

16. The method as claimed in claim 1 wherein the electrolyzer comprises a cation selective membrane to separate the anode and cathode.

17. The method as claimed in claim 1 wherein the electrolyzer is a membrane free electrolyzer.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) Further aspects of the magnesium silicate processing methods and uses of the processed magnesium silicate products will become apparent from the following description that is given by way of example only and with reference to the accompanying drawings in which:

(2) FIG. 1 shows an overall process for recovery of magnesium hydroxide Mg(OH).sub.2 with additional products being silica, iron oxy(oxide), magnesium oxide MgO, hydrogen gas H.sub.2 and acids from magnesium silicate hydrolysis, digestion, electrolysis, and/or dehydration;

(3) FIG. 2 shows a simplified carbon sequestration process using magnesium hydroxide. Note that the Mg carbonation product is likely to be hydrated magnesium carbonate such as nesquehonite, dypingite, or hydromagnesite, but magnesium carbonate MgCO.sub.3 formation is also possible depending on reaction conditions;

(4) FIG. 3 illustrates the characterization of recovered material wherein a) shows a scanning electron microscope SEM image of recovered magnesium hydroxide Mg(OH).sub.2 from electrolysis of olivine digestion solution, b) shows thermo gravimetric analysis TGA of recovered magnesium hydroxide Mg(OH).sub.2 from olivine, and c) shows x-ray powder diffraction XRD of recovered magnesium hydroxide Mg(OH).sub.2 and of recovered silica (B: brucite, L: lizardite, F: forsterite);

(5) FIG. 4 shows a thermo gravimetric analysis TGA for magnesium hydroxide exposed to air (containing CO.sub.2) at ambient temperatures and pressures. The small peak at approximately 500? C. indicates the presence of a magnesium carbonate compound due to the reaction between the magnesium hydroxide and the air. The other sample shown is the unreacted magnesium hydroxide which does not show a carbonate peak.

(6) FIG. 5 Illustrates the distribution of ultramafic rocks and olivine lifetime estimates. The general distribution of ultramafic rocks (including peridotites and serpentinites) is shown worldwide. Ultramafic rock distributions are based on location data from art publications. The billions of tonnes of olivine per year needed to sequester all anthropogenic carbon dioxide CO.sub.2, and reduce global atmospheric CO.sub.2 as well as the lifetime supply of olivine from the Red Hills (New Zealand) and Semail Ophiolite (Oman) is shown based on calculations completed by the inventors;

(7) FIG. 6 shows cumulative heat measurements related to hydration reactions for the CMgO+SF and RMgO+RS over a period of 72 hours using isothermal calorimetry; and

(8) FIG. 7 shows XRD patterns hydrated samples (P: Periclase, B: Brucite, L: Lizardite, grey bands correspond to M-S-H).

WORKING EXAMPLES

(9) The above described magnesium silicate processing methods and uses of the processed magnesium silicate products are now described by reference to specific examples.

Example 1

(10) The production and synthesis of Mg(OH).sub.2 described herein uses olivine, a mineral commonly present in ultramafic (e.g., peridotites and dunites) and mafic (e.g., basalt) rocks. FIG. 1 is a reaction diagram of the process. FIG. 2 shows the reaction when magnesium hydroxide is used in carbon sequestration. These reaction processes are used hereafter below.

(11) Olivine is a nesosilicate with most mineral compositions represented in the system MgOFeOSiO.sub.2. With acidification, olivine solubility increases, thereby, increasing Mg release rates and its concentration into solution. In the inventor's investigations, powdered forsteritic ((Mg.sub.0.9Fe.sub.0.1).sub.2SiO.sub.4) or Mg-rich olivine (?100 g with a mean particle size of 28 ?m) was combined with 500 mL of 2 M HCl. This resulted in a solution containing MgCl.sub.2, FeCl.sub.2, and SiO.sub.2. Strong acids such as HCl and/or H.sub.250.sub.4 accelerate hydrolysis.

(12) Mg.sup.2+ concentration in the digested solution was ?24 g L.sup.?1 as determined by complexometric titration. The Mg concentration was found to be, much higher than Mg in seawater. Therefore, the Mg extraction efficiency is improved using an HCl digestion.

(13) Following the initial digestion, the solution was allowed to settle for ?1 hour and then it was decanted to separate Mg, Fe and Si ions from any remaining olivine (FIG. 1). Silica was precipitated using an acid swing process by increasing the pH to >3.5 by adding 1 g of Mg(OH).sub.2.

(14) Silica was produced through hydrolysis, polymerization, and condensation of silicic acid (Si(OH).sub.4). The solution pH was again increased to ?7 using 0.32 g of NaOH to precipitate iron in solution. Silica and iron were separated using a centrifuge in this example. Please note that industrial filtration system or precipitate flotation could be used in lieu of centrifuging. Direct filtration or some suitable form of solid separation is possible but may require vacuum or pressure.

(15) The remaining MgCl.sub.2 in solution underwent electrolysis in an H-cell with a carbon anode and platinum cathode where Cl.sub.2 gas at the anode and H.sub.2 gas at the cathode formed. Mg(OH).sub.2 formed at the cathode and pH of solution became (?9.5). Mg(OH).sub.2 from the cathode was placed in a drying oven (?100? C. for 1 day) and the dried product was assessed using scanning electron microscope (SEM), thermo gravimetric analysis (TGA) and X-ray powder diffraction (XRD). For the commercial production of Mg(OH).sub.2, H.sub.2 and Cl.sub.2 can be combined to produce HCl that can be reused for Mg silicate processing.

(16) Results

(17) From 100 g of olivine, 35 g of Mg(OH).sub.2, 35 g of amorphous silicate and ?5 grams of iron oxide was produced. Only 1 g of Mg(OH).sub.2 was added in the silica precipitation stage. The SEM image and TGA graph provided in FIG. 3 a,b confirm that the material recovered after electrolysis was Mg(OH).sub.2. XRD analyses provided in FIG. 3c. show the recovered silica was amorphous SiO.sub.2 (note: some residual olivine was present) and that the product following electrolysis was Mg(OH).sub.2. Compositions of the olivine sand, recovered Mg(OH).sub.2, and recovered silica from olivine were determined via X-ray fluorescence (XRF) and these results are provided in Table 1.

(18) TABLE-US-00001 TABLE 1 Elemental compositions of raw olivine sand and recovered Mg(OH).sub.2 and silica from olivine, determined by XRF analysis. (Cite the table) SiO.sub.2 Al.sub.2O.sub.3 Fe.sub.2O.sub.3 CaO MgO Na.sub.2O K.sub.2O LOI Total Wt. % Raw 39.6 0.38 10.7 0.73 45.0 0.14 0 3.2 100.0 olivine Mg(OH).sub.2- 0.1 0.11 6.4 0.30 60.4 0.03 <0.01 32.3 99.9 olivine Silica- 63.2 0.23 4.7 0.57 13.9 0.03 0.06 16.4 99.0 olivine

(19) Our example and approach provides a highly efficient and nearly closed system for the production of Mg(OH).sub.2 with the only additions being olivine, HCl, and minor amounts of NaOH. In addition to the recovered Mg(OH).sub.2, secondary materials (e.g., silica and iron hydroxide) provide useful products, such as a partial replacement for Portland cement and a high purity iron ore.

(20) CO.sub.2 and Energy Implications of Mg-Hydroxide Extraction

(21) Transforming olivine into Mg(OH).sub.2 produces no direct CO.sub.2 emissions. Total energy required, including mining and processed, to produce Mg(OH).sub.2 from olivine was determined to be 6.28 GJ tonne.sup.?1. Further work may allow greater use of Mg(OH).sub.2 produced for pH control and further reduce the energy and NaOH currently required. The Mg(OH).sub.2 could be further processed into MgO but for CO.sub.2 sequestration; however, Mg(OH).sub.2 is known to be a faster reactant than MgO. Although, MgO, presumably a slower reactant, does provide potential as a route to carbon sequestration.

(22) For one tonne of CO.sub.2 to be sequestered as a Mg-carbonate, including a variety of carbonate phases including nesquehonite and hydromagnesite, 1.3 tonnes of Mg(OH).sub.2 is needed and requires an energy consumption 8.17 GJ tonne.sup.?1 of CO.sub.2. Carbon sequestration benefits could be improved if low carbon energy sources were used.

(23) Using the recovered Mg(OH).sub.2, a Mg(OH).sub.2 water slurry was pressurized with concentrated CO.sub.2 (4 bar). Over a 48 hour period >50% of the Mg(OH).sub.2 was converted to a hydrated Mg-carbonate, demonstrating the CO.sub.2 conversion into a solid. Reactivity of Mg(OH).sub.2 was rapid.

(24) Additionally, a slurry of magnesium hydroxide and DI water was exposed to atmosphere concentrations of CO.sub.2 at ambient temperatures and pressures and under a humidity of approximately 90%. After a period of one week of exposure a sample was collected and assessed using thermogravimetric analysis (TGA). FIG. 4 shows the differential (DTGA) results for the unreacted material prior to exposure together with a curve for the material after one week exposure. The peak just past 500? C., of the reacted material, indicates the formation of a magnesium carbonate containing compound and verifies the potential suitability of the material for carbon sequestration.

(25) Olivine Resources and Implications

(26) Olivine-rich deposits (FIG. 5) are globally abundant and primarily present within populated areas near convergent margins such as the Circum-Pacific region. There is enough accessible olivine to sequester all anthropogenic CO.sub.2 and further reduce global atmospheric CO.sub.2 levels for thousands of years as shown in FIG. 5 which focuses on only two olivine-rich deposits that occur in Oman and New Zealand. Bulk transport of Mg(OH).sub.2 to be used for onsite point source emissions control is economically feasible and would still result in significant anthropogenic CO.sub.2 reductions.

Example 2

(27) Olivine, a Mg-rich nesosilicate and sourced from Red Hills, New Zealand, was processed/ground to an average particle size of 30 ?m. Processed olivine was combined with 2M HCl in a ratio of 1:10 (% W/V), heated to 60? C. and continuously stirred; allowing for 2 hours of digestion. The mixture rested for one hour and the solution was decanted to remove the remaining olivine. Using ICP-MS, the solution was determined to consist of silica, magnesium, iron and chloride ions. Products were separated by using a pH swing process in several steps. Magnesium hydroxide (0.2% w/v) was added in the solution to increase the pH to >3 to condense and polymerize silicic acid (Si(OH).sub.4), thereby, producing silica gel. Polymerized silica was filtered and rinsed with water to remove excess acids and chlorides. Following this step, 2M NaOH was added to the filtered solution, thereby, raising the pH to 7 in order precipitate and remove (via filtration or centrifuging) iron hydroxide. The remaining solution underwent electrolysis where magnesium hydroxide formed at the cathode. ?35 g of Mg(OH).sub.2 was produced from 100 g of olivine, of which 1 g was added during the silica precipitation stage as discussed in Example 1. Hydrogen gas (cathode) and chlorine gas (anode) was produced and could be recombined to produce HCl for recycling into the initial digestion reactions. Mg(OH).sub.2 obtained was calcined for 1 hour at 500? C. for 1 hour, thereby, producing MgO.

(28) Recovered silica (RS) and recovered MgO (RM) from olivine using the procedure discussed above was assessed as a binder and compared with commercially available MgO and silica fume using several methods including isothermal calorimetry, XRD, FTIR and SEM.

(29) Materials and Experiments

(30) Commercial MgO (CM) and silica fume (SF) from Sibelco Australia and Sika New Zealand, respectively, were assessed and compared to the RS and RM described above. Chemical compositions of recovered and commercial MgO and SiO.sub.2 are provided in Table 2. Loss on ignition (LOI) of recovered silica is high and can be explained by water loss in the polymerized silica. The recovered and commercial MgO both had similar MgO concentrations and include uncalcined brucite. Recovered silica, as shown in FIG. 3, was primarily amorphous silica with some minor amounts of unreactive olivine.

(31) TABLE-US-00002 TABLE 2 Chemical composition of raw materials (%/100 g) (REF) % CM RM SF RS SiO.sub.2 2.11 0.19 94.85 63.18 Al.sub.2O.sub.3 0.15 0.16 0.57 0.23 Fe.sub.2O.sub.3 0.35 11.45 0.33 4.65 CaO 2.8 0.5 0.27 0.57 MgO 84.16 81.66 0.47 13.92 Na.sub.2O <0.01 <0.01 0.33 0.03 K.sub.2O <0.01 <0.01 0.76 0.06 LOI 10.66 5.63 1.94 16.37 SSA (m.sup.2/g) 37 23 17 93

(32) A MgOSiO.sub.2 binder was prepared by combining and mixing MgO and SiO.sub.2 at 1:1 ratio by mass. Mix compositions between recovered and commercial products were investigated and are shown in Table 3. Paste samples with a water to binder (w/b) ratio of one were used; these mixes provided comparable workability and intrinsic hydration behaviours without the use of external agents like superplasticisers.

(33) Paste samples (?20 g) were placed in an isothermal calorimeter at 20? C. (Calmetrix I-Cal Flex) and heat evolved related to hydration for a variety of mixes was measured. Remaining pastes were cured at 20? C. in PVC vials (Diameter 20 mm, Height 100 mm).

(34) TABLE-US-00003 TABLE 3 Mix design proportion [ref] Commercial Silica Recovered Recovered Notation MgO Fume MgO Silica CM-SF 0.50 0.50 RM-SF 0.50 0.50 CM-RS 0.50 0.50 RM-RS 0.50 0.50

(35) Paste samples, dried and ground, after 3, 7 and 28 days were assessed for hydration via XRD (Rigaku SmartLab Diffractometer), FTIR (Bruker Spectrometer Alpha II), scanning electron microscopy (SEM), electron microscopy (JEOL 6400 in secondary electron mode).

(36) Compressive strengths for mortar samples were measured where a binder to sand ratio of 1:3 was used. Water to binder ratios with the addition of a superplasticizer addition was varied. Cube samples (50?50?50 mm) were cast and maintained at 20? C. and 60% relative humidity for 24 hours. Samples were stored at 20? C.

(37) Results & Discussion

(38) Heat Evolution

(39) FIG. 6 shows cumulative heat measurements related to hydration reactions for the CMgO+SF and RMgO+RS over a period of 72 hours using isothermal calorimetry. Recovered MgO (RMgO)+recovered silica (RS) showed a faster and higher heat of hydration compared to Commercial MgO (CMgO)+silica fume (SF) mixes.

(40) Hydration Products

(41) Qualitative XRD Analysis

(42) XRD analyses of the hydrated samples is shown in FIG. 7. Peaks corresponding to brucite, related to MgO hydration, are present in all the mixes. Differences between the intensity of peaks in SF and RS mixes are present. The brucite peaks in RS mixes support a rapid reaction of RS with brucite to form M-S-H. Broad peaks representing amorphous M-S-H (shown as grey bands in FIG. 7) were present in the recovered silica mixes. Higher reactivity of RS is attributed to its high surface area. The CM-SF mix peaks that demonstrate unreacted MgO support an overall lower extent of hydration. For the CM-RS mix, brucite depletion by RS would require increased magnesium ions in pore solutions, which then furthers MgO consumption. This is supported by the absence of unreacted MgO in the CM-RS mix. XRD analyses suggest that the RM-SF mix provides faster dissolution of RM compared to CM. Additionally, the peak intensities of brucite compared to the prominent broad M-S-H peak in the RM-SF mix supports a faster brucite reaction with silica to form M-S-H. The analyses confirms that the materials recovered from olivine could be effectively used to produce MgOSiO.sub.2 binder.

(43) Compressive Strength

(44) Compressive strengths of mortar samples (3, 7 and 28 days) is shown in Table 4 and provides a comparison of the material characteristics. The compressive strength of the CM+RS mix at 3 days was ?60% higher compared to the CM+SF mix, despite a higher w/b ratio, possibly due to the rapid formation of M-S-H. This observation agrees with calorimetry results. Higher compressive strengths were observed after 7 days and the slower reacting SF continued to form M-S-H, thereby increasing its strength, after 28 days. The RM-RS mix had ?20% higher strength compared to CM-SF mix at 3 days despite a 50% higher w/b ratio. Strength development traits displayed by the RM-RS mixes supports that the MgOSiO.sub.2 binder using magnesium silicate minerals, such as olivine, is viable.

(45) TABLE-US-00004 TABLE 4 Summary of compressive strength (MPa) of mixes CM-SF CM-RS RM-RS (W/B: 0.50) (W/B: 0.58) (W/B: 0.75) 3 Days 13.7 21.9 16.5 7 Days 30.1 31.4 21.6 28 Days 49.1 36.0 22.0
Sustainability Assessment

(46) Portland cement production is energy intensive and it requires that calcination of limestone which releases CO.sub.2. The MgO:SiO.sub.2 ratio of 1:1 used in this example, have shown effective binder systems can be produced with MgO:SiO.sub.2 ratios from 0.4 to 0.6. The total energy is may be higher for the MgO-silica binder system sourced from Mg silicates, such as olivine, there is no chemical release of CO.sub.2 from the raw materials during the manufacturing process, therefore, CO.sub.2 generated from this process is dependent on the source of electricity.

(47) Conclusions

(48) This example demonstrates that MgOSiO.sub.2 binders can be produced from Mg silicates, such as olivine, and compares well with binders made from commercially available materials. Recovered MgO and SiO.sub.2 were more reactive compared to commercial materials where the use of recovered silica enhanced the hydration rates. M-S-H formation during hydration was confirmed in binder produced using recovered materials. Compressive strengths for the CM-RS mix and CM-SF mix was comparable, where 28 day mortar strength of >20 MPa was determined for the RM-RS mix.

Example 3

(49) The recovered and dried silica, as outlined in Example 2, was further tested to determine its potential use as partial replacement (SCM) for Portland cement. A binder was created using 30% recovered silica along with 70% Portland cement. The binder was mixed with sand at a ratio of 1:3 and a water/cement ratio of 0.5 to create a mortar cubes. The samples containing the recovered silica were found to have a 28 day compressive strength more than 10% greater than the control samples which were made using a 100% Portland cement binder. The results show that the recovered silica is capable of replacing at least 30% of the Portland cement while at the same time increasing the overall compressive strength.

Example 4

(50) A solution of 1.5M magnesium sulphate was subject to electrolysis, whereby, after 2 hours magnesium hydroxide was recovered at the cathode (pH 10.1) and sulphuric acid was created at the anode (pH 1.56). These results show that the alternative magnesium salt of magnesium sulphate may be processed in a similar manner to other salts. The example also illustrates the way that magnesium hydroxide inherently increases solution pH.

(51) Aspects of the magnesium silicate processing methods and uses of the processed magnesium silicate products have been described by way of example only and it should be appreciated that modifications and additions may be made thereto without departing from the scope of the claims herein.