Titanium Silicate Porous Materials for Carbon Dioxide Adsorption

Ind. Eng. Chem. Res. 2010, 49, 7515–7523

7515

Titanium Silicate Porous Materials for Carbon Dioxide Adsorption: Synthesis Using a Structure Directing Agent, Detemplation and Inclusion of Alkaline Earth Metal Cations Jose´ N. Primera-Pedrozo, Brenda D. Torres-Cosme, Meghan E. Clardy, Milton E. Rivera-Ramos, and Arturo J. Hernańdez-Maldonado* Department of Chemical Engineering, UniVersity of Puerto RicosMayaguëz Campus, Mayaguëz, Puerto Rico 00681-9000

A titanium silicate variant named UPRM-5 was prepared using tetraethylammonium hydroxide as a structuredirecting agent (SDA). Characterization of the material by scanning electron microscopy (SEM) and X-ray diffraction (XRD) techniques showed a homogeneous and crystalline solid powder phase, with unit cell dimensions of a ) 6.8 Å, b ) 11.7 Å, c ) 13.4 Å, R ) 102.9°, β ) 92.8°, and γ ) 91.1°. Successful detemplation was achieved via ion exchange with NH4Cl as evidenced by thermal gravimetric analysis (TGA) and Fourier transform infrared (FT-IR) spectroscopy data. Effective functionalization was obtained after ion exchanging the detemplated material using SrCl2 and BaCl2. These ion exchanged variants were also characterized using XRD and porosimetry techniques. Thermal-vacuum activation of Sr-UPRM-5 at 90 °C resulted in a material with a surface area of ca. 240 m2/g, while activation at higher temperatures resulted in low surface areas and plausible structural distortion. On the other hand, the barium variant exhibited the best thermal stability, with an average surface area on the order of 250 m2/g after employing activation temperatures up to 180 °C. The differences in thermal stability may be a result of structurally coordinated water. Adsorption of CO2 at 25 °C in Sr- and Ba-UPRM-5 materials activated at different temperatures also co-corroborated the aforementioned thermal stability observations in addition to what appears to be cation relocation. Fitting of the CO2 adsorption data with the Dubinin-Astakhov model revealed a heterogeneous surface, which was corroborated by isosteric heats of adsorption estimated from the uptake data. For low partial pressures, the observed CO2 adsorption capacities increased as follows: NH4-UPRM-5 < Sr-UPRM-5 < Ba-UPRM-5. Both the Sr- and Ba-UPRM-5 materials exhibited outstanding selectivity for CO2 over CH4, N2, and O2. Introduction Environmental contamination due to high emission levels of greenhouse gases such as carbon dioxide (CO2) continues to escalate. Sources include industrial fuel processes and fossil fuel burning, which remain the largest contributors despite implementations of gas capture technologies to remove CO2.1-6 The quandary is also reaching levels that are more complex as the world energy crisis has called the attention of many scientists and engineers to develop alternatives to petroleum-based fuels. One of these options includes increasing natural gas production, but the presence of considerable amounts of nitrogen (N2) and CO2 in the natural gas effluents has raised questions regarding energy density when compared to other alternatives. The latter contaminant could be removed via chemical adsorption, pressure swing adsorption, membrane separation, and cryogenic distillation, but these strategies either are energy intensive or involve expensive operational processes.1,7-12 The challenges associated with CO2 removal from gas mixtures are not exclusive to energy-related applications. With the prospective space exploration long-term missions, the need for better CO2 ultrapurification systems is going to be a critical aspect. Although the breathing air that ought to be supplied to the spacecraft cabins will require ultralow CO2 concentrations,13 the most important challenge is to find a process that provides uninterrupted air purification with minimal onboard energy and work impact. One possible solution to all of the aforementioned problems is the use of enhanced adsorption processes, including * To whom correspondence should be addressed. Tel.: 787-832-4040 x3748. Fax: 787-834-3655. E-mail: [email protected].

development of sorbent materials with pore-window tailoring capabilities, with large pore volume, and with outstanding CO2 selectivity under physisorption level interactions. The last criterion is vital for ease of regeneration, especially in spacerelated applications. For many years, zeolite-like materials have been considered for CO2 removal from gas mixtures.2,7-9,13-20 These materials include one synthesized by Kuznicki and co-workers during the 1990s called Engelhard titanium silicate variant no. 4 (ETS-4), which was obtained from hydrothermal synthesis and exhibited a pore size between 3 and 4 Å.21,22 Several studies have found that the structure of ETS-4 is similar to the one of exhibited by the zorite mineral,23 the key difference being the chemical composition.24-27 ETS-4 also contains tetrahedrally coordinated silicon and both square-pyramidal/octahedrally coordinated titanium framework atoms24,28-31 and was the first molecular sieve ever to exhibit such atomic arrangement.32 Depending on the activation (degas) temperature, a structural contraction or shrinkage is obtained. The final pore dimensions depend on the amount of structural water that has been removed from the sorbent material.8,25,27,33-35 It should be mentioned that this material has been used for CO2 adsorption from natural gas mixtures,12 and it has been reported that it has a substantial selectivity of CO2 over CH4, N2, O2, and Ar.8 Although the adsorption and flexible features of ETS-4 are extraordinary, the available pore volume requires additional cycles during pressure swing adsorption applications.8,27,36,37 In an attempt to provide a solution to this problem, here we present a new titanosilicate called University of Puerto Rico at Mayaguëz variant no. 5 (UPRM-5) prepared using tetraethyl-

10.1021/ie100331q  2010 American Chemical Society Published on Web 07/11/2010

7516

Ind. Eng. Chem. Res., Vol. 49, No. 16, 2010

ammonium hydroxide ((TEA)OH) as a molecular structuredirecting agent (SDA). Although the use of (TEA)OH is known to have a considerable effect on the nucleation, crystallization, and void volume of materials,38,39 there are no reports on its use to prepare flexible titanium silicates. In addition, we present a nondestructive detemplation process and two sorbent variants based on strontium (Sr2+) and barium (Ba2+) cations. UPRM-5 and the ion exchanged variants have been characterized here via powder X-ray diffraction (XRD), Fourier transform infrared (FT-IR) spectroscopy, thermal gravimetric analysis (TGA), scanning electron microscopy (SEM), and porosimetry techniques, all aimed at elucidating the structural properties relevant to the material adsorption performance. Adsorption isotherms for CO2, CH4, N2, and O2 as well as CO2 isosteric heat of adsorption on the Sr2+ and Ba2+ variants are also presented. Experimental Section Materials Synthesis. A UPRM-5 crystalline material was prepared via hydrothermal synthesis using (TEA)OH. In a typical synthesis, 1.88 g of sodium hydroxide pellets (Sigma ultra, 98%, Sigma-Aldrich) were added to 10.8 mL of sodium silicate solution (27 wt % SiO2, Sigma-Aldrich), and the resulting mixture was manually agitated until the sodium hydroxide pellets fully dissolved. Afterward, 1.52 g of potassium fluoride dihydrate (98 wt %, Aldrich) was added to the mixture and agitated until a homogeneous solution was attained. About 5.5 mL of titanium(III) chloride solution (20% in HCl, Aldrich) was added to the mixture dropwise. The resulting dark colored gel was stirred continuously using a magnetic stirrer at room temperature for 1.5 h. Finally, 16.3 mL of (TEA)OH (40% (TEA)OH, Fluka) was added and the new mixture stirred for 30 min at room temperature until homogeneous (13.80 < pH < 13.95). The resulting composition formula for the gel was 3.4: 7.3:1.2:1.3:10.0:200.0 (TEA)2O:Na2O:K2O:TiO2:SiO2:H2O. The gel was transferred to a 50 mL Teflon lined stainless-steel autoclave and maintained at 180 °C under autogenous pressure conditions for 12 days. A clear stagnant liquid and a white solid paste were obtained. The solid was filtered and washed with copious amounts of distilled/deionized water and dried in a forced convection oven at 60 °C for 18 h. ETS-4 was prepared following the procedure described elsewhere.22,40 A gel containing sodium silicate, sodium hydroxide, potassium fluoride dihydrate, titanium(III) chloride, and distilled/deionized water was prepared and heated in a Teflon lined autoclave at 150 °C for 7 days to obtain the ETS-4 crystalline phase. The initial gel composition after mixing all the source reactants was 0.9:0.18:1.0:0.18:10-20 Na2O:K2O: SiO2:TiO2:H2O. Recovery of the solids was achieved via filtration followed by washing with copious amounts of distilled/ deionized water. SDA Removal and Ion Exchanges. To remove the SDA, the as-synthesized sorbent material (TEA-UPRM-5) was ion exchanged with a 1 M solution of NH4Cl (100.6%, Fisher) at 80 °C for 24 h.41 The detemplated material (NH4-UPRM-5) was functionalized with strontium and barium cations, respectively, by also using liquid-phase ion exchange methods. For strontium ion exchange, a three-stage method using SrCl2 (ACS reagent, 99% SrCl · 6H2O, Sigma-Aldrich) solution was used.30 For barium ion exchange, a mixture containing 6 mequiv/g BaCl2 (99.999% BaCl2, Aldrich) and 10 g of H2O/(g of detemplated material) was mixed and heated at 80 °C for 24 h.18 Each ion exchanged material (i.e., NH4-, Sr-, or Ba-UPRM-5) was filtered and washed with copious amounts of distilled/deionized water and dried in a forced convection oven at 60 °C for 18 h. Sr-

ETS-4 was prepared by ion-exchanging Na,K-ETS-4 with SrCl2 following the procedure described elsewhere.22,30,33,40 Characterization. X-ray powder diffraction patterns were recorded using a Rigaku Ultima III θ-θ goniometer unit outfitted with a Cu KR target (λ ) 1.5418 Å). The anode was operated at 40 kV and 44 mA. A scanning speed of 1°/min and a step of 0.02° were used to obtain the diffraction patterns for each sample in the 5-50° 2θ range. Both, the as-synthesized and the ion exchanged materials were analyzed using XRD, but only the former was used for indexing. Unit cell indexing was estimated using the CRYSFIRE and the TREOR-90 indexing routine.42,43 Scanning electron micrographs were obtained using a JEOL-JSM-6930LV scanning electron microscope and employing an accelerating voltage of 10 kV. The images were obtained following standard procedures for low conductive samples. Elemental compositional data were obtained using several methods. Evaluation of the as-synthesized material silicon, titanium, sodium, and potassium content was completed using inductively coupled plasma mass spectrometry (ICP-MS). Meanwhile, an induction furnace and Kjeldahl methods were used to estimate the atomic concentrations of carbon and nitrogen, respectively. These tests were performed at Galbraith Laboratories, Inc. in Knoxville, TN. Tests to estimate the strontium and barium content were performed using an energy dispersive X-ray (EDX) spectrometer available on the SEM instrument previously mentioned. To ensure that the data collected were representative of the bulk sample, scans were performed at several different randomly selected spots, and the average was used for unit cell related calculations. The water content of the samples was estimated from thermal gravimetric analysis. Weight percent loss measurements for all samples were performed using a high-resolution TA-Q500 thermogravimetric analyzer with a constant helium flow of 60 mL/min as the carrier gas, while heating from room temperature to 600 °C at a rate of 5 °C/min. All gases were pretreated with presorbers to remove any possible traces of water or other contaminants. For all samples, FT-IR spectra were obtained in the 600-4000 cm-1 range with a resolution in absorbance of 4 cm-1 using a Bruker Optics IFS 66v/S spectrometer coupled to a Hyperion II IR microscope with 159 magnification. An MCT detector and a potassium bromide (KBr) beam splitter were employed during the FT-IR measurements. Porosimetry, Adsorption Measurements, and Isosteric Heat of Adsorption. Surface area analyses were performed for Sr- and Ba-UPRM-5 samples using a Micromeritics ASAP 2020 static volumetric adsorption system equipped with turbomolecular pumps. The instrument pressure measurement unit has a reading accuracy of ca. 0.12%. Back-filling was achieved with helium (high purity grade, Linde) as a back-fill gas. The samples were degassed/activated in vacuum at less than 10 µmHg and a temperature of 90, 120, 150, or 180 °C before each analysis. Heating and vacuum rates used were 5 °C/min and 50 mmHg/s, respectively. Samples were loaded into vials fitted with isolation valves, which allowed transfer from degassing to analysis stages without exposing the adsorbent material to ambient. The materials were also tested for carbon dioxide adsorption at 25 °C and pressures up to 1 atm. The analysis was performed using the volumetric adsorption apparatus previously mentioned and CO2 gas (ultrahigh purity grade, Praxair). Before each analysis, the Sr- and Ba-UPRM-5 sorbents were activated following the procedure described above. NH4-UPRM-5 samples


7517

were activated only at 90 °C. Equilibrium adsorption isotherms of other light gases including methane (ultrahigh purity grade, Praxair), nitrogen (ultrahigh purity grade, Linde), and oxygen (ultrahigh purity grade, Linde) were gathered only for Sr- and Ba-UPRM-5 sorbents at 25 °C. Pure CO2 equilibrium adsorption data for Sr- and Ba-UPRM-5 were fitted with the Langmuir-Freundlich (L-F)44,45 and Dubinin-Astakhov (D-A)45,46 models, repectively. The L-F equation is given by q (bP)1/nLF ) qo 1 + (bP)1/nLF

(1)

while the D-A equation is given by

[ ( ( )) ]

Ps q ) exp - C ln qo P

nDA

(2)

where C is defined by C)

RT βE

and q is the adsorbed amount, qo is the saturated adsorbed amount, b is an interaction parameter, P is partial pressure, Ps is the saturated vapor pressure of the adsorbate, R is the universal ideal gas constant, T is temperature, β is the affinity coefficient, and E is the characteristic adsorption energy. The constants nLF and nDA are related to the material heterogeneity. Carbon dioxide isosteric heats of adsorption (∆Hads) in Srand Ba-UPRM-5 were estimated using equilibrium adsorption data at different temperatures (0, 25, and 50 °C) and evaluated at constant surface loading. The calculations were done using the classic Clausius-Clapeyron equation. Results and Discussion Sample Crystallinity and Morphology. As-synthesized UPRM-5 has a crystalline structure that deviates considerably from that of titanosilicates that are polymorphs of zorite22,27 and even others that are not.47-54 This is evident in the XRD patterns shown in Figure 1, which differs considerably from that of ETS-4 and other titanosilicates previously referenced. A preliminary indexing analysis for UPRM-5 indicates that the unit cell dimensions could be described by a Triclinic system with a ) 6.8 Å, b ) 11.7 Å, c ) 13.4 Å, R ) 102.9°, β ) 92.8°, and γ ) 91.1°. Although the indexing resulted in relatively high figures of merit, further tests will be required to properly refine the structure, and the results will be reported elsewhere. XRD patterns for the detemplated and ion exchanged materials (see Figure 1) suggest that the inclusion of different extraframework species induces some structural changes. Although the crystallinity is generally preserved in both the NH4and Sr-UPRM-5 samples, some small peak displacement was observed in the 25-35° range. Since the peaks are shifting in both directions (i.e., higher and lower d spacing), the results suggest that the inclusion of different cations affect the coordination of titanium and/or structural water species, resulting in some degree of structural distortion. Ba-UPRM-5 samples meanwhile exhibit considerable loss in peak intensity (10-25° range) when compared to that of the as-synthesized material. All of these observations could be the direct result of the ion exchange temperature and/or pH. In addition, they could be attributed to the replacement of the as-synthesized material

Figure 1. X-ray powder diffraction patterns of as-synthesized and ion exchanged materials. Data obtained at room temperature.

cations with NH4+, Sr2+, or Ba2+, something that has been observed before in ETS-4 variants.34,55 Figure 2 shows SEM micrographs for all of the materials. For as-synthesized UPRM-5, the micrographs show a homogeneous crystalline phase composed of particles of flat rectangular shape and with a particle size distribution in the ∼5 to ∼15 µm range. This morphological shape was generally conserved after each ion exchange (i.e., NH4- and Sr-UPRM5), with only minor changes in the particle size probably due to pitting from mechanical stress induced by the ion exchange methods employed here. For the case of Ba-UPRM-5 some morphological changes were observed, and that may be related to the XRD pattern observations previously discussed. TGA, FT-IR, and Framework Composition. Weight loss profiles (Figure 3) were gathered for all of the UPRM-5 variants. In the case of the as-synthesized material, the profile shows an inflection between 25 and 160 °C which corresponds to water weakly bound to the structure. A secondary inflection present in the 175-525 °C range was been assigned to the SDA thermal elimination.56 Although this region is absent in the other variants, which is a strong indication of successful detemplation, the observed difference in total weight loss among the detemplated UPRM-5 variants is probably due to the amount of water molecules coordinated to the framework and to extraframework species, in addition to ammonia in the case of NH4-UPRM-5. It has been reported that coordination water plays a critical role in the thermal stability of titanosilicates.27 Further corroboration of successful detemplation was achieved via FT-IR spectroscopy analyses. Quaternary ammonium salts such as (TEA)OH cations are characterized by NC4 skeletal stretching, and this observation is present in the IR spectrum of the as-synthesized UPRM-5 (see Figure 4). The bands observed in the 1030-1060 and 666-672 cm-1 regions correspond to antisymmetric and symmetric NC4 stretching, respectively, confirming the presence of the SDA moiety. Additional bands corresponding to the template are those observed at 914 and 1000 cm-1 for C-N stretching and 1132 cm-1 for C-N-C antisymmetric stretching. Meanwhile, the IR spectra of NH4-, Sr-, and Ba-UPRM-5 samples still show bands in the 650-1400 cm-1 region but with much less

7518


Figure 4. FT-IR spectra for as-synthesized and ion exchanged UPRM-5 materials. Table 1. Unit Cell Composition of As-Synthesized UPRM-5 and Ion Exchanged Materials sorbent TEA-UPRM-5 NH4-UPRM-5 Sr-UPRM-5 Ba-UPRM-5

Figure 2. SEM micrographs for as-synthesized and ion exchanged UPRM-5 materials.

Figure 3. Thermal gravimetric analysis profiles for as-synthesized and ion exchanged UPRM-5 materials.

intensity. These are due to symmetric and asymmetric stretching of the TO4 tetrahedra57 and not due to the SDA molecules, which confirms detemplation. The bands observed in the 2800-3500 and 1400-1440 cm-1 regions of the NH4-UPRM-5 IR spectrum are due to the stretching and deformation vibrations of the NH4+ cations, respectively. These regions disappear when the NH4-UPRM-5 material is ion exchanged with strontium or barium cations. It should be mentioned that the bands present

unit cell +

+

|TEA0.5 K Na5+|[Si9Ti4O37]:9H2O |(NH4+)2.4K0.2+Na3.9+|[Si9Ti4O37] |Sr2.52+K0.1+Na1.4+|[Si9Ti4O37] |Ba1.82+K0.1+Na2.9+|[Si9Ti4O37]

at 2340 and 1640 cm-1 are related to CO2 and water, respectively, present in the materials.58 Elemental compositional data of the samples were obtained from a combination of experimental methods, including TGA, ICP, and EDX, and the results are compiled in Table 1. The as-synthesized material has about 0.5 SDA molecules/(unit cell) (see Table 1), which are protonated and offer counterbalancing of negative charges brought by the framework. Nevertheless, the titanium-silicate framework still requires additional counterbalancing charges which are supplied here by sodium and potassium cations. According to the data shown in Table 1, the as-synthesized UPRM-5 Si/Ti ratio is close to the value reported for ETS-4.26,30 In fact, the unit cell also suggests the presence of titanium centers of nontetrahedral coordination, which is also typical of ETS-4, but further studies are required to elucidate the actual coordination states and the presence of other extraframework species such as hydroxyl groups. Table 1 also shows the compositional data for the detemplated and strontium/ barium exchanged materials. Since ion exchanges are not expected to change the framework metal content, it was assumed that the Si and Ti amounts remained constant. In general, the ion exchange procedures resulted in complete removal of the SDA and partial exchange of the alkali metal cations present in the as-synthesized UPRM-5. Effect of Activation Temperature on Surface Area. One particular feature of ETS-4 materials is the structure shrinkage upon increasing the activation or degassing temperature.33 For instance, the reported surface areas for the as-synthesized (hydrated) and thermally activated (at 200 °C) ETS-4 materials are 19 and 12 m2/g, respectively.21 A similar but more prominent feature was observed in Sr-UPRM-5 materials, where the surface area decreases from 244 to 11 m2/g when the activation temperature is increased to about 180 °C (see Figure 5). However, it is important to notice that the surface areas reported here were obtained from N2 adsorption isotherms gathered at -196 °C tests and hence the values correspond to what such


Figure 5. Surface area of Sr- and Ba-UPRM-5 materials activated at different temperatures. Values estimated from nitrogen adsorption data gathered at -196 °C.

probe molecule is able to access. In other words, if the pore dimensions are close or less than the probe molecule kinetic diameter, the reported area will be perhaps smaller than the actual value. For the Ba-UPRM-5 variant, however, it appears that the inclusion of barium cations enhances the stability of the material in a considerable fashion. In fact, the surface area increases an average of 7% upon increasing the activation

7519

temperature. This behavior is probably a result of stronger water coordination59 as evident from the TGA tests (see Figure 3). Increasing the temperature to ca. 240 °C results in a Ba-UPRM-5 surface area of ca. 45 m2/g (result not shown in Figure 5), which can be attributed to nearly complete framework shrinkage. It should be mentioned that the as-synthesized UPRM-5 sorbent has a surface area of ca. 61 m2/g after activation at 90 °C, which further supports that the SDA is present on the structure and that it occupies most if not all of the pore volume. Adsorption Isotherms and Isosteric Heat of Adsorption. Pure component CO2 adsorption isotherms were measured on Sr- and Ba-UPRM-5 at 25 °C after activation of the sorbents at different temperatures, and the data are summarized in Figure 6. Careful attention was given to data gathering at low partial pressures in order to analyze the sorbate-sorbent interaction levels in a qualitative way. Given the rectangular shape of the isotherms, which is an excellent indicator of strong interactions, it was necessary to plot the data also in log-log scale in order to appreciate the differences in loading amounts among the metal exchanged UPRM-5 variants. For the case of Sr-UPRM-5, the low-pressure region does not show appreciable difference in loading amounts, with the exception of data for the material activated at 180 °C. It appears that at this activation temperature the pore width of the strontium-based material has been reduced dramatically, blocking or offering considerable resistance to diffusion of CO2. A similar behavior was expected after employing an activation temperature of 150 °C since the material

Figure 6. Adsorption isotherms for CO2 on Sr- and Ba-UPRM-5 at 25 °C. Adsorbents activated at different temperatures.

7520


Table 2. Isotherm Models Parameters for CO2 Adsorption on Ion Exchanged UPRM-5 Materials Langmuir-Freundlich sorbent

activation temp (°C)

adsorption temp (°C)

Sr-UPRM-5

90

Ba-UPRM-5

180

0 25 0 25

Dubinin-Astakhov

qo (mmol g-1)

nLF

b (atm-1)

std. dev.*

qo (mmol g-1)

nDA

C

std. dev.*

2.96

1.96 1.66 4.17 3.29

19.32 6.42 9.13 9.68

(0.019 (0.015 (0.016 (0.018

2.76

2.59 2.80 1.40 1.93

0.128 0.140 0.108 0.109

(0.015 (0.026 (0.009 (0.018

2.85

2.35

* Standard deviation calculated on the basis of residuals between the observed and calculated equilibrium loading amounts for the complete pressure range.

surface area as reported by N2 probe molecules is ca. 12 m2/g. One possible explanation for the observed higher CO2 loading at this condition could be the result of the smaller kinetic diameter of this sorbate when compared to that of N2. Analysis of the higher pressure range clearly shows that there is a pore volume contraction in Sr-UPRM-5 when activated at different temperatures, which is typical in many titanosilicates with multicoordinated framework titanium.18,33 It is concluded after the analysis that the best activation temperature for SrUPRM-5 in order to achieve the highest CO2 adsorption capacities is 90 °C. For selectivity considerations, it is required to study the adsorption of other light gases before concluding which activation temperature is best suited, and this will be discussed shortly. CO2 adsorption analysis at low pressure for the barium-based materials activated at different temperatures indicates that the ultimate sorbate-sorbent interactions are affected by the sorbent pretreatment temperature choice. As shown in Figure 6, an increase in activation temperature results in higher loadings in the prescribed pressure region. One possible explanation for this behavior is the presence of framework-coordinated water after activation of the sorbent at lower temperatures. TGA data (see Figure 3) for Ba-UPRM-5 indicates that the moisture content of the material is released in a stepwise fashion, which indeed is evidence for the presence of water with different coordination stages. This explains why an activation temperature of 180 °C results in higher CO2 loadings at low pressure and 25 °C. It is plausible that a considerable amount of water molecules was coordinated to Ba2+ cations, which required higher activation temperatures in order to allow the metal to be exposed to interact with other hosts. In addition, the ultimate cation location is known to be a function of activation temperature and hence will affect the degree of exposure of a particular adsorption site.34 Adsorption of CO2 at higher pressure shows almost no difference in equilibrium loading for each sorbent activation temperature, this being evidence of a remarkable thermal stability brought by the presence of barium ions. It is concluded after the analysis that the best activation temperature for BaUPRM-5 in order to achieve the highest CO2 adsorption capacities should be approximately 180 °C. In an attempt to understand the adsorption mechanism and perhaps the sorbent structural characteristics, the Sr- and BaUPRM-5 CO2 adsorption equilibrium data were fitted with both the L-F and D-A models. These adsorbents were preactivated in vacuum at 90 and 180 °C, respectively. A simple Langmuir model fitting with no heterogeneity parameter (i.e., nLF ) 1) was also tested, but the fitting was not adequate for the prescribed pressure range (results not shown here). The L-F and D-A isotherms model data are summarized in Table 2. Only the 0 °C (data not shown here) and 25 °C adsorption isotherms were employed for these calculations. The saturation loadings were estimated at 0 °C and assumed constant for the rest of the calculations. This is a reasonable approach since sorbate-sorbate interactions become negligible at lower temperatures and loading

saturation amounts could be achieved at pressures near the 1 atm range. Although Table 2 shows that the L-F fit resulted in small standard deviations, one must proceed with caution when using such a model because it does not predict adsorption at low partial pressures properly. Furthermore, given the microporous nature of the adsorbents the surface should interact with the volume of adsorbates rather than a monolayer, something the L-F model does not take into account. On the other hand, the D-A model does take all these arguments into consideration45 and, therefore, will be the focus of the discussion presented next. Table 2 shows that the D-A model heterogeneity factor varies from 1.4 to 2.8, depending on the sorbent variant. Values smaller than 3 usually correlate with heterogeneous systems45 and for the system under consideration could correlate to the presence of more than one unique cation site along the unit cell. Surprisingly, the estimated saturation loading amounts for each sorbent variant are not that different from one to another, which indicates that perhaps the completely dehydrated structure pore volume is similar in both cases. Meanwhile, the C constant, which is inversely proportional to the characteristic energy or interaction between the sorbate and the sorbent was found to be smaller for the Ba-UPRM-5 case. In other words, the sorbate-sorbent interactions are stronger when barium is present in UPRM-5. Figure 7 compares CO2 equilibrium isotherms for all the detemplated variants thermally treated at their corresponding optimum activation temperature. In addition, the figure includes data for CO2 adsorption on Sr-ETS-4 material activated at 90 °C. Comparison of the adsorption data between Sr-UPRM-5 and Sr-ETS-4 is suitable since the materials densities are similar (2.27 vs 2.20 g/cm,3 respectively).60,61 It should be mentioned that although the ETS-4 sorbent activation temperature was chosen to establish a direct comparison, one should keep in mind that the amount of structural or tenacious water in Sr-ETS-4 at this condition is considerable and could affect the adsorption behavior. In fact, an analysis by Marathe et al. concluded that the diffusional hindrance observed in Sr-ETS-4 decreases with activation temperature.62 Notwithstanding, data recently reported by Rodrigues and co-workers for CO2 equilibrium isotherms on Sr-ETS-4 at pressures up to ca. 6 atm and gathered using different activation temperatures indicate that the loading amounts at low partial pressure were almost invariant to the choice of activation temperature.63 According to the data shown in Figure 7, the CO2 interaction at low partial pressures increases as follows: NH4-UPRM-5 < Sr-ETS-4 < Sr-UPRM-5 < Ba-UPRM-5. Although the bariumbased UPRM-5 sorbent offers remarkable sorption capacity at low pressures, the strontium counterpart excels at saturation capacity, plausibly due to water molecules displacement by the CO2. According to Figure 3 the amount of coordination water available in Sr-UPRM-5 is greater than that of the barium-based counterpart, especially at temperatures equal to or greater than 90 °C, which probably explains the observed CO2 isotherm


7521

Figure 7. Adsorption isotherms for CO2 on NH4-UPRM-5, Sr-UPRM-5, Ba-UPRM-5, and Sr-ETS-4 at 25 °C. All adsorbents activated at 90 °C except Ba-UPRM-5, which was activated at 180 °C.

Figure 8. Isosteric heats of adsorption of CO2 in Sr- and Ba-UPRM-5. Adsorbents activated at 90 and 180 °C, respectively.

behavior. Comparison of the CO2 loadings among the strontiumbased sorbents indicates that the use of (TEA)OH during the synthesis of the titanium silicate indeed has an effect on the ultimate framework characteristics, including the presence of a larger void volume. However, further studies would be required to determine the UPRM-5 framework atomic positions and their correlation to any templating effect brought by the SDA employed here. CO2 isosteric heats of adsorption profiles for Sr- and BaUPRM-5 activated at 90 and 180 °C, respectively, are shown in Figure 8. These were used to estimate the interaction range as well as the degree of heterogeneity. Both sorbents show physisorption level interactions with CO2 with maximum values of 43 and 52 kJ/mol for Sr-UPRM-5 and Ba-UPRM-5, respectively, and a high degree of heterogeneity. Using the data shown in Table 1, the observed peak energy values correspond to about 0.4 and 0.7 CO2 molecules adsorbed per strontium and barium cation, respectively. As the sorbate loading is increased, interactions probably occur with the remaining cationic species, including sodium and potassium. For the Sr-UPRM-5-case, this evidences that not all of the strontium ions are available for interaction, probably due to the presence of metal coordinated water molecules. It should be mentioned that similar CO2 adsorption energy trends have been observed, reported by Hernańdez-Maldonado and co-workers for strontium- and

barium-based SAPO-34 sorbents.9,13 They also concluded from DFT studies that such values correspond to a bond of ionic character. Single-component equilibrium adsorption isotherms for CH4, N2, and O2 were also measured at 25 °C for the Sr- and BaUPRM-5 sorbents activated at 90 and 180 °C, respectively. Figure 9 shows that both materials were quite selective toward CO2, which could be due to a combination of size exclusion and surface interaction principles. All four gases have electrostaticrelated properties, which will lead to similar nonspecific level interactions.9,64 Carbon dioxide, however, has a strong quadrupole moment that could interact with the cation site electric field and enhance the interactions.8,14 The quadrupole moment value for CO2 is -4.3 × 10-26 erg1/2 cm5/2,39 which is three times larger than that of N2 and an order of magnitude larger than that of O2. Methane does not have a quadrupole moment. Also, the observed CH4 adsorbed amounts are larger compared to those obtained for N2 and O2 and could be attributed to the induced octopole moment that CH4 exhibits when in contact with the electric field generated by the cations8,65-67 and its slightly large polarizability.64,68 In terms of adsorbent regeneration, the CO2 uptake in both the strontium and barium variants is completely reversible upon application of vacuum (data not shown here). This correlates with the isosteric heats of adsorption, which are in the physisorption range. At this stage, it appears that these materials could be suitable for CO2 removal applications without being energy intensive. However, further studies would be required to estimate multicycle adsorption capacities and elucidate the long-term reusability of these materials. Conclusions (TEA)OH was used to prepare a novel porous titanium silicate material, UPRM-5, with porosimetry characteristics that differ considerably from those of ETS-4. Due to the as-synthesized UPRM-5 framework limited thermal stability, detemplation was achieved via ion exchange with NH4Cl. Further ion exchange with either strontium or barium cations greatly affects the material thermal stability, with the latter case resulting in the most stable variant. For the strontium-exchanged material, a porosimetry analysis indicated that the material surface area diminishes considerably after vacuum treatment at 180 °C. These

7522


Figure 9. Pure component adsorption isotherms for CO2, CH4, N2, and O2 on Sr- and Ba-UPRM-5. Adsorbents activated at 90 and 180 °C, respectively.

stability observations are plausibly a result of elimination of framework-coordinated water, which depends strongly on activation temperature as indicated by the TGA profiles. Both the Sr- and Ba-UPRM-5 materials exhibit considerable CO2 adsorption capacities at low pressures due to considerable sorbate-sorbent interactions. Analysis of the equilibrium isotherms and isosteric heats of adsorption indicated that the surface of these materials is rather heterogeneous. This could be correlated to the presence of more than one type of cation location. In addition, the estimated CO2 isosteric heats of adsorption suggest that the sorbate-sorbent interaction is of a physisorption type, which is adequate for the materials regeneration. Acknowledgment This material is based upon work supported by the National Science Foundation (NSF) under Grant No. HRD 0833112 (CREST Program). Partial support was also provided by the National Aeronautics and Space Administration under Grant No. NNX09AV05A. We would also like to acknowledge the support provided by the NSF Research Experiences for Undergraduates under Grant No. DMR-0552673. Literature Cited (1) Yong, Z.; Mata, V.; Rodrigues, A. E. Adsorption of Carbon Dioxide at High Temperature: A Review. Sep. Purif. Technol. 2002, 26, 195.

(2) Ko, D.; Siriwardane, R.; Biegler, L. T. Optimization of Pressure Swing Adsorption and Fractionated Vacuum Pressure Swing Adsorption Processes for CO2 Capture. Ind. Eng. Chem. Res. 2005, 44, 8084. (3) Herzog, H. J. Peer Reviewed: What Future for Carbon Capture and Sequestration. EnViron. Sci. Technol. 2001, 35, 148A. (4) Idem, R.; Tontiwachwuthikul, P. Preface for the Special Issue on the Capture of Carbon Dioxide from Industrial Sources: Technological Developments and Future Opportunities. Ind. Eng. Chem. Res. 2006, 45, 2413. (5) Coti, J.; Sweetnam, G. E. Emissions of Greenhouse Gases in the United States 2008-DOE/EIA-05.; Energy Information Administration, Office Integrated Analysis and Forecasting, U.S. Department of Energy: Washington, D.C., 2009. (6) Keith, D. W. Why Capture CO2 from the Atmosphere? Science 2009, 325, 1654. (7) Stevens, R. W.; Siriwardane, R. V.; Logan, J. In Situ Fourier Transform Infrared (FTIR) Investigation of CO2 Adsorption onto Zeolite Materials. Energy Fuel 2008, 22, 3070. (8) Pillai, R. S.; Peter, S. A.; Jasra, R. V. Adsorption of Carbon Dioxide, Methane, Nitrogen, Oxygen and Argon in NaETS-4. Microporous Mesoporous Mater. 2008, 113, 268. (9) Rivera-Ramos, M. E.; Ruiz-Mercado, G. J.; Hernandez-Maldonado, A. J. Separation of CO2 from Light Gas Mixtures Using Ion-Exchanged Silicoaluminophosphate Nanoporous Sorbents. Ind. Eng. Chem. Res. 2008, 47, 5602. (10) Abanades, J. C.; Rubin, E. S.; Anthony, E. J. Sorbent Cost and Performance in CO2 Capture Systems. Ind. Eng. Chem. Res. 2004, 43, 3462. (11) Baker, R. W. Future Directions of Membrane Gas Separation Technology. Ind. Eng. Chem. Res. 2002, 41, 1393. (12) Mitariten, M. New Technology Improves Nitrogen-Removal Economics. Oil Gas J. 2001, 99, 42. (13) Are´valo-Hidalgo, A. G.; Santana, J. A.; Fu, R.; Ishikawa, Y.; Hernańdez-Maldonado, A. J. Separation of CO2 from Light Gas Mixtures Using Nanoporous Silicoaluminophosphate Sorbents: Effect of MultipleStep Ion Exchange and Adsorption Mechanism via Computational Studies. Microporous Mesoporous Mater. 2010, 130, 142. (14) Aguilar-Armenta, G.; Hernandez-Ramirez, G.; Flores-Loyola, E.; Ugarte-Castaneda, A.; Silva-Gonzalez, R.; Tabares-Munoz, C.; JimenezLopez, A.; Rodriguez-Castellon, E. Adsorption Kinetics of CO2, O2, N2, and CH4 in Cation-Exchanged Clinoptilolite. J. Phys. Chem. B 2001, 105, 1313. (15) Liu, X.; Li, J.; Zhou, L.; Huang, D.; Zhou, Y. Adsorption of CO2, CH4 and N2 on Ordered Mesoporous Silica molecular Sieve. Chem. Phys. Lett. 2005, 415, 198. (16) Liu, X.; Zhou, L.; Fu, X.; Sun, Y.; Su, W.; Zhou, Y. Adsorption and Regeneration Study of the Mesoporous Adsorbent SBA-15 Adapted to the Capture/Separation of CO2 and CH4. Chem. Eng. Sci. 2007, 62, 1101. (17) Ribeiro, R. P.; Sauer, T. P.; Lopes, F. V.; Moreira, R. F.; Grande, C. A.; Rodrigues, A. E. Adsorption of CO2, CH4, and N2 in Activated Carbon Honeycomb Monolith. J. Chem. Eng. Data 2008, 53, 2311. (18) Anson, A.; Lin, C. C. H.; Kuznicki, S. M.; Sawada, J. A. Adsorption of Carbon Dioxide, Ethane, and Methane on Titanosilicate Type Molecular Sieves. Chem. Eng. Sci. 2009, 64, 3683. (19) Shen, S. C.; Chen, X. Y.; Kawi, S. CO2 Adsorption over Si-MCM41 Materials Having Basic Sites Created by Postmodification with La2O3. Langmuir 2004, 20, 9130. (20) Ghoufi, A.; Gaberova, L.; Rouquerol, J.; Vincent, D.; Llewellyn, P. L.; Maurin, G. Adsorption of CO2, CH4 and Their Binary Mixture in Faujasite NaY: A Combination of Molecular Simulations with GravimetryManometry and Microcalorimetry Measurements. Microporous Mesoporous Mater. 2009, 119, 117. (21) Armaroli, T.; Busca, G.; Milella, F.; Bregani, F.; Toledo, G. P.; Nastro, A.; De Luca, P.; Bagnasco, G.; Turco, M. A Study of ETS-4 Molecular Sieves and of Their Adsorption of Water and Ammonia. J. Mater. Chem. 2000, 10, 1699. (22) Kuznicki, S. M. Preparation of Small-Pored Cristalline Titanium Molecular Sieve Zeolites. U.S. Patent 4,938,939, 1990. (23) Sandomirsky, P. A.; Belov, N. V. The OD Structure of Zorite. Kristallografiya 1979, 24, 1198. (24) Du, H.; Zhou, F.; Pang, W.; Yue, Y. Synthesis and Charaterization of Titanium Silicate Molecular Sieves with Zorite-Type Structure. Microporous Mater. 1996, 7, 73. (25) Valtchev, V.; Mintova, S.; Mihailova, B.; Konstantinov, L. Comparison of Physicochemical Properties of Zorite and ETS-4. Mater. Res. Bull. 1996, 31, 163. (26) Cruciani, G.; De Luca, P.; Nastro, A.; Pattison, P. Rietveld Refinement of the Zorite Structure of ETS-4 Molecular Sieves. Microporous Mesoporous Mater. 1998, 21, 143.

Ind. Eng. Chem. Res., Vol. 49, No. 16, 2010 (27) Philippou, A.; Anderson, M. W. Structural Investigation of ETS4. Zeolites 1996, 16, 98. (28) Nair, S.; Tsapatsis, M.; Toby, B. H.; Kuznicki, S. M. A Study of Heat-Treatment Induced Framework Contraction in Strontium-ETS-4 by Powder Neutron Diffraction and Vibrational Spectroscopy. J. Am. Chem. Soc. 2001, 123, 12781. (29) Mihailova, B.; Valtchev, V.; Mintova, S.; Konstantinov, L. Vibrational Spectra of ETS-4 and ETS-10. Zeolites 1996, 16, 22. (30) Braunbarth, C.; Hillhouse, H. W.; Tsapatsis, M.; Burton, A.; Lobo, R. F.; Jacubinas, R. M.; Kuznicki, S. M. Structure of Strontium IonExchanged ETS-4 Microporous Molecular Sieves. Chem. Mater. 2000, 12, 1857. (31) Nair, S.; Jeong, H. K.; Chandrasekaran, A.; Braunbarth, C. M.; Tsapatsis, M.; Kuznicki, S. M. Synthesis and Structure Determination of ETS-4 Single Crystals. Chem. Mater. 2001, 13, 4247. (32) Kuznicki, S. M.; Jacubinas, R. M.; Langner, T. W. PolymorphEnriched ETS-4. U.S. Patent 6,464,957 B1, 2002. (33) Kuznicki, S. M.; Bell, V. A.; Nair, S.; Hillhouse, H. W.; Jacubinas, R. M.; Braunbarth, C. M.; Toby, B. H.; Tsapatsis, M. A Titanosilicate Molecular Sieve with Adjustable Pores for Size-Selective Adsorption of Molecules. Nature 2001, 412, 720. (34) Marathe, R. P.; Mantri, K.; Srinivasan, M. P.; Farooq, S. Effect of Ion Exchange and Dehydration Temperature on the Adsorption and Diffusion of Gases in ETS-4. Ind. Eng. Chem. Res. 2004, 43, 5281. (35) Naderi, M.; Anderson, M. W. Phase Transformation of Microporous Titanosilicate ETS-4 into Narsarsukite. Zeolites 1996, 17, 437. (36) Jayaraman, A.; Hernandez-Maldonado, A. J.; Yang, R. T.; Chinn, D.; Munson, C. L.; Mohr, D. H. Clinoptilolites for Nitrogen/Methane Separation. Chem. Eng. Sci. 2004, 59, 2407. (37) Kuznicki, S. M.; Bell, V. A.; Jacubinas, R. M.; Nair, S.; Braunbarth, C. M.; Hillhouse, H. W.; Tsapatsis, M. ETS-4 Pore Contraction: The Molecular Gate Effect. Abstracts of Papers of the American Chemical Society 2001, 221, U726. (38) Tuel, A.; Taarit, Y. B. Synthesis of TS-1 from Titanosilicate Gels Containing TPAOH/TEAOH and TPAOH/NH4OH Mixtures. Microporous Mater. 1993, 1, 179. (39) Yang, R. T., Adsorbents: Fundamentals and Applications; Wiley: New York, 2003. (40) Kuznicki, S. M.; Bell, V. A.; Petrovic, I.; Desai, B. T. Small-Pored Crystalline Titanium Molecular Sieve Zeolites and Their Use in Gas Separation Processes. US Patent 6,068,682, 2000. (41) Takewaki, T.; Beck, L. W.; Davis, M. E. Zincosilicate CIT-6: A Precursor to a Family of *BEA-Type Molecular Sieves. J. Phys. Chem. B 1999, 103, 2674. (42) Shirley, R. The CRYSFIRE System for Automatic Powder Indexing: User’s Manual; The Lattice Press: Guildford, Surrey, England., 1999. (43) Werner, P.-E.; Eriksson, L.; Westdahl, M. A Semi-Exhaustive Trialand-Error Powder Indexing Program for All Symmetries. J. Appl. Crystallogr. 1985, 18, 367. (44) Sips, R. On the Structure of a Catalyst Surface. J. Chem. Phys. 1948, 16, 490. (45) Do, D. D. Adsorption Analysis: Equilibria and Kinetics; Imperial College Press: London, 1998. (46) Rege, S. U.; Yang, R. T.; Buzanowski, M. A. Sorbents for Air Prepurification in Air Separation. Chem. Eng. Sci. 2000, 55, 4827. (47) Brandaõ, P.; Valente, A.; Philippou, A.; Ferreira, A.; Anderson, M. W.; Rocha, J. A Novel Large-Pore Framework Titanium Silicate Catalyst. J. Mater. Chem. 2002, 12, 3819. (48) Nandi, M.; Bhaumik, A. Highly Active Ti-Rich Ordered Mesoporous Titanium Silicate Synthesized Under Strong Acidic Condition. Chem. Eng. Sci. 2006, 61, 4373. (49) Zhao, Q.; Li, P.; Li, D. Q.; Zhou, X. G.; Yuan, W. K.; Hu, X. J. Synthesis and Characterization of Titanium Silicate-1 Supported on Carbon Nanofiber. Microporous Mesoporous Mater. 2008, 108, 311.

7523

(50) Salehirad, F.; Aghabozorg, H. R.; Manoochehri, M.; Aghabozorg, H. Synthesis of Titanium Silicalite-2 (TS-2) from Methylamine-Tetrabutylammonium Hydroxide Media. Catal. Commun. 2004, 5, 359. (51) Liu, H.; Lu, G.; Hu, H. Synthesis, Characterization and Catalytic Performance of Titanium Silicalite-1 Prepared in the Presence of Nonionic Surfactants. Mater. Chem. Phys. 2006, 100, 162. (52) Kuznicki, S. M. Large-Pored Crystalline Titanium Molecular Sieve Zeolites. U.S. Patent 4,853,202, 1989. (53) Men’shikov, Y. P.; Krivovichev, S. V.; Pakhomovsky, Y. A.; Yakovenchuk, V. N.; Ivanyuk, G. Y.; Mikhailova, J. A.; Armbruster, T.; Selivanova, E. A. Chivruaiite, Ca4(Ti, Nb)5 [(Si6O17)2 [(OH, O)5] · 1314H2O, a New Mineral from Hydrothermal Veins of Khibiny and Lovozero Alkaline Massifs. Am. Mineral. 2006, 91, 922. (54) Lin, C. C. H.; Sawada, J. A.; Wu, L.; Haastrup, T.; Kuznicki, S. M. Anion-Controlled Pore Size of Titanium Silicate Molecular Sieves. J. Am. Chem. Soc. 2008, 131, 609. (55) Kuznicki, S. M.; Bell, V. A.; Petrovic, I.; Blosser, P. W. Separation of Nitrogen from Mixtures thereof with Methane Utilizing Barium Exchanged ETS-4. U.S. Patent 5,989,316, 1999. (56) Kasunic, M.; Legisa, J.; Meden, A.; Logar, N. Z.; Beale, A. M.; Golobic, A. Crystal Structure of Pure-Silica ZSM-12 with Tetraethylammonium Cations from X-Ray Powder Diffraction Data. Microporous Mesoporous Mater. 2009, 122, 255. (57) Akolekar, D. B.; Bhargava, S.; van Bronswijk, W. Fourier Transform Raman Spectroscopy of Novel Aluminophosphate Molecular Sieves. Appl. Spectrosc. 1999, 53, 931. (58) Lin-Vien, D.; Colthup, N. B.; Fateley, W. G.; Grasselli, J. G. The Handbook of Infrared and Raman Characteristic Frequencies of Organic Molecules; Academic Press: San Diego, 1991. (59) Turta, N. A.; Veltri, M.; Vouno, D.; De Luca, P.; Bilba, N.; Nastro, A. Effect of Crystallization Temperature on the Synthesis of ETS-4 and ETS-10 Titanosilicates. J. Porous Mater. 2008, 16, 527. (60) Ferreira, T. R.; Lopes, C. B.; Lito, P. F.; Otero, M.; Lin, Z.; Rocha, J.; Pereira, E.; Silva, C. M.; Duarte, A. Cadmium(II) Removal from Aqueous Solution Using Microporous Titanosilicate ETS-4. Chem. Eng. J. 2009, 147, 173. (61) Lopes, C. B.; Lito, P. F.; Otero, M.; Lin, Z.; Rocha, J.; Silva, C. M.; Pereira, E.; Duarte, A. C. Mercury Removal with Titanosilicate ETS-4: Batch Experiments and Modelling. Microporous Mesoporous Mater. 2008, 115, 98. (62) Marathe, R. P.; Farooq, S.; Srinivasan, M. P. Effects of Site Occupancy, Cation Relocation, and Pore Geometry on Adsorption Kinetics in ETS-4. J. Phys. Chem. B 2005, 109, 3257. (63) Cavenati, S.; Grande, C. A.; Lopes, F. V. S.; Rodrigues, A. E. Adsorption of Small Molecules on Alkali-Earth Modified Titanosilicates. Microporous Mesoporous Mater. 2009, 121, 114. (64) Rivera-Ramos, M. E.; Hernandez-Maldonado, A. J. Adsorption of N2 and CH4 by Ion-Exchanged Silicoaluminophosphate Nanoporous Sorbents: Interaction with Monovalent, Divalent, and Trivalent Cations. Ind. Eng. Chem. Res. 2007, 46, 4991. (65) Snook, I. K.; Spurling, T. H. Intermolecular Interactions in Methane, Carbon Tetrafluoride and Sulphur Hexafluoride. J. Chem. Soc., Faraday Trans. 2 1972, 68, 1359. (66) Sinai, J. J. Octopole Moment of Methane. J. Chem. Phys. 1964, 40, 3596. (67) Allen, M. P. A Molecular Dynamics Simulation Study of Octopoles in the Field of a Planar Surface. Mol. Phys. 1984, 52, 717. (68) Marathe, R. P.; Farooq, S.; Srinivasan, M. P. Modeling Gas Adsorption and Transport in Small-Pore Titanium Silicates. Langmuir 2005, 21, 4532.

ReceiVed for reView February 11, 2010 ReVised manuscript receiVed May 25, 2010 Accepted June 21, 2010 IE100331Q

Titanium Silicate Porous Materials for Carbon Dioxide Adsorption

Recommend Documents