Journal of CO2 Utilization 1 (2013) 5059

Contents lists available at SciVerse ScienceDirect

Journal of CO2 Utilization

journal homepage: www.elsevier.com/locate/jcou

A new catalytic opportunity for waste materials: Application of waste slag based catalyst in CO2 xation reaction
Yasutaka Kuwahara a,b, Hiromi Yamashita a,*
a b

Division of Materials and Manufacturing Science, Graduate School of Engineering, Osaka University, 2-1 Yamada-oka, Suita, Osaka 565-0871, Japan National Institute of Advanced Industrial Science and Technology (AIST), Research Institute for Innovation in Sustainable Chemistry, Heterogeneous Catalysis Group, 16-1 Onogawa, Tsukuba, Ibaraki 305-8569, Japan

A R T I C L E I N F O

A B S T R A C T

Article history: Received 15 January 2013 Received in revised form 13 March 2013 Accepted 13 March 2013 Available online 6 April 2013 Keywords: Blast furnace slag CO2 xation Heterogeneous catalysis Layered compound Material recycling

Iron and steel industry is one of the most energy-intensive industries, consuming 56% of the worlds total energy consumption and emitting 57% of total world CO2 gas emissions. Large volume of CO2 emitted from iron and steel industry makes a signicant contribution to global warming and climate change. Blast furnace slag (BFS), a high volume mineral waste discharged from iron and steel making processes, can potentially be used as an abundant and low-cost precursor for synthesizing active materials of catalytic interest that utilize the emitted CO2 in synthetic chemistry. In this study, we demonstrate that a Ca-based layered double hydroxide (LDH) compound can be synthesized from BFS through a facile synthetic procedure, and the thus synthesized material acts as a solid base catalyst that efciently catalyzes the cycloaddition reaction of epoxides with atmospheric pressure of CO2 to give vemembered cyclic carbonates. The strategy proposed in this study would be one of the possible approaches that contribute to both efcient utilization of CO2 and waste management problems lying in iron and steel industry. 2013 Elsevier Ltd. All rights reserved.

1. Introduction Iron and steel industry is one of the most energy-intensive industries, consuming 56% of the worlds total energy consumption and emitting 57% of total world CO2 gas emissions [15]. In iron and steel making processes, CO2 is mainly produced as a result of (i) reduction of iron ore with coke in a blast furnace to produce pig iron as a main product and (ii) decarbonisation of limestone (CaCO3) and dolomite (MgCO3) which are added as uxing materials together with coke for stripping the oxygen and other impurities from iron ore [1,2]. Production of every ton of steel produces about 2200 kg of CO2 on a world average (1800 kg of CO2 for many developed countries) [1,2]. Given a global crude steel production of 1400 Mt per year, the total CO2 emissions at least amount to approximately 2.5 Gt, and the CO2 emission from this industry is anticipated to further increase as this trend continues [6,7]. As witnessed over the last century, the increasing CO2 concentration in the atmosphere has been regarded as a leading contributor to global warming and climate change. The atmospheric CO2 level is anticipated to increase over the current concentration of approximately 395 ppm in the near future as

* Corresponding author. Tel.: +81 6 6879 7457; fax: +81 6 6879 7457. E-mail address: yamashita@mat.eng.osaka-u.ac.jp (H. Yamashita). 2212-9820/$ see front matter 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.jcou.2013.03.001

fossil fuels remain the major source utilized to meet global energy demand and unless advanced CO2 mitigation technologies are rapidly deployed [3,4,811]. Reduction of the CO2 emission into the atmosphere has been an urgent and critical concern in both academic and industrial research communities, and accordingly signicant research has recently focused on development of materials and technologies/processes suited for CO2 capture, storage and utilization (CCSU) and for improving the energy utilization efciency [1219]. In iron and steel industry, technologies for reducing CO2 gas emission and for enhancing energy efciency over the iron and steel making processes have extensively been developed; for example, in the short-term, energy-saving or recycling technologies (e.g. recycling steel scraps as secondary sources, recycling slag and sludge as uxing materials in steel making process to replace limestone) as well as increasing the use of renewable energy (such as biomass fuels) instead of the conventional reducing agents (coke, coal, etc.) [1,2]. In the long-term, in-process and postcombustion CO2 capture and its permanent or semi-permanent storage (CO2 capture and storage; CCS) have currently been considered as effective mitigation technologies for climate change [14], and the technical feasibility and the economic and environmental assessments of the processes have been believed to be promising [20]. However, CCS technology is still under development [21], and in an ideal CCS technology currently

Y. Kuwahara, H. Yamashita / Journal of CO2 Utilization 1 (2013) 5059 Table 1 Types and amounts of products generated from coal-red power plants and iron and steel making plants. Plant Input Iron and steel making plant Iron ore Limestone (CaCO3), dolomite (MgCO3) (uxing material) Coke, coal (reducing reagent) Pig iron (1026 Mt globally in 2010)a Crude steel (1417 Mt globally in 2010)a Blast furnace slag (230280 Mt globally in 2008)b Converter/rening slag (130200 Mt globally in 2008)b 2.5 Gt globally in 2009c (57% of total CO2 emission) Coal-red power plant Coal, oil, natural gas, biomass

51

Output Byproduct CO2 emission

a b c d

Electricity Fly ash (430 Mt globally in 2003)d 11.8 Gt globally in 2005 (40% of total CO2 emission)

Data from Ref. [6]. Data from Ref. [25]. Data from Ref. [7]. Data from Ref. [36].

proposed, the captured CO2 is nally sequestered in a reservoir, such as depleted oil and gas reservoirs, deep oceans, and deep aquifers, for an indenite period without any benets [4,12,20,22]. In terms of efcient utilization of CO2, the extracted and concentrated CO2 generated from such large stationary sources should be subjected into on-site usages and should be used for more productive purposes, such as feeding greenhouses, feeding enclosed algae installations for biofuel production, and supplying as a C1 feedstock for chemical and polymer production, where continuous supply of a large volume of concentrated CO2 gas is required [5,12]. Thus, further breakthrough technologies that efciently capture and utilize CO2 have been required in steel industry. Iron and steel making industry plants produce not only huge amount of CO2 but also high volume of byproduct, together with pig iron and steel as main products. Types and amounts of products generated from iron and steel making plants are tabulated in Table 1, together with those from coal-red power plants as comparative data. The solid byproducts generated during iron and steel manufacturing are mainly in the form of slag. There are different types of steel slag, each one named for the process from which they are generated, e.g. blast furnace slag (BFS), basic oxygen furnace slag (BOF), electric arc furnace acid slag (EAF) and ladle furnace basic slag (LF), the former one is also called ironmaking slag and the latter three are also called converter/rening slag [23,24]. Approximately 290 kg of BFS is generated per ton of pig iron, and approximately 110 kg of BOF is generated to produce one ton of crude steel. The annual world production of slag from iron and steel industries reaches almost 400500 Mt, which accounts for approximately one third of the global emissions of inorganic waste, and is anticipated to further increase as this trend continues [23,25]. Due to their mineral compositions, iron and steel making slags have been mostly recycled as secondary sources for hydraulic cement, concrete aggregate, and pavement materials in civil engineering work, or in part disposed of by landlling and dumping into sea without any prots [2327]. However, the heavy metal ions in the slags may have adverse effects on human health, aquatic life and the overall ecosystem via a potential seawater/ groundwater contamination, and the urgent problems such as the shortage of storage sites and severe environmental regulations are matters of great concern as well. Hence, development of new and

advanced recycling processes for waste slags in an energy-saving and environmentally-friendly way has been required to promote sustainable development of iron and steel industry [28]. In recent years, driven by both economic and environmental considerations, studies on recycling processes for waste materials have been attracting much attention. High volume waste materials resulting from large scale industrial sources particularly afford opportunities in terms of catalysis [29]. To date, a number of reports detailing the application of wastes to the preparation of catalysts have been reported, in which they are directly applied, in some instances, as catalysts themselves, in others they are used as precursors or alternative sources for the synthesis of active catalysts [3035]. For example, y ash, a solid waste residue produced from coal, oil and biomass combustion in fossil fueled power plants, has extensively been investigated to be utilized as adsorbent, heterogeneous catalyst, catalyst support as well as feedstock for zeolite synthesis exploiting its high silica and alumina content (see also Table 2) [3641]. One of the crucial drawbacks for recycling y ash is the potential inuence of compositional variation on the performance of the products, because the chemical compositions of y ashes are strongly dependent upon the fuel sources being burned (coal, oil, natural gas and biomass, etc.) and the combustion technology applied [36]. Compared to y ashes, the chemical composition of iron and steel making slags is less variable and is especially rich in CaO that by nature show strong basicity. These fundamental backgrounds of iron and steel making slags potentialize them to be used as lowcost and abundant raw materials for preparing a solid base catalyst [33,34]. Motivated by the above situations, we herein propose a new approach that would potentially contribute to both the efcient CO2 utilization and waste management problems associated with iron and steel industry. The strategy developed in this study is comprised of two parts: (i) conversion of waste slag into a Ca-based solid base catalyst and (ii) application of the slag-made catalyst in CO2 xation reaction (Scheme 1). Among several kinds of waste slags discharged from iron and steel making processes, BFS was chosen as a raw material primarily because of its larger amount produced than those of other types of slags [23,25]. As listed in Table 2, BFS mainly consists of CaO, SiO2, Al2O3 and MgO as well as slight amount of metallic elements such as Fe, Ti and Mn, whereas

Table 2 The chemical composition of BFS and y ash (wt%). Waste Blast furnace slaga Fly ashb
a b c

CaO 40.09 1.7

SiO2 34.58 61.8

Al2O3 14.78 27.9

MgO 5.29 0.3

FeO 1.53 2.6

TiO2 0.78 1.0

MnO 0.27

LOIc 2.0

Others 2.68 (S, Na2O, etc.) 4.7 (P2O5, Na2O, K2O, etc.)

BFS used in this study. Determined by ICP analysis. Data from Ref. [26]. May vary depending on the fuel sources being burned (coal, oil, natural gas and biomass, etc.) and the combustion technology applied. Loss on ignition.

52

Y. Kuwahara, H. Yamashita / Journal of CO2 Utilization 1 (2013) 5059

Scheme 1. Recycling of blast furnace slag (BFS) as a solid base catalyst and its applications in CO2 xation reaction.

converter/rening slags contain relatively larger iron oxide contents. As BFS is especially rich in divalent alkali earth metals (Ca2+ and Mg2+) as well as in trivalent element (Al3+), it was used as a low-cost and abundant precursor for preparing a layered double hydroxide (LDH) compound [42]. LDH compounds are synthetic or natural crystalline clays with the general formula of [M1x2+Mx3+(OH)2]x+(An)x/nmH2O (abbreviated as M2+M3+ An), where M2+ and M3+ are a divalent and a trivalent metal cation 6-fold coordinated to OH, respectively, and An is a chargecompensating anion species [43]. They consist of brucite (Mg(OH)2)-like positively charged two-dimensional hydroxide layers in which divalent and trivalent cations are randomly arranged, and the positive charge on the layers is compensated by intercalated anions. This type of material has recently been receiving rapidly increasing attention for a wide variety of applications in environmental remediation, catalysis, pharmaceutical and biological industries owing to its (i) anion exchange ability in the interlayer, (ii) easy accommodation of metal cations in the hydroxide layer and (iii) strong basicity derived from its surface hydroxyl groups [4349]. Exploiting the inherent strong basicity of the slag-made LDH compound, it was applied in CO2 xation reaction as a solid base catalyst. Although CO2 had been recognized as an unreactive molecule, recent advances in catalysis have realized the use of CO2 as a chemical feedstock in synthetic chemistry [5053]. Among various types of approaches for utilizing CO2 in synthetic chemistry, catalytic coupling of CO2 with epoxides to provide ve-membered cyclic carbonates was examined in this study (Scheme 2). Aggressive use of waste slags for such a productive purpose not only contributes to the waste management problems encountered in iron and steel industry but also affords new opportunities for the waste slags in CCSU technology. The structures of the LDH material synthesized from BFS (slagLDH) were veried by using X-ray diffraction (XRD), eld emission-scanning electron microscopy (FE-SEM), thermogravimetric analysis and nitrogen adsorptiondesorption measurement. The surface basicity that substantially affects the catalytic activity was investigated by CO2-temperature programmed desorption (TPD) measurement. The catalytic performances of the synthesized solid products were examined by the

cycloaddition reaction of epoxides with CO2 and were compared with those of pure LDH analogues. 2. Experimental 2.1. Materials The chemical composition of the raw slag (BFS) is listed in Table 2. The BFS mainly consists of amorphous CaO, SiO2, Al2O3 and MgO with a molar ratio of approximately Ca/Al = 2.5, which is fairly common in BFS produced in Japanese iron making plants. Fe2O3, TiO2 and MnO account for the remaining small proportion. These minor components are inevitably included in every BFS. All experiments were performed using the same batch of BFS. All commercially available chemical reagents for synthesis, solvents and organic compounds for catalytic reactions were puried using standard procedures. 2.2. Synthesis of slagLDH The slag-made LDH compound (slagLDH) was synthesized according to the method we previously reported [42]. Typically, 10.0 g of ball-milled BFS with a grain size of <1.0 mm was completely dissolved in 200 mL of 3 N HCl aqueous solution (Wako Pure Chemical Ind. Ltd.) and was stirred at 100 8C for 2 h to polymerize Si component. The obtained suspension was then ltered and washed with 200 mL of deionized water to separate into hydrated silica gel (SiO2 content > 92 wt% as a dried product) and leaching-solution containing other elements. To this leaching solution, 2 N NaOH aqueous solution (Wako Pure Chemical Ind. Ltd.) was added dropwise and pH was adjusted to 11.5 0.1. The suspension was vigorously stirred for 1 h and then aged at 100 8C for another 18 h to carry out crystallization, followed by ltration, washing with a copious amount of deionized water and drying at 100 8C overnight to yield 6.2 g of pale brown-colored slagLDH. The samples calcined at T 8C in air were named slagLDH(T). To provide a comparison, CaAlCl LDH (Ca2Al(OH)6Cl2H2O) was also prepared by conventional coprecipitation method [54]. In a typical method, 2 N NaOH aqueous solution was slowly added to an aqueous solution containing calcium chloride (CaCl2, Wako Pure Chemical Ind. Ltd.) and aluminum chloride (AlCl36H2O, Wako Pure Chemical Ind. Ltd.) with Ca2+/Al3+ molar ratio of 2.0 at 65 8C under vigorous stirring. The metal concentration in the solution was adjusted to 0.66 mol/L, and pH of the solution was nally adjusted to 11.5 0.1. The resulting gel was aged for 18 h at the same temperature, followed by ltration, washing with a copious amount of deionized water and drying at 100 8C overnight to yield CaAlCl LDH as a white powder.

O R

cat. CO2 R

O O O

Scheme 2. Cycloaddition reaction of epoxides with CO2.

Y. Kuwahara, H. Yamashita / Journal of CO2 Utilization 1 (2013) 5059

53

2.3. Characterization Powder X-ray diffraction (XRD) patterns were recorded using a Rigaku Ultima IV diffractometer with CuKa radiation ). The X-ray patterns were recorded over the 2u (l = 1.54056 A range of 10708 with a step size of 0.028. The morphology and particle size of the crystals were observed by a eld emissionscanning electron microscopy (FE-SEM) in a JEOL, JSM-6500 equipped with an energy dispersive X-ray uorescence spectrometer (EDX), by which the elemental analysis was performed. Thermal analysis was carried out with differential thermal analysisthermogravimetry (DTATG) (MAC Science Co. Ltd., TGDTA2000S) from room temperature to 1000 8C at a heating rate of 10 8C/min in N2 ow of 50 cm3/min using a-Al2O3 as a standard. Nitrogen adsorptiondesorption isotherms were measured at 196 8C using BELSORP-max system (BEL Japan, Inc.). For layered double hydroxide compounds, the samples were degassed under vacuum at 200 8C to eliminate physisorbed water molecules prior to the measurements, whereas oxide compounds were degassed under vacuum at 350 8C. Specic surface areas were calculated by the BET (BrunauerEmmettTeller) method using N2 adsorption data ranging from P/P0 = 0.05 to 0.25. The basicity of solids was studied by temperature programmed desorption using CO2 as probe molecule (CO2-TPD) using BELCAT-B system (BEL Japan, Inc.). Samples (50 mg) were preheated under a He ow (100 mL/min) for 6 h at the maximum temperature at which sample calcination was performed (as-synthesized slagLDH was exceptionally pretreated at 200 8C), subsequently were allowed to cool down to 40 8C and exposed to owing CO2 (50 mL/min) for 30 min. Then, the system was purged at 40 8C for 30 min with He in order to eliminate weakly adsorbed CO2. The CO2-TPD was carried out between 40 and 850 8C under a He ow (30 mL/min) with a pumping rate of 10 8C/min, and CO2 was detected by an on-line mass spectrometry. 2.4. Cycloaddition reaction of epoxides with CO2 A typical experimental procedure for the cycloaddition reaction of epoxide with 1 atm CO2 is as follows. Catalyst (0.5 g) was added

to a mixture containing epoxide (4 mmol) and anhydrous N,Ndimethylformamide (DMF, 3 mL) in a Pyrex glass reactor equipped with a reux condenser, after which time the reactor was immediately sealed to minimize any contamination due to moisture. When calcined samples were used as catalysts, they were added into the reaction solution immediately after the calcination procedure. The air in the reactor was replaced with 1 atm of CO2 using a balloon and the mixture was heated at 100 8C with magnetic stirring. A portion of the reaction mixture was collected by ltration after appropriate period of reaction time, and the amounts of organic substrates and products were quantied using a gas chromatograph (GC; Shimadzu GC-14B) with a ame ionization detector equipped with a HiCap-CBP 20 capillary column. To examine the reusability of catalyst, the spent catalyst was four times repeatedly used after washing with acetone and drying at 100 8C in air. 3. Results and discussion 3.1. Synthesis of LDH compound from BFS The structure of the slagLDH was identied by the combination of XRD, FE-SEM, nitrogen adsorptiondesorption and thermogravimetric analyses. Fig. 1A shows the XRD patterns of assynthesized slagLDH, together with those of pure CaAlCl LDH and raw BFS as references. BFS is generally obtained by rapidly or slowly quenching molten slag from a blast furnace in water or air, and its particle size and structure alter depending on the cooling speed. The BFS used in this study is amorphous (Fig. 1A(a)) and has a relatively large surface area (SBET = 8.1 m2/g), from which the slag-derived metal species (Ca2+, Al3+, Mg2+, etc.) were efciently extracted via an acid-leaching process using 3 N HCl aqueous solution and the subsequent separation procedure. In the acidleaching procedure, HCl or HNO3 can be used as acids, whereas other inorganic acids such as sulfonic acid and phosphoric acid are unacceptable due to the formation of the corresponding insoluble calcium salts [55], and acid solutions with higher concentrations (over 2.0 N) and with higher temperatures are preferred to facilitate a rapid polymerization of Si4+ species. The solid product

Fig. 1. (A) X-ray diffraction patterns of (a) raw BFS, (b) slagLDH and (c) CaAlCl LDH (Ca2Al(OH)6Cl2H2O). (B) X-ray diffraction patterns of slagLDH calcined at different temperatures: (a) as-synthesized, (b) 400 8C, (c) 600 8C and (d) 800 8C. The main crystallographic planes are indicated for pure CaAlCl LDH. Unindexed peaks seen in calcined samples are identical to those of mayenite (Ca12Al14O33).

54

Y. Kuwahara, H. Yamashita / Journal of CO2 Utilization 1 (2013) 5059

Fig. 2. (A) FE-SEM image and (B) EDX spectrum of slagLDH.

obtained by coprecipitation of the extracted metal species (for detailed synthesis procedures, see Section 2) showed a XRD pattern similar to that of CaAlCl LDH (Ca2Al(OH)6Cl2H2O; hydrocalumite) [54], and no other dominant crystalline phases were observed (Fig. 1A(b) and (c)), proving that a single-phase Ca AlCl LDH compound could be successfully synthesized from BFS. As displayed in Fig. 2, the FE-SEM micrograph of as-synthesized slagLDH showed a plate-like morphology with the particle size of 0.12.0 mm (Fig. 2A), which well resembles that typical of pure Ca AlCl LDH, and the EDX spectrum revealed that slagLDH contains not only Ca, Al and Cl but also the other slag-derived elements such as Mg, Fe, Mn, Si and Ti as minor constituents (Fig. 2B). The nominal chemical composition of slagLDH determined based on elemental and TG analyses can be represented by Ca2.03Mg0.49Fe0.046Mn0.019Si0.025Ti0.026Al1.00(OH)7.15Cl1.01(CO3)0.112.05H2O, revealing the generation of stoichiometric CaAlCl LDH compound (Ca: Al: Cl = 2: 1: 1). The intercalated Clions are derived from hydrochloric acid used for the acid-leaching. The minor metal cations such as Mg2+, Fe3+, Si4+, Ti4+ and Mn4+ (Fe and Mn may contain multi-valent species) are considered to be incorporated in the hydroxide layers by isomorphorously substituting either Ca2+ or Al3+ sites depending on their valence states (tetravalent metals may be present as oxoanion species in the interlayer or as aggregated species on the crystallite surface) [42]. It is worth noting that CaAlCl LDH possesses hexagonal lattice with an R-3 rhombohedral symmetry where layers consisting of Ca(OH)6H2O and Al(OH)6 octahedra are alternated and the Cl anions are placed between interlayers for chargecompensation [54,56]. Two important structural parameters in this type of material, basal spacings and the lattice parameters which represent the interlayer distance and the average cation cation distance within the hydroxide layers, respectively [54,56] calculated using XRD data are listed in Table 3. It was found that both structural parameters of the slagLDH were slightly smaller
Table 3 Textural properties of slag-made LDH compound and CaAlCl LDH. Sample XRD ) Basal spacing (A
a

than the data for pure CaAlCl LDH. Furthermore, the slagLDH showed diffraction peaks with relatively weak intensities compared to those of pure CaAlCl LDH (cf. Fig. 1A(b) and (c)). These results suggest a low regularity of the layers due to the interfusion of the slag-derived impurity elements with different ionic radius. In particular, the slagLDH was found to contain a considerable ) than that of Ca2+ quantity of Mg2+ with smaller ionic radius (0.69 A ) replacing the center of Ca(OH)6 octahedra, thereby (1.00 A resulting in the decreased structural parameters. Additionally, the BET surface area of the slagLDH was determined to be 21.1 m2/ g, which is nearly twice as large as that of pure CaAlCl LDH (11.9 m2/g), suggesting a modest impact of the impurity elements on surface areas (Table 3). The slagLDH showed a drastic crystallographic change along with calcination. Fig. 1B shows the XRD patterns of slagLDH compounds calcined at different temperatures in air. Above 400 8C, the diffraction patterns assignable to the LDH structure totally disappeared and instead showed a broad peak at 2u = 25408 (Fig. 1B(b)), indicating a phase transformation from LDH structure to amorphous Ca(Al)O mixed oxide. Calcination above 600 8C caused the generation of new crystalline phases assignable to CaO, MgO and mayenite (Ca12Al14O33) (Fig. 1B(c) and (d)). These crystallographic changes along with calcination agree well with the results for pure CaAlCl LDH previously reported elsewhere [54,57]. In these calcined samples, the minor impurity elements are believed to be present in their lattice structures isomorphorously substituting either Ca2+ or Al3+ sites [33,34]. Upon calcination, the BET surface area gradually decreased in the order of: as-synthesized slagLDH (21.1 m2/g) > slagLDH(400) (18.9 m2/ g) > slagLDH(600) (5.9 m2/g) > slagLDH(800) (2.8 m2/g). Fig. 3 shows the weight loss curve and the corresponding rst derivative of the slagLDH. The slagLDH exhibited three main weight losses over the temperature ranges of 25 8C T 200 8C, 200 8C T 400 8C and T 400 8C; the rst weight loss seen at

N2 adsdes ) Lattice parameter a (A

b

Nominal chemical compositione Vtotal (cm /g) 0.157 0.126 Ca2.03Mg0.49Fe0.046Mn0.019Si0.025Ti0.026 Al1.00(OH)7.15Cl1.01(CO3)0.112.05H2O Ca2Al(OH)6Cl2H2O
d 3

SBET (m /g) 21.1 11.9

slagLDH CaAlCl LDH

a b c d e

7.858 7.869

5.747 5.753

Calculated from d0 0 3. Dened by a = 2d1 1 0. Calculated by the BET method using N2 adsorption data. Amount of N2 adsorbed at P/P0 = 0.99. Determined based on ICP and TG analyses.

Y. Kuwahara, H. Yamashita / Journal of CO2 Utilization 1 (2013) 5059

55

0.4

(A)
0
-1

(B)
LDH Amorphous mixed oxide CaO+MgO +Mayenite

Weight loss / wt%

10.4 wt% -10

First derivative / wt% K

0.3 317 132 0.2

-20

17.1 wt%

0.1 668 0.0

-30 0 200 400 600 800 1000 0 200 400

600

800

1000

Temperature / C

Temperature / C

Fig. 3. (A) Weight loss curve and (B) its rst derivative of slagLDH.

below 200 8C corresponds to dehydration of 2 water molecules intercalated in the interlayer, and the second one centered at around 317 8C corresponds to dehydroxylation of the hydroxyls losing the mass fraction of 3 water molecules (Ca2Al(OH)6Cl ! Ca2AlO3Cl + 3H2O) [54,57]. An additional peak seen at 550750 8C is ascribed to the transition from amorphous Ca(Al) O mixed oxide into CaO and mayenite binary phases involving anion decomposition [54,57]. These thermal decomposition dynamics are typical of LDH compounds and coincide well with the crystallographic changes observed by XRD, again verifying a LDH formation. Thus a single-phase CaAlCl LDH compound could be synthesized using BFS as a sole material source, while it contains appreciable amount of the slag-derived impurity elements, such as Mg, Fe, Ti and Mn, in its structure. The material balance evaluation analysis shows that in our synthesis pathway about 55 wt% of BFS can be nally converted into LDH compound, giving an atom economy of 56% (based on metals). This value seems quite low, however, total atom economy would reach 90% when the separated hydrated silica, a gel product recovered during the synthesis by ltration, is fully utilized. As more than 92 wt% of the separated hydrated silica consists of SiO2, it can potentially be utilized as an alternative low-cost silica source for benecial purposes, such as chemical feedstock for zeolite synthesis [42]. Particular consideration needs to be given to the potential inuence of compositional variation of BFS upon the quality of the nal product. When BFS with a signicantly small Ca (or Al) content is used as a raw material, another makeup Ca (or Al) source needs to be added to produce a stoichiometric CaAlCl LDH, otherwise, poorly-crystallized calciumaluminate hydroxide is produced. Although a further analysis on water discharges and energy consumption over the process needs to be done for substantiating the practicability of this conversion process, our synthesis pathway can replace the existing LDH synthesis route that requires multiple pure metal-containing chemical sources and would meet the strong demand for the effective utilization of waste slags in iron and steel industry. 3.2. Cycloaddition reaction of epoxides with atmospheric CO2 Among various types of approaches for utilizing CO2 in synthetic chemistry [5053], catalytic coupling of CO2 with epoxides to give ve-membered cyclic carbonates is considered

to be the most powerful approach, because the products are commercially valuable for the production of engineering plastics as well as for the synthesis of pharmaceuticals and ne chemicals [58]. To date, a variety of homogeneous and heterogeneous catalytic systems that can catalyze this reaction have been reported [59,60]; however there have been few reports on this reaction under 1 atm of CO2 and, when examined, toxic solvents, homogeneous additives and/or reactive epoxides are required in these catalyst systems [61]. Among a number of reports concerning this catalysis, Yamaguchi et al. have reported that Mg(Al)O mixed oxides obtained from MgAl hydrotalcites via a thermal decomposition act as efcient solid base catalysts for this reaction without any toxic solvents and additives [62]. Although this catalyst can catalyze various epoxides to the corresponding vemembered cyclic carbonates under relatively mild reaction conditions, it still requires a high CO2 pressure (5 atm) for achieving high catalytic efciency. A worthwhile challenge for this reaction is to develop efcient heterogeneous catalyst that can provide a high catalytic efciency under atmospheric pressure of CO2. CaAlCl LDH with nominal chemical formula of Ca2Al(OH)6Cl2H2O is a structurally close analogue to MgAl hydrotalcite. Structurally considering, CaAlCl LDH and its calcined derivatives (i.e. Ca(Al)O mixed oxides obtained from CaAlCl LDH via a thermal decomposition) as well as those synthesized from BFS can be alternative solid base catalysts active for this reaction, or may provide superior catalytic activities. We herein examined the catalytic activity of slagLDH in the coupling reaction of CO2 with epoxides under a variety of reaction conditions using styrene oxide as a model substrate and at atmospheric pressure of CO2 (Scheme 3). In this reaction, use of highly polar solvents, such as DMF and acetonitrile, were particularly effective for achieving a high catalytic efciency probably due to high solubility of CO2 into the solvent, while nonpolar solvents resulted in a poor reactivity. Fig. 4 shows the temperature dependence on the conversion of styrene oxide and selectivity of the products when slagLDH(600) was used as a catalyst. Coupling reaction of CO2 with styrene oxide yielded styrene carbonate as a main product, with styrene glycol and trace of polymerized esters as minor byproducts. A higher reaction rate was attained by simply increasing the reaction temperature, however, increasing the reaction temperature over 120 8C resulted in low selectivity of styrene carbonate due to the increased formation of styrene glycol as an undesirable byproduct.

56

Y. Kuwahara, H. Yamashita / Journal of CO2 Utilization 1 (2013) 5059

O O cat. CO2 (1 atm)

Scheme 3. Cycloaddition reaction of styrene oxide with atmospheric pressure of CO2.

O O

100

80

60

40

20

0 80 100 120 140

Reaction temperature / C
Fig. 4. Effect of reaction temperature on the coupling reaction of styrene oxide and 1 atm CO2 using slagLDH(600) as a catalyst; conversion of styrene oxide (bar), selectivity of styrene carbonate (*) and selectivity of styrene glycol (~). Reaction conditions: catalyst (0.5 g), styrene oxide (4 mmol), DMF (3 mL), CO2 pressure = 1 atm, reaction time = 24 h.

In order to further optimize the catalytic efciency, slagLDH and pure CaAlCl LDH calcined at various temperatures were tested for this reaction under the reaction conditions of 1 atm CO2 and at 100 8C. As shown in Fig. 5, over a series of slagLDH, the catalytic

100

slagLDH Ca-Al-Cl LDH

80

60

40

20

0 200 400 600 800

Calcination temperature / C
Fig. 5. Effect of calcination temperature on the coupling reaction of styrene oxide and 1 atm CO2 using slagLDH and pure CaAlCl LDH as catalysts; gray bar and lled symbols (*) show styrene oxide conversion and styrene carbonate selectivity over slagLDH catalysts, respectively, and white bar and open symbols (*) show those over CaAlCl LDHs. Reaction conditions: catalyst (0.5 g), styrene oxide (4 mmol), DMF (3 mL), 100 8C, CO2 pressure = 1 atm, reaction time = 24 h.

performance strongly depended on the calcination temperature; higher conversion rates were attained as increasing calcination temperature, while calcination at 800 8C led to an appreciable decrease in styrene carbonate selectivity due to the phase separation into CaO, MgO and mayenite phases. Among the catalysts we examined, slagLDH(800) which is mainly comprised of crystalline CaO and mayenite phases was found to be the best, affording styrene carbonate in 85.3% yield and with 90% selectivity after 24 h of reaction time even under atmospheric pressure of CO2, whereas slagLDH(200) retaining a LDH structure afforded styrene carbonate in 56.2% yield and with 87.5% selectivity under the same reaction conditions. The catalytic activity afforded by slagLDH(800) is considerably higher than that given by the previously reported Mg(Al)O mixed oxides obtained from the conventional MgAl hydrotalcite (2344% carbonate yield was provided by Mg(Al)O mixed oxides with Mg2+/Al3+ molar ratio of 35). This is primarily because of the stronger basicity derived from Ca atoms than that from Mg. The catalytic results over a series of pure CaAlCl LDH also showed a similar trend, albeit to a lesser degree (Fig. 5). Such increased catalytic performances upon thermal treatment can be strongly associated with the changes in surface basicity of the solids derived from the phase transition. Fig. 6A and B displays the CO2-TPD proles of slagLDH and those of pure CaAlCl LDH calcined at different temperatures, respectively. As-synthesized LDH samples show several desorption peaks in the low temperature region, around 100 and 300 8C (Fig. 6(a)), which are attributable to weak basic sites related to surface hydroxyl groups and medium basic sites derived from O2 ions adjacent to the surface hydroxyl groups, respectively [57]. The samples calcined at 400600 8C exhibited dened CO2 desorption peaks at around 600800 8C (Fig. 6(b) and (c)), reecting the generation of strongly basic sites on the Ca(Al)O mixed oxides. For both samples, calcination at 800 8C afforded a single CO2 desorption band at around 450650 8C, which is identical to that of CaO. These drastic changes in basicity upon thermal decomposition are fairly consistent with the crystallographic changes conrmed by XRD and TG measurements (see Figs. 1 and 3). A quantitative calculation based on the CO2-TPD and nitrogen physisorption measurements indicates that the surface density of basic sites, dened as the number of moles of CO2 adsorbed per surface area of solids, of slagLDH increases upon thermal treatment in the order of: slagLDH(400) (85.4 mmol/m2) < slagLDH(600) (91.4 mmol/m2) < slagLDH(800) (202.3 mmol/m2). This order is fairly consistent with the trend seen in the catalytic reaction, hence the increased catalytic activity can be attributed to the increased number of accessible basic sites and/or the stronger basic sites afforded by the thermal decomposition [57]. It is very important to mention that the surface densities of basic sites of slagLDH materials were overall higher than those of the corresponding CaAlCl LDH materials; CaAlCl LDH(400) (73.6 mmol/m2), CaAlCl (600) (77.1 mmol/m2) and CaAlCl LDH(800) (89.7 mmol/m2). Although slagLDH and CaAlCl LDH showed similar crystallographic changes along with the thermal decomposition, the former (and its calcined derivatives) overall provided higher catalytic activities than do the corresponding Ca AlCl LDH materials (Fig. 5). This result indicates a positive role of the slag-derived impurity elements in the catalysis, that is, the slag-derived impurity elements incorporated within the Ca(Al)O mixed oxide matrix have an ability to increase the number of accessible basic sites, thereby leading to the improved catalytic activities. It is also worth mentioning that neither CaO alone nor a physical mixture of CaO and Al2O3 provided such high activities (17% and 16% conversion of styrene oxide after 24 h of reaction time, respectively). This is because the reaction involves the cooperative effect of both acid and basic sites, in which unreactive

Conversion or selectivity / %

Conversion or selectivity / %

Y. Kuwahara, H. Yamashita / Journal of CO2 Utilization 1 (2013) 5059

57

(A)
97 298 98

(B)
300

(a)

730

(a)
700

Intensity (m/z = 44) / a.u.

(b)

716

Intensity (m/z = 44) / a.u.

(b)
730

(c)
580

(c)
541

(d)
0 200 400 600 800 0

(d)
200 400 600 800

Temperature / C

Temperature / C

Fig. 6. CO2-TPD proles of (A) slagLDH and (B) CaAlCl LDH calcined at different temperatures: (a) as-synthesized, (b) 400 8C, (c) 600 8C and (d) 800 8C.

CO2 is activated on the Lewis basic sites to form a carbonate species and epoxide is coordinated on the adjacent acid sites to be ringopened by a nucleophilic attack of the carbonate species independently [62,63], i.e. the proximity of basic sites (attributed to the lattice oxygen atoms) and acid sites (attributed to the Al sites) to each other given by the formation of CaOAl bonds is the key essential for achieving a high catalytic efciency in this catalytic system. Table 4 shows the results of catalytic coupling of various epoxides to the corresponding cyclic carbonates under 1 atm of CO2 and at 100 8C using slagLDH(600) as a catalyst, which successfully converted a variety of terminal epoxides, possessing aromatic, aliphatic, and both electron-donating and -withdrawing substituents, into the corresponding cyclic carbonates with good carbonate selectivity. Epichlorohydrin reacted with CO2 to form the corresponding carbonate in 92.3% yield after 12 h (entry 1). Glycidylphenylether showed as a similar reaction rate as that of styrene oxide (cf. entry 25). Aliphatic epoxides such as 1,2epoxyhexane and 1,2-epoxyoctane gave 93.6 and 92.1% yields for hexane carbonate and octane carbonate, respectively, after 48 h of reaction time (entries 7 and 9).

To evaluate the reusability of the catalyst, slagLDH(600) was subjected to the cycloaddition reaction of styrene oxide with 1 atm CO2 over four repeated cycles. As shown in Fig. 7, the slagLDH catalyst could be easily recovered from the reaction mixture by simple ltration, and was at least four times reusable without any appreciable loss of activity, selectivity and catalyst weight; conversions of 98% and selectivities of over 90% could be retained. The spent catalyst showed X-ray diffractions almost similar to those of a fresh catalyst (Fig. 8), and no signicant difference in chemical compositions was observed. Additionally, ICP analysis of the ltrate conrmed that the elution of the slag-derived metals (Fe, Mn, Ti) was negligible. These results ensure the ne durability and stability of the catalyst. It was demonstrated that the slagLDH and its calcined derivatives work as efcient heterogeneous base catalysts for the cycloaddition reaction of epoxides with 1 atm CO2, where the slag-derived impurity elements were found to function as catalyst promoters. The chemical xation of CO2 into synthetically benecial compounds is of great interests in terms of green and

Table 4 Cycloaddition reaction of various epoxides with CO2 using slagLDH(600).a Entry 1 2 3 4 5 6 7 8 9 Substrate t (h) 12 24 48 24 48 24 48 24 48 Conversion (%) 99.0 82.2 97.5 89.1 >99 71.6 >99 70.2 95.0 Yield of carbonate (%) 92.3 79.0 90.0 82.4 90.8 70.0 93.6 67.8 92.1

80

80

60

60

40

40

20

20

fresh

2nd

3rd

4th

Number of cycles
Fig. 7. Reusability of slagLDH(600) catalyst in cycloaddition reaction of styrene oxide with 1 atm CO2. Reaction conditions: catalyst (0.5 g), styrene oxide (4 mmol), DMF (3 mL), 100 8C, CO2 pressure = 1 atm, reaction time = 24 h.

a Reaction conditions: catalyst (0.5 g), epoxide (4 mmol), DMF (3 mL), CO2 pressure = 1 atm, 100 8C. Corresponding glycols were produced as major byproducts.

Selectivity of styrene carbonate / %

100

100

Conversion of styrene oxide / %

58

Y. Kuwahara, H. Yamashita / Journal of CO2 Utilization 1 (2013) 5059

operated on a sufciently large scale and in commercializable levels. Although the economics of the overall process have hardly been studied, the strategy proposed in this study would be one of the promising approaches that contribute to both the efcient utilization of CO2 and the waste management problems encountered in iron and steel industry. Acknowledgements This work was nancially supported by the Global COE Program for Chemistry Innovation (Project: Center of Excellence for Advanced Structural and Functional Materials Design), and the Grant-in-Aid for Scientic Research (KAKENHI) from the Ministry of Education, Culture, Sports, Science and Technology (No. 40200688). The authors appreciate Mr. Keita Tsuji at BEL Japan, Inc. for assistance with CO2-TPD measurements. References
[1] C.-Q. Hu, L.-Y. Chen, C.-X. Zhang, Y.-H. Qi, R.-Y. Yin, J. Iron Steel Res. Int. 13 (2006) 3842. [2] C. Xu, D.Q. Cang, J. Iron Steel Res. Int. 17 (2010) 17. [3] J.T. Houghton, B.A. Callander, S.K. Varney (Eds.), Climate Change 1992: The Supplementary Report to the IPCC Scientic Assessment, Cambridge University Press, Cambridge, UK, 1992. [4] B. Metz, O. Davidson, H. de Coninck, M. Loos, L. Meyer (Eds.), Intergovernmental Panel on Climate Change Special Report on Carbon Capture and Storage, Cambridge University Press, Cambridge, UK, 2005. [5] A.J. Hunt, E.H.K. Sin, R. Marriott, J.H. Clark, ChemSusChem 3 (2010) 306322. [6] Steel Statistical Yearbook 2011, World Steel Association, Worldsteel Committee on Economic Studies, Brussels, 2011. [7] K. Tanaka, Energy Policy 51 (2012) 578585. [8] Summary for Policymakers, S. Solomon, D. Qin, M. Manning, Z. Chen, M. Marquis, K.B. Averyt, M. Tignor, H.L. Miller (Eds.), Climate Change 2007: The Physical Science Basis. Contribution of Working Group 1 to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change, Cambridge, UK, 2007. [9] The International Energy Outlook 2008, Energy Information Administration, Washington, 2008. [10] D.P. Schrag, Science 315 (2007) 812813. [11] S. Chu, Science 325 (2009) 1599. [12] P. Markewitz, W. Kuckshinrichs, W. Leitner, J. Linssen, P. Zapp, R. Bongartz, A. ller, Energy Environ. Sci. 5 (2012) 72817305. Schreiber, T.E. Mu [13] S. Choi, J.H. Drese, C.W. Jones, ChemSusChem 2 (2009) 796854. [14] C.W. Jones, Annu. Rev. Chem. Biomol. Eng. 2 (2011) 3152. [15] Y. Kuwahara, D.-Y. Kang, J.R. Copeland, N.A. Brunelli, S.A. Didas, P. Bollini, C. Sievers, T. Kamegawa, H. Yamashita, C.W. Jones, J. Am. Chem. Soc. 134 (2012) 1075710760. [16] Y. Kuwahara, D.-Y. Kang, J.R. Copeland, P. Bollini, C. Sievers, T. Kamegawa, H. Yamashita, C.W. Jones, Chem.-Eur. J. 18 (2012) 1664916664. [17] Z.-Z. Yang, Y.-N. Zhao, L.-N. He, RSC Adv. 1 (2011) 545567. [18] Z.-Z. Yang, L.-N. He, Y.-N. Zhao, B. Li, B. Yu, Energy Environ. Sci. 4 (2011) 3971 3975. [19] A.-H. Liu, R. Ma, C. Song, Z.-Z. Yang, A. Yu, Y. Cai, L.-N. He, Y.-N. Zhao, B. Yu, Q.-W. Song, Angew. Chem. Int. Ed. 51 (2012) 1130611310. [20] D. Gielen, Y. Moriguchi, Energy Policy 30 (2002) 849863. [21] A.S. Bhown, B.C. Freeman, Environ. Sci. Technol. 45 (2011) 86248632. [22] R.S. Haszeldine, Science 325 (2009) 16471652. [23] The annual report of statistics of iron and steel slag 2008, The Japan Iron and Steel Federation, Nippon Slag Association; see the website, http://www.slg.jp. [24] B. Das, S. Prakash, P.S.R. Reddy, V.N. Misra, Resour. Conserv. Recycl. 50 (2007) 40 57. [25] H.G.V. Oss, Iron and Steel Slag Ch. in 2008 Minerals Yearbook, U.S. Department of the Interior, U.S. Geological Survey, 2010, pp. 69.169.8. [26] S. Kumar, R. Kumar, A. Bandopadhyay, Resour. Conserv. Recycl. 48 (2006) 301 314. [27] Y. Huang, R.N. Bird, O. Heidrich, Resour. Conserv. Recycl. 52 (2007) 5873. [28] Z. Huaiwei, H. Xin, Resour. Conserv. Recycl. 55 (2011) 745754. [29] M. Balakrishnan, V.S. Batra, J.S.J. Hargreaves, I.D. Pulford, Green Chem. 13 (2011) 1624. [30] K. Hosni, E. Srasra, Appl. Clay Sci. 43 (2009) 415419. [31] Y. Kuwahara, T. Ohmichi, T. Kamegawa, K. Mori, H. Yamashita, J. Mater. Chem. 19 (2009) 72637272. [32] Y. Kuwahara, T. Ohmichi, T. Kamegawa, K. Mori, H. Yamashita, Chem. Lett. 38 (2009) 626627. [33] Y. Kuwahara, K. Tsuji, T. Ohmichi, T. Kamegawa, K. Mori, H. Yamashita, Catal. Sci. Technol. 2 (2012) 18421851. [34] Y. Kuwahara, K. Tsuji, T. Ohmichi, T. Kamegawa, K. Mori, H. Yamashita, ChemSusChem 5 (2012) 15231532. [35] T. Li, H. Liu, Y. Fan, P. Yuan, G. Shi, X.T. Bi, X. Bao, Green Chem. 14 (2012) 3255 3259.

Fig. 8. X-ray diffraction patterns of (a) fresh and (b) used slagLDH(600) catalyst. Unindexed peaks are identical to those of mayenite (Ca12Al14O33).

sustainable chemistry, because CO2 can be considered as an inexpensive, nontoxic and abundant C1 feedstock. The results shown here, therefore, open up the possibility of the slag-made LDH compound to be used as a highly-valuable and environmentally-friendly catalyst. The catalytic performance of the slag-made LDH catalyst is, of course, still lower than those of the highlyelaborated, state-of-the-art catalysts reported elsewhere [59,60]. However, the merit of this catalyst lies in the fact that it was prepared from the waste slags through an economical and energysaving route. It is believed that the slag-made LDH compound can be a low-cost and abundant alternative catalyst for the CO2 xation reaction and that its practical application will surely contribute to the establishment of sustainable chemical processes. 4. Conclusions We demonstrated that BFS can be used as an inexpensive and abundant precursor for the preparation of LDH compound, and the synthesized LDH compound, especially in the form of mixed oxide, works as an efcient heterogeneous base catalyst for the cycloaddition reaction of epoxides with CO2 to provide vemembered cyclic carbonates. The reaction proceeded under relatively mild reaction conditions, i.e. at atmospheric pressure of CO2 and at 100 8C, which can directly be supplied on-site from the exhaust gas and exhaust heat emitted from iron and steel making plants, respectively. The use of BFS as alternative sources of chemical components for LDH compounds not only provides a lowcost, energy-saving and environmentally-friendly route to fabricate active materials of particular catalytic interest, but also provided an unexpected, productive role in the catalysis, that is, the slag-derived impurity elements acted as catalyst promoters. With the strong basicity derived from its large CaO content, the slagLDH and its activated form can nd another application in CCS technology as CO2 adsorbents to reduce the CO2 emission from iron and steel manufacturing plants. (Assessment of the CO2 adsorption properties of this type of material has already been undertaken in other literature [64,65].) Aggressive use of waste slag overproduced in iron and steel industry for such productive purposes facilitates its recycling, which would lead to the reduction of the environmental impact associated with its disposal. To be an advanced and practicable technology suited for efcient carbon capture, storage and utilization (CCSU), the technology needs to be

Y. Kuwahara, H. Yamashita / Journal of CO2 Utilization 1 (2013) 5059 [36] S. Wang, Environ. Sci. Technol. 42 (2008) 70557063. [37] C.-F. Lin, H.-C. Hsi, Environ. Sci. Technol. 29 (1995) 11091117. [38] C. Amrhein, G.H. Haghnia, T.S. Kim, P.A. Mosher, R.C. Gagajena, T. Amanios, L. de la Torre, Environ. Sci. Technol. 30 (1996) 735742. ova , Fuel 78 (1999) 1225 [39] G.G. Hollman, G. Steenbruggen, M. Janssen-Jurkovic 1230. [40] H.-L. Chang, W.-H. Shih, Ind. Eng. Chem. Res. 39 (2000) 41854191. [41] M. Inada, Y. Eguchi, N. Enomoto, J. Hojo, Fuel 84 (2005) 299304. [42] Y. Kuwahara, T. Ohmichi, T. Kamegawa, K. Mori, H. Yamashita, J. Mater. Chem. 20 (2010) 50525062. ` , A. Vaccari, Catal. Today 11 (1991) 173301. [43] F. Cavani, F. Triro [44] A. Guida, M.H. Lhouty, D. Tichit, F. Figueras, P. Geneste, Appl. Catal. A: Gen. 164 (1997) 251264. [45] V. Rives, M.A. Ulibarri, Coord. Chem. Rev. 181 (1999) 61120. [46] K. Kaneda, K. Yamaguchi, K. Mori, T. Mizugaki, K. Ebitani, Catal. Surv. Japan 4 (2000) 3138. [47] M.J. Climent, A. Corma, S. Iborra, K. Epping, A. Velty, J. Catal. 225 (2004) 316326. [48] S. Kannan, Catal. Surv. Asia 10 (2006) 117137. [49] D.P. Debecker, E.M. Gaigneaux, G. Busca, Chem.-Eur. J. 15 (2009) 39203935. [50] T. Sakakura, J.-C. Choi, H. Yasuda, Chem. Rev. 107 (2007) 23652387. [51] K.M.K. Yu, I. Curcic, J. Gabriel, S.C.E. Tsang, ChemSusChem 1 (2008) 893899.

59

[52] E.A. Quadrelli, G. Centi, J.-L. Duplan, S. Perathoner, ChemSusChem 4 (2011) 1194 1215. hler, W. Kuckshinrichs, W. Leitner, P. Markewitz, T.E. Mu ller, [53] M. Peters, B. Ko ChemSusChem 4 (2011) 12161240. ho, Chem. Mater. 15 [54] L. Vieille, I. Rousselot, F. Leroux, J.-P. Besse, C. Taviot-Gue (2003) 43614368. [55] A.V. Radha, P. Vishnu Kamath, C. Shivakumara, Solid State Sci. 7 (2005) 11801187. ho, F. Leroux, P. Le one, P. Palvadeau, J.-P. Besse, J. Solid [56] I. Rousselot, C. Taviot-Gue State Chem. 167 (2002) 137144. pez-Salinas, M.E.L. Serrano, M.A.C. Ja come, I.S. Secora, J. Porous Mater. 2 [57] E. Lo (1995) 291297. [58] A.-A.G. Shaikh, Chem. Rev. 96 (1996) 951976. [59] M. Yoshida, M. Ihara, Chem.-Eur. J. 10 (2004) 28862893. [60] M. North, R. Pasquale, C. Young, Green Chem. 12 (2010) 15141539. [61] K. Mori, Y. Mitani, T. Hara, T. Mizugaki, K. Ebitani, K. Kaneda, Chem. Commun. (2005) 33313333. [62] K. Yamaguchi, K. Ebitani, T. Yoshida, H. Yoshida, K. Kaneda, J. Am. Chem. Soc. 121 (1999) 45264527. [63] M. North, R. Pasquale, Angew. Chem. Int. Ed. 48 (2009) 29462948. [64] S. Wang, S. Yan, X. Ma, J. Gong, Energy Environ. Sci. 4 (2011) 38053819. [65] F.-C. Yu, N. Phalak, Z. Sun, L.-S. Fan, Ind. Eng. Chem. Res. 51 (2012) 21332142.