Journal of Environmental Engineering

A novel liquid-impregnated clay sorbent [R. V. Siriwardane, U.S. Patent No. 6,908,497 B1 (2003)] was developed for carbon dioxide $\left(\mathrm{C}{\mathrm{O}}_2\right)$ removal in the temperature range of ambient to $60°\mathrm{C}$ for both fixed-bed and fluidized-bed reactor applications. The sorbent is regenerable at $80–100°\mathrm{C}$ . A $20\text{-}\text{cycle}$ test conducted in an atmospheric reactor with simulated flue gas with moisture demonstrated that the sorbent retains its $\mathrm{C}{\mathrm{O}}_2$ sorption capacity with $\mathrm{C}{\mathrm{O}}_2$ removal efficiency of about 99% during the cyclic tests. The sorbents suitable for fluidized-bed reactor operations showed required delta $\mathrm{C}{\mathrm{O}}_2$ capacity requirements for sorption of $\mathrm{C}{\mathrm{O}}_2$ at $40°\mathrm{C}$ and regeneration at $100°\mathrm{C}$ . The parameters such as rate of sorption, heat of sorption, minimum fluidization velocities, and attrition resistance data that are necessary for the design of a reactor suitable for capture and regeneration were also determined for the sorbent. A $20\text{-}\text{cycle}$ test conducted in the presence of flue-gas pollutant sulfur dioxide— $\mathrm{S}{\mathrm{O}}_2$ (20 parts per million)—indicated that the sorbent performance was not affected by the presence of $\mathrm{S}{\mathrm{O}}_2$ .

This study examines the use of ion-exchange fibers (IX fibers) to permanently sequester carbon dioxide present in flue gas into an aqueous phase as calcium or magnesium alkalinity while concurrently softening hard water. The only process inputs besides carbon dioxide (or flue gas) are snowmelt (or rainwater); no other chemicals are required for the regeneration of the IX fibers. Importantly, the process is not energy intensive and carbon dioxide does not need to be compressed to excessive pressures $(>150\mathrm{psi})$ for efficient use. Sources of carbon dioxide do not require concentration and, therefore, the use of raw flue gas ( $\sim 17\%$ $\mathrm{C}{\mathrm{O}}_2$ ) is feasible with the rate of sequestration governed only by the partial pressure of carbon dioxide. While valid for flue gas obtained from any combustion process (e.g., coal, oil, natural gas, etc.), emissions from oil or gas combustion may be more appropriate for use in the described process due to the absence of mercury and particulates. It should also be noted that the presence of sulfur dioxide in flue gas would not adversely affect the process and may even enhance regeneration efficiency. The only product of the proposed process is an environmentally benign regenerant stream containing calcium and/or magnesium alkalinity. The unique property of IX fibers that makes the proposed process both environmentally sustainable and economically feasible is amenability to efficient regeneration with carbon dioxide and harvested snowmelt. Low intraparticle diffusional resistance is the underlying reason why IX fibers are amenable to efficient regeneration using snowmelt sparged with carbon dioxide; 95% calcium recovery was attained at a $\mathrm{C}{\mathrm{O}}_2$ partial pressure of $6.8\mathrm{atm}$ . On the contrary, commercial weak-acid ion-exchange resins in spherical bead forms are ineffective for regeneration with carbon-dioxide-sparged snowmelt due to extremely slow ion-exchange kinetics involving counter-transport of ${\mathrm{Ca}}^{2+}$ (or ${\mathrm{Mg}}^{2+}$ ) and ${\mathrm{H}}^{+}$ . The energy balance for a typical electric utility shows that up to 1% of carbon dioxide emitted during combustion would be sequestered in the softening process.

The effects of steam addition on the stability and $\mathrm{C}{\mathrm{O}}_2$ capture and regeneration properties of ${\mathrm{Li}}_2\mathrm{Zr}{\mathrm{O}}_3$ , K-doped ${\mathrm{Li}}_2\mathrm{Zr}{\mathrm{O}}_3$ , ${\mathrm{Na}}_2\mathrm{Zr}{\mathrm{O}}_3$ , and ${\mathrm{Li}}_4\mathrm{Si}{\mathrm{O}}_4$ at relevant sorption enhanced steam methane reforming conditions has been investigated. It has been observed that the presence of water in the form of steam enhances the capture and regeneration rates. However, a large decay in the capacity is observed when compared to the performance of the acceptors in dry conditions. The reasons for the improved kinetics are not clear, but a higher mobility of the alkaline ions is suggested. The losses in capacity could be due to sintering under high pressure of steam, vaporization of alkali metals, and/or phase segregation.

Sintering during calcination/carbonation may introduce substantial economic penalties for a $\mathrm{C}{\mathrm{O}}_2$ looping cycle using limestone/dolomite-derived sorbents. Here, cyclic carbonation and calcination reactions were investigated for $\mathrm{C}{\mathrm{O}}_2$ capture under fluidized bed combustion (FBC) conditions. The cyclic carbonation characteristics of $\mathrm{Ca}\mathrm{C}{\mathrm{O}}_3$ -derived sorbents were compared at various calcination temperatures $(700–925°\mathrm{C})$ and different gas stream compositions: pure ${\mathrm{N}}_2$ and a realistic calciner environment where high concentrations of $\mathrm{C}{\mathrm{O}}_2>80–90\%$ (and the presence of $\mathrm{S}{\mathrm{O}}_2$ ) are expected. The conditions during carbonation employed here were $700°\mathrm{C}$ and 15% $\mathrm{C}{\mathrm{O}}_2$ in ${\mathrm{N}}_2$ and 0.18% or 0.50% $\mathrm{S}{\mathrm{O}}_2$ in selected tests, i.e., typically expected for a carbonator. Up to 20 calcination/carbonation cycles were conducted using a thermogravimetric analyzer (TGA) apparatus. Three Canadian limestones were tested: Kelly Rock, Havelock, and Cadomin, using a prescreened particle size range of $400–650\mu \mathrm{m}$ . In addition, calcined Kelly Rock and Cadomin samples were hydrated by steam and examined. Sorbent reactivity was reduced whenever $\mathrm{S}{\mathrm{O}}_2$ was introduced to either the calcining or carbonation streams. The multicyclic capture capacity of CaO for $\mathrm{C}{\mathrm{O}}_2$ was substantially reduced at high concentrations of $\mathrm{C}{\mathrm{O}}_2$ during the sorbent regeneration process and carbonation conversion of the Kelly Rock sample obtained after $20\text{cycles}$ was only 10.5%. Hydrated sorbents performed better for $\mathrm{C}{\mathrm{O}}_2$ capture, but also showed significant deterioration following calcination in high $\mathrm{C}{\mathrm{O}}_2$ gas streams. This indicates that high $\mathrm{C}{\mathrm{O}}_2$ and $\mathrm{S}{\mathrm{O}}_2$ levels in the gas stream lead to lower CaO conversion because of enhanced sintering and irreversible formation of $\mathrm{Ca}\mathrm{S}{\mathrm{O}}_4$ . Such effects can be reduced by separating sulfation and carbonation and by introducing steam to avoid extremely high $\mathrm{C}{\mathrm{O}}_2$ atmospheres, albeit at a higher cost and/or increased engineering complexity.

A hybrid process for the capture of $\mathrm{C}{\mathrm{O}}_2$ and ${\mathrm{H}}_2$ from a treated fuel gas mixture is presented. It consists of two hydrate crystallization stages operating at $273.7\mathrm{K}$ and 3.8 and $3.5\mathrm{MPa}$ , respectively. The $\mathrm{C}{\mathrm{O}}_2$ -lean stream from the first stage is directed to a membrane separation unit whereas the $\mathrm{C}{\mathrm{O}}_2$ -rich one is directed to the second hydrate stage. These operating pressures at the crystallization stages are possible by adding 2.5% by mole propane. Propane enables the reduction in the hydrate formation pressure and thus reduces the cost associated with the compression of the fuel gas. The two hydrate stages would operate at 7.5 and $3.5\mathrm{MPa}$ without adding propane. This work provides the relevant kinetic data, as well as the separation efficiency and recoveries achieved.

Experiments of $\mathrm{C}{\mathrm{O}}_2$ capture and sorbent regeneration characteristics of limestone, dolomite, and $\mathrm{Ca}\mathrm{O}∕{\mathrm{Ca}}_{12}{\mathrm{Al}}_{14}{\mathrm{O}}_{33}$ at high temperature were investigated in a thermogravimetric analyzer (TGA) and a fluidized bed reactor. The effect of reactivity decay of limestone, dolomite, and $\mathrm{Ca}\mathrm{O}∕{\mathrm{Ca}}_{12}{\mathrm{Al}}_{14}{\mathrm{O}}_{33}$ sorbents on $\mathrm{C}{\mathrm{O}}_2$ capture and sorbent regeneration processes was studied. The experimental results indicated that the operation time of high efficient $\mathrm{C}{\mathrm{O}}_2$ capture stage declined continuously with increasing of the cyclic number due to the loss of the sorbent activity, and the final $\mathrm{C}{\mathrm{O}}_2$ capture efficiency would remain nearly constant, due to the sorbent already reaching the final residual capture capacity. After the $\mathrm{C}{\mathrm{O}}_2$ capture step, the Ca-based sorbents need to be regenerated to be used for a subsequent cycle, and the multiple calcination processes of Ca-based sorbent under different calcination conditions are studied and discussed. Reactivity loss of limestone, dolomite and $\mathrm{Ca}\mathrm{O}∕{\mathrm{Ca}}_{12}{\mathrm{Al}}_{14}{\mathrm{O}}_{33}$ sorbents from a fluidized bed reactor at both mild and severe calcination conditions was compared with the TGA data. At mild calcination conditions, TGA results of sorbent reactivity loss were similar to the experimental results of fluidized bed reactor for three sorbents at $850°\mathrm{C}$ calcination temperature, and this indicated that TGA experimental results can be used as a reference to predict sorbent reactivity loss behavior in fluidized bed reactor. At severe calcination condition, sorbent reactivity loss behavior for limestone and dolomite from TGA compare well with the result from a fluidized bed reactor.

Different carbon materials were tested as precursors for the production of $\mathrm{C}{\mathrm{O}}_2$ adsorbents. The chemical modification of the surface of the prepared adsorbents was studied by means of three different approaches: impregnation with amines, electrophilic aromatic substitution, and heat treatment in the presence of ammonia. The samples were chemically characterized and the porous texture was evaluated from the ${\mathrm{N}}_2$ adsorption isotherms at $-196°\mathrm{C}$ . The $\mathrm{C}{\mathrm{O}}_2$ adsorption capacities of the adsorbents at 25 and $100°\mathrm{C}$ were evaluated in a thermogravimetric analyzer. In general, the incorporation of basic nitrogen functionalities enhanced the $\mathrm{C}{\mathrm{O}}_2$ capture capacities of the modified carbons, but this increase depended on the textural properties of the support and the surface modification methodology. $\mathrm{C}{\mathrm{O}}_2$ adsorption capacities of up to $111\mathrm{mg}$ $\mathrm{C}{\mathrm{O}}_2∕\mathrm{g}$ at room temperature were attained. All the tested samples were completely regenerated when subjected to heat treatment at $100°\mathrm{C}$ under inert atmosphere.

Bauxite residue (red mud), an alkaline slurry from alumina refining, is produced in large volumes and disposed of in large surface impoundments. The objectives of this study were to measure the extent of neutralization of bauxite residue by carbon dioxide as a function of $\mathrm{C}{\mathrm{O}}_2$ partial pressure and to determine the geochemical reactions responsible for carbon sequestration. Bauxite residue was exposed to carbon dioxide $\left(\mathrm{C}{\mathrm{O}}_2\right)$ at partial pressures ranging from ${10}^{-3.5}\text{to}1\mathrm{atm}$ and the residue pore-water pH was measured until a steady state pH was achieved. Using pure $\mathrm{C}{\mathrm{O}}_2$ $(P_{\mathrm{C}{\mathrm{O}}_2}=1\mathrm{atm})$ , a steady state pH of 7.7 was established in $1\text{day}$ while it took $9\text{days}$ to reach a pH of 9.8 at $P_{\mathrm{C}{\mathrm{O}}_2}={10}^{-3.5}\mathrm{atm}$ . The pH rebounded to 9.9 after $1\text{day}$ when bauxite residue first neutralized at $P_{\mathrm{C}{\mathrm{O}}_2}$ of 1, 0.1, or $0.01\mathrm{atm}$ for $10\text{days}$ was subsequently exposed to $P_{\mathrm{C}{\mathrm{O}}_2}={10}^{-3.5}\mathrm{atm}$ , indicating that the pH change during short-term carbonation $(t⩽10\text{days})$ was due to reactions of carbonic acid and $\mathrm{O}{\mathrm{H}}^{-}$ in the pore water. A longer reaction time $\left(30\text{days}\right)$ at $P_{\mathrm{C}{\mathrm{O}}_2}=1\mathrm{atm}$ indicated carbonation of tricalcium aluminate (C3A solid) and conversion to calcite. This was confirmed in aged field samples that had been carbonated in air $(P_{\mathrm{C}{\mathrm{O}}_2}={10}^{-3.5}\mathrm{atm})$ for $30\text{years}$ . The maximum solid-phase $\mathrm{C}{\mathrm{O}}_2$ sequestration (as calcite) potential of red mud can be realized at $P_{\mathrm{C}{\mathrm{O}}_2}={10}^{-3.5}\mathrm{atm}$ ; however, the sequestration rate is limited by slow dissolution of C3A. The $\mathrm{C}{\mathrm{O}}_2$ sequestration potential is small (0.029–0.057 MMT $\mathrm{C}{\mathrm{O}}_2∕\text{year}$ ) compared to annual $\mathrm{C}{\mathrm{O}}_2$ emissions $(110\mathrm{MMT}∕\text{year})$ for bauxite refining and red mud production of $30\mathrm{MMT}∕\text{year}$ .

Measurements of the distribution of carbon and nitrogen isotopes in soils are needed due to their potential to improve our understanding of soil $\mathrm{C}{\mathrm{O}}_2$ emissions and sequestration under varying climatic conditions and land management technologies. Organic carbon and nitrogen isotopic and elemental composition was measured in baseline forest soils as well as in anthropogenically and intermittently flooded soils in the southern Appalachian Mountains. For the undisturbed forest soils, the consistent relationship between elemental and isotopic composition in the soil column, across soil organic matter pools, was supportive of the hypothesis that isotopic signature is reflective of microbial induced carbon turnover and nitrogen decomposition. A climatologic analysis suggested that isotopic indicators of soil organic matter turnover were influenced by mean annual temperature at our study sites, as well as in 10 other regional and global soil carbon and nitrogen isotopic studies. Intermittently flooded soils showed carbon and nitrogen distributions that reflected successive high magnitude events where sediments were deposited upon a developing forest floor. Measurements from soils disturbed by mining, agriculture, and recreation suggest the potential for using carbon and nitrogen isotopes to indicate the degree of soil organic matter turnover induced by different land management scenarios.

Carbon capture and storage is one family of technologies that could be used to significantly reduce global carbon dioxide $\left(\mathrm{C}{\mathrm{O}}_2\right)$ emissions. This paper reviews the likely flexibility of power plants with postcombustion capture, with a focus on an improved characterization of the dynamic performance of power plants with $\mathrm{C}{\mathrm{O}}_2$ capture. The literature has focused on design and optimization for steady state operation of power plants with capture, often at a single design point. When dynamic behavior is considered, it is possible that designs should be altered for best overall plant performance. Economic trade-offs between improving transport and storage scheme flexibility and constraining power plant operations should also be carefully analyzed, particularly if the captured $\mathrm{C}{\mathrm{O}}_2$ is to be used in another process such as enhanced oil recovery. Another important aspect of real plant operation will be adhering to legislative requirements. Further work is required to identify mechanisms that allow flexible operation without undermining any targets set for storing $\mathrm{C}{\mathrm{O}}_2$ and/or restricting global $\mathrm{C}{\mathrm{O}}_2$ emissions.

The modification of CaO sorbent with organic alumina precursor to enhance the sorption capacity during cyclic capture is demonstrated in this study. The results indicate that during the $10\min$ sorption time for each cycle, the $\mathrm{C}{\mathrm{O}}_2$ sorption capacity of original CaO sorbent is significantly decreased from 0.27 to 0.19 and $0.12\mathrm{g}$ $\mathrm{C}{\mathrm{O}}_2∕\mathrm{g}$ sorbent after 5 and $60\text{cycles}$ , respectively. On the other hand, the organic alumina modified CaO has a fresh capacity of $0.22\mathrm{g}$ $\mathrm{C}{\mathrm{O}}_2∕\mathrm{g}$ sorbent, it increases to $0.24\mathrm{g}$ $\mathrm{C}{\mathrm{O}}_2∕\mathrm{g}$ sorbent after $5\text{cycles}$ , and then decreases to around $0.15\mathrm{g}$ $\mathrm{C}{\mathrm{O}}_2∕\mathrm{g}$ sorbent after $60\text{cycles}$ . When increasing the sorption time to $60\min$ at the 66th cycle, the sorption capacity of original CaO is 0.2, whereas it is $0.26\mathrm{g}$ $\mathrm{C}{\mathrm{O}}_2∕\mathrm{g}$ sorbent for organic alumina modified CaO. The results demonstrate that by adding only 5% by weight of ${\mathrm{Ca}}_{12}{\mathrm{Al}}_{14}{\mathrm{O}}_{33}$ species into the CaO sorbent, the $\mathrm{C}{\mathrm{O}}_2$ sorption capacity can be enhanced up to $\sim 27\%$ by weight.

Anthropogenic releases of carbon dioxide are being studied with respect to the global carbon dioxide cycle. Concrete, the most used building material in the world, contributes to these releases significantly in the manufacturing of cement, both from energy use and from the calcination chemical process. The reverse process, carbonation, in which the hydrated cement reacts with carbon dioxide to form calcite, is being studied under many different circumstances for its potential to balance this release by sequestering carbon dioxide. This research focuses on other carbon dioxide sequestration species within concrete. It is hypothesized, and has been shown in some cases, that many other dissolved carbon dioxide complexes exist within the cement paste matrix, as they do in many aqueous-based systems and also that carbonate species other than calcite may exist. It is shown that carbon dioxide species, other than calcite, do exist in cement paste samples in amounts that are significant with respect to the stoichiometric potential for reversing the calcination process.

We investigated the effects of carbonation and regeneration temperature on the $\mathrm{C}{\mathrm{O}}_2$ capture characteristics using SorbKX35, a potassium-based solid sorbent in a bubbling fluidized-bed reactor. A dry sorbent, SorbKX35 consists of ${\mathrm{K}}_2\mathrm{C}{\mathrm{O}}_3$ for absorption and supporters for mechanical strength. We also measured the physical properties of the sorbent, such as pore size, pore volume, and surface area after carbonation or regeneration, to confirm the extent of the reaction. With ${\mathrm{H}}_2\mathrm{O}$ vapor pretreatment, nearly complete $\mathrm{C}{\mathrm{O}}_2$ removal was initially achieved and maintained for about $10\min$ within a temperature range of $333.15–363.15\mathrm{K}$ with $2\mathrm{s}$ gas residence time. At lower temperature, $\mathrm{C}{\mathrm{O}}_2$ capture was more effective during $1\mathrm{h}$ of carbonation. From the results of temperature programmed desorption and gas adsorption method (BET), we found that the regeneration of carbonated SorbKX35 was not complete at $473.15\mathrm{K}$ . The results obtained in this study can be used as basic data for designing and operating a large-scale $\mathrm{C}{\mathrm{O}}_2$ capture process with two fluidized-bed reactors.