Subscriber access provided by UNIVERSITY OF THE SUNSHINE COAST
Article
Wollastonite Carbonation in Water-bearing Supercritical CO2: Effects of Particle Size Yujia Min, Qingyun Li, Marco Voltolini, Timothy J Kneafsey, and Young-Shin Jun Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.7b04475 • Publication Date (Web): 02 Oct 2017 Downloaded from http://pubs.acs.org on October 7, 2017
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Environmental Science & Technology is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 33
Environmental Science & Technology
Wollastonite Carbonation in Water-bearing Supercritical CO2: Effects of Particle Size
Yujia Min,† Qingyun Li,† Marco Voltolini,‡ Timothy Kneafsey,‡ Young-Shin Jun*† †
Department of Energy, Environmental and Chemical Engineering, Washington University in St. Louis, MO 63130
‡
Earth and Environmental Science, Lawrence Berkeley National Laboratory, CA 94720
E-mail:
[email protected] Phone: (314) 935-4539 Fax: (314) 935-7211 http://encl.engineering.wustl.edu/ Submitted: August 2017
Environmental Science & Technology
*To Whom Correspondence Should be Addressed
ACS Paragon Plus Environment
Environmental Science & Technology
1
ABSTRACT
2
The performance of geologic CO2 sequestration (GCS) can be affected by CO2 mineralization and
3
changes in the permeability of geologic formations resulting from interactions between water-
4
bearing supercritical CO2 (scCO2) and silicates in reservoir rocks. However, without understanding
5
size effects, the findings in previous studies using nanometer or micrometer size particles cannot
6
be applied to the bulk rock in field sites. In this study, we report the effects of particle sizes on the
7
carbonation of wollastonite (CaSiO3) at 60 oC and 100 bar in water-bearing scCO2. After
8
normalization by the surface area, the thickness of the reacted wollastonite layer on the surfaces
9
was independent of particle sizes. After 20 hours, the reaction was not controlled by the kinetics
10
of surface reactions, but by the diffusion of water-bearing scCO2 across the product layer on
11
wollastonite surfaces. Among the products of reaction, amorphous silica, rather than calcite,
12
covered the wollastonite surface and acted as a diffusion barrier to water-bearing scCO2. The
13
product layer was not highly porous, with 10 times smaller specific surface area than the altered
14
amorphous silica formed at the wollastonite surface in aqueous solution. These findings can help
15
us evaluate the impacts of mineral carbonation in water-bearing scCO2.
1 ACS Paragon Plus Environment
Page 2 of 33
Page 3 of 33
16
Environmental Science & Technology
TOC Art
17
2 ACS Paragon Plus Environment
Environmental Science & Technology
18
INTRODUCTION
19
To mitigate the adverse impacts of global climate change, anthropogenic CO2 can be
20
captured from point sources and sequestered in subsurface environments, such as deep saline
21
aquifers or depleted oil and gas reservoirs.1 The safety and efficiency of CO2 trapping in these
22
storage sites are related to interactions among supercritical CO2 (scCO2), water, and rock.2 For
23
example, the carbonation of silicates can trap CO2 in mineral phases and cause changes in the
24
wettability and porosity of reservoir rocks, which can in turn alter their permeability. Furthermore,
25
a change of permeability in geologic formations can influence the transport of CO2 and potentially
26
affect CO2 leakage. For similar reasons, scCO2–water–rock interactions can be important in other
27
subsurface CO2 injection processes, including scCO2-enhanced oil recovery and scCO2-energized
28
hydraulic fracturing.3-7
29
While most previous studies on scCO2–water–rock chemical interactions focused on
30
mineral dissolution and precipitation in the presence of bulk aqueous phase,2 recent studies have
31
revealed the following importance of interactions between scCO2 and minerals in the absence of
32
aqueous solutions.8-10 First, because injected scCO2 will displace brine, most of the contact areas
33
between scCO2 and caprocks and a certain portion of the contact areas between scCO2 and
34
formation rocks, are in the absence of bulk aqueous solution.11 Second, scCO2 has higher
35
diffusivity, lower viscosity, and thus lower capillary entry pressure than aqueous phase fluid (i.e.,
36
CO2-dissolved brine).12 Therefore, scCO2 can enter smaller pore throats which brine cannot enter.
37
Third, recent studies reported that water dissolved in scCO2 has higher reactivity on mineral
38
surfaces than that of water in the aqueous phase.8, 12, 13 Based on these considerations, the reactions
39
of minerals with water-bearing scCO2 are at least equally, if not more, important than reactions of
40
mineral with aqueous solutions.12, 14
3 ACS Paragon Plus Environment
Page 4 of 33
Page 5 of 33
Environmental Science & Technology
41
Our current understanding of water-bearing scCO2–mineral interactions focuses on the
42
carbonation of various Ca-, Mg-, or Fe-bearing minerals, including wollastonite,15-18 forsterite,11,
43
14, 19-22
44
water saturation percentage of CO2,14, 18, 19, 21, 25 temperature,17, 18, 20, 23 and organic ligands,22 on
45
carbonation reactions, we need a better understanding of the dependency of water-bearing scCO2–
46
silicates reactions on particle sizes. There are several knowledge gaps: First, previous studies
47
mostly used micrometer or even nanometer size particles,14-23 a far cry from the bulk rock in field
48
sites. Without understanding the effects of particle sizes, we cannot know whether the outcomes
49
obtained with small particles are applicable to bulk rock in field sites. Second, findings from
50
different studies are not directly comparable due to the different particle sizes used. Thus, results
51
from previous studies on water-bearing scCO2–silicate reactions cannot be combined. Third,
52
investigating the effects of particle sizes can reveal the mechanisms of mineral carbonation
53
reactions. For example, a recent study of metal oxide carbonation (i.e., CaO) in dry CO2 found that
54
the reacted fraction is constant for different particle sizes and concluded that the reaction extent of
55
CaO carbonation is determined by the porosity of CaO particles.26 Similarly, investigating the
56
effects of particle sizes on silicate carbonation in water-bearing scCO2 can inform us about whether
57
the reaction is controlled by the porosity of silicates or other limiting factors, such as the kinetics
58
of silicate dissolution in water films. However, to the best of the authors’ knowledge, there have
59
been no studies on the effects of particle sizes on water-bearing scCO2–silicate interactions.
fayalite,23 dolomite,24 and phlogopite.8 While these studies investigated the effects of the
60
The goal of this study is, therefore, to understand the effects of particle size on the
61
carbonation of wollastonite in water-bearing scCO2. We chose wollastonite because it has been
62
frequently used as a representative of silicates in previous studies on silicate reaction in scCO2 and
63
in aqueous solutions.17, 18, 27, 28 We hypothesize that the reaction is controlled by the kinetics of
4 ACS Paragon Plus Environment
Environmental Science & Technology
64
wollastonite hydrolysis. If anisotropic reactivity of different crystal faces is not significant, the
65
reaction extent of different size particles should be similar after normalization to surface area. To
66
test this working hypothesis, particles with five different size ranges were reacted at 60 oC and 100
67
bar, conditions closely relevant to GCS. The information provided can contribute to better
68
understanding the water-bearing scCO2–silicate reactions and scaling up findings in laboratory
69
studies using small size mineral particles to the larger scale bulk rock formations.
70
EXPERIMENTAL METHODS
71
Minerals
72
Natural wollastonite particles with five different size ranges were purchased from NYCO
73
Company (Willsboro, NY). By using X-ray powder diffraction (XRPD, Bruker D8), its structure
74
was identified to be wollastonite-1A, the most common polymorph in the natural environment.29
75
The size distribution and shape for each size range were determined by laser diffraction and image
76
analysis (Microtrac Inc.) performed by NYCO Company. Most particles were in cylindrical shape
77
(Figure S1). The spherical equivalent volumetric mean diameters were 3.8, 5.2, 11.1, 17.8, and
78
82.0 µm, and the respective typical aspect ratios were 3:1, 3:1, 3:1, 4:1, and 15:1. The detailed size
79
distribution is available in the Supporting Information (Figure S1). These size ranges were chosen
80
because they can provide detectable and significantly different reaction extents. Using N2 as
81
adsorbate and 11 points on the isotherm in the BET method (AX1C-MP-LP, Quantachrome
82
Instruments), the specific surface areas were determined to be 4.46±0.01, 3.61±0.01, 1.95±0.01,
83
1.45±0.01, and 0.54±0.01 m2/g, respectively, for 3.8, 5.2, 11.1, 17.8, and 82.0 µm particles. The
84
error ranges are based on duplicate measurements. X-ray fluorescence (Table S1) showed that the
85
Ca/Si ratio was 0.959. Thermogravimetric analysis (TGA, Q5000IR, TA Instruments) showed
5 ACS Paragon Plus Environment
Page 6 of 33
Page 7 of 33
Environmental Science & Technology
86
0.45 % mass loss between 150–780 oC. As shown by the XRD patterns in Figure S5, the unreacted
87
sample contains aragonite, which is estimated to be 1 wt%, based on 0.45% mass loss in TGA. No
88
further treatment was performed after the wollastonite were obtained from the vendor.
89
Carbonation in water-bearing scCO2 at simulated GCS conditions
90
Carbonation experiments were conducted in a 300 mL high pressure and high temperature
91
reactor (Parr Instruments, Moline, IL) modified from the reactor used in our previous studies.30-34
92
A schematic diagram of the reaction system setup is available in the Supporting Information
93
(Figure S2). The conditions (60 oC and 100 bar CO2) are similar to typical conditions at GCS sites:
94
65 oC and 150 bar in the Frio formation,35 37 oC and 100 bar at the Sleipner site,36 and 63 oC and
95
140 bar at the Weyburn field site.37 To investigate carbonation of wollastonite in the absence of
96
aqueous phase, wollastonite particles were placed in 10 ml Teflon tubes, while ultra-purified
97
deionized water (Barnstead, > 18.2 MΩ·cm) was added outside the tubes. After scCO2 was injected
98
into the vessel, water dissolved in the scCO2 and generated water-bearing scCO2. Five PTFE tubes
99
were placed in the reactor during each test. Considering the volume taken up by tubes and liners
100
in the reactor, the volume of CO2 during the experiment was 201 mL. The tubes were capped, but
101
have small holes allowing contact between the wollastonite and water-bearing scCO2. The
102
solubility of water in CO2 was predicted using Spycher’s model.38 At 35 and 60 oC, the mole
103
fraction of 100% saturated water in 100 bar CO2 is 0.41 and 0.49%, which indicates 234 and 117
104
µL water in the 201 mL scCO2. The solubility of water in scCO2 is affected by the salinity of water.
105
As a starting point, ultra-purified water was used for simplicity. This result can serve as an
106
important underpinning for understanding the effects of salinity on silicate carbonation in the
107
future. Each sample contained 0.3 g particles. To check the possible influence of particle stacking
108
inside the tubes on the reaction, results were compared with samples using 0.05 g particles.
6 ACS Paragon Plus Environment
Environmental Science & Technology
Page 8 of 33
109
Selected samples were analyzed using XRPD. In this study, no attempt of XRPD quantitative
110
analysis was made. Instead, the reaction extent was determined by TGA analysis. A previous study
111
has shown that the results from XRPD quantitative analysis are comparable with the results from
112
TGA analysis.18
113
Determination of the reaction extent
114
The reacted fraction of wollastonite was determined based on TGA results. After the
115
sample was recovered from the reactor, the samples were stored at atmospheric pressure and
116
analyzed within 24 hours. The powder sample was briefly mixed using a plastic spatula, and
117
approximately 10 mg was used for each TGA analysis. At least duplicate samples were analyzed.
118
Samples were heated to 900 oC with a ramp of 20 oC/min under N2 flow (25 ml/min). Wollastonite
119
and amorphous silica were stable below 900 oC, while calcium carbonate completely decomposed
120
into CaO and CO2.18 The amount of CO2 was measured by the mass loss between 150-780 oC
121
during TGA (Figure 1B). Given the amount of CO2, the reacted fraction of the original wollastonite
122
was derived using this equation:
123 124
(1) (2)
125
where mL is the mass loss during TGA, which is the mass of CO2 from CaCO3 decomposition.
126
According to equation (2), the mass loss is equal to the mass of CO2 consumed during reaction
127
with water-bearing scCO2, so the mass of wollastonite consumed as reactant can be calculated
128
using mL and the molecular weight of CO2 and wollastonite; mo is the mass of the sample reacted
129
with water-bearing scCO2, which includes the mass of the original wollastonite (wollastonite
130
before reaction with CO2) and the mass of sorbed CO2 on wollastonite (equal to mL); and M is the
7 ACS Paragon Plus Environment
Page 9 of 33
Environmental Science & Technology
131
molecular weight of different species. Given the 0.45% mass loss in unreacted wollastonite, the
132
detection limit of this method is estimated to be 1%. The reacted fraction of the largest particle
133
size used (82 µm) is 7.8%, so the uncertainty of the reaction extent determined using this method
134
is no more than 15%.
135
Calculation of the reacted thicknesses of wollastonite samples
136
Based on the reacted fraction, we calculated the thicknesses of the reacted layer by using a
137
shrinking core model. The schematic diagram in Figure 2A shows the geometry. Before the
138
reaction, each particle was assumed to be cylindrical in shape, with aspect ratio R and diameter D,
139
calculated using equation (3) based on the spherical equivalent diameter (Deq) measured by laser
140
diffraction by the NYCO Company. During the reaction, a shell with thickness L was consumed
141
as the reactant, and according to the model, the wollastonite particle shrank to a cylindrical shape
142
core with diameter D-2L. The volume fraction of the shell is the reacted fraction of the particle.
143
By integrating all the size intervals shown in the size distribution (Figure S1), the reacted fraction
144
of each size range was calculated. The reacted thickness L was obtained by solving equation (4),
145
which equated the calculated reacted fraction and the reacted fraction determined by TGA.
146
147
(3) ∑
(4)
∑
148
In Equations 3 and 4, L is the thickness of the reacted layer, F is the reacted fraction, Deq is the
149
spherical equivalent diameter of each size range from laser diffraction results provided by NYCO
150
Company, D is the cylindrical equivalent diameter of each size range, and R is the aspect ratio. L
151
was obtained by solving this equation. For small particle size ranges, in which (R × D – 2 × L)