Direct C[O.sub.2] sequestration onto alkaline modified oil shale fly ash.
1. IntroductionThe study is part of a continuous investigation to find new applications for oil shale ash, which until now has been disposed of as an industrial waste. The current work focuses on the potential of alkaline modified OSFA for C[O.sub.2] capture and storage (CCS). As a pre-treatment of OSFA, hydrothermal activation was applied [1]. In the treatment process, the silica in the original OSFA is converted mainly into calcium-silica-aluminum hydrates (75 wt% Al-substituted crystalline 1.1 nm tobermorite - [Ca.sub.5][Si.sub.5]Al(OH) [O.sub.17] x 5 [H.sub.2]O and 9 wt% katoite - [Ca.sub.3][Al.sub.2][(Si[O.sub.4].sub.kx][(OH).sub.4x], where 3 [less than or equal to] x [less than or equal to] 2.67) [2, 3].
Tobermorites are reactive with C[O.sub.2] gas. Siauciunas et al. observed a complete destruction of tobermorite, but noted that the carbonization reaction requires the presence of water for tobermorite to react with H2C[O.sub.3] forming CaC[O.sub.3] and Si[O.sub.2] gel [4]. In the initial stage of indirect tobermorite carbonization, CaC[O.sub.3] modification, vaterite and sometimes, particularly at higher temperatures, metastable aragonite are formed. Over the extended carbonization process, intermediates rearrange into calcite. Siauciaunas et al. concluded in their study that in the absence of water vapor, C[O.sub.2] chemisorption was not observed in the temperature range of 25-75[degrees]C.
Touze et al. conducted a study on the direct carbonization of tobermorite [5]. Plausible reaction stoichiometry for 1.1 nm tobermorite and C[O.sub.2] in dry conditions at normal temperature and pressure was as follows:
[Ca.sub.5][Si.sub.6][H.sub.11][O.sub.22.5] [right arrow] 5C[O.sub.2] - 5CaC[O.sub.3] + 6Si[O.sub.2] + 5.5H2O (1)
Further, the reactivity of 1.1 nm tobermorite was also calculated at 150[degrees]C with pressures 1 bar and 100 bar. Touze et al. pointed out that the pressure is more important in terms of reactivity than temperature. For example, at 150[degrees]C and 100 bar, the reactivity of tobermorites is high, unlike at 150[degrees]C and 1 bar, at which no reactivity was observed. The researchers concluded that the direct carbonation process of tobermorites is possible, but it is still unclear how or under which conditions it would be feasible. In the current study we investigated the direct C[O.sub.2] chemisorption reaction with activated OSFA samples from electrostatic precipitators of the oil shale fueled power plant, containing ca 75 wt% of crystalline tobermorite [2].
2. Materials and methods
2.1. Activated oil shale ash
Activation of oil shale fly ash was conducted as described in the authors' previous study [6], except the reaction temperature was 145[degrees]C and 5 M NaOH solution was used. Ash samples were collected from the 1st and 4th units of electrostatic precipitators (ESPs) of the Estonian Power Plant's boiler (Narva Power Plants Ltd.) operated as a circulating fluidized bed.
2.2. Analysis
The specific surface area of samples (nitrogen physisorption) was measured with a Micrometrics TriStar 3000 sorptometer (Micrometrics Instrument Corp., USA). The samples (ca 0.3 g) were degassed at a temperature of 120[degrees]C for 5 hours and before placing in the sorptometer. Isotherm data, BET surface area, t-plot and pore volumes were processed with TriStar 3000 v. 6.07 software.
Average elemental composition for original, activated and chemisorbed ash samples was determined by an energy-dispersive X-ray spectroscope (EDS, Jeol JSM-6400) mounted on a field emission scanning electron microscope (FESEM, Zeiss FE-SEM Ultra Plus).
In chemisorption experiments the samples were weighted using a Mettler AE 163 electronic balance (precision 0.0001 g).
XRD analyses were performed in a Bruker d8 Advance instrument in 6-6 mode, with an optical configuration consisting of a primary Gobel mirror and a Vantec-1 detector. Continuous scans were applied on the sample. By adding repeated scans, the total data collection lasted for 6 hours. The PDF2 databank (ICDD, Newtown Square, PA, 2004) together with Bruker software was used to analyze the diffraction patterns.
[sup.29]Si MAS-NMR spectra of the original and activated samples were recorded on a Bruker AMX-360 spectrometer at 8.5 T external magnetic field, using a bespoke MAS probe and 10 mm od zirconia rotors (rotation speed 5 kHz, simple 90-degree pulse excitation). About 400 accumulations with the recycle time of 200 seconds were used to get a reasonable signal to noise ratio.
2.3. C[O.sub.2] chemisorption
Chemisorption of C[O.sub.2] onto activated OSFA samples was conducted in a high-pressure bench top reactor (200 mL) provided with a Parr 4848 temperature controller. The reactor was connected to a C[O.sub.2] (AGA, analytical grade) gas cylinder.
The samples were dried and degassed in a Carbolite tube furnace under nitrogen atmosphere at 180[degrees]C for 4 hours to remove crystallized and physisorbed water molecules. The sample was then cooled down to room temperature in a calcium chloride dessicator and weighted with a Mettler 168 analytical balance and placed in the reactor bomb in a glass vessel.
The reactor was sealed securely and flushed with a C[O.sub.2] gas stream to remove air and to provide pure C[O.sub.2] atmosphere in the reactor. Then the reactor was pressurised up to 56 bar and checked for possible leaks, before heating. Upon heating to 150[degrees]C, the reactor pressure reached 100 bar.
After 24 hours of reaction, heating was turned off, the pressure was lowered and the reaction vessel was removed. Before weighting, the sample was dried and degassed in a Carbolite tube furnace under a nitrogen atmosphere at 180[degrees]C for 4 hours, in order to remove the water product of reaction. The weight change was calculated and the solid reaction product characterized using XRD, NMR and imaging.
3. Results and discussion
3.1. Specific surface area
For characterizing original and activated OSFA as well as the material subjected to reaction with C[O.sub.2], the specific surface area was calculated from the N2 absorption-desorption isotherm (P/P0 = 0.025-0.999) using the B.E.T. equation [7] (Table 1). Nitrogen was chosen as the adsorbate because it afforded the opportunity to compare specific surface area measurements with those from previous studies [2]. In addition, t-plots, BJH cumulative pore surface areas, cumulative pore volumes, and average pore diameters were all calculated from the isotherm data.
The [BET.sub.N2] specific surface area of activated OFSA samples did not change significantly, decreasing only slightly, during the chemisorption reaction, though they were an order of magnitude higher than the values for the original OFSA. Consistent with this, the t-plot micropore volumes did not significantly change during reaction either.
3.2. C[O.sub.2] chemisorption analysis by weighting
When the activated OSFA reacts with carbon dioxide, the weight of the sample is expected to increase (Equation (1)). C[O.sub.2] chemisorption onto the activated oil shale ash was characterized by weighting the dry samples before and after reaction. The uptake of C[O.sub.2] may be calculated from this mass change.
The reaction with C[O.sub.2] is a surface reaction taking place at particular active sites, and due to the low specific surface area of activated OSFA samples (65-68 [m.sup.2]/g, Table 1), most of the minerals contained in the samples are inaccessible for reaction with C[O.sub.2]. The complete stoichiometric reaction of 1.1 nm tobermorite with C[O.sub.2] would lead to a 16% mass uptake according to Equation (1). On the other hand, the activated OSFA contains other minerals which may have also reacted with C[O.sub.2], but this aspect was not examined in the present study.
The activated oil shale ash from the 1st electrostatic precipitator showed good C[O.sub.2] chemisorption capacity. The results of experiments are presented in Table 2.
3.3. X-Ray diffraction (XRD) analysis
The X-Ray diffraction patterns of original, activated and chemisorbed ash samples from the 1st and 4th ESP are presented in Figures 1 and 2, respectively. The PDF2 databank (ICDD, Newtown Square, PA, 2004) together with Bruker software identified the peaks of lime (L), quartz (Q), anhydrite (A) and calcite (C) in the original ash sample. In the activated oil shale ash sample, the quartz peak has disappeared and those of tobermorite (T), katoite (K) and gibbsite (G) have appeared, with some amount of calcite remaining in the material.
In chemisorbed ash samples tobermorite appears to be converted to calcite, whilst katoite and gibbsite have remained almost intact during chemisorption according to XRD diffractions (Figs. 1 and 2).
3.4. Magic Angle Spinning Nuclear Magnetic Resonance (MASNMR) analysis
[sup.29]Si MAS-NMR spectra of original, activated and chemisorbed ash samples from the 1st and 4th ESP as well as the deconvolutions of these spectra are presented in Figures 3 and 4, respectively. The spectrums of activated oil shale ash samples present five resonance lines at -80, -82.5, -85.9 and -92.3 that can be assigned to the silicon sites [Q.sup.1], [Q.sup.2](1Al), [Q.sup.2](0Al) and [Q.sup.3](1Al) of 1.1 nm tobermorite, respectively [8]. The absence of peak [O.sup.3] refers to that the activation at given reaction parameters is not complete and silicate chains are not wholly bound together [6].
Some quantitative characteristics of samples are given in Table 3, in which the mean length n of Si[O.sub.4]/Al[O.sub.4] chains and the ratio of Al/Si were evaluated using the formulas presented in [9].
The data reveal that the mean lengths n of Si[O.sub.4]/Al[O.sub.4] chains in the activated oil shale ash from the 1st and 4th ESP are similar and increase with chemisorption. Also the ratio of Al/Si in the tobermorite structure apparently increases with chemisorption, being 0.5 and 0.6 in the activated 1st and 4th ESP ash samples, becoming 1.0 and 2.5 for chemisorbed samples, respectively (Table 3). The latter is mainly caused by an increase of the intensity of the [Q.sup.2](1Al) site and decrease of the intensity of the [Q.sup.1] site. In chemisorbed ash samples, the silicon sites [Q.sup.3](1Al) have been reacted away, due to the partial depolymerization of the tobermorite framework (Table 3).
3.5. Field-emission scanning electron microscope analysis coupled with energy-dispersive X-ray spectroscopy (FESEM/EDS) analysis
Imaging results of field-emission scanning electron microscope (Zeiss FESEM Ultra Plus) for original, activated and chemisorbed ash samples from the 1st ESP are presented in Figure 5.
Energy-dispersive X-ray spectroscopy (Jeol JSM-6400) coupled with FESEM (Zeiss FE-SEM Ultra Plus) allows determining elemental compositions of the scanned samples. The samples were analyzed at many different spots and the average elemental composition was calculated. The data are presented in Table 4. The content of residual Na from the activation process remained below 2 wt% in the products. The content of sulphur, potassium and cloride was not detected in hydrothermally treated ash samples either. These elements are apparently flushed away during washing of activated OSFA after the treatment process.
4. Conclusions
The chemisorptive reaction of C[O.sub.2] under dry conditions, at a temperature of 150[degrees]C and C[O.sub.2] partial pressure of 100 bar resulted in a weight increase of 5.7 [+ or -] 0.4% and 2.2 [+ or -] 0.4% in the activated OSFA from the 1st and 4th ESP, respectively. In the XRD patterns the characteristic tobermorite peaks are seen to have largely disappeared, and katoite and calcite peaks increased after the C[O.sub.2] chemisorption reaction.
The Al/Si ratio increases with C[O.sub.2] chemisorption, from 0.5 and 0.6 in the activated ash samples of the 1st and 4th ESP to 1.0 and 2.5 in the chemisorbed samples, respectively. In the chemisorbed ash samples, the silicon site [Q.sup.3](1Al) is almost gone, which indicates the depolymerization of the tobermorite framework and also explains the apparent change in the elemental ratio.
Encouraging results were obtained and a further study addressing the reaction mechanism of tobermotite carbonization will be implemented.
doi: 10.3176/oil.2014.1.08
Acknowledgments
This work was supported by the Estonian Ministry of Education and Research (grants SF 0690001s09 and SF 0690034s09AP). The Bio4Energy program, the Knut and Alice Wallenberg Foundation, as well as the Kempe Foundations, all Sweden, are acknowledged. Also, the Micre (Micro Energy to Rural Enterprise) project under the auspices of the Northern Periphery Program by the EU is acknowledged.
REFERENCES
[1.] Holler, H., Wirsching, U. Zeolites formation from fly ash. Fortschr. Mineral., 1985, 63, 21-43.
[2.] Reinik, J., Heinmaa, I., Kirso, U., Kallaste, T., Ritamaki, J., Bostrom, D., Pongracz, E., Huuhtanen, M., Larsson, W., Keiski, R., Kordas, K., Mikkola, J. P. Alkaline modified oil shale fly ash: Optimal synthesis conditions and preliminary tests on C[O.sub.2] adsorption. J. Hazard. Mater., 2011, 196, 180-186.
[3.] Rivas Mercury, J. M., Pena, P., De Aza, A. H., Turrillas, X., Sobrados, I., Sanz, J. Solid-state 27Al and 29Si NMR investigations on Si-substituted hydrogarnets. Acta Mater., 2007, 55(4), 1183-1191.
[4.] Siauciunas, R., Rupsyte, E., Kitrys, S., Galeckas, V. Influence of tobermorite texture and specific surface area on C[O.sub.2] chemisorption. Colloid. Surface. A, 2004, 244(1-3), 197-204.
[5.] Touze, S., Bourgeois, F., Baranger, P., Durst, P. Analyse bilantielle de procedes ex situ de sequestration du C[O.sub.2]. Rapport Final BRGM/RP-53290-FR, 2004 (in French).
[6.] Reinik, J., Heinmaa, I., Mikkola, J.-P., Kirso, U. Hydrothermal alkaline treatment of oil shale ash for synthesis of tobermorites. Fuel, 2007, 86(5-6), 669676.
[7.] Brunauer, S., Emmett, P. H., Teller, E. Adsorption of gases in multimolecular layers. J. Am. Chem. Soc., 1938, 60(2), 309-319.
[8.] Wieker, W., Grimmer, A.-R., Winkler, A., Magi, M., Tarmak, M., Lippmaa, E. Solid-state high-resolution 29Si NMR spectroscopy of synthetic 14A, 11A and 9A tobermorites. Cement Concrete Res., 1982, 12(3), 333-339.
[9.] Andersen, M. D., Jakobsen, H. J., Skibsted, J. Characterization of white Portland cement hydration and the C-S-H structure in the presence of sodium aluminate by 27Al and 29Si MAS NMR spectroscopy. Cement Concrete Res., 2004, 34(5), 857-868.
Presented by E. Suuberg
Received April 17, 2013
JANEK REINIK (a)*, IVO HEINMAA (a), JOHANNES RITAMAKI (b,c), DAN BOSTROM (d), EVA PONGRACZ (b,e), MIKA HUUHTANEN (b), WILLIAM LARSSON (c), RIITTA KEISKI (b), KRISZTIAN KORDAS (c,f), JYRI-PEKKA MIKKOLA (c,g)
(a) National Institute of Chemical Physics and Biophysics, Akadeemia tee 23, 12618 Tallinn, Estonia
(b) Mass and Heat Transfer Process Laboratory, P.O.Box 4300, FI-90014 University of Oulu, Finland
(c) Technical Chemistry, Department of Chemistry, Chemical-Biological Center, Umea University, SE-90187, Umea, Sweden
(d) Energy Technology and Thermal Process Chemistry ETPC, Department of Chemistry and the Department of Applied Physics and Electronics, ChemicalBiological Center, Umea University, SE-90187, Umea, Sweden
(e) Thule Institute, NorTech Oulu, University of Oulu, P.O.Box 8000, FI-90014
University of Oulu, Finland
(f) Microelectronics and Materials Physics Laboratories, EMPART Research Group of Infotech Oulu, University of Oulu, P.O.Box 4500, FI-90014 Oulu, Finland
(g) Industrial Chemistry and Reaction Engineering, Process Chemistry Center, Abo Akademi University, FI-20500, Abo-Turku, Finland
* Corresponding author: e-mail [email protected]
Table 1. [BET.sub.N2] specific surface area and physisorption
characteristics of original, activated and chemisorbed oil
shale ash samples
Sample BET-surface t-Plot t-Plot
area, micropore external
[m.sup.2]/g area, area,
[m.sup.2]/g [m.sup.2]/g
OSFA 1st ESP 7.1 0.6 6.5
Activated
1st ESP
Chemsiorbed 67.9 13.1 54.8
activated
1st ESP 65.7 10.2 55.5
OSFA 4th ESP 6.3 0.8 5.6
Activated
4th ESP 64.6 12.2 52.3
Chemsiorbed
activated
4th ESP 62.6 12.1 50.5
Sample BJH BJH
surface average
area, pore
[m.sup.2]/g diameter,
nm
OSFA 1st ESP 8.4 18.1
Activated
1st ESP
Chemsiorbed 63.5 16.1
activated
1st ESP 63.1 16.7
OSFA 4th ESP 7.3 20.0
Activated
4th ESP 58.6 15.6
Chemsiorbed
activated
4th ESP 57.0 18.5
Table 2. Weighting results of chemisorption experiments
Set Before After
chemisorption, chemisorption, Difference,
g g %
Activated OSFA
1st ESP 0.898 0.949 5.7 [+ or -] 0.4
Activated OSFA
4th ESP 1.039 1.062 2.2 [+ or -] 0.4
Table 3. Relative intensities of lines in the [sup.29]Si MAS NMR
spectra of activated and chemisorbed ash samples from the
1st and 4th electrostatic precipitators at PF and CFB oil
shale boilers
Sample Data from Figures 3 and 4
[I.sub.2]
[I.sub.1] (1Al) [I.sub.2]
Activated
ash 1st ESP 22.3 54.0 54.8
Chemisorbed
activated
ash 1st ESP 9.6 90.0 82.6
Activated
ash 4th ESP 11.7 42.5 64.4
Chemisorbed
activated
ash 4th ESP 1.2 124.0 48.0
Sample Data from Evaluated from
Figures 3 and 4 Equation
presented in [9]
[I.sub.3]
(1Al) [I.sub.x] n Al/Si
Activated
ash 1st ESP 11.1 283 15.7 0.6
Chemisorbed
activated
ash 1st ESP 0.3 240 47.4 1.0
Activated
ash 4th ESP 13.0 366 27.2 0.5
Chemisorbed
activated
ash 4th ESP 0.0 370.0 399 2.5
[I.sub.1] /[I.sub.3] are the intensities of lines assigned to silicon
sites [Q.sup.1] /[Q.sup.3], respectively. [I.sub.x] is the intensity
of the broad background line arising from amorphous silicon sites.
Table 4. FESEM/EDS average elemental composition (wt%)
of original, activated and chemisorbed ash samples
Element OSFA 1st ESP
Original Activated Chemisorbed
O 44.0 46.6 44.4
Ca 29.2 24.3 23.7
Si 11.0 19.3 17.8
Al 4.3 4.2 4.2
K 3.4 n.d. n.d.
S 2.5 n.d. n.d.
Fe 2.7 1.5 2.1
Mg 1.9 2.0 2.7
Cl 0.6 n.d. n.d.
Na n.d. 1.9 1.8
C n.d. n.d. 2.9
Total 99.6 99.8 99.6
Element OSFA 4th ESP
Original Activated Chemisorbed
O 42.9 45.6 45.5
Ca 25.6 23.6 23.9
Si 12.7 14.4 14.7
Al 4.7 4.5 4.6
K 2.8 n.d. n.d.
S 3.3 n.d. n.d.
Fe 3.5 2.8 3.1
Mg 3.5 3.0 3.8
Cl 1.1 n.d. n.d.
Na n.d 1.1 0.9
C n.d. 4.5 3.3
Total 100.1 99.5 99.8
n.d.--not determined
 Printer friendly
 Cite/link
 Email
 Feedback
| |
| Author: | Reinik, Janek; Heinmaa, Ivo; Ritamaki, Johannes; Bostrom, Dan; Pongracz, Eva; Huuhtanen, Mika; Larss |
|---|---|
| Publication: | Oil Shale |
| Article Type: | Report |
| Geographic Code: | 4EUFI |
| Date: | Mar 1, 2014 |
| Words: | 3184 |
| Previous Article: | Influence of retorting conditions on the pyrolysis of Yaojie oil shale. |
| Next Article: | Oil shale ash based stone formation hydration, hardening dynamics and phase transformations. |
| Topics: | |