Comparative Study of the Effect of Defects on ... - ACS Publications

Subscriber access provided by Warwick University Library

Article

A Comparative Study of the Effect of Defects on Selective Adsorption of Butanol from Butanol/Water Binary Vapor Mixtures in Silicalite-1 Films Amirfarrokh Farzaneh, Robert F. DeJaco, Lindsay Ohlin, Allan Holmgren, J. Ilja Siepmann, and Mattias Grahn Langmuir, Just Accepted Manuscript • DOI: 10.1021/acs.langmuir.7b02097 • Publication Date (Web): 02 Aug 2017 Downloaded from http://pubs.acs.org on August 3, 2017

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.

Langmuir 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 30

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

Langmuir

A Comparative Study of the Effect of Defects on Selective Adsorption of Butanol from Butanol/Water Binary Vapor Mixtures in Silicalite-1 Films Amirfarrokh Farzaneha, Robert F. DeJacob, Lindsay Ohlina, Allan Holmgrena, J. Ilja Siepmannb and Mattias Grahna* a

Chemical Technology, Luleå University of Technology, SE-971 87 Luleå, Sweden

b

Department of Chemical Engineering and Materials Science and Department of Chemistry and

Chemical Theory Center, University of Minnesota, Minneapolis, Minnesota 55455 *e-mail address: [email protected]

ABSTRACT A promising route for sustainable 1-butanol (butanol) production is ABE (acetone, butanol, ethanol) fermentation. However, recovery of the products is challenging because of the low concentrations obtained in the aqueous solution, thus hampering large-scale production of biobutanol. Membrane and adsorbent-based technologies using hydrophobic zeolites are interesting alternatives to traditional separation techniques (e.g., distillation) for energy-efficient separation of butanol from aqueous mixtures. To maximize the butanol over water selectivity of the material, it is important to reduce the number of hydrophilic adsorption sites. This can, for

ACS Paragon Plus Environment

1

Langmuir

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 30

instance, be achieved by reducing the density of lattice defect sites where polar silanol groups are found. The density of silanol defects can be reduced by preparing the zeolite at neutral pH instead of using traditional synthesis solutions with high pH. In this work, binary adsorption of butanol and water in two silicalite-1 films was studied using in situ attenuated total reflectance−Fourier transform infrared (ATR−FTIR) spectroscopy under equal experimental conditions. One of the films was prepared in fluoride medium, whereas the other one was prepared at high pH using traditional synthesis conditions. The amount of water and butanol adsorbed from binary vapor mixtures of varying composition were determined at 35 and 50 oC and the corresponding adsorption selectivities were also obtained. Both samples showed very high selectivities (100 – 23000) towards butanol under the conditions studied. The sample having low density of defects, in general, showed ca a factor 10 times higher butanol selectivity than the sample having a higher density of defects at the same experimental conditions. This difference was due to a much lower adsorption of water in the sample with low density of internal defects. Analysis of molecular simulation trajectories provides insights on the local selectivities in the zeolite channel network and at the film surface.

INTRODUCTION Biobutanol is a promising alternative to bioethanol for blending in gasoline since it possesses a higher specific energy content, is less hygroscopic, and its lower vapor pressure and lower flash point make it safer to handle and less prone to cause problems.1 1–Butanol (hereafter referred to as only butanol) can be produced using the acetone-butanol-ethanol (ABE) fermentation process.2 Like other alcohols produced during biomass conversion, butanol must be separated from the other components present in the fermentation broth before being used as a fuel. Traditional thermal separation processes for recovery of butanol from aqueous mixtures usually


2

Page 3 of 30

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

Langmuir

require considerable energy in the form of heat or are not very selective.3-7 However, adsorptionbased recovery processes and separation by membranes have been ranked as the most energyefficient and selective techniques for recovery of butanol.4,8,9 Several hydrophobic materials, for example zeolites, activated carbon and zeolite-like materials, have been reported as promising adsorbents for separation of alcohols from dilute aqueous mixtures.10-17 Among the zeolites, most research has been done on high-silica MFI which has a non-polar nature18,19 and thus selectively adsorbs alcohols from dilute aqueous solutions.9,20-26 Moreover, mixed matrix- and zeolite MFI membranes have also been evaluated for this separation,27-30 both high flux and separation factors were reported for thin high-silica MFI membranes recently.28 Traditionally, synthesis of silicalite-1, the pure silica analogue of MFI, is carried out at high pH, and it is well-known that a substantial number of internal defects in the form of silanol groups are formed at these synthesis conditions.31-33 The polar nature of these silanol groups makes the zeolite more hydrophilic and consequently decreases the selectivity for alcohol separation from aqueous mixtures.9,16,34,35 On the other hand, zeolites prepared in fluoride (F⁻) medium show significantly lower concentration of internal silanol defects.35-37 Recently, we studied singlecomponent adsorption of butanol and water in two different silicalite-1 films, one prepared in hydroxide (OH⁻)38 and the other one in fluoride (F⁻)39 medium. Single-component adsorption isotherms of butanol and water were measured at different temperatures using in situ ATR-FTIR spectroscopy. Adsorption of water was significantly lower in the silicalite-1(F⁻) film with the lowest uptake ever reported for MFI-type zeolites. The low adsorption of water in silicalite-1(F⁻) was ascribed to the low density of defects in this sample as evidenced by

29

Si NMR

spectroscopy.


3

Langmuir

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 30

The ideal adsorbed solution theory (IAST) has been widely used to predict mixed gas adsorption using only single-component isotherms as input.40,41 However, the capability of IAST to accurately predict the adsorbed amounts for hydrogen bonding systems, like alcohol-water mixtures, is doubtful due to the effect of hydrogen bonding between water and alcohol molecules resulting in e.g. clustering of adsorbed molecules.26,42,43 Moreover, Monte Carlo simulations have proven to be a powerful tool for studying adsorption in zeolites,21,44-46 but most of these simulations on the prediction of alcohol-water mixtures were performed without considering the effect structural defects on adsorption.21,44,47 Therefore, experimental investigations, in particular on adsorption from gas mixtures, are of great importance for further adsorbent and membrane development as well as for further refining the modeling tools. There are only a few reports in the literature on the adsorption of alcohol/water mixtures in silicalite-1 (in particular for silicalite-1(F⁻)) from vapor phase.21,34,35,38,39,44 Particularly, we are not aware of any experimental data reported on butanol and water adsorption from binary vapor mixtures. However, Zhang et al.35 estimated the ideal ethanol/water adsorption selectivities for different high- and pure-silica MFI-type zeolites using single-component adsorption isotherms determined from vapor phase. They reported significantly higher ethanol/water selectivity for a silicalite-1 sample synthesized using F⁻ as mineralizing agent compared to silicalite-1 and ZSM5 samples prepared at high pH. The amount of adsorbed water reported by Zhang et al. for the silicalite-1 sample prepared at neutral pH was the lowest water uptake reported for MFI-type zeolites. Recently, DeJaco et al.21 studied binary adsorption of butanol and water in all-silica MFI-type zeolites from liquids using Monte Carlo simulations. They reported selectivities higher than 30 000 for adsorption of butanol from very dilute (less than 1 g/L) butanol/water solutions. They performed their simulations on a “perfect” silicalite-1 crystal without any defects or


4

Page 5 of 30

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

Langmuir

external surface, so their results should resemble an upper bound for experiments for real crystals with low defect density. In both the latter papers, the values obtained from simulations of binary adsorption of alcohol/water mixtures in MFI-type zeolites would be attainable only for very large and defect free single crystals. Such large crystals can possibly be promising as adsorbents for alcohol; however, such large crystals are unsuitable for producing high-flux zeolite membranes.27 In the present work we study binary adsorption of butanol and water in two silicalite-1 films prepared in either F⁻ or OH⁻ media, i.e. in samples having different densities of internal defects. Thus, we can study the effect of these defects on the butanol/water adsorption selectivity. Molecular simulations are performed to investigate the influence of silanol groups on the external surface of a film and to provide molecular-level information on the non-ideal interactions of these adsorbed mixtures of butanol and water. We know of no previous systematic experimental study of binary adsorption of butanol and water in MFI-type zeolites, and this work together with our previous papers38,39 can not only be of guidance for further developing both the materials and their application as adsorbents or membranes, but also for refining the simulation tools available.

EXPERIMENTAL The in situ ATR-FTIR adsorption experiments and the method for quantification of spectral data together with the synthesis and characterization of the films have been comprehensively described previously38,39,48 and are only briefly described briefly below. A more detailed description is also included in the Supporting Information. Silicalite-1 films were synthesized on trapezoidal ATR crystals (ZnS, with the dimension of 50 × 20 × 2 mm3 and 45° cut edges, Crystran, Ltd.) using a seeding method were synthesis of seeds


5

Langmuir

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 30

and films were performed either in fluoride or hydroxide media.[38,39] The silicalite-1 coated ATR crystal was mounted into a temperature-controlled flow cell connected to a gas delivery system and dried under a constant flow rate of helium (AGA, 99.999%) for 4 h at 300 °C with a heating and cooling rate of 0.9 °C/min.49 A single beam background spectrum of the dried film was recorded under a flow of helium at the temperature of the adsorption experiment using a Bruker IFS 66v/S FTIR spectrometer equipped with a deuterated triglycine sulfate (DTGS) detector. After the background spectrum was recorded, binary adsorption experiments were carried out by exposing the silicalite-1 films to butanol/water vapor mixtures with different compositions and at different temperatures and recording infrared spectra by averaging 256 scans at a resolution of 4 cm-1 after adsorption equilibrium was achieved. The composition of the feed to the cell was controlled by two digital mass flow controllers and two parallel saturators (one containing butanol (Sigma-Aldrich, 99.9%) and the other one distilled water) to obtain the desired fractions of water and butanol in the gas stream while using helium as the carrier gas. The adsorbed amount of water was calculated from the infrared spectra using equation (1). A more detailed description of the procedure for determining the adsorbed amounts from infrared spectra can be found in the Supporting information.48,50-52 2 A n21E0 d pC −2d = ε 1− e a N 2 cos θ

(

dp

)

(1)

Where A is the integrated absorbance of a band of interest, N is the number of total reflections in the ATR element (20 reflection within the gasket sealing the cell), n21 is the ratio of the reflective indices of the ATR element and the silicalite-1 film, E0 is the amplitude of the electric field at the interface of silicalite-1 film and ATR element, C is the concentration in the adsorbed phase, θ is the angle of incidence, ε is the molar absorptivity, da is the thickness of the film and


6

Page 7 of 30

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

Langmuir

dp is the penetration depth. The penetration depth is deﬁned as the distance required for the electric field amplitude to fall to e-1 of its value at the surface (equation 2): dp =

(2)

λ1 2 12 2π (sin θ − n21 ) 2

In this equation, λ1 is the wavelength of the infrared radiation in the ATR crystal. Because the molar absorptivity for butanol adsorbed in silicalite-1 is not reported yet, it was assumed that the maximum absorbance determined for butanol from infrared spectra of the butanol loaded silicalite-1 films corresponds to the maximum adsorbed loading of butanol determined previously by volumetric adsorption measurements on silicalite-1 powder.38,39

SIMULATION METHODS Lennard-Jones (LJ) 12—6 and Coulomb potentials were used to describe sorbate— sorbate and sorbent—sorbate intermolecular interactions. As in preceding studies,26,53 the parameters for water, butanol, and zeolite frameworks were adapted from the TIP4P model54 and the transferable potentials for phase equilibria (TraPPE) force field for alcohols55 and zeolites,56 respectively. The LJ parameters for the hydroxyl groups on external silanols as well as the bending and torsional degrees of freedom of silanol hydrogens were adapted from those in the TraPPE force field for alcohols.55 Following previous work,57 the charge on the silanol hydrogen was +0.435e (i.e., the same as the TraPPE alcohol model) and the charge on the silanol hydroxyl group oxygen was −0.739e (this choice yields charge neutrality). LJ parameters for all unlike interactions are obtained using Lorentz—Berthelot combining rules.58 The LJ sorbate—sorbate and sorbate—silanol interactions are spherically truncated at a distance of 14 Å with analytical tail corrections to estimate interactions beyond this distance.59 Coulomb interactions are


7

Langmuir

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 30

described using the Ewald summation method59 with a screening parameter of κ = 3.2/rcut and an upper bound of the reciprocal space summation at Κmax = ⌈κLbox⌉. For computational efficiency, the 200 nm film was modeled as a weighted average of a 4 nm nanosheet (two unit cell thickness) and a defect-free bulk zeolite crystal. In doing this, therefore, separate simulations at each state point were performed for both the nanosheet (NS) and perfect crystal (PC). The PC was represented in a simulation box by replicating the unit cell60 by factors of 2, 2, and 3, in the a, b, and c-directions, respectively. The NS was represented in a simulation box by taking the same 2×2×3 supercell and extending the x-direction to 8 nm, i.e., the periodic simulation box contains the 4 nm NS and a 4 nm vapor region, where all undercoordinated surface Si atoms are converted to silanol groups (i.e., by appending hydroxyl groups). The host—guest interactions for a rigid, charge-neutral portion of the film were pretabulated using a grid spacing of 0.1 Å and used for interpolation during simulation.53,61 The silanol groups retained their dihedral rotation and angle bending degrees of freedom. The loadings at a given state point in the 200 nm film were calculated as a weighted average of those in the NS and PC. The weights for the NS and PC were determined by the amount of each that would be required to construct a 200 nm film with surface silanols, resulting in weights of 0.02 and 0.98, respectively. The isobaric—isothermal version of the Gibbs ensemble62,63 was employed with three simulation boxes to represent an adsorbent, a water gas-phase reservoir, and a butanol gas-phase reservoir. The gas-phase reservoirs had applied pressures corresponding to the desired partial pressures of water and butanol, and the reservoirs were implemented as ideal (i.e., intermolecular interactions were turned off, but intramolecular interactions were considered). In all simulations, the total number of water molecules was Nwater = 100. The bulk simulations and nanosheet


8

Page 9 of 30

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

Langmuir

simulations had a total of NButOH = 150 and NButOH = 250 molecules, respectively. The temperature of the system was either 35 or 50 °C. For the simulations probing adsorption on the NS, the number of adsorbed molecules was determined by subtracting the number of molecules that would be found in a 4×4×4 nm3 region with the vapor density from the total number of molecules found in the simulation box containing the NS. The Monte Carlo moves employed included volume moves on the butanol and water gasphase reservoirs, as well as rigid-body center-of-mass translations and rotations on butanol and water and coupled-decoupled configurational-bias Monte Carlo (CD-CBMC) moves64,65 for regrowth (i.e., changing conformation) of part of butanol molecules and of the hydrogen atoms belonging to silanol groups. CD-CBMC swap66,67 moves were implemented to improve sampling of particle transfers between simulation boxes exploring multiple positions and orientations for water and also internal degrees of freedom for butanol. Equilibration and production periods consisted of 100,000—200,000 and 100,000—300,000 Monte Carlo Cycles (MCCs). A MCC consists of Ntot randomly selected trial moves where Ntot is the sum of Nwater, NButOH, and the number of silanol groups. 8 independent simulations were carried out at each state point. Statistical uncertainties are reported as the 95% confidence intervals estimated by multiplying the standard error of the mean with a factor of 2.4.

RESULTS AND DISCUSSION The silicalite-1 films used in this work were characterized by scanning electron microscopy (SEM) and X-ray diffraction (XRD) previously.38,39 However, a short summary of the characterization is given here. The XRD data confirmed the MFI phase of the silicalite-1 films, and that the films were composed of b-oriented crystals. Observations by SEM also support the XRD data and showed that ATR crystals were covered by uniform b-oriented silicalite-1 films


9

Langmuir

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 10 of 30

with average thicknesses of 750 nm and 200 nm for the silicalite-1(OH⁻) and silicalite-1(F⁻) films respectively. Moreover,

29

Si MAS NMR spectroscopy revealed that the density of

structural defects in the form of silanol groups was significantly lower in the silicalite-1(F⁻) sample compared to the silicalite-1(OH⁻) sample.37,39


10

Page 11 of 30

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

Langmuir

Figure 1 shows the IR spectra of butanol and water adsorbed in the two silicalite-1 films at 35 °C from binary vapor mixtures (red and blue) as well as reference spectra obtained from the singlecomponent adsorption of butanol and water in silicalite-1(OH⁻) (black, top and bottom, respectively). The spectra were recorded when the silicalite-1 films were exposed to a binary vapor mixture with 0.09 kPa and 0.79 kPa partial pressures of butanol and water, respectively (corresponding to a butanol mole fraction of ~ 0.1). The reference spectra were recorded at partial pressures close to those in the mixture. In the reference spectrum for pure butanol adsorbed in the film, the characteristic CH3 and CH2 stretching vibration bands between 3000 and 2800 cm-1 are evident, and the bands at 1380 and 1465 cm−1 were assigned to the CH3 and CH2 bending vibrations, respectively.68 In the wavenumber region above 3100 cm-1 two bands appear—these are assigned to the stretching vibration of the OH function of butanol. The two bands show maximum intensities at 3620 cm-1 and 3445 cm-1. The peak intensity of the latter band is observed near 3300 cm-1 in spectra of pure liquid butanol due to intermolecular hydrogen bonding, whilst a band at 3644 cm-1 of liquid butanol in a dilute CCl4 solution has been assigned to non-hydrogen bonded butanol monomers.69 It may be noticed that these two bands remain when butanol/water vapor mixtures are adsorbed in a silicalite-1(OH⁻) film as well as in a silicalite-1(F⁻) (Figure 1) indicating that butanol is either hydrogen-bonded to silanol groups in the two silicalite-1 films or to other adsorbates.43 In the reference spectrum of adsorbed water, there is a band at ca. 1620 cm−1 assigned to the bending vibration of water, whereas the broad band in the 3700-2700 cm−1 region with a peak maximum at about 3370 cm-1 is assigned to the O-H stretching vibration of adsorbed hydrogenbonded water.70 This band shows up as a shoulder in the spectra of butanol and water adsorbed from the binary mixtures (Figure 1, middle traces). In the spectra recorded from the binary


11

Langmuir

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 30

adsorption experiments, both of the characteristic band of water at 1620 cm−1 and the band assigned to butanol at 1465 cm−1 are visible and clearly separated. As may be expected, the intensity of the bands assigned to adsorbed water were lower in the spectra of the binary mixture compared to the reference spectrum of pure water adsorption at similar partial pressures because of strong competitive adsorption of butanol during mixture adsorption. The intensity of the bands in the 3000-2800 cm-1 region assigned to adsorbed butanol were lower for silicalite-1(F⁻) compared to those for silicalite-1(OH⁻). This difference between the intensities is due to the slightly higher butanol uptake of the silicalite-1(OH⁻) film, which is most likely the result of a slightly larger micropore volume in silicalite-1 (OH⁻) compared to silicalite-1(F⁻) due to the higher density of internal defects in the former sample.37,39


12

Page 13 of 30

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

Langmuir

0.2 AU

3800

3300

2800

2300

1800

1300

-1

Wavenumber (cm ) Figure 1. Infrared spectra of butanol and water adsorbed from a binary mixture in the silicalite-1(F⁻) (red, second from bottom) and in the silicalite-1(OH⁻) (blue, second from top) films at 35 °C with a mole fraction of butanol of 0.10 in the vapor phase, together with reference spectra of pure component butanol (top) and water (bottom) adsorbed in silicalite-1(OH⁻). Total pressure was 1 atm with helium as balance. (Because the film thicknesses were different, the spectra were normalized).

The concentrations of water and butanol adsorbed from the binary mixtures were extracted from the infrared spectra by integrating the area of the water deformation band at 1620 cm−1 or the characteristic CH3 and CH2 stretching vibration bands at 3000-2800 cm-1 using the method presented in our previous work (and in the Supporting information).38,39 Figures 2 a and b show the concentrations of water adsorbed from butanol/water binary mixtures in silicalite-1(F⁻) and silicalite-1 (OH⁻) films, respectively. The total pressure of the feed was always 1 atm, and the partial pressure of water was held constant at either PW1 = 0.79 kPa or PW2 = 1.9 kPa while the partial pressure of butanol was increased step by step. Considering that concentration of butanol


13

Langmuir

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 30

in a fermentation broth is usually low,7 a maximum butanol partial pressure of 0.12 kPa (in the binary mixtures) was applied in our measurements. For both sililcalite-1 films, the concentration of water in the adsorbed phase decreased with increasing partial pressure of butanol in the feed. However, at 50 °C and a partial pressure of water of 0.79 kPa (PW1), the concentration of adsorbed water was almost constant for both films which is probably due to the very low concentration of water in the adsorbed phase. In the single-component adsorption results reported in our previous work,39 it was shown that water adsorption was very low in the silicalite-1(F⁻) sample, comparable to the lowest reported value for silicalite-1. Consequently, a very low adsorption of water from the binary mixtures was expected for the sililcalite-1(F⁻) film. The concentration of water adsorbed in the silicalite-1 (OH⁻) film from binary mixtures was also lower than the single-component water adsorption.38 However, it was significantly higher than the concentration of adsorbed water in the silicalite-1 (F⁻) film. The silicalite-1(F) film has less structural defects than the silicalite-1(OH-) sample resulting in less adsorption of water in the former. Moreover, the external surface of the silicalite-1(F-) film is more hydrophobic than the external surface of the silicalite-1(OH-) sample as evidenced by water contact angle measurements, 85º and 63º respectively as reported in our previous work. [39] This most likely also contributed to a higher butanol selectivity being obtained for the silicalite-1(F-) film than for the silicalite-1(OH-) film. For example, Kosinov et al. showed that zeolite MFI and MEL membranes with the external surface modified such that some of the external silanol groups were preplaced with Si-F moieties showed higher ethanol/water selectivities than the as –prepared samples with only silanol groups at the external surface.71


14

Page 15 of 30

a) Water Loading (mmol/g)

35 οC & PW2

0.30

50 οC & PW2

0.025

0.25

35 οC & PW1 50 οC & PW1

0.020

0.20 0.015

0.15 0.010

0.10 0.00

0.03

0.06

0.09

0.12

0.05 0.00 0.00

0.02

0.04

0.06

0.08

0.10

0.12

ButOH P (kPa)

b) Water Loading (mmol/g)

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

Langmuir

35 οC & PW2

0.30

50 οC & PW2 35 οC & PW1

0.25

50 οC & PW1

0.20 0.15 0.10 0.05 0.00 0.00

0.02

0.04

0.06

0.08

0.10

0.12

ButOH P (kPa) Figure 2. Concentration of water adsorbed from butanol/water binary vapor mixtures in silicalite-1(F⁻) (a) and silicalite-1(OH⁻) (b) films as a function of partial pressure of butanol in the feed at 35 °C and 50 °C and at two constant partial pressures of water in the feed (PW1 = 0.79 kPa and PW2 = 1.9 kPa).


15

Langmuir

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 30

The adsorption isotherms for butanol in the two films from the binary vapor mixtures are presented in Figure 3 and are compared with the single-component butanol adsorption isotherms determined at the same experimental conditions. These isotherms show that the presence of water in the binary mixtures had a smaller effect on the adsorption of butanol in the silicalite1(OH⁻) film than in the sililcalite-1(F⁻) film. Almost the same amount of butanol was adsorbed from binary mixtures as from pure butanol in the silicalite-1(OH⁻) when comparing at the same partial pressures of butanol in the feed. However, for the silicalite-1(F⁻) film, the amount of butanol adsorbed from the binary mixtures were on an average 10-15 % lower than when adsorbed from a pure butanol feed. This difference between the two films is probably due to the contribution of defects in adsorption of butanol in silicalite-1(OH⁻) where butanol molecules may be hydrogen bonded to water molecules adsorbed on the silanol groups or directly to silanol groups.


16

Page 17 of 30

ButOH Loading (mmol/g)

a) 1.8 1.6 1.4 1.2 Single at 35 oC PW1 & 35 oC

1.0 0.8

PW2 & 35 oC

0.6

Single at 35 oC PW1 & 35 oC

0.4 0.2 0.00

PW2 & 35 oC

0.02

0.04

0.06

0.08

0.10

0.12

ButOH P (kPa)

b) ButOH Loading (mmol/g)

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

Langmuir

1.8 1.6 1.4 1.2 Single at 50 oC PW1 & 50 oC

1.0 0.8

PW2 & 50 oC

0.6

Single at 50 oC PW1 & 50 oC

0.4 0.2 0.00

PW2 & 50 oC

0.02

0.04

0.06

0.08

0.10

0.12

ButOH P (kPa) Figure 3. The concentration of butanol adsorbed from butanol/water binary vapor mixtures (symbols) in silicalite1(F⁻) (blue) and silicalite-1(OH⁻) (black) films at 35 °C (a) and 50 °C (b) and at two constant partial pressures of water in the feed (PW1 = 0.79 kPa and PW2 = 1.9 kPa) together with the butanol single-component isotherms (solid lines) at 35 °C and 50 °C


17

Langmuir

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 30

For adsorption from binary vapor mixtures, the butanol/water adsorption selectivity can be defined as:

α BuOH H

2O

=

X BuOH X H 2O

(3)

YBuOH YH 2 O

where X is the mole fraction of an adsorbate in the adsorbed phase and Y is the mole fraction of an adsorbate in the vapor phase at equilibrium with the adsorbed phase. Figure 4 shows the adsorption selectivities of butanol as a function of mole fraction and partial pressure of butanol in the feed for silicalite-1(OH⁻) and silicalite-1(F⁻) films at two constant partial pressures of water (PW1 in Figure 4a and PW2 in Figure 4b) at 35 oC and 50 oC. Both silicalite-1 samples are very selective towards butanol. However, the silicalite-1(F⁻) film showed significantly higher butanol selectivity values than the silicalite-1(OH⁻) film as expected from the single-component adsorption measurements.38,39 In general, the selectivity for the silicalite-1 (F⁻) film was ca 10 times higher than the corresponding values for the silicalite-1(OH⁻) film. The highest selectivity of ~23000 was obtained at 50 °C and partial pressures of water and butanol in the feed of 1.87 kPa and 0.23 kPa, respectively. For all experiments, the selectivity decreases with increasing partial pressures of butanol. This is mainly due to the shape of the isotherm where at these quite high-adsorbed loadings of butanol, the concentration of adsorbed butanol is approaching its saturation value, and therefore the concentration of butanol adsorbed does not change as fast as the partial pressure of butanol in the feed. At the same time, the adsorption of water is more sensitive to temperature than butanol,38,39 thus higher selectivity values were obtained for both of the silicalite-1 films at 50 °C compared to 35 °C when comparing at the same feed composition.


18

Page 19 of 30

a) ButOH P (kPa) 0.00

0.02

ButOH Selectivity

0.04

0.05

0.07

0.09

0.11

0.13

F- film, 50 oC F- film, 35 oC OH- film, 50 oC OH- film, 35 oC

Pwater = PW1

10000

1000

100 0

2

4

6

8

10

12

14

0.12

0.14

ButOH Mole Fraction (%)

b) ButOH P (kPa) 0.00

0.02

0.04

0.06

0.08

0.10

F- film, 50 oC F- film, 35 oC OH- film, 50 oC OH- film, 35 oC

Pwater = PW2

10000

ButOH Selectivity

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

Langmuir

1000

100 0

1

2

3

4

5

6

7

ButOH Mole Fraction (%) Figure 4. Butanol/water adsorption selectivities as a function of mole fraction and partial pressure of butanol in the gas phase at two constant partial pressure of water of PW1 = 0.79 kPa (a) and PW2 = 1.9 kPa (b) at 35°C and 50 °C.


19

Langmuir

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 30

Figure 5 compares the adsorption selectivities for the silicalite-1(F-) film determined by experiment and simulation. The difference in selectivity between the two methods can be explained by the difference in loading both of butanol and of water at a given state point (see Figures S1 and S2 in the Supplementary Information) of the binary isotherms. The simulated loading of butanol is greater than the experimental loading by a value nearly constant at the same Pw and T. The water isotherms observed for the silicalite-1(F-) film by simulation and experiment, on the other hand, have much different shapes. At high partial pressures of butanol, the water loading observed by simulation is the same or lower than that of experiment. In combination with the higher butanol loading, this leads to a higher selectivity for simulation than experiment at high partial pressures of butanol. As the partial pressure of butanol is decreased, the water loading from simulation increases more than that from experiment, resulting in an experimental selectivity that increases more than simulation with decreasing partial pressure. With increasing temperature at constant PW, the simulated loadings of butanol and water at a given state point decrease slightly. This results in simulation having a selectivity that decreases with increasing temperature at low partial pressure of butanol, but becomes insignificant at larger partial pressures due to a large uncertainty in the loading of water. This joint approach with experiment and simulation is useful in aiding the understanding of binary mixture adsorption.


20

Page 21 of 30

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

Langmuir

Figure 5. Adsorption selectivity of butanol over water as a function of butanol partial pressure, determined by simulation and experiment for a 200 nm silicalite-1(F-) film. The simulated isotherms at PW1 = 0.79 kPa and T = 35 °C, PW2 = 1.9 kPa and T = 35 °C, PW1 = 0.79 kPa and T = 50 °C, as well as PW2 = 1.9 kPa and T = 50 °C are represented by cyan down triangles, magenta up triangles, blue left triangles, and red right triangles, respectively. For comparison, the experimental isotherms (shown also in Figure 4) are depicted with the same colors as simulation but are represented as solid lines connecting neighboring points.

Along this route, the molecular simulations are used to shed some light on the local selectivity on a surface and in the bulk of a silicalite-1 film. Positions of the butanol and water oxygens, obtained from simulation snapshots, are used to obtain a two-dimensional profile for the adsorption selectivity of butanol over water at PW = PW2 and PB = 0.0043 kPa and 35°C. These are shown for the simulated nanosheet and perfect crystal in Figure 6a and Figure 6b, respectively. The highest selectivity in both cases is observed near the intersections (as opposed to the channels, showing a markedly lower selectivity), which is likely due to increased availability for hydrogen bonding in the former region. Figure 6c shows that the nanosheet and perfect crystal share a peak in the histogram of their local selectivity profiles (colors) at ln(S) of around 9 (corresponding to S of 8000), which occurs in the channels. Since only one peak occurs in the perfect crystal, this peak is not specific to the zig-zag or straight channels, implying no


21

Langmuir

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 30

strong preference for the channel topology at this loading. The nanosheet, unlike the perfect crystal, has a second peak at a lower selectivity of ln(S) around 4 (corresponding to a selectivity of 50). This less selective peak is colored blue and is located at the nanosheet surface. Interestingly, a local selectivity of even 100 (corresponding to ln(S) of 4.6) at this state point corresponds to a local composition of 18 mol % butanol, implying that the nanosheet surface is quite enriched in water.

Figure 6. Two-dimensional color maps and histogram showing selectivity of butanol oxygen over water oxygen for simulated (a) nanosheet (NS) and (b) perfect crystal (PC) at PW = PW2 and PB = 0.0043 kPa and T = 35°C. The selectivity was computed using the average number densities of oxygen atoms in rectangular cuboids with dimensions of a/20, b/19, and 3c (where a, b, and c are the unit cell dimensions of MFI). Volume rendering was performed with splatting.72 The orientations depicted for both (a) and (b) are viewed in the positive z-direction, with straight channels and zig-zag channels horizontal and vertical, respectively, to the plane-of-view. Regions of no color (white) correspond to regions of low density, where the local density of the oxygen position of either component is less than 0.015 molec / nm3.(c) Color-bar legend with histogram of the natural logarithm of selectivity values for NS and PC. The histogram values for NS and PC were obtained from those shown as colors (i.e., not including the low-density regions shown in white) in (a) and (b), respectively, and were normalized to yield an integral of unity.

CONCLUSIONS In situ ATR-FTIR spectroscopy was successfully used for measuring binary adsorption of butanol and water in silicalite-1 films quantitatively. Adsorbed concentrations and selectivities


22

Page 23 of 30

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

Langmuir

were determined at different feed compositions and two temperatures for two different silicalite1 films with different adsorption properties. Both films were highly butanol selective with, butanol/water selectivities varying between ca 100-23000. The silicalite-1(F⁻) film showed excellent selectivities due to its low density of internal defects giving it a hydrophobic nature with very small amounts of water adsorbed. Force-field based simulations provided a molecularlevel interpretation of the selectivity of butanol over water. The presence of defects in the lattice structure, in combination with partial exchange of external silanol groups by Si-F moieties, were found to be the main reasons for difference in hydrophobic property and consequently the butanol adsorption selectivity of silicalite-1 films studied in the present work.

ACKNOWLEDGMENT The authors at Luleå University of Technology acknowledge the Swedish Research Council (VR, under Grant 2011-4060) for financially supporting this work. We acknowledge the Foundation in memory of J. C. and Seth M. Kempe for funding a Bruker IFS 66v/S FTIR spectrometer at LTU. The molecular simulation work was supported as part of the Catalysis Center for Energy Innovation, an Energy Frontier Research Center funded by the U.S. Department of Energy, Office of Science, Basic Energy Sciences, under Award DE-SC0001004. RFD and JIS thank the Minnesota Supercomputing Institute for part of the computer resources used in this work.


23

Langmuir

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 24 of 30

REFERENCES [1] D.C. Rakopoulos, C.D. Rakopoulos, R.G. Papagiannakis, and D.C. Kyritsis. Combustion heat release analysis of ethanol or n-butanol diesel fuel blends in heavy-duty DI diesel engine. Fuel, 90 (2011) 1855. [2] J. Li, X. Chen, B. Qi, J. Luo, X. Zhuang, Y. Su, Y. Wan. Continuous Acetone, Butanol, Ethanol (ABE) Fermentation with in Situ Solvent Recovery by Silicalite-1 Filled PDMS/PAN Composite Membrane. Energy Fuels, 28 (2014) 555. [3] A.B. Thompson, S. Cope J., T.D. Swift, and J.M. Notestein. Adsorption of n-Butanol from Dilute Aqueous Solution with Grafted Calixarenes. Langmuir, 27 (2011) 11990. [4] A. Oudshoorn, L.A.M. van der Wielen, and A.J.J. Straathof. Assessment of Options for Selective 1-Butanol Recovery from Aqueous Solution. Ind Eng Chem Res, 48 (2009) 7325. [5] T.C. Ezeji, N. Qureshi, and H.P. Blaschek. Acetone butanol ethanol (ABE) production from concentrated substrate: reduction in substrate inhibition by fed-batch technique and product inhibition by gas stripping. Appl.Microbiol.Biotechnol., 63 (2004) 653. [6] I.S. Maddox, N. Qureshi, and K. Roberts-Thomson, Production of acetone-butanol-ethanol from concentrated substrate using clostridium acetobutylicum in an integrated fermentationproduct removal process, Process Biochem., 30 (1995) 209. [7] N. Qureshi, H.P. Blaschek. Recovery of butanol from fermentation broth by gas stripping. Renewable Energ., 22 (2001) 557. [8] Y. Long, H. Jiang, and H. Zeng. Sorbate/Framework and Sorbate/Sorbate Interaction of Organics on Siliceous MFI Type Zeolite. Langmuir, 13 (1997) 4094. [9] T.C. Bowen, L.M. Vane. Ethanol, Acetic Acid, and Water Adsorption from Binary and Ternary Liquid Mixtures on High-Silica Zeolites. Langmuir, 22 (2006) 3721. [10] N. Abdehagh, F.H. Tezel, and J. Thibault. Separation techniques in butanol production: Challenges and developments. Biomass Bioenergy, 60 (2014) 222. [11] N. Abdehagh, P. Gurnani, F.H. Tezel, and J. Thibault. Adsorptive separation and recovery of biobutanol from ABE model solutions. Adsorption, 21 (2015) 185. [12] N. Abdehagh, F.H. Tezel, and J. Thibault. Multicomponent adsorption modeling: isotherms for ABE model solutions using activated carbon F-400. Adsorption, 22 (2016) 357. [13] J. Cousin Saint Remi, G. Baron, and J. Denayer. Adsorptive separations for the recovery and purification of biobutanol. Adsorption, 18 (2012) 367.


24

Page 25 of 30

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

Langmuir

[14] J. Cousin Saint Remi, T. Rémy, V. Van Hunskerken, S. van de Perre, T. Duerinck, M. Maes, D. De Vos, E. Gobechiya, C.E.A. Kirschhock, G.V. Baron, J. Denayer. Biobutanol Separation with the Metal–Organic Framework ZIF-8. ChemSusChem, 4 (2011) 1074. [15] K. Zhang, R.P. Lively, C. Zhang, R.R. Chance, W.J. Koros, D.S. Sholl, S. Nair. Exploring the Framework Hydrophobicity and Flexibility of ZIF-8: From Biofuel Recovery to Hydrocarbon Separations. J.Phys.Chem.Lett., 4 (2013) 3618. [16] M. Trzpit, M. Soulard, J. Patarin, N. Desbiens, F. Cailliez, A. Boutin, I. Demachy, A.H. Fuchs. The Effect of Local Defects on Water Adsorption in Silicalite-1 Zeolite: A Joint Experimental and Molecular Simulation Study. Langmuir, 23 (2007) 10131. [17] H. Zhang, D. Liu, Y. Yao, B. Zhang, and Y.S. Lin. Stability of ZIF-8 membranes and crystalline powders in water at room temperature. J.Membr.Sci., 485 (2015) 103. [18] D.H. Olson, W.O. Haag, and W.S. Borghard. Use of water as a probe of zeolitic properties: interaction of water with HZSM-5. Micropor. Mesopor. Mater., 35–36 (2000) 435. [19] V. Bolis, C. Busco, and P. Ugliengo. Thermodynamic Study of Water Adsorption in HighSilica Zeolites. J Phys Chem B, 110 (2006) 14849. [20] A. Faisal, A. Zarebska, P. Saremi, D. Korelskiy, L. Ohlin, U. Rova, J. Hedlund, M. Grahn. MFI zeolite as adsorbent for selective recovery of hydrocarbons from ABE fermentation broths. Adsorption, 20 (2014) 465. [21] R.F. DeJaco, P. Bai, M. Tsapatsis, and J.I. Siepmann. Adsorptive Separation of 1-Butanol from Aqueous Solutions Using MFI- and FER-Type Zeolite Frameworks: A Monte Carlo Study. Langmuir, 32 (2016) 2093. [22] A. Oudshoorn, L.A.M. van der Wielen, and A.J.J. Straathof. Adsorption equilibria of biobased butanol solutions using zeolite. Biochem.Eng.J., 48 (2009) 99. [23] V. Saravanan, D.A. Waijers, M. Ziari, and M.A. Noordermeer. Recovery of 1-butanol from aqueous solutions using zeolite ZSM-5 with a high Si/Al ratio; suitability of a column process for industrial applications. Biochem.Eng.J., 49 (2010) 33. [24] B. Sowerby, B.D. Crittenden, Vapour phase separation of alcohol-water mixtures by adsorption onto silicalite, Gas Sep. Purif., 2 (1988) 177. [25] C. Xue, F. Liu, M. Xu, I. Tang, J. Zhao, F. Bai, S.-T. Yang. Butanol production in acetonebutanol-ethanol fermentation with in situ product recovery by adsorption. Bioresour.Technol., 219 (2016) 158. [26] P. Bai, M. Tsapatsis, and J.I. Siepmann. Multicomponent adsorption of alcohols onto silicalite-1 from aqueous solution: Isotherms, structural analysis, and assessment of ideal adsorbed solution theory. Langmuir, 28 (2012) 15566.


25

Langmuir

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 30

[27] D. Korelskiy, T. Leppäjärvi, H. Zhou, M. Grahn, J. Tanskanen, and J. Hedlund. High flux MFI membranes for pervaporation. J.Membr.Sci., 427 (2013) 381. [28] L. Yu, D. Korelskiy, M. Grahn, and J. Hedlund. Very high flux MFI membranes for alcohol recovery via pervaporation at high temperature and pressure. Sep. Purif. Techn., 153 (2015) 138. [29] N. Qureshi, M.M. Meagher, and R.W. Hutkins. Recovery of butanol from model solutions and fermentation broth using a silicalite/silicone membrane1. J.Membr.Sci., 158 (1999) 115. [30] F. Liu, L. Liu, and X. Feng. Separation of acetone–butanol–ethanol (ABE) from dilute aqueous solutions by pervaporation. Sep. Purif. Techn., 42 (2005) 273. [31] Z. Qin, L. Lakiss, L. Tosheva, J.-. Gilson, A. Vicente, C. Fernandez, V. Valtchev. Comparative Study of Nano-ZSM-5 Catalysts Synthesized in OH − and F− Media Adv. Funct. Mater., 24 (2014) 257. [32] M. Hunger, J. Karger, H. Pfeifer, J. Caro, B. Zibrowius, M. Bulow, R. Mostowicz. Investigation of internal silanol groups as structural defects in ZSM-5-type zeolites. J.Chem.Soc. Farad. T.1, 83 (1987) 3459. [33] M. Hunger, D. Freude, H. Pfeifer, and W. Schwieger. MAS NMR studies of silanol groups in zeolites ZSM-5 synthesized with an ionic template. Chemical Physics Lett. 167 (1990) 21. [34] M.E. Dose, K. Zhang, J.A. Thompson, J. Leisen, R.R. Chance, W.J. Koros, B. Mc Cool, R.P. Lively. Effect of Crystal Size on Framework Defects and Water Uptake in Fluoride Mediated Silicalite-1. Chem.Mater., 26 (2014) 4368. [35] K. Zhang, R.P. Lively, J.D. Noel, M.E. Dose, B.A. McCool, R.R. Chance, W.J. Koros. Adsorption of Water and Ethanol in MFI-Type Zeolites. Langmuir, 28 (2012) 8664. [36] E.M. Flanigen, R.L. Patton, Silica polymorph and process for preparing same. , Patent US4073865 A, 1978. [37] H. Zhou, J. Mouzon, A. Farzaneh, O. Antzutkin, M. Grahn, and J. Hedlund. Colloidal defect-free silicalite-1 single crystals: preparation, structure characterization, adsorption and separation properties for alcohol/water mixtures. Langmuir, 31 (2015) 8488. [38] A. Farzaneh, M. Zhou, E. Potapova, Z. Bacsik, L. Ohlin, A. Holmgren, J. Hedlund, M. Grahn. Adsorption of Water and Butanol in Silicalite-1 Film Studied with in Situ Attenuated Total Reflectance-Fourier Transform Infrared Spectroscopy. Langmuir, 31 (2015) 4887. [39] A. Farzaneh, M. Zhou, O.N. Antzutkin, Z. Bacsik, J. Hedlund, A. Holmgren, J. Hedlund, M. Grahn. Adsorption of Butanol and Water Vapors in Silicalite-1 Films with a Low Defect Density. Langmuir, 32 (2016) 11789.


26

Page 27 of 30

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

Langmuir

[40] A.L. Myers, J.M. Prausnitz. Thermodynamics of mixed-gas adsorption. AIChE J., 11 (1965) 121. [41] D.W. Hand, S. Loper, M. Ari, and J.C. Crittenden. Prediction of multicomponent adsorption equilibria using ideal adsorbed solution theory. Environ.Sci.Technol., 19 (1985) 1037. [42] R. Krishna, J.M. van Baten. Highlighting pitfalls in the Maxwell-Stefan modeling of wateralcohol mixture permeation across pervaporation membranes. J.Membr.Sci., 360 (2010) 476. [43] R. Krishna, J.M. van Baten. Hydrogen Bonding Effects in Adsorption of Water & Alcohol Mixtures in Zeolites and the Consequences for the Characteristics of the Maxwell-Stefan Diffusivities. Langmuir, 26 (2010) 10854. [44] J. Gutierrez-Sevillano, S. Calero, and R. Krishna. Separation of benzene from mixtures with water, methanol, ethanol, and acetone: highlighting hydrogen bonding and molecular clustering influences in CuBTC. Phys.Chem.Chem.Phys., 17 (2015) 20114. [45] R. Krishna, J.M. van Baten, Using molecular simulations for screening of zeolites for separation of CO2/CH4 mixtures, Chem. Eng. J., 133 (2007) 121. [46] M. Coppens, A.T. Bell, and A.K. Chakraborty, Dynamic Monte-Carlo and mean-field study of the effect of strong adsorption sites on self-diffusion in zeolites, Chem. Eng. Sci., 54 (1999) 3455. [47] J.Z. Yang, Q.L. Liu, and H.T. Wang. Analyzing adsorption and diffusion behaviors of ethanol/water through silicalite membranes by molecular simulation. J.Membr.Sci., 291 (2007) 1. [48] M. Grahn, A. Holmgren, and J. Hedlund. Adsorption of n-Hexane and p-Xylene in Thin Silicalite-1 Films Studied by FTIR/ATR Spectroscopy. J.Phys.Chem.C, 112 (2008) 7717. [49] M. Grahn, Z. Wang, M. Lidström-Larsson, A. Holmgren, J. Hedlund, and J. Sterte. Silicalite-1 coated ATR elements as sensitive chemical sensor probes. Micropor. Mesopor. Mater., 81 (2005) 357. [50] L. Ohlin, P. Bazin, F. Thibault-Starzyk, J. Hedlund, and M. Grahn. Adsorption of CO2, CH4, and H2O in zeolite ZSM-5 studied using in situ ATR-FTIR spectroscopy. J. Phys. Chem. C, 117 (2013) 16972. [51] H.G. Tompkins. The Physical Basis for Analysis of the Depth of Absorbing Species Using Internal Reflection Spectroscopy. Appl.Spectrosc., 28 (1974) 335. [52] F.M. Mirabella, Internal reflection spectroscopy: Theory and applications, CRC Press, 1992.


27

Langmuir

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 30

[53] P. Bai, M.Y. Jeon, L. Ren, C. Knight, M.W. Deem, M. Tsapatsis, and J. I. Siepmann. Discovery of optimal zeolites for challenging separations and chemical transformations using predictive materials modeling. Nature Commun., 6 (2015) 5912. [54] W.L. Jorgensen, J. Chandrasekhar, J.D. Madura, R.W. Impey, and M.L. Klein. Comparison of simple potential functions for simulating liquid water. J.Chem.Phys., 79 (1983) 926. [55] B. Chen, J.J. Potoff, and J.I. Siepmann. Monte Carlo Calculations for Alcohols and Their Mixtures with Alkanes. Transferable Potentials for Phase Equilibria. 5. United-Atom Description of Primary, Secondary, and Tertiary Alcohols. J Phys Chem B, 105 (2001) 3093. [56] P. Bai, M. Tsapatsis, and J.I. Siepmann. TraPPE-zeo: Transferable Potentials for Phase Equilibria Force Field for All-Silica Zeolites. J.Phys.Chem.C, 117 (2013) 24375. [57] J.L. Rafferty, J.I. Siepmann, and M.R. Schure. Influence of bonded-phase coverage in reversed-phase liquid chromatography via molecular simulation: I. Effects on chain conformation and interfacial properties. J. Chromatogr. A, 1204 (2008) 11. [58] G. Maitland, M. Rigby, E. Smith, and W.A. Wakeham, Intermolecular Forces: Their Origin and Determination, Oxford University Press, 1987. [59] D.J. Tildesley and M.P. Allen, Computer Simulation of Liquids, Oxford, Clarendon, 1987. [60] H. Van Koningsveld, H. Van Bekkum, and J.C. Jansen. On the location and disorder of the tetrapropylammonium (TPA) ion in zeolite ZSM-5 with improved framework accuracy. Acta Crystallogr. Sect. B, 43 (1987) 127. [61] R.L. June, A.T. Bell, and D.N. Theodorou. Prediction of low occupancy sorption of alkanes in silicalite. J.Phys.Chem., 94 (1990) 1508. [62] A.Z. Panagiotopoulos, N. Quirke, M. Stapleton, and D.J. Tildesley. Phase equilibria by simulation in the Gibbs ensemble. Mol.Phys., 63 (1988) 527. [63] A.Z. Panagiotopoulos. Direct determination of phase coexistence properties of fluids by Monte Carlo simulation in a new ensemble. Mol.Phys., 61 (1987) 813. [64] G.- C.- A.- M. Mooij, D. Frenkel and B. Smit. Direct simulation of phase equilibria of chain molecules. J. Phys.: Condens. Mat., 4 (1992) L255. [65] J.I. Siepmann, D. Frenkel. Configurational bias Monte Carlo: a new sampling scheme for flexible chains. Mol.Phys., 75 (1992) 59. [66] M.G. Martin and J.I. Siepmann. Novel Configurational-Bias Monte Carlo Method for Branched Molecules. Transferable Potentials for Phase Equilibria. 2. United-Atom Description of Branched Alkanes. J Phys Chem B, 103 (1999) 4508.


28

Page 29 of 30

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

Langmuir

[67] J.J. De Pablo, Simulation of phase equilibria for chain molecules, Fluid Phase Equil., 104 (1995) 195. [68] V.D. Spasojević, B.D. Djordjević, S.P. Šerbanović, I.R. Radović, and M.L. Kijevčanin. Densities, Refractive Indices, Viscosities, and Spectroscopic Study of 1-Amino-2-propanol + 1Butanol and + 2-Butanol Solutions at (288.15 to 333.15) K. J. Chem. Eng. Data, 59 (2014) 1817. [69] A.V. Stuart, G.B.B.M. Sutherland. Effect of Hydrogen Bonding on the Deformation Frequencies of the Hydroxyl Group in Alcohols. J. Chem. Phys., 24 (1956) 559. [70] A. Jentys, G. Warecka, M. Derewinski, and J.A. Lercher. Adsorption of water on ZSM 5 zeolites. J. Phys. Chem. 93 (1989) 4837. [71] N. Kosinov, V.G.P. Sripathi, and E.J.M. Hensen, Improving separation performance of high-silica zeolite membranes by surface modification with triethoxyfluorosilane, Micropor. Mesopor. Mater. 194 (2014) 24. [72] H. Childs, E. Brugger, B. Whitlock, J. Meredith, S. Ahern, D. Pugmire, K. Biagas, M. Miller, C. Harrison, G.H. Weber, H. Krishnan, T. Fogal, A. Sanderson, C. Garth, E.W. Bethel, D. Camp, O. Rubel, M. Durant, J.M. Favre, P. Navratil. VisIt: An End-User Tool For Visualizing and Analyzing Very Large Data., High Performance Visualization--Enabling Extreme-Scale Scientific Insight, Boca Raton, FL: Chapman & Hall, 2013, 357- 372.


29

Langmuir

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

158x109mm (120 x 120 DPI)


Page 30 of 30

Comparative Study of the Effect of Defects on ... - ACS Publications

Recommend Documents