into a thermodynamically stable solid, MgC03, using Mg-bearing minerals such as ... ucts: high surface area silica, iron oxide, and magnesium carbonate, while ...
Journal of Nanoscience and Nanotechnology
Copyright © 2009 American Scientific Publishers All rights reserved Printed in the United States of America
Vol. 9, 1--6. 2009
Synthesis of Iron-Based Chemical Looping Sorbents Integrated with pH Swing Carbon Mineral Sequestration Hyung Ray Kim 1 , Dong Hyun Lee 2, Liang-Shih Fan 1 , and Ah-Hyung Alissa Park 3•* 1
Chemical and Biomolecular Engineering, The Ohio State University, Ohio, USA 2 Chemical Engineering, Sungkyunkwan University, Suwon, Korea 'Earth and Environmental Engineering & Chemical Engineering, Columbia University, New York, USA
(,
I oo
The previously developed pH swing carbon mineral sequestration immobilizes the gaseous C02 into a thermodynamically stable solid, MgC03 , using Mg-bearing minerals such as serpentine. This mineral carbonation technology is particularly promising since it generates value-added solid products: high surface area silica, iron oxide, and magnesium carbonate, while providing a safe and permanent storage option for C02 • By carefully controlling the pH of the system, these solids products can be produced with high purity. This study focuses on the synthesis of iron oxide particles as a chemical looping sorbent in order to achieve the integration between carbon capture and storage technologies. Since the solubility of Fe in aqueous phase is relatively low at neutral pH, the effect of the weak acid and chelating agents on the extraction of Fe from serpentine was investigated. The synthesized iron-based chemical looping sorbent was found to be as effective as commercially available iron oxide nanoparticles at converting syngas into high purity H2 , while producing a sequestrationwready C02 stream.
Keywords: Carbon Capture and Storage (CCS), Chemical Looping Sorbents, Carbon Mineral Sequestration, pH Swing, Chelating Agents, Hydrogen Production, Iron Oxide Particles.
JJ
m en
~
1. INTRODUCTION Since the industrial revohllion, the amount of C02 in the atmosphere has risen from 280 ppm in 1800 to 370 ppm in 2000, mainly due to the consumption of fossil fuels. l-.l More than half of the energy used in the United States comes from the use of coa1, 4 and it is mostly used to generate electricity. Unfortunately, C02 is one of the greenhouse gases that is considered to be responsible for global warming. Moreover, the increased atmospheric C02 concentration will acidify the ocean and will change the chemistry of the surface ocean, leading to a potentially detrimental impact on the ecosystem. In order to meet the everincreasing global energy demands, while stabilizing the atmospheric C0 2 level, current carbon emissions should be significantly reduccd. 5 There have been significant research and development activities in the area of carbon capture and storage (CCS), including a number of integrated technologies (e.g., chemical looping processes) to combine C0 2 capture with electricity/chemical/fuel production. 5 Chemical looping processes involve a sorbent, typically a metal, or more ·Author to whom correspondence should be addressed.
J. Nanosci. Nanotechnol. 2009, Vol. 9, No. xx
likely a low oxidation state metal oxide that can be oxidized in ait: n. 7 The oxide is reduced by
~ (")
r
m
•
Synthesis of Tron~Based Chemical Looping Sorhcnts Integrated with pH Swing Carbon Mineral Sequestration
silica layer on the serpentine particles. This phenomenon was also observed during the dissolution of .Mg from serpentineL'~. lx and Park and Fan 16 have already developed an in-situ physical activation scheme that can improve the overall dissolution of serpentine by refreshing the surface of serpentine particles during dissolution. However, in this study, the physical activation method was not used in order
to investigate the isolated clfcct of chemical activation via the addition of the chelating agents and weak acid on the dissolution of serpentine in terms of Fe content The dissolution of Fe was superiorly higher ln the presence of citric acid than the other chelating agents and this may be due to the number of chelation sites. Citric acid has four chelation sites while glycine and IDA have three chelation sites. Although acetic acid has only one binding site, it performed as well as glycine and IDA. This may be due to the size of the molecule. Since acetic acid is much smaller than other chelating agents, it can access the surface of the serpentine particles more easily. Since the mineral dissolution is surface reaction, the dissolution rate would be a function of both chclating power and size of the ligand.
3.2. Synthesis of Iron Oxide Particles ''ia pH Swing Process Afler the extraction of Fe into the aqueous phase, the pH was increased up to 8.2 by adding NH4 OH, and Fe( OH)_, was precipitated out. After the precipitation step, filtered Fe(OH)_, was treated with air at 600 "C to produce the highest oxidation state of iron, Fe 2 0~. The XRD analysis of Sample 1 confii·rned the formation of Fc2 0J after the heat treatment of the precipitated solid with air at 600 oc. The dehydrated sample (Sample I) was then sintered at 900 "C and directly reduced with H2 in a TGA at 840 "C. Figure 5 shows the weight loss of Sample 1 during the reduction process. Initially, the sample was heated to 840 oc and stabilized for 30 seconds in N 2 environment. Once the TGA reached 840 "C, the diluted H, was 100y-~~-----------r==========~ -Pure Fe 2 0 3 • • Fe,P3 synthesized from Fe extracted from serpentine
90
..,
Kim et al.
introduced into the TGA for the reduction reaction. In the N 2 environment, the dried sample showed no weight change with a great thermal stability, but as soon as H2 was introduced, Fe 2 0 3 was undergone the reduction by losing 28.1% of its weight. The weight percent of the synthesized Fc 2 0~ stabilized at 69.7%, which indicates about 2.2 wt% of impurities (e.g., serpentine) in Sample/, which is rather insignificant. Overall, the TGA resulls shows that the iron-based chemical looping .sorhent synthesized from serpentine during the pH swing carbon mineral sequestration has similar oxygen carrying capacity as commercially available Fe1 0J nanoparticles. The reduced particles were labeled as Sample 3. Next, the morphological structures or Samples I and 2 synthesized from 1'vfethods I and 2, respectively, were examined via the SEM analysis and the results arc shown in Figure 6. Tn case of Fe2 0J produced without the presence of supports, individual grains were in relatively spherical shape and the size of single grain was generally less than I 0 ,um with a wide size distribution. As shown in Figure 6(a), small amount of pores were observed throughout the sorbent particle. On the other hand, Sample 2, which was prepared by precipitating Fc2 0J directly onto the supports, shows uniformly dispersed iron oxide with no porosity. The size of Fe2 0 3 grains in Figure 6(h) seems to be smaller and more irregular than the grains in Figure 6(a) and this confirms that iron oxide was precipitated within the pores of the support particles and maximized the surbent loading. The EDX analysis of Sample 2 confirmed a great loading efficiency of Method 2.
3.3. Fixed Bed Redox Reactions of ll'on Oxide Particles for H 2 Production Finally, the fixed bed experiments were conducted to further validate the concept of the chemical looping process integrated with the pH swing carbon mineral sequestration. The chemical looping process illustrated in Figure 1 was carried out using syngas as the carboneous fuel. The involved reactions are: Fuel reactor: 3CO + Fe2 0,--> 3C02 + 2Fe 3H2 +Fe2 0 3 --> 3H 2 0+2Fe
.E 80
.Q>
~ 70
60-!-------~·
0
2
3
Time (min) Fig. 5.
Reduction of pure Fe 2 0~ .
4Fep4 +0 2 --> 6Fe2 0 1
.c 0.6
B
Figure 7 shows the normalized compositions of the exiting gas during the reduction of Fe20, in Sample 4 by syngas. Initially, both CO and H 2 were completely oxidized by Fe2 0 3 and produced sequestration-ready pure C02 • The maximum C02 concentration appeared as 99.8% at 7 minutes. As Fe 2 0 3 phased out via syngas reduction, limited oxidation of syngas was observed before the final breakthrough point at I 00 minutes. This step change in exit gas compositions of the fixed bed reactor is due to the multiple oxidation states of iron (i.e., Fe5 0 4 , FcO, and Fe). Next, the reduced Fc2 0 3 particles were regenerated via the oxidation with steam. Figure 8 shows the conversion of steam to H 2 based on the microGC measurements of the exit gas stream. Before reaching breakthrough, the conversion of steam to H2 was found to be rv80%. A small amount of CO was detected during oxidation; however, the amount was negligible (0. I ~0.2 mol%) compared to the H 2 concentrations. CO in the exit steam is originated from carbon deposited during the previous syngas reduction step. Fluctuatlons seen in the H2 concentration were due to the unsteady vaporization of the water injected into the reactor. It is interesting that the breakthrough points for thy oxidation and reduction steps were quite similar, which imply that a similar reactor size can be used for hath reducer and oxidizer. In other words, two reactors can be used in parallel and chemical looping can be achieved by simply switching inlet gases. After the oxidation with steam, the chemical looping sorbents in the fixed bed were further oxidized with air. It was found that the iron-based
• f." ....
~"'.
0
">c 0
"E "85
0.3
\..
0.0 0
20
40
60
80
100
120
140
Time (min) Fig. 8. H1 production during the oxidmion of the i'>ynthesized chemical looping .sorbcnts with steam.
sorbents were fully re-oxidizcd to its highest oxidation state, Fc2 0].
4. CONCLUSIONS The ironMbased chemical looping sorbcnt \Vas successfully synthesized from Fe recovered from the pH swing carbon mineral sequestration process. This was achieved by chemically activating the dissolution of serpentine. In order to maximize the extraction of Fe content in serpentine, a number of chelating agents were tested and citric acid was selected as the suitable iron extraction enhancer for scrpenM tine. \Vith the presence of supports during the precipitation step, iron oxide sorbent was produced with higher metal oxide loading compared to the sorbent produced via dry mixing. The synthesized iron-based chemical looping sorbent was found to be very effective at producing H 2 , while generating scqucstrationMrcady C0 2 stream.
Acknowledgments: This work was partially supported
Sequestration-ready C0 2
. ..
c
"§
by ORICA Ltd .
•,
m
en
~
:IJ (')
:I: )>
~ (')
r
'
•••
m
References and Notes
-
I
Time (min) Fig. 7, Generation of sequestration-ready C02 during the reduction of Fe 2 0 3 with syngas.
J. Nanosci. Nanotechnol. 9, 1-6, 2009
:IJ
1. IPCC (Intergovernmental Panel on Climate Change). Carbon Dioxide C11pture and Storage, Cambridge University Press, Cambridge (2005). 2. CCSP (U.S. Climate Change Science Program and the Subcommittee on Global Change Research), Abrupt Climate Change: Final Report, Synthesis and Assessment Product 3.4, U.S. Geological Survey, Reston, VA (2008). 3, IPCC (Intergovernmental Panel on Climate Change), Climate Change 2007: Synthesis Report, Cambridge University Press, Cambridge (2008). 4. ETA (Energy Information Administration), Annual Energy Review 2007, Office of Energy ivlarkets and End Use, U.S. Department of Energy: Washington DC, Report# DOE/EIA-0384 {2008). 5. A.-H. A. Park, K. S. Lackner, and L.-S. Fan, Hydrogen Fuel: Production, Transport, and Storage. edited hy R. H. Gupta, CRC Press, Boca H.aton (2008), p. 569. 6. R. Leithner, Int. 1. Ene1;~y Teclmnl. Policy 5, 340 (2007). 7, M. Ishida, L. Takeshita, K. Suzuki, and T. Ohba, Eni!IXY ami Fuels 19, 25t4 (2005).
5
Synthesis of Iron~ Based Chemical Looping Sorbcnts Integrated with pH Swing Carbon Mineral Sequestration 8. T. ~·fattisson, ivf. Johansson, and A. Lyngfeit, Eneq~y and Fuels 18, 628 (2004). 9. L.~S. Fan and P. Gupta, U.S.A. PCT Patent Application 2007 W.O. 2007082089 (2007). 10. K. S. Lackner, Sdence 300, 1677 (2003). 11. F. Goff and K. S. Lackner, Envimn. Geosci. 5, R9 (1998). 12. K. S. Lackner, C. H. Wendt, D. P. Bull, E. L. Joyce, and D. H. Sharp, E11e1:~y 20, 1153 (1995). 13. L. Penner, W. K. O'Connor, S. Gerdcmann, and D. C. Dahlin, Proceeding of 20th l'ittshurKh Coal Cm!ference, Pittsburgh, PA (2003).
Kim eta!.
14. M. Hanchen, V. Prigiobbe, G. Storti, T. M. Seward. and 1\1. ivfazzotti. Geochim. Cosmm·him. Ac. 70, 4403 (2006). 15. A.~H. A. Park, R. Jadhav, and L.~S. Fan, Cmwdian 1. 4 Chem. Eng. 81, 885 (2003). 16. A.~H. A. Park and L~S. Fan. Clwn Eng. Sci. 59. 5241 (2004). 17. T. Furia, CRC Handbook ufFoml Additives, Chemical Rubber Co., Cleveland (1968). 18. H. Barat, M. J. McKelvy, A. V. G. Chizmeshya, D. Gormley, R. Nunez, R. W. Carpenter, K. Squires, and G. H. Wolf. F.urimn. Sci. Tedmol. 40, 4802 (2006).
Received: 8 October 2008. Accepted: 10 February 2009.
w u
.J
lr
