Simulation of the Adsorption and Transport of CO2 ... - ACS Publications

Subscriber access provided by NEW MEXICO STATE UNIV

Article 2

Simulation of the Adsorption and Transport of CO on Faujasite Surfaces Jennifer Clare Crabtree, Marco Molinari, Stephen C. Parker, and John Ashley Purton J. Phys. Chem. C, Just Accepted Manuscript • Publication Date (Web): 01 Oct 2013 Downloaded from http://pubs.acs.org on October 2, 2013

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

The Journal of Physical Chemistry C is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 36

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Simulation of the Adsorption and Transport of CO2 on Faujasite Surfaces Jennifer C. Crabtree†‡, Marco Molinari‡, Stephen C. Parker‡*, John A. Purton§. †

Centre for Sustainable Chemical Technologies, University of Bath, Claverton Down, Bath,

BA2 7AY, UK. ‡Department of Chemistry, University of Bath, Claverton Down, Bath, BA2 7AY, UK. §Scientific Computing Department, STFC, Daresbury Laboratory, Keckwick Lane, Warrington, WA4 4AD, UK.

1 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 36

ABSTRACT

We have investigated the effect of surfaces on the adsorption and transport of CO2 with faujasite (FAU) using molecular dynamics. We modeled the {111}, {011} and {100} surfaces of FAU. The {011} and {100} surfaces have incomplete sodalite cages, which adsorb CO2 more favorably than the most stable {111} surface where the sodalite cages are intact. The surfaces of siliceous, sodium and potassium FAU were modeled to compare the effect of zeolite composition. The results show that CO2 diffusion through the surface is intermediate between diffusion in the zeolite and in bulk CO2 above the surface. In siliceous FAU the diffusion of bulk CO2 is reduced by 42 % in the surface region and 61 % in the zeolite. CO2 diffusion is reduced by up to 83 % inside aluminosilicate zeolites compared to siliceous. However, the surface adsorption of CO2 is more affected by the surface structure than the composition and at the surface there are dense layers of adsorbed CO2 indicating sites of enhanced adsorption with reduced diffusion across them, particularly associated with the incomplete sodalite cages. Thus we suggest that spherical particles with these surface sites are likely to be more effective sorbents than {111} facetted particles.

KEYWORDS: Zeolite, diffusion, MD, interface


Page 3 of 36

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1 – Introduction Carbon dioxide is a greenhouse gas and its increasing levels in the atmosphere are contributing to global warming and climate change.1 Recently CO2 levels in the atmosphere at Mauna Loa, Hawaii were recorded above 400 ppm for the first time since measurements began in 1958.2 One method of reducing the emissions of CO2 to the atmosphere is capturing it at source from the flue gases of power stations and industry.3 Amine scrubbing is the traditional method but this has several disadvantages such as the high energy costs.4 An alternative is the use of solid sorbents

5-6

and the zeolite-structured microporous alumino-

silicates are promising candidates.7-8 There is a significant amount of work, both computational and experimental, on the adsorption of CO2 in zeolites. This has covered many aspects of the adsorption properties 9-21 and diffusion of CO2

22-31

in bulk zeolites. However, little work has been published on the

effect of surfaces on these processes. Surface effects are very important as in industrial applications it can be beneficial to use membranes made up of nanosized microporous crystals i.e. not exceeding 100 nm.32 In these cases, the diffusion through the surface region may have a great effect on the overall transport in the system as the external surface area to volume ratio is high. There are several computational studies of the surfaces of LTA including Slater et al.

33

calculating the stable surfaces of siliceous LTA and Greń et al. 34 investigating the structuring of water at the LTA surfaces for both the siliceous and sodium aluminosilicate forms. Combariza and Sastre

35

studied the influence of the siliceous {100} LTA surface on the

adsorption of methane using molecular dynamics (MD) simulations. They found that increasing the methane loading, the adsorption free energy barrier decreases while the surface permeability increases. Thompho et al. explored the effect of silanol groups at the zeolite surface on the permeation of methane through the {010} silicalite-1 surface.36 The surface 3 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 36

permeation was found to be lower in the presence of silanol groups. The adsorption mechanism of n-butane and isobutane at the surface of silicalite was probed using MD and Monte Carlo (MC) simulations by Chandross et al.37 The molecules first adsorb in a layer at the surface, where they hop between preferred adsorption sites before entering a pore mouth. The time spent in this adsorption layer is usually around 10 ps. The interfacial region of silicalite was also found to be important for the adsorption of hexadecane in a study by Webb et al.,38 particularly when adsorbed molecules blocked the surface adsorption sites. The sorption of molecules to a zeolite surface was also investigated experimentally. Jentys et al. 39 used fast time-resolved IR spectroscopy to study the sorption and transport of aromatics such as benzene from the gas phase onto hydroxyl groups on the surface and in the pores of H/ZSM-5. The surface permeability of nanoporous particles was determined experimentally using pulsed field gradient NMR.40-41 Newsome and Sholl 42 compared two methods (dual control volume grand canonical molecular dynamics, DVC-GCMD and Local Equilibrium Flux Method, LEFM) for estimating the resistance to mass transfer as molecules diffuse through the gas-zeolite interfaces of membranes. They found that their new LEFM method is a good approximate technique to determine the importance of the interface resistance before using the computationally expensive DVC-GCMD simulations. Zimmermann et al.

43-44

investigated

the effect of surfaces on the transport of methane and ethane in zeolite AFI. They calculated the surface barriers to diffusion and their implication for the overall flux across a membrane using a variety of molecular simulation techniques (MD, Monte Carlo and Reactive Flux) and tracer molecules. The molecules were found to follow paths close to the pore walls both in the zeolite interior and at the surface. They accurately predicted the surface permeability and found that it does not follow the behavior of the diffusion coefficients and can be very low, particularly at low temperatures and loading. The work was extended to cover zeolites with


Page 5 of 36

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


different structures, comparing AFI, LTL and MFI.45 The impact of the surface barriers for gases is important for thin zeolite layers only (up to around 100 unit cells deep) and after that can be assumed to be insignificant. The barriers are also found to depend strongly on the nanopore type, in particular the surface barriers increase in significance with smoother nanopores. There are a number of experimental studies on the permeability of small molecules such as H2, CO2, N2 and CH4 through zeolite membranes and the selectivity to specific molecules,46-54 many of which focus on silicalite-1 membranes and their modifications. The surface structure of zeolites has been shown to be important for selectivity of binary gas mixtures. Mizukami et al. 55 modeled the diffusion of a binary mixture of CO2/N2 through thin (30 Å) membranes of NaY and NaA. The surfaces were hydroxylated and the zeolite structures were fixed throughout the MD simulations. The simulations monitored the diffusion of a 1:1 CO2/N2 mixture permeating NaY and NaA {111} membranes and a NaA {100} membrane and the separation ratios across the three membranes. The CO2 selectivity decreased in the order NaA {100} > NaY {111} > NaA {111} and the difference in selectivity between membranes was attributed to the difference in pore diameters at the surface. Jee et al. 56 took the investigation into the impact of surface structure one step further by actively modifying the structure of a silicalite surface with a thin layer of amorphous silica. This improved the H2/CH4 selectivity without an unacceptable drop in H2 flux through the membrane. These are all important studies of the effect of surfaces on diffusion and transport of molecules; however many of them focus purely on siliceous zeolites, and none compare silicates and aluminosilicates of the same framework structure. In the present study we investigate the effect of composition and surface structure of FAU on CO2 adsorption and transport, comparing structural surface features such as the most favorable CO2 adsorption



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 36

sites and the CO2 structure at the surface and dynamical properties like the diffusion coefficient of CO2 through the surface region.

2.1 – Theoretical Methods Faujasite has a cubic structure with space group Fd-3m. The framework is constructed from sodalite cages (β-cages) joined together by double six-rings (d6R) to form a threedimensional network of large 12-membered ring pores. In this work a siliceous (Si-FAU) and two aluminosilicates zeolites with a Si:Al ratio of 1, one with Na+ cations (Na-FAU) and the other with K+ (K-FAU), were simulated using potential based techniques. The input zeolite framework structure was fully dehydrated sodium Zeolite Y, from Seo et al.57 This was modified to give both the siliceous form and the aluminosilicates with Si:Al ratio 1 by changing the identity of the T-site atoms, always ensuring Lowenstein’s rule was obeyed. The position of the cations within the bulk structure was determined using energy minimization as implemented in the METADISE code.58 The minimized bulk structure was cut along Miller indices to obtain the {111}, {011} and {100} surfaces which are the low Miller index surfaces with the {111} surface being the dominant surface in the crystal morphology.59 Several terminations are possible for each surface and we wanted to use the same termination for each composition to allow direct comparison of the change in composition. Therefore we selected the surface terminations based on those that were most stable for Si-FAU. The surfaces were hydroxylated so that all surface Si and Al were saturated. The stability was determined by calculating the surface energy, which is defined as the energy to cleave the surface in water. We followed the same procedure as Greń et al.34, using a water correction factor of -2.5 eV. The water correction factor differs from that used by Greń et al. as we are using a different potential model to simulate the zeolites. The calculated surface energies for the lowest energy terminations are shown in figure 1. For each


Page 7 of 36

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Miller index we chose the most stable of the symmetrical slabs. In the case of the {100} Miller index the two surfaces that would give symmetrical slabs have similar surface energies, so we selected to use the surface that had only one hydroxyl group per Si atom at the surface. Our model correctly identified the {111} surface as the most stable overall.

Figure 1. Surface energies of Si-FAU for the lowest energy terminations of the {111}, {011} and {100} surfaces (colored red, green and blue, respectively). Black outlines: the terminations that were used in this work; black dashed outlines: the other symmetrical slab terminations. The simulation cells are periodic in all three dimensions and comprise slabs of approximately 50 Å thickness and a gap of at least 40 Å filled with CO2. MD was carried out in the DL_POLY code

60

where the forces between atoms consist of long-range Coulombic

terms and short-range potentials in the Lennard Jones 12-6 form. All atoms were allowed to move during the simulations, including the cations. The electrostatic interactions of the system are evaluated using the Ewald method to a precision of 10-5 and the potential cut-off was 8.6 Å. All simulations were run for a total of 5 ns at 300 K with a timestep of 1 fs and with the Nosé-Hoover thermostat/barostat. Equilibration was performed for 1 ns in the NVT ensemble and 1 ns was then run in the NσT ensemble allowing relaxation of the cell dimensions but keeping the angles fixed, to ensure that the CO2 reached its desired density. The surface dimensions after this equilibrium are listed in table 1. A final 3 ns run in the 7 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 36

NVT ensemble was performed but data were collected only for the final 2 ns (2,000,000 timesteps).

Table 1. Surface area cell parameters of the slabs after NσT equilibration a (Å)

b (Å)

Si-FAU {111}

34.22 29.59

Si-FAU {011}

24.19 34.21

Si-FAU {100}

24.17 24.24

Na-FAU {111}

34.18 29.76

Na-FAU {011}

24.30 34.23

Na-FAU {100}

24.25 24.22

K-FAU {111}

34.61 29.97

K-FAU {011}

24.52 34.70

K-FAU {100}

24.48 24.53

2.2 – Validation of Potentials In the literature there are several potential models derived for zeolites.14, 61-64 We needed our potential model to be able to deal with CO2 in different zeolite structures, including different cations and hydroxyl groups for the generation of hydroxylated slabs, which none of the literature potential models that were derived for zeolites could provide. Therefore to meet all the criteria the potential parameters were chosen from the CLAYFF model.65 The Lennard-Jones parameters and the partial charges for all interactions and species are listed in table 2 and will be discussed in the following. Bushuev and Sastre 66 successfully used CLAYFF to model zeolites with an addition of OSi-O and Si-O-Si bend terms to accurately reproduce the zeolite structures. We also included three-body terms in our simulations (table 2) but we chose a different harmonic force


Page 9 of 36

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


constant (

) from Bushuev and Sastre. The EPM2 model

67

was used for CO2 and the

Lorentz-Berthelot mixing rules used to combine the forcefields. The combination of these two forcefields was used successfully by Kerisit et al.

68

in simulations of scCO2/H2O

mixtures above a forsterite surface. The CO2-CO2 interactions we use are purely the EMP2 parameters as they reproduce the equation of state. Modifications to the CO2-zeolite interactions were introduced to get a better representation of the experimental isotherms for Si-FAU, K- and Na- FAU. Although forcefields can be adapted to modeling chemisorbed CO2,69 experimental work on aluminosilicates such as zeolites and clay systems

70

does show that physical adsorption

dominates allowing us to neglect carbonate formation. Grand canonical Monte Carlo as implemented in the DL_MONTE code 71 was used to evaluate the isotherms of zeolites with different structures, Si:Al ratios and cations. A more detailed description of the derivation of the potentials can be found in Purton et al.71

Table 2. Lennard-Jones potential parameters and partial charges. Subscript Z refers to zeolite, H refers to hydroxyl, s refers to siliceous zeolite and a refers to aluminosilicate zeolite. Three body terms apply to the following bonds: O-Si-O, OH-Si-O, OH-Si-OH, Si-OHH, O-Al-O, OH-Al-O, OH-Al-OH, Al-OH-H. Charge

ε (eV)

r (Å)

Si

2.1

7.98817x10-8

3.7061

Al

1.575

7.98817x10-8

3.8564

OZs

-1.05

0.006739

3.5532

OZa

-1.16875

0.007501

3.5532

OHs

-0.950

0.006739

3.5532

OHa

-1.0095

0.007501

3.5532



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Na

1.0

0.005641

2.2421

K

1.0

0.004336

3.1810

H

0.425

-

-

Page 10 of 36

CO2 - CO2 OC

-0.3256

0.006930

3.4044

C

0.6512

0.002421

3.0946

CO2 - zeolite OC

-0.3256

0.006739

3.5532

C

0.6512

0.002421

3.3191

Three-body terms (eVÅ-2) 12.1

(°) 109.47

3 – Results and Discussion In this section we discuss the effect of surfaces on CO2 adsorption and diffusion. We report the calculated diffusion coefficients, density, residence time and coordination number of CO2 to build up a comprehensive picture of how the density and diffusion of CO2 vary between the zeolite, the surface and the region above the surface (specified regions are defined in section 3.1, figure 3) and how it is affected by different compositions and surface structures. First (section 3.1), we assess the effect of composition by focusing on CO2 diffusion and transport at the {111} surface of the siliceous, sodium and potassium forms of faujasite (Si-FAU, Na-FAU and K-FAU). Next, we consider the effect of surface structure by comparing the {111}, {011} and {100} surfaces of Si-FAU in section 3.2.

3.1 – The effect of surface composition The {111} surface is the most stable surface of faujasite and therefore dominates the morphology of the crystal.59 To determine the effect of the surface composition on the 10 ACS Paragon Plus Environment

Page 11 of 36

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


adsorption of CO2 we therefore studied the same hydroxylated termination (4.7 OH groups per nm2) of the {111} surface of Si-FAU, Na-FAU and K-FAU as illustrated for Si-FAU in figure 2. The most stable termination of the {111} surface was determined to be the one that cuts through the double six rings (d6R) between sodalite cages (β-cages) meaning no sodalite cages are broken and accessible to CO2.

Figure 2. Structure of the {111} surface of siliceous faujasite (Si-FAU). The density of the CO2 in the zeolite structure can be shown using the z-density. The simulation cell is divided into slices of 0.3 Å along the z-direction, the slices are therefore parallel to the zeolite surface (fig 2). In each slice the average density of CO2 is calculated and plotted. The z-density of CO2 in {111} slabs of Si-FAU, Na-FAU and K-FAU are presented in figure 3. Region 1 (fig 3a) is the zeolite bulk and exhibits a periodic pattern of maxima and minima due to the regular structure of channels and cages in the zeolite that are accessible to CO2. Zero is defined as the left hand edge of this region. Region 3 is the interlayer space, which corresponds to bulk CO2 and region 2 is the surface region which is defined in a similar way to that used by Webb and Grest.38 The surface region is 9 Å thick for all of the slabs, with the upper limit as the first minimum in CO2 density above the surface. In all cases it also covers the minimum below the surface. The density values in figure 3 are normalized values with respect to the CO2 density in the interlayer space.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 36

Figure 3. Normalized z-density plots of CO2 on (a) Si-FAU {111} slab, (b) Na-FAU {111} slab and (c) K-FAU {111} slab. 1 = zeolite region, 2 = surface region and 3 = interlayer space (see text for definitions). Si-FAU has three main oscillations in the zeolite region (region 1, fig 3a) while Na-FAU and K-FAU have four (fig 3b,c). This is due to the thickness of the slabs; the aluminosilicate slabs are greater in depth than the siliceous one but the thickness of the slab does not affect the behavior of CO2 at the surface, which is the focus of this study. Disregarding the number


Page 13 of 36

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


of oscillations there is a clear difference between the CO2 density peaks in the bulk region of Si-FAU compared to Na- and K-FAU. In Si-FAU each maximum is a single peak with a shoulder on each side while in the aluminosilicate zeolites it splits into two peaks of equal (or similar) magnitude indicating that the presence of the cations disrupts the redistribution of the CO2 within the pores. There is on average a higher relative density of CO2 in the aluminosilicates than in Si-FAU and the density in Na-FAU is greater than K-FAU. This is due to the enhanced attraction between CO2 and the cations with greater attraction for Na+-CO2 than for K+-CO2.72 The density of CO2 in the interlayer space (region 3) is 0.01 CO2/Å3 in all three systems. This is equivalent to working at elevated pressures such as those experienced in pre-combustion carbon capture.73 Kerisit et al., working under supercritical conditions, recorded values of CO2 density in the order of 0.05 CO2/Å3 above a forsterite surface .68 For clarity figure 4 depicts an enlargement of the surface region (region 2) of the z-density plots (fig 3), with the density of the zeolite oxygen and hydrogen atoms shown to mark the location of the zeolite surface. The densities have been scaled for clarity, so comparison can only been made between the positions of the different peaks but not their heights. For all systems, in the surface region there are two peaks of enhanced CO2 adsorption, one just below and one just above the terminating hydroxyl groups at the surface. The ratio of these two peaks varies between the three zeolites. For Si-FAU the peak above the surface is higher than the one below, as above the surface there is a greater available volume for the CO2 to adsorb. In Na-FAU the two peaks are approximately of equal magnitude due to two competing phenomena. As in Si-FAU, above the surface there is greater available volume for adsorption but in this case below the surface the interaction between CO2 and Na+ ions increases the level of adsorption giving rise to two peaks of similar magnitude. In K-FAU the situation is different again as the peak below the surface is higher than the one above the



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 36

surface. Some of the K+ ions have diffused out of the zeolite bulk as can be seen from the density of K in figure 4c, leaving a greater volume available for CO2 adsorption just below the surface compared to the case of Na-FAU.

Figure 4. z-density plots of surface region of (a) Si-FAU, (b) Na-FAU and (c) K-FAU {111} slabs. Densities have been scaled for clarity, so peak heights should not be compared, just their positions. CCO2: blue, K+/Na+: purple dashed, H: green dashed, Oz: red dashed. In general, the CO2 z-density plots indicate areas of high and low density due to available volume in the zeolite, but from them we cannot tell how closely packed the CO2 is without


Page 15 of 36

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


accurately knowing the available volume as a function of the z direction, which is not easy to calculate. An alternative way of viewing the density of CO2 in the slabs is the time averaged (over the duration of the simulation) density plots shown in figure 5. These pictures give qualitative information not just about how dense the CO2 is within different regions of the simulation cell but also the favored CO2 adsorption sites. We cannot use these plots to get numerical values for the density of CO2, but they can be used to compare densities between different systems. The red color denotes areas where CO2 is more likely to be adsorbed and therefore more densely packed, while the blue color denotes the areas where the CO2 prefers not to adsorb and thus is rarer. The interlayer spaces contain a uniform bulk density of CO2, while in the zeolite region there are white areas of zero CO2 density corresponding to the overlap between the d6Rs and the sodalite cages where CO2 is unable to access (fig 6). Inspection of figure 5 shows that the adsorption in Na-FAU is stronger than Si-FAU, which is likely to be due to the attraction between CO2 and Na+ ions and increased enthalpy of adsorption that is observed in sodium faujasite than siliceous faujasite.14 In Si-FAU the adsorption sites within the zeolite region are ordered as the CO2 is free to fill the whole of the pores whereas in Na-FAU the location of these sites is more disordered due to the Na+ ions in the pores which vary the available volume for CO2 adsorption and the CO2 molecules clustering around the cations. While Si-FAU has many regions of intermediate density suggesting that the CO2 is moving relatively freely throughout the structure, Na-FAU only has regions of high and low density, which indicates that the CO2 is anchored more firmly into adsorption sites. K-FAU has a similar profile to Na-FAU but with slightly weaker adsorption sites compared to Na-FAU. The time averaged density plots (fig 5) also confirm the argument we presented on the ratio of adsorption peaks at the {111} surfaces for the different FAU compositions in figure 4. The strength of the CO2 adsorption sites above and below the surface of Si-FAU (fig 5a) is similar



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 36

but above the surface there are a greater number of adsorption sites available due to the greater available volume. This explains why the adsorption peak above the surface of Si-FAU (fig 4a) is higher than the peak below the surface. In Na-FAU (fig 5b) the density at each adsorption site below the surface is greater than the density above the surface, but due to less available volume for adsorption below the surface, the result is two adsorption peaks of equal magnitude. K-FAU has a similar time averaged density profile to Na-FAU, however K+ diffuses out of the zeolite bulk as shown by the z-density (fig 4c), resulting in the CO2 adsorption peak below the surface having higher intensity compared to the one above (fig 4c).

Figure 5. Time averaged density plots of (a) Si-FAU {111} slab, (b) Na-FAU {111} slab and (c) K-FAU {111} slab.

Figure 6. Faujasite structure, with red denoting the cross-section of the ‘channels’ of zero CO2 density running through the zeolite. By analyzing radial distribution functions (RDF) in the 3 different regions we are able to determine how the average distance between carbon atoms of CO2 molecules varies, as listed 16 ACS Paragon Plus Environment

Page 17 of 36

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


in table 3. In all cases the distance between CO2 molecules is shorter in the zeolite region than in the interlayer space (a contraction of 1.7 - 4.4 %), with the value in the surface region in between. The contraction in distance is greater for the aluminosilicates than for the siliceous zeolite. In general shorter C-C distances give lower diffusion, as more closely packed molecules diffuse more slowly. The diffusion coefficient in the zeolite region of SiFAU, Na-FAU and K-FAU is 61 %, 95 % and 93 % lower than the interlayer space, respectively (table 3). To explain the lower diffusion coefficient in the K-FAU zeolite region than in Na-FAU, we suggest that the interaction between CO2 and Na+ is stronger than between CO2 and K+, and this is likely to dominate over the fact that the bulkier K+ ions are able to coordinate to more CO2 molecules (4.7 compared to 4.1 for Na+), holding them closer together to sterically hinder their diffusion.

Table 3. Average C-C distances between CO2 molecules and average Dxyz diffusion coefficients in the zeolite, surface and interlayer space. The errors were calculated at 95 % confidence level. Average Dxyz (10-9 m2s-1)

Average C-C distance (Å) Zeolite

Surface

Interlayer

Zeolite

Surface

Interlayer

Si-FAU

4.05 ±0.01

4.07 ±0.03

4.12 ±0.01

4.44 ±0.11

6.57 ±1.07

11.43 ±0.06

Na-FAU

4.02 ±0.03

4.09 ±0.06

4.14 ±0.01

0.75 ±0.02

4.22 ±1.68

14.29 ±0.08

K-FAU

3.95 ±0.04

4.03 ±0.06

4.13 ±0.01

0.95 ±0.04

3.83 ±1.27

12.70 ±0.07

The diffusion coefficient was calculated from the mean-square displacement (MSD) of the CO2 molecules in the directions perpendicular and parallel to the surface as a function of distance. The diffusion coefficient was evaluated as for the z-density, dividing the simulation cell along the z direction into slices and calculating the three components (Dx, Dy and Dz) of the diffusion coefficient of CO2 in each slice.74-75 For our purpose only Dz (equation 1) is 17 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 36

shown as it represents the diffusion in the direction perpendicular to the surface. CO2 molecules have to move along this direction to cross the surface and enter the zeolite. (1) We consider the trajectory of a particle, correlating its first position z(0) with its final position z(t) and assign the trajectory to the slice in which the particle resides at z(0). There are therefore two critical parameters in the calculation of the diffusion profiles, the correlation time and the width of the slice. We tested a range of correlation times from 0.5 to 25 ps and found that the shape of the diffusion profiles remains unaffected, however the magnitude of the diffusion coefficients did show some variation. We therefore tested the same correlation times on a box of pure CO2 with a density similar to that above the slab (0.01 CO2/Å3) and found that correlation times between 2.5 and 5 ps gave the same magnitude for the x, y and z components of the diffusion coefficient in each slice as the global diffusion coefficient for the system. We have also tested changing the width of the slices between 0.2 and 1 Å for these correlation times and found that the magnitude of the diffusion coefficient was unaffected. Thus we chose slices of width 0.5 Å and a correlation time of 5 ps as they resulted in smooth profiles. Figure 7 shows the change in Dz as a function of the z coordinate in Si-FAU {111}. The Dx and Dy (diffusion in the plane of the surface) are similar and do not show any particular features and therefore are not shown. The z-density of CO2 of Si-FAU {111} is plotted for comparison. The profiles for Na-FAU and K-FAU are similar to Si-FAU, although with much lower diffusion rates in the zeolite region, and are therefore not displayed. Generally, as the density of CO2 increases the diffusion coefficient decreases. In the surface region the dense layers of adsorbed CO2 reduce the z component of the diffusion, impeding the diffusion of CO2 through the surface into the zeolite bulk. However, at no point is the diffusion in the surface region lower than in the zeolite itself and in the surface region the


Page 19 of 36

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


diffusion increases in between the adsorption peaks to give an average that is higher than in the zeolite (see also table 3).

Figure 7. Dz (red) and CO2 density (blue) as a function of distance along z in Si-FAU {111}. We use the residence time (τr) to determine the average length of time that a CO2 molecule spends in contact with each T-site atom (Si and Al atoms of the zeolite). The τr was calculated from the residence time correlations function, which is defined as: 76 (2)

where

is the number of CO2 molecules within a 5 Å radius of a T-site atom and

is the

Heaviside function, which is 1 if the ith CO2 molecule is in the 5 Å radius at time t and 0 otherwise. A CO2 molecule was only judged to have left the 5 Å radius if it did so for at least 2 ps, which allowed molecules that temporarily left then reentered to be included in τr. Equation 2 is integrated to calculate τr: (3)

Figure 8 illustrates the residence time of OC atoms (oxygen of CO2) around the T-site, and the average coordination number (CN), which is the average number of OC atoms within the 5 Å radius. The images display all of the T-site atoms, colored according to their τr and CN. Red and blue correspond to smaller and larger τr and CN respectively. The side and top views 19 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 36

are both shown. The dashed line in the side view defines the boundary between the zeolite bulk and the surface region. Si-FAU has a homogeneous environment within the zeolite as can be seen from the uniform CN of OC around Si atoms (fig 8b). CN is greater at the surface, 4.8 compared to 3.0 in the zeolite, as the surface atoms are more accessible to CO2. In the zeolite and surface regions τr only varies from 5.2 to 7.5 ps, in a 2 ns simulation, indicating that the diffusion is rapid and the CO2 diffuse freely without any obstruction. τr is shorter at the surface (top layer of red atoms in the side view and top view for fig 8a) suggesting more rapid diffusion occurring at the surface. Just below the surface there is a region with longer τr, which agrees with the data from the diffusion profile in figure 7 where at the surface there are two minima above and below the surface, with a diffusion maximum in between. The τr profiles of CO2 in Na-FAU and K-FAU (fig 8c,e) are different from Si-FAU (fig 8a); τr values are much longer (up to 0.5 ns) as the diffusion of CO2 in Na- and K-FAU is very slow due to the strong adsorption. Again we see a lower τr at the surface suggesting faster diffusion in this region. Na-FAU and K-FAU have similar CO2 τr although the times are shorter in K-FAU. The majority of the CO2 molecules have a τr of 50 ps in K-FAU and 90 ps in Na-FAU as is highlighted by the peaks in the graphs (fig 8). The CN profiles of CO2 around the T-site atoms of Na-FAU and K-FAU (fig 8d,f) are similar to Si-FAU, and the CNs are on average slightly higher in Na-FAU than K-FAU (4.3 compared to 3.6), suggesting that the CO2 molecules are more closely packed. Again we see a higher coordination number at the surface, 7.0 OC atoms per Si atom in Na-FAU and 6.5 in K-FAU.


Page 21 of 36

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 8. T-sites (Si and Al atoms) in Si-FAU {111} slab (a,b), Na-FAU {111} slab (c,d) and K-FAU {111} slab (e,f), colored according to average residence time in ps (a,c,e) and average coordination number (b,d,f) of OC within a 5 Å radius of Si. Yellow dashed lines represent the boundary between the zeolite and surface regions. White vertical lines on the graphs represent the range of the scale; anything above this line is in blue. 3.2 – Effect of surface structure in Si-FAU This section will explore the differences in CO2 behavior in the surface regions with a change of surface structure. For this purpose we chose the {111}, {011} and {100} slabs of Si-FAU, considering only the most stable hydroxylated symmetrical surface termination in each case. The structures of the surfaces are presented in figure 9; the {111} surface was also shown previously in figure 2, but is presented here with the sodalite cage highlighted. The coverage of hydroxyl groups is 4.7, 5.8 and 8.2 OH/nm2 for the {111}, {011} and {100} surfaces respectively.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 36

Figure 9. Hydroxylated surface structures and time averaged density plots of CO2 in siliceous faujasite, a: {111}, b: {011}, c: {100}. O: red, Si: grey, H: white. Blue dashed rings indicate the sodalite cages at the surface. (d): how these surfaces cut through a sodalite cage. The most striking difference between the three surface structures is whether any sodalite cages are broken by the surface cut. Figure 9d indicates how the sodalite cages are cut by the three surfaces. As discussed in section 3.1, the most stable termination of the {111} surface does not cut through any sodalite cages so they remain inaccessible to CO2. The {011} surfaces cut the sodalite cages in half across the middle of the six-ring window (s6R) (fig 9d), to give a zigzag channel along the surface plane, where CO2 adsorbs strongly. In the case of the {100} surface, the sodalite cage is cut parallel to and just below the s4R face (fig 9d), meaning that the cage is accessible to CO2 and acts as a strong adsorption site. The strong adsorption sites in the sodalite cages can be seen in the time averaged density plots in figure 9a-c. In the image of the {011} surface (fig 9b), two adsorption sites are visible at the surface, which correspond to the sodalite cages to the left and the right of the zigzag channel. At the {100} surface (figure 9c), the CO2 is also allowed into the sodalite cages, but in this case a larger volume of the cage is accessible so the CO2 penetrates deeper. For the {111} surface, the most favorable adsorption sites are above the surface, as all of the sodalite cages 22 ACS Paragon Plus Environment

Page 23 of 36

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


are intact. The adsorption sites at the {011} and {100} surfaces can also be observed in the τr distributions in figure 10 as areas with longer τr, and the zigzag channels on the {011} surface can also be more clearly seen. The longest τr observed at the Si-FAU {111} surface is 7.5 ps (fig 8a), while we see τr values at the {011} and {100} surfaces of 8.3 ps and 8.5 ps, respectively, confirming that CO2 binds more strongly at these surfaces in the sodalite cage adsorption sites.

Figure 10. Si atoms in (a) Si-FAU {011} slab and (b) Si-FAU {100} slab, colored according to average residence time (ps) of OC within a 5 Å radius of Si, shown from the side and from the top (looking down on the surface). Open sodalite cages on the surface indicated by dashed yellow circles. Table 4 lists density, diffusion and C-C distance data for the 10 Å above the surface hydrogen atoms of the three surfaces. We see a small increase in the density of CO2 and decrease in the C-C distance above the {100} surface, but the significant difference that we see between the three surfaces is a decrease in the CO2 diffusion coefficient above the {100} surface. Therefore although the CO2 density may not be significantly greater, the {100} surface is having an impact on the CO2 transport compared to the other surfaces.

Table 4. Density, average C-C distance and average Dxyz in the 10 Å above the surface hydrogen atoms. Errors calculated at 95% confidence level.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Density (NCO2/Å3)

Page 24 of 36

Average C-C Average Dxyz distance (Å) (x 10-9 m2s-1)

{111} 0.012

4.10 ±0.03

10.23 ±0.85

{011} 0.012

4.12 ±0.03

9.26 ±1.23

{100} 0.014

4.08 ±0.03

6.61 ±0.34

The change in CO2 behavior above the {100} surface was also observed for the Na-FAU and K-FAU slabs. We conclude therefore that the {100} surface slows the transport of CO2, and as such would improve the adsorption in zeolite micropores. The experimental morphology of faujasite is octahedral with expression of the {111} surfaces. Spherical particles have been synthesized,77 but this is still not a common morphology. However, if a synthetic route could be found to enhance expression of the {100} surfaces in the nanocrystals, we predict that CO2 uptake could be improved.

4 – Conclusions We have presented a comprehensive study on the effect of the composition and surface structure of FAU on the adsorption and transport of CO2. The first was achieved by comparing the CO2 adsorption on the {111} surface of three different FAU (Si-FAU, NaFAU and K-FAU) and the second by comparing the CO2 behavior on the {100}, {011} and {111} surfaces of Si-FAU. We have used the CLAYFF potential model for the zeolite and the EPM2 for CO2, modifying the potential parameters where needed (details in Purton et al. 71

). The model correctly predicts the {111} surface as the most stable (fig 1) and therefore the

most dominant in the morphology. Three regions were identified: bulk zeolite, surface and interlayer regions. We found that there is always structuring of CO2 at the surfaces. When comparing the same surface for different FAU compositions we found that all three surfaces have two large CO2 adsorption


Page 25 of 36

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


peaks, just above and just below the surface, but the ratio in the height of these peaks varies between the three FAU compositions, depending on the balance between the strength of the CO2-zeolite interaction and the volume available for CO2 to adsorb. When comparing the three different surfaces for the same composition we found that the adsorption sites at the surface were governed by the surface structure, in particular whether or not any sodalite cages were cut by the surface. The {011} and {100} surfaces both have broken sodalite cages, which act as strong adsorption sites for CO2 compared to the {111} surface. Above the {100} surface CO2 diffused more slowly than above the other surfaces. This gives the potential of improved CO2 adsorption by hindering CO2 flow through zeolite micropores. Overall we have seen that CO2 behavior at the surface is affected more by the surface structure than the zeolite composition. In keeping with other published work 55-56 the surface structure is seen to be an important factor for adsorption. Furthermore, as we predict the enhanced adsorption on surfaces with incomplete sodalite units, which are most stable when {111} surfaces are not present, we suggest that best adsorption properties will be achieved with spherical particles rather than annealed and facetted crystals expressing a preponderance of {111} surfaces.

ASSOCIATED CONTENT Corresponding Author *[email protected] Author Contributions



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 26 of 36

The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript. ACKNOWLEDGMENT The authors thank EPSRC for funding (DTC studentship for J.C.C.) and EPSRC grant no. EP/J010480 and CCP5 for support. The SCARF computing resources provided by STFC’s eScience facility.

REFERENCES (1)

VijayaVenkataRaman, S.; Iniyan, S.; Goic, R., A Review of Climate Change,

Mitigation and Adaptation. Renew. Sust. Energy. Rev. 2012, 16, 878-897. (2)

National Oceanic and Atmospheric Administration; United States Department of

Commerce. Carbon Dioxide at NOAA's Mauna Loa Observatory Reaches New Milestone: Tops

400

ppm.

http://researchmatters.noaa.gov/news/Pages/CarbonDioxideatMaunaLoareaches400ppm.aspx (accessed 23/05/2013). (3)

Plasynski, S. I.; Litynski, J. T.; McIlvried, H. G.; Srivastava, R. D., Progress and New

Developments in Carbon Capture and Storage. Crit. Rev. Plant Sci. 2009, 28, 123-138. (4)

Rochelle, G. T., Amine Scrubbing for CO2 Capture. Science 2009, 325, 1652-1654.

(5)

Choi, S.; Drese, J. H.; Jones, C. W., Adsorbent Materials for Carbon Dioxide Capture

from Large Anthropogenic Point Sources. ChemSusChem 2009, 2, 796-854. (6)

D'Alessandro, D. M.; Smit, B.; Long, J. R., Carbon Dioxide Capture: Prospects for

New Materials. Angew. Chem.-Int. Edit. 2010, 49, 6058-6082.


Page 27 of 36

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(7)

Smit, B.; Maesen, T. L. M., Molecular Simulations of Zeolites: Adsorption, Diffusion,

and Shape Selectivity. Chem. Rev. 2008, 108, 4125-4184. (8)

Bonenfant, D.; Kharoune, M.; Niquette, P.; Mimeault, M.; Hausler, R., Advances in

Principal Factors Influencing Carbon Dioxide Adsorption on Zeolites. Sci. Technol. Adv. Mat. 2008, 9, 013007-1 - 013007-7. (9)

Goj, A.; Sholl, D. S.; Akten, E. D.; Kohen, D., Atomistic Simulations of CO2 and N2

Adsorption in Silica Zeolites: The Impact of Pore Size and Shape. J. Phys. Chem. B 2002, 106, 8367-8375. (10) Plant, D. F.; Maurin, G.; Deroche, I.; Gaberova, L.; Llewellyn, P. L., CO2 Adsorption in Alkali Cation Exchanged Y Faujasites: A Quantum Chemical Study Compared to Experiments. Chem. Phys. Lett. 2006, 426, 387-392. (11) Plant, D. F.; Maurin, G.; Jobic, H.; Llewellyn, P. L., Molecular Dynamics Simulation of the Cation Motion upon Adsorption of CO2 in Faujasite Zeolite Systems. J. Phys. Chem. B 2006, 110, 14372-14378. (12) Hirotani, A.; Mizukami, K.; Miura, R.; Takaba, H.; Miya, T.; Fahmi, A.; Stirling, A.; Kubo, M.; Miyamoto, A., Grand Canonical Monte Carlo Simulation of the Adsorption of CO2 on Silicalite and NaZSM-5. Appl. Surf. Sci. 1997, 120, 81-84. (13) Maurin, G.; Llewellyn, P.; Poyet, T.; Kuchta, B., Influence of Extra-Framework Cations on the Adsorption Properties of X-faujasite Systems: Microcalorimetry and Molecular Simulations. J. Phys. Chem. B 2005, 109, 125-129. (14) Maurin, G.; Llewellyn, P. L.; Bell, R. G., Adsorption Mechanism of Carbon Dioxide in Faujasites: Grand Canonical Monte Carlo Simulations and Microcalorimetry Measurements. J. Phys. Chem. B 2005, 109, 16084-16091. 27 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 28 of 36

(15) Maurin, G.; Belmabkhout, Y.; Pirngruber, G.; Gaberova, L.; Llewellyn, P., CO2 Adsorption in LiY and NaY at High Temperature: Molecular Simulations Compared to Experiments. Adsorpt.-J. Int. Adsorpt. Soc. 2007, 13, 453-460. (16) Pham, T. D.; Liu, Q. L.; Lobo, R. F., Carbon Dioxide and Nitrogen Adsorption on Cation-Exchanged SSZ-13 Zeolites. Langmuir 2013, 29, 832-839. (17) Bulanek, R.; Frolich, K.; Frydova, E.; Cicmanec, P., Microcalorimetric and FTIR Study of the Adsorption of Carbon Dioxide on Alkali-Metal Exchanged FER Zeolites. Top. Catal. 2010, 53, 1349-1360. (18) Montanari, T.; Busca, G., On the Mechanism of Adsorption and Separation of CO2 on LTA Zeolites: An IR Investigation. Vib. Spectrosc. 2008, 46, 45-51. (19) Dunne, J. A.; Mariwals, R.; Rao, M.; Sircar, S.; Gorte, R. J.; Myers, A. L., Calorimetric Heats of Adsorption and Adsorption Isotherms .1. O2, N2, Ar, CO2, CH4, C2H6 and SF6 on Silicalite. Langmuir 1996, 12, 5888-5895. (20) Dunne, J. A.; Rao, M.; Sircar, S.; Gorte, R. J.; Myers, A. L., Calorimetric Heats of Adsorption and Adsorption Isotherms .2. O2, N2, Ar, CO2, CH4, C2H6, and SF6 on NaX, HZSM-5, and Na-ZSM-5 Zeolites. Langmuir 1996, 12, 5896-5904. (21) Fischer, M.; Bell, R. G., Influence of Zeolite Topology on CO2/N2 Separation Behavior: Force-Field Simulations Using a DFT-Derived Charge Model. J. Phys. Chem. C 2012, 116, 26449-26463. (22) Papadopoulos, G. K.; Jobic, H.; Theodorou, D. N., Transport Diffusivity of N2 and CO2 in Silicalite: Coherent Quasielastic Neutron Scattering Measurements and Molecular Dynamics Simulations. J. Phys. Chem. B 2004, 108, 12748-12756.


Page 29 of 36

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(23) Babarao, R.; Jiang, J. W., Diffusion and Separation of CO2 and CH4 in Silicalite, C168 Schwarzite,and IRMOF-1: A Comparative Study from Molecular Dynamics Simulation. Langmuir 2008, 24, 5474-5484. (24) van den Bergh, J.; Ban, S. A.; Vlugt, T. J. H.; Kapteijn, F., Modeling the Loading Dependency of Diffusion in Zeolites: The Relevant Site Model. J. Phys. Chem. C 2009, 113, 17840-17850. (25) van den Bergh, J.; Ban, S. A.; Vlugt, T. J. H.; Kapteijn, F., Modeling the Loading Dependency of Diffusion in Zeolites: the Relevant Site Model Extended to Mixtures in DDRType Zeolite. J. Phys. Chem. C 2009, 113, 21856-21865. (26) van den Bergh, J.; Ban, S. A.; Vlugt, T. J. H.; Kapteijn, F., Diffusion in Zeolites: Extension of the Relevant Site Model to Light Gases and Mixtures Thereof in Zeolites DDR, CHA, MFI and FAU. Sep. Purif. Technol. 2010, 73, 151-163. (27) Makrodimitris, K.; Papadopoulos, G. K.; Theodorou, D. N., Prediction of Permeation Properties of CO2 and N2 through Silicalite via Molecular Simulations. J. Phys. Chem. B 2001, 105, 777-788. (28) Selassie, D.; Davis, D.; Dahlin, J.; Feise, E.; Haman, G.; Sholl, D. S.; Kohen, D., Atomistic Simulations of CO2 and N2 Diffusion in Silica Zeolites: The Impact of Pore Size and Shape. J. Phys. Chem. C 2008, 112, 16521-16531. (29) Beerdsen, E.; Dubbeldam, D.; Smit, B., Understanding Diffusion in Nanoporous Materials. Phys. Rev. Lett. 2006, 96, 044501-1 - 044501-4. (30) Krishna, R.; van Baten, J. M.; Garcia-Perez, E.; Calero, S., Incorporating the Loading Dependence of the Maxwell-Stefan Diffusivity in the Modeling of CH4 and CO2 Permeation Across Zeolite Membranes. Ind. Eng. Chem. Res. 2007, 46, 2974-2986. 29 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 30 of 36

(31) Deroche, I.; Maurin, G.; Borah, B. J.; Yashonath, S.; Jobic, H., Diffusion of Pure CH4 and Its Binary Mixture with CO2 in Faujasite NaY: A Combination of Neutron Scattering Experiments and Molecular Dynamics Simulations. J. Phys. Chem. C 2010, 114, 5027-5034. (32) Saengsawang, O.; Nanok, T.; Vasenkov, S.; Fritzsche, S., Relationship Between Sorbate Transport Inside and at the Margins of Zeolite Crystals. Soft Materials 2012, 10, 202215. (33) Slater, B.; Titiloye, J. O.; Higgins, F. M.; Parker, S. C., Atomistic Simulation of Zeolite Surfaces. Curr. Opin. Solid State Mat. Sci. 2001, 5, 417-424. (34) Gren, W.; Parker, S. C.; Slater, B.; Lewis, D. W., Structure of Zeolite A (LTA) Surfaces and the Zeolite A/Water Interface. J. Phys. Chem. C 2010, 114, 9739-9747. (35) Combariza, A. F.; Sastre, G., Influence of Zeolite Surface in the Sorption of Methane from Molecular Dynamics. J. Phys. Chem. C 2011, 115, 13751-13758. (36) Thompho, S.; Chanajaree, R.; Remsungnen, T.; Hannongbua, S.; Bopp, P. A.; Fritzsche, S., The Permeation of Methane Molecules through Silicalite-1 Surfaces. J. Phys. Chem. A 2009, 113, 2004-2014. (37) Chandross, M.; Webb, E. B.; Grest, G. S.; Martin, M. G.; Thompson, A. P.; Roth, M. W., Dynamics of Exchange at Gas-Zeolite Interfaces I: Pure Component n-Butane and Isobutane. J. Phys. Chem. B 2001, 105, 5700-5712. (38) Webb, E. B.; Grest, G. S., Interfaces Between Silicalite Surfaces and Liquid Hexadecane: A Molecular Dynamics Simulation. J. Chem. Phys. 2002, 116, 6311-6321. (39) Jentys, A.; Tanaka, H.; Lercher, J. A., Surface Processes during Sorption of Aromatic Molecules on Medium Pore Zeolites. J. Phys. Chem. B 2004, 109, 2254-2261.


Page 31 of 36

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(40) Krutyeva, M.; Yang, X.; Vasenkov, S.; Kärger, J., Exploring the Surface Permeability of Nanoporous Particles by Pulsed Field Gradient NMR. J. Magn. Reson. 2007, 185, 300307. (41) Krutyeva, M.; Vasenkov, S.; Yang, X.; Caro, J.; Kärger, J., Surface Barriers on Nanoporous Particles: A New Method of Their Quantitation by PFG NMR. Microporous Mesoporous Mat. 2007, 104, 89-96. (42) Newsome, D. A.; Sholl, D. S., Predictive Assessment of Surface Resistances in Zeolite Membranes using Atomically Detailed Models. J. Phys. Chem. B 2005, 109, 72377244. (43) Zimmermann, N. E. R.; Smit, B.; Keil, F. J., On the Effects of the External Surface on the Equilibrium Transport in Zeolite Crystals. J. Phys. Chem. C 2010, 114, 300-310. (44) Zimmermann, N. E. R.; Smit, B.; Keil, F. J., Predicting Local Transport Coefficients at Solid-Gas Interfaces. J. Phys. Chem. C 2012, 116, 18878-18883. (45) Zimmermann, N. E. R.; Balaji, S. P.; Keil, F. J., Surface Barriers of Hydrocarbon Transport Triggered by Ideal Zeolite Structures. J. Phys. Chem. C 2012, 116, 3677-3683. (46) Kapteijn, F.; Bakker, W. J. W.; van de Graaf, J.; Zheng, G.; Poppe, J.; Moulijn, J. A., Permeation and Separation Behaviour of a Silicalite-1 Membrane. Catal. Today 1995, 25, 213-218. (47) Algieri, C.; Bernardo, P.; Golemme, G.; Barbieri, G.; Drioli, E., Permeation Properties of a Thin Silicalite-1 (MFI) Membrane. J. Membr. Sci. 2003, 222, 181-190.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 32 of 36

(48) Lindmark, J.; Hedlund, J.; Wirawan, S. K.; Creaser, D.; Li, M.; Zhang, D.; Zou, X., Impregnation of Zeolite Membranes for Enhanced Selectivity. J. Membr. Sci. 2010, 365, 188197. (49) Lindmark, J.; Hedlund, J., Modification of MFI Membranes with Amine Groups for Enhanced CO2 Selectivity. J. Mater. Chem. 2010, 20, 2219-2225. (50) Lavorgna, M.; Sansone, L.; Scherillo, G.; Gu, R.; Baker, A. P., Transport Properties of Zeolite Na-X-Nafion Membranes: Effect of Zeolite Loadings and Particle Size. Fuel Cells 2011, 11, 801-813. (51) Lito, P. F.; Santiago, A. S.; Cardoso, S. P.; Figueiredo, B. R.; Silva, C. M., New Expressions for Single and Binary Permeation Through Zeolite Membranes for Different Isotherm Models. J. Membr. Sci. 2011, 367, 21-32. (52) Wirawan, S. K.; Creaser, D.; Lindmark, J.; Hedlund, J.; Bendiyasa, I. M.; Sediawan, W. B., H2/CO2 Permeation Through a Silicalite-1 Composite Membrane. J. Membr. Sci. 2011, 375, 313-322. (53) Tawalbeh, M.; Tezel, F. H.; Letaief, S.; Detellier, C.; Kruczek, B., Separation of CO2 and N2 on Zeolite Silicalate-1 Membrane Synthesized on Novel Support. Sep. Sci. Technol. 2012, 47, 1606-1616. (54) Lara-Medina, J. J.; Torres-Rodriguez, M.; Gutierrez-Arzaluz, M.; Mugica-Alvarez, V., Separation of CO2 and N2 with a Lithium-Modified Silicalite-1 Zeolite Membrane. Int. J. Greenh. Gas Control 2012, 10, 494-500. (55) Mizukami, K.; Kobayashi, Y.; Morito, H.; Takami, S.; Kubo, M.; Belosludov, R.; Miyamoto, A., Molecular Dynamics Studies of Surface Difference Effect on Gas Separation


Page 33 of 36

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


by Zeolite Membranes. Jpn. J. Appl. Phys. Part 1 - Regul. Pap. Short Notes Rev. Pap. 2000, 39, 4385-4388. (56) Jee, S. E.; McGaughey, A. J. H.; Sholl, D. S., Molecular Simulations of Hydrogen and Methane Permeation Through Pore Mouth Modified Zeolite Membranes. Mol. Simul. 2009, 35, 70-78. (57) Seo, S. M.; Kim, G. H.; Lee, H. S.; Ko, S.-O.; Lee, O. S.; Kim, Y. H.; Kim, S. H.; Heo, N. H.; Lim, W. T., Single-crystal Structure of Fully Dehydrated Sodium Zeolite Y (FAU). Analytical Sciences: X-ray Structure Analysis Online 2006, 22, 209-210. (58) Watson, G. W.; Kelsey, E. T.; de Leeuw, N. H.; Harris, D. J.; Parker, S. C., Atomistic Simulation of Dislocations, Surfaces and Interfaces in MgO. J. Chem. Soc. Faraday T. 1996, 92, 433-438. (59) IZA Commission on Natural Zeolites. Faujasite Data Sheet. http://www.izaonline.org/natural/Datasheets/Faujasite/faujasite.htm (accessed 27th February). (60) Smith, W.; Forester, T. R., DL_POLY_2.0: A General-Purpose Parallel Molecular Dynamics Simulation Package. J. Mol. Graphics 1996, 14, 136-141. (61) Garcia-Sanchez, A.; Ania, C. O.; Parra, J. B.; Dubbeldam, D.; Vlugt, T. J. H.; Krishna, R.; Calero, S., Transferable Force Field for Carbon Dioxide Adsorption in Zeolites. J. Phys. Chem. C 2009, 113, 8814-8820. (62) Jaramillo, E.; Chandross, M., Adsorption of Small Molecules in LTA Zeolites. 1. NH3, CO2, and H2O in Zeolite 4A. J. Phys. Chem. B 2004, 108, 20155-20159.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 34 of 36

(63) Babarao, R.; Hu, Z. Q.; Jiang, J. W.; Chempath, S.; Sandler, S. I., Storage and Separation of CO2 and CH4 in Silicalite, C-168 Schwarzite, and IRMOF-1: A Comparative Study from Monte Carlo Simulation. Langmuir 2007, 23, 659-666. (64) Akten, E. D.; Siriwardane, R.; Sholl, D. S., Monte Carlo Simulation of Single- and Binary-Component Adsorption of CO2, N2, and H2 in Zeolite Na-4A. Energy Fuels 2003, 17, 977-983. (65) Cygan, R. T.; Liang, J. J.; Kalinichev, A. G., Molecular Models of Hydroxide, Oxyhydroxide, and Clay Phases and the Development of a General Force Field. J. Phys. Chem. B 2004, 108, 1255-1266. (66) Bushuev, Y. G.; Sastre, G., Atomistic Simulations of Water and Organic Templates Occluded During the Synthesis of Zeolites. Microporous Mesoporous Mat. 2010, 129, 42-53. (67) Harris, J. G.; Yung, K. H., Carbon Dioxide's Liquid-Vapor Coexistence Curve And Critical Properties as Predicted by a Simple Molecular Model. J. Phys. Chem. 1995, 99, 12021-12024. (68) Kerisit, S.; Weare, J. H.; Felmy, A. R., Structure and Dynamics of ForsteritescCO2/H2O Interfaces as a Function of Water Content. Geochim. Cosmochim. Acta 2012, 84, 137-151. (69) Allen, J. P.; Marmier, A.; Parker, S. C., Atomistic Simulation of Surface Selectivity on Carbonate Formation at Calcium and Magnesium Oxide Surfaces. J. Phys. Chem. C 2012, 116, 13240-13251. (70) Rother, G.; Ilton, E. S.; Wallacher, D.; Hauβ, T.; Schaef, H. T.; Qafoku, O.; Rosso, K. M.; Felmy, A. R.; Krukowski, E. G.; Stack, A. G., et al., CO2 Sorption to Subsingle


Page 35 of 36

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Hydration Layer Montmorillonite Clay Studied by Excess Sorption and Neutron Diffraction Measurements. Environ. Sci. Technol. 2012, 47, 205-211. (71) Purton, J. A.; Crabtree, J. C.; Parker, S. C. DL_MONTE: A General Purpose Program for Parallel Monte Carlo Simulation. Mol. Simul. [Online early access]. DOI: 10.1080/08927022.2013.839871

Published

Online:

Sep

25,

2013.

http://www.tandfonline.com/doi/full/10.1080/08927022.2013.839871#.UkWq0IbEOql (accessed Sep 25, 2013). (72) Barrer, R. M.; Gibbons, R. M., Zeolitic Carbon Dioxide - Energetics and Equilibria in Relation to Exchangeable Cations in Faujasite. Trans. Faraday Soc. 1965, 61, 948-961. (73) Feron, P. H. M.; Hendriks, C. A., CO2 Capture Process Principles and Costs. Oil Gas Sci. Technol. 2005, 60, 451-459. (74) Marrink, S.-J.; Berendsen, H. J. C., Simulation of Water Transport Through a Lipid Membrane. J. Phys. Chem. 1994, 98, 4155-4168. (75) Kerisit, S.; Parker, S. C., Free Energy of Adsorption of Water and Metal Ions on the {1014} Calcite Surface. J. Am. Chem. Soc. 2004, 126, 10152-10161. (76) Impey, R. W.; Madden, P. A.; McDonald, I. R., Hydration and Mobility of Ions in Solution. J. Phys. Chem. 1983, 87, 5071-5083. (77) Hincapie, B. O.; Garces, L. J.; Gomez, S.; Ghosh, R.; Suib, S. L., Direct Synthesis of Copper Faujasite. Catal. Today 2005, 110, 323-329.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 36 of 36

TOC Image:


Simulation of the Adsorption and Transport of CO2 ... - ACS Publications

Recommend Documents