A continuous process for manufacture of magnesite and silica from olivine, CO2 and H2O

A continuous process for manufacture of magnesite and silica from olivine, CO2 and H2O

Available online at www.sciencedirect.com Energy Procedia (2009) 4891–4898 Energy Procedia Procedia100 (2008) 000–000 www.elsevier.com/locate/proced...

1MB Sizes 2 Downloads 30 Views

Available online at www.sciencedirect.com

Energy Procedia

(2009) 4891–4898 Energy Procedia Procedia100 (2008) 000–000 www.elsevier.com/locate/procedia www.elsevier.com/locate/XXX

GHGT-9

A continuous process for manufacture of magnesite and silica from olivine, CO2 and H2O Ingrid Anne Munza*, Jan Kihlea, Øyvind Brandvolla, Ingo Machenbacha, James William Careyb, Tove Anette Haugc, Harald Johansena, Nils Eldrupd a Institute for Energy Technology, P.O.Box 40, NO-2027 Kjeller, Norway Earth and Environmental Sciences, Los Alamos National Laboratory, Los Alamos, NM, USA c Department of Geology and Mineral Resources Engineering, Norwegian University of Science and Technology, Sem Saelandsvei 1, NO-7491 Trondheim, Norway d Tel-Tek, Kjølnes Ring 30, NO-3918 Porsgrunn, Norway b

Elsevier use only: Received date here; revised date here; accepted date here

Abstract Mineral carbonation is based on the reaction of CO2 with metal oxide bearing materials to form solid carbonates. Further technology development and cost reduction are however needed for an industrial realisation of mineral carbonation. Added value products are clearly one factor, which may change the cost estimates. Separation of reaction products and sufficient product quality must be demonstrated. A concept of using CO2 and water for reaction with olivine in a continuous process with separation of reaction products has been investigated. The reaction products, magnesite and silica, are of potential commercial interest. The process consists of three steps: 1) dissolution of olivine; 2) precipitation of magnesite and 3) precipitation of silica. Separation and precipitation of the reaction products do not require chemical additives, such as acids or bases, and there will thus be few requirements of chemical reclamation. A semi-continuous set of laboratory-scale experiments including process steps 1 and 2 have been carried out. Experimental conditions were in the range 100-150 bar and 130-250 ºC. Process step 3 has been tested separately, using process water from step 2. The results show a congruent dissolution of olivine with reaction rates comparable to known kinetic models in the lower end of the temperature range. Precipitation of magnesite and silica has different dependence on pH and temperature, and detailed reaction mechanisms are addressed through the experiments. Magnesite precipitation takes place at high temperature (180-250 ºC). A magnesite with very low iron-content can be precipitated as the only product in the second reaction step.

c 2009

Ltd. AllAll rights reserved. © 2008Elsevier Elsevier Ltd. rights reserved

Keywords: Mineral carbonation, olivine, added value, CO2

* Corresponding author. Tel.: +47 63 80 60 00; fax: +47 63 81 55 53 E-mail address: [email protected] doi:10.1016/j.egypro.2009.02.319

2

4892

I.A. Munz et /al. / Energy Procedia (2009)000–000 4891–4898 Author name Energy Procedia 00 1(2008)

1. Introduction Mineral carbonation is based on the reaction of CO2 with metal oxide bearing materials to form solid carbonates, with calcium and magnesium being the most attractive metals. The reaction products are stable minerals, occurring in nature. The carbon storage capacity on a global scale and the characteristic storage time for mineral carbonation are far higher than for other storage methods (Lackner, 2003). Since the CO2 is stored as stable minerals, the concerns around safety and the requirements for monitoring are few. At present, this technology is still viewed as being in the research phase (IPCC, 2005). A recent IEA review (IEA-GHG, 2005) concludes that further technology development and cost reduction are needed for mineral CO2 sequestration to become part of a broad portfolio of employable CO2 sequestration technologies. The main advantage for the mineral carbonation is the few concerns around long-term safety of the repository. Energy assessment of a one-step process shows a net sink for CO2, and on average 76 % of the CO2 can be removed, when olivine is used as the mineral resource (O’Connor et al., 2005). At present, mineral carbonation is viewed as a feasible niche option for CO2 storage, where a CO2 point source coincides with sources of mineral reactants (O’Connor et al., 2004). The possibility of added value products from mineral carbonation has been discussed (e.g. Goff and Lackner, 1998), but process design, energy and economics has not yet been evaluated. Added value products are clearly a factor, which may change the cost estimates for mineral carbonation. Another important factor is new, less energy intensive process design. In this paper, we present a preliminary evaluation of a process based on olivine, with magnesite and silica as possible added value products. The principles of separation are discussed, and a techno-economical study of a possible industrial concept has been carried out. 2. Carbonation of mafic minerals International research on mineral carbonation has been directed towards rocks of high Mg or Ca content. In particular, experiments have been concentrated on serpentine and olivine-bearing rocks. The overall reaction with olivine can be formulated as follows: Mg2SiO4 (s) + 2 CO2 ļ 2 MgCO3 (s) + SiO2 (s) Reaction may occur with CO2 gas phase (dry carbonation) or with CO2 dissolved in water (wet carbonation). Two different approaches of wet carbonation have been used: x Direct aqueous carbonation, where CO2 ± water is used in direct reaction with solids (e.g. Fauth et al., 2000; O’Connor et al. 2002, 2004, 2005). x Indirect carbonation, where the rock/minerals are dissolved in acids or bases and CO2 is used in reaction with a solution in a second step (e.g. Druckenmiller and Maroto-Valer, 2005; Maroto-Valer et al., 2005a,b; Park and Fan, 2004). 2.1. Direct aqueous carbonation The approach to direct carbonation has been by the use of batch reactors (Fauth et al., 2000; O’Connor et al. 2002, 2004, 2005). In some of these experiments CO2 has been added during the run, in order to maintain a constant CO2 pressure when carbonates are precipitating. Chemicals, such as NaHCO3, have also been added in the experiments. Flow out of the system has however not been applied. The sequence of reactions taking place within a single reactor is dissolution of CO2 in water to form carbonic acid. The silicates dissolve in reaction with the carbonic acid, and magnesium carbonate precipitate from solution. O’Connor et al. (2002, 2005) found the optimal reaction conditions for carbonation of olivine to be 185 ºC and 150 bar in a aqueous solution with NaCl and NaHCO3.

I.A. Author Munz etname al. / Energy Procedia 1 (2009) 4891–4898 / Energy Procedia 00 (2008) 000–000

4893

3

2.2. Indirect carbonation The reaction rate of olivine dissolution is pH dependent (e.g. Hänchen et al., 2006). Indirect carbonation, applying a strong acid in the first process step, will therefore optimize the reaction rate of dissolution. The first process step does not need to be pressurized. However, precipitation of carbonates requires higher pH (Park and Fan, 2004; Chen et al., 2006). A pH increase from the dissolution to the carbonation step is therefore necessary. 3. Materials and methods 3.1. Materials The olivine samples originate from a dunite deposit in western Norway, and have been supplied by North Cape Minerals. In most of the experiments a commercial dried sand grade quality, AFS 50, has been used. Quantitative XRD analysis shows an olivine content of more than 93 wt %. Orthopyroxene, chlorite, amphibole and chromite are the main constituents of the remaining 7 %. The olivine has a forsterite content of 93 %. Orthopyroxene, chlorite and amphibole show a range in XMg=Mg/(Mg+Fe) of 0.93, 0.94-0.97 and 0.94-0.96 respectively. The samples were slightly crushed and sieved, and two fractions; < 75 μm and 75 -150 μm were used for the experiments. 3.2. Experiments Flow-through column and continuously stirred reactor experiments have been carried out. The experimental setup is shown in Figure 1. A system of two or three interconnected reactors has been used. The feeding system of the reactors consisted of an HPLC-pump for H2O, and an ISCO syringe pump for CO2. Water and CO2 was continuously fed with a constant flow rate into the first reactor for dissolution and mixing of CO2 in water at the temperature and pressure conditions used for dissolution of minerals. The H2O-CO2 solution from reactor 1 was then fed into a second reactor containing the olivine sample. The second reactor was either a column or a continuously stirred reactor. The outlet solution from reactor 2 could be fed into reactor 3 for precipitation. The temperature of each reactor as well as the connecting flow-lines can be operated individually. Typical experimental conditions would be a constant temperature for the lines and reactors until reactor 3. Reactor 3 and the remaining lines would then be set at a different temperature. A number of by-pass lines are mounted to ensure flexibility of the system (Figure 1).The design allows to run the system for dissolution only or for combined dissolution-precipitation. After finishing the experiments, the columns were vacuum impregnated with epoxy in order to conserve the in situ textures, cut and polished for thin sections. Specific surface area of the starting materials was measured using the BET method with N2 absorption. The particle size analysis was performed on a Coulter LS 230, using laser diffraction. The mineral chemistry was analysed on a Cameca SX100 electron microprobe. Quantitative XRay Diffraction (XRD) was used for analyses of mineral contents. ICP has been used for cation analyses of the water. 3.3. Geochemical modelling Geochemical modelling was carried out using PHREEQC (Parkhurst and Appelo, 1999). The solubility of CO2 was calculated using a Henry’s Law method based on the Lichtner et al. (2003) modified approach of Crovetto (1991) with the addition of a Poynting correction for pressure.

4

4894

I.A. Munz et /al. / Energy Procedia (2009)000–000 4891–4898 Author name Energy Procedia 00 1(2008) Cond

FRACTION SAMPLER

Gas supply

BPR

Autoclave II

Autoclave I

V-2

Crystallizator Liquid supply SOLID CHARGE

Figure 1: Experimental set-up. A column has been used as the second reactor in some experiments.

4. Direct carbonation – possibility for product separation In a direct aqueous carbonation approach, it is possible to keep process chemicals to a minimum. Large mass flows of acid or other chemical waste may be avoided. If, however, added value products shall be possible to make, separation of reaction products is necessary. Geochemical modelling with PHREEQC has been used to analyse the possibility of product separation.

0.02

B J H

0.018

1 3

Si or Mg/2 (mol/kg)

0.016

Silica - 50 bar CO2

J

Silica - 100 bar CO2

H

Silica - 150 bar CO2

F

Magnesite - 50 bar CO2

3

Magnesite - 100 bar CO2

1

Magnesite - 150 bar CO2

B J H

F

0.014

B

0.012

B J H

0.01

B J H 0.008

B J H 1 3 F

0.006 0.004

B J H

0.002 0 0

50

1 3 F 1 3 F

1 3 F

100 150 200 Temperature (°C)

1 3 F 250

Figure 2: Solubility of amorphous silica and magnesite, estimated with PHREEQC (The magnesite solubility is scaled, i.e.Mg/2, in order to reflect forsterite stoichiometry).

I.A. Author Munz etname al. / Energy Procedia 1 (2009) 4891–4898 / Energy Procedia 00 (2008) 000–000

Saturation index

0

J J J J

J

-0.5

J B B B

-1

B

-1.5

J B B

SI magnesite

J

SI silica

B

-3

T=50 °C

B

-3.5

Saturation index

5

J B

-2 -2.5

0

4895

0

0.002

0.004

0.006

0.008

B B B B J J J J

B

-0.2 -0.4

0.01

0.012

0.014

0.016

0.018

B J

0.02

B J

-0.6 -0.8

B

-1

J

J

T=150 °C

-1.2 0

0.002

0.004

0.006

0.008 0.01 0.012 0.014 Forsterite consumed (mol/kg)

0.016

0.018

0.02

Figure 3: Reaction progress calculations at 150 bar CO2

The solubility estimations of magnesite and amorphous silica is shown in Figure 2. Whereas silica has a prograde solubility independent of CO2 pressure, the magnesite solubility is retrograde and pH dependent. If forsterite dissolves at low temperature, i.e. below the intersection between the two solubility curves shown in Figure 2 at approximately 100 °C, precipitation of silica will occur before magnesite. In fact, at low temperature precipitation of hydrated magnesium carbonates will occur (Hänchen et al., 2008), but these are for simplicity not included in the modelling. At high temperature, magnesite will reach saturation before silica during forsterite dissolution. This principle is shown by forsterite reaction progress calculations at 150 bar CO2 and temperatures of 50 and 150 °C (Figure 3). 5. Experimental results 5.1. Column experiments Flow-through column experiments have been used for evaluation of reaction mechanisms from textural and mineral chemistry data. The experiments only utilized a two reactor system; a first reactor for mixing of CO2, and H2O and the second reactor was a packed column with olivine. The temperature range of the experiments was 75250 °C. Pressure was mostly kept at 150 bar. The results show extensive dissolution in top of the column (Figure 4). The olivine is completely dissolved in the top as shown in Figure 4. The relic olivine has a homogeneous composition. Reaction zones at olivine rims with different Mg:Si ratios were not found. Iron hydroxides, however, have been precipitated on grain boundaries. Magnesite was only found in experiments run at high temperature (>180 °C) in the lower part of the columns. Silica precipitates were not observed.

6

4896

I.A. Munz et /al. / Energy Procedia (2009)000–000 4891–4898 Author name Energy Procedia 00 1(2008)

Figure 4: Overview of a column experiment (left). To the right: Microscope photo (upper left) and back scatter images of top part of a column experiment. Microphotograph, crossed nicols (upper left) – the brown lines represent grain boundaries consisting of iron hydroxides. The olivine is completely dissolved, but a few grains of other silicates (greyish white) remain. Back-scatter images – Iron hydroxides are seen as bright lines. The grains (more dark grey) are orthopyroxenes and amphiboles. The olivine is completely dissolved and only outlines of previous grain boundaries remain.

5.2. Dissolution and precipitation in separate reactors Flow-through stirred reactor experiments were carried out with two or three reactors (Figure 1). The entire process line was kept at constant pressure (150 bar). The dissolution of olivine in reactor 2 was kept at a lower temperature (130 °C) than the precipitation (250 °C) in reactor 3. The outlet fluid from reactor 2 had a Mg:Si composition close to the olivine stoichiometry. At the applied conditions, no precipitation of magnesium carbonates takes place in reactor 2. Geochemical calculations indicate a magnesite saturation index of approximately 0 in reactor 2. If the temperature is increased to 250 °C, the estimated saturation index of magnesite is above 2, indicating a supersaturated fluid in reactor 3. The experimental results also show that magnesite is precipitated in reactor 3 (Figure 5). The mineral chemical analyses show a pure magnesite end member, with FeO contents <0.05 wt %. Dissolved silica occurs in the solution from reactor 3. Precipitations or leached layers of silica were not observed in any of the reactors.

I.A. Author Munz etname al. / Energy Procedia 1 (2009) 4891–4898 / Energy Procedia 00 (2008) 000–000

4897

7

Figure 5: Magnesite crystals from reactor 3. Microphotograph to the left and back scatter image to the right.

6. Discussion A congruent dissolution of olivine is shown from the fluid composition. This is also confirmed by the complete dissolution of olivine in the column tops and the lack of extensive silica enriched rims of olivine relics. Both congruent and incongruent dissolution behaviour of olivine as well as other silicates has previously been described. Congruent dissolution behaviour of olivine was shown in flow-through experiments (Oelkers, 2001; Hänchen et al., 2006) and in batch experiments (Giammar et al., 2005). These dissolution experiments were kept at conditions far from equilibrium. Giammar et al. (2005) show that the aqueous silica concentration is ultimately controlled by the solubility of amorphous silica. In other batch experiments a silica enriched layer forms on the surface of olivine or other silicates, and the formation of this layer is considered to be a rate limiting step (e.g. Bearat et el. 2006; Schulze et al., 2004). The results of this study demonstrate the possibility of separating olivine dissolution and precipitation of reaction products in several different process steps. There is a good agreement between the experimental results and the predicted behaviour. The dissolution of olivine takes place at conditions far from equilibrium with a forsterite saturation index of the order of –6 to –7. The solution never reaches saturation of amorphous silica, which may explain the congruent dissolution behaviour. Formation of a passivating silica layer will thus not be a rate limiting factor. The solubility of amorphous silica and magnesite does however impose important constraints on the mass balances. The dissolution requires a minimum ratio of water to dissolved olivine determined by the least soluble precipitate – in our experiments using temperatures above 100 °C – this precipitate has been magnesite. The amounts of water also represent constraints on possibilities and effectiveness of precipitation in subsequent process steps. The increase of temperature from 130 in reactor 2 to 250 °C in reactor 3 leads to a change in saturation index from approximately 0 to above 2, and enables precipitation of magnesite. The need of supersaturation for magnesite nucleation agrees well with the conclusions of Giammar et al. (2005). The outlet solution needed evaporation for silica precipitation. Breaking down the mineral carbonation reactions into several continuous process steps, as we do in this study, an improved understanding of complex processes taking place in batch reactor experiments can be achieved. Furthermore, the results have useful implications for interpretation of natural processes, such as observations of carbonation/silification of oceanic crust or other basic or ultrabasic rock types. The laboratory-scale experiments have also been used for a preliminary techno-economical evaluation of possible up-scaling to an industrial process

8

4898

I.A. Munz et /al. / Energy Procedia (2009)000–000 4891–4898 Author name Energy Procedia 00 1(2008)

based on olivine. The high investment costs related to the need of pressurized reactors was the most important factor in the economical sensitivity analysis. The dependence of reaction rates on pressure is thus an important issue. 7. Conclusions This paper has shown some principles of separating the dissolution of olivine in carbonated water and the precipitation magnesite and silica into different process steps. The dissolution of olivine takes place at conditions far from equilibrium. The reaction is congruent. Magnesite is shown to precipitate in a second reaction step. The solubility of amorphous silica and magnesite impose important constraints on the mass balances in both dissolution and precipitation steps. Acknowledgements We greatly appreciate the funding of two strategic institute programs from the Research Council of Norway (No: 158916/i30 and 181710/i30). Are Korneliussen at the Geological Survey of Norway has been a great inspiration and source of information on Norwegian mineral resources. We also thank North Cape Minerals for supplying samples. References Bearat H, McKelvy MJ, Chizmeshya AVG, Gormley D, Nunez R, Carpenter RW, Squires K and Wolf GH (2006) Carbon sequestration via aqueous olivine mineral carbonation: Role of passivating layer formation. Environmental Science & Technology 40 (15): 4802-4808 Chen Z-Y, O’Connor WK and Gerdemann SJ (2006) Chemistry of aqueous mineral carbonation for carbon sequestration and explanation of experimental results. Environmental Progress 25: 161-166 Crovetto R (1991) Evaluation of solubility data of the system CO2-H2O from 273 K to the critical point of water. Journal of Physical and Chemical Reference Data 20: 575-589 Druckenmiller ML and Maroto-Valer MM (2005) Carbon sequestration using brine of adjusted pH to form mineral carbonates. Fuel Processing Technology 86: 1599-1614. Giammar DE, Bruant Jr RG and Peters CA (2005) Forsterite dissolution and magnesite precipitation at conditions relevant for deep saline aquifer storage and sequestration of carbon dioxide. Chem. Geol. 217: 257-276 Hänchen M, Prigiobbe V, Baciocchi R and Mazzotti M (2008) Precipitation in the Mg-carbonate system – effects of temperature and CO2 pressure. Chem. Eng. Sci. 63: 1012-1028 Hänchen M, Prigiobbe V, Storti G, Seward TM and Mazzotti M (2006) Dissolution kinetics of forsteritic olivine at 90-150 °C including effects of the presence of CO2. Geochim. Cosmochim. Acta 70: 4403-4416 IEA Greenhouse R&D programme (2005) Carbon dioxide storage by mineral carbonation. Report no. 2005/11 Lichtner PC, Carey JW, O’Connor WK, Dahlin DC and Guthrie GD Jr (2003) Geochemical mechanisms and models of olivine carbonation. In: Proceedings of the 28th International Technical Conference on Coal Utilization & Fuel systems, March 9-13, 2003. Maroto-Valer MM, Fauth DJ, Kuchta ME, Zhang Y, Andrésen JM (2005a) Activation of magnesium rich minerals as carbonation feedstock materials for CO2 sequestration. Fuel Processing Technology 86: 1627-1645 Maroto-Valer MM, Zhang Y, Kutcha ME, Andresen JM and Fauth DJ (2005b) Process for sequestering carbon dioxide and sulphur dioxide. US Patent Application 20050002847 O’Connor WK, Dahlin DC, Rush GE, Dahlin CL and Collins WK (2002) Carbon dioxide sequestration by direct mineral carbonation: process mineralogy of feed and products. Minerals & Metallurgical Processing 19: 95-101 O’Connor WK, Dahlin DC, Rush GE, Gerdemann SJ and Penner LR (2004) Energy and economic evaluation of ex situ aqueous mineral carbonation. IEA-GHGT-7, Vancouver, Canada, 5-9 September 2004. O’Connor WK, Dahlin DC, Rush GE, Gerdemann SJ, Penner LR and Nilsen DN (2005) Final report – Aqueous mineral carbonation. Mineral availability, pre-treatment, reaction parametrics and process studies. DOE/ARC-TR-04-002 Oelkers EH (2001) An experimental study of forsterite dissolution rates as a function of temperature and aqueous Mg and Si concentrations. Chem. Geol. 175: 485-494 Park A-HA and Fan L-S (2004) CO2 mineral sequestration: physically activated dissolution of serpentine and pH swing process. Chem. Eng. Sci. 59: 5241-5247. Parkhurst DL and Appelo CAJ (1999) User’s guide to PHREEQC (Version 2) – A computer program for speciation, batch-reaction, ondimentional transport and inverse geochemical calculations. USGS, water-resources investigations report 99-4259 Schulze RK, Hill MA, Field RD, Papin PA, Hanrahan RJ and Byler DD (2004): Characterization of carbonated serpentine using XPS and TEM. Energy Conversion and Management 45 (20): 3169-3179 Summers CA, Dahlin DC, Rush GE, O’Connor and Gerdemann SJ (2005) Grinding methods to enhance the reactivity of olivine. Minerals & metallurgical processing 22: 140-144