Production of Composite Membranes by Coupling Coating and Melt

Jan 11, 2017 - Department of Chemical Engineering, Laval University, Quebec, Canada G1V 0A6. Ind. Eng. Chem. Res. , 2017, 56 (5), pp 1306–1315...
1 downloads 4 Views 6MB Size
Subscriber access provided by University of Newcastle, Australia

Article

Production of composite membranes by coupling coating and melt extrusion/salt leaching Kazem Shahidi, and Denis Rodrigue Ind. Eng. Chem. Res., Just Accepted Manuscript • DOI: 10.1021/acs.iecr.6b04362 • Publication Date (Web): 11 Jan 2017 Downloaded from http://pubs.acs.org on January 17, 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.

Industrial & Engineering Chemistry Research 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 43

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

Industrial & Engineering Chemistry Research

Production of composite membranes by coupling coating and melt extrusion/salt leaching Kazem Shahidi, Denis Rodrigue* Department of Chemical Engineering, Laval University, Quebec, Canada, G1V 0A6 *Corresponding author Tel.: 1 (418) 656-2903; Fax: 1 (418) 656-5993. E-mail address: [email protected]

1 ACS Paragon Plus Environment

Industrial & Engineering Chemistry Research

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 43

Abstract

This work presents a simple and efficient method to produce a flat sheet composite membrane (FSCM) with minimum resistance in the support layer. The process is also low cost and uses a limited amount of solvent (water). In particular, a series of polydimethylsiloxane/low-density

polyethylene

(PDMS/LDPE)

FSCM

were

produced

by coating an active PDMS layer on a microporous LDPE support via continuous extrusion and salt leaching (68% wt. NaCl) using immersion in hot water (50°C). The membranes were then characterized before and after leaching in terms of morphology, porosity and pore size distribution, as well as thermal properties. The results showed that the microporous structure is highly correlated to the salt used and the amount leached out from the polymer structure. The resulting FSCM were finally used for gas separation of C3H8 from CO2, CH4, N2, and H2 in terms of permeability, solubility and diffusivity. The results show that for an upstream pressure of 120 psi, the FSCM is about 6, 16, 23, and 53 times more permeable to C3H8 than CO2, CH4, H2, and N2 respectively, confirming the outstanding separation performance of these membranes for industrial applications.

Keywords:

Composite

membrane;

PDMS;

LDPE;

permeability; diffusivity.

2 ACS Paragon Plus Environment

salt

leaching;

solubility;

Page 3 of 43

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

Industrial & Engineering Chemistry Research

1. Introduction Composite membranes are usually fabricated by coating a permselective layer on a porous substrate which is prepared by phase inversion1 or non-solvent induced phase separation2

processes.

But

these

methodologies

use

large

amounts

of

harmful,

expensive, and flammable organic solvents which must be removed after the porous structure is

created

through

numerous

washing steps,

making this

approach

not

environmentally friendly and not economical (solvent recuperation and purification, health and safety hazards, etc.). These methods also have low production rates due to the slow kinetics of liquid-liquid phase separation3. For these reasons, solvent free technologies were developed. Very

few

solvent

free

approaches

for

the

fabrication

of

porous

structures

for

membranes application can be found in the literature4–6. The most interesting ones are probably the stretching and melt-spinning techniques which are based on the meltextrusion of neat semi-crystalline polymers to create the precursors, followed by axial stretching of these precursors to create a porous network. This method is only useful for semi-crystalline polymers, but following mechanical stretching, several thermal post-treatments

are

necessary

to

stabilize

the

crystalline

structure

and

prevent

membrane shrinkage7. Some development on the preparation of open-cell foams based on

polymers/fillers

polyethylene

compounds

(LDPE),

using

polypropylene

leachable (PP),

particles

(salts)

polysulfone,

with

low-density

polymethylmetacrylate,

polystyrene, and polyurethane rigid foams were presented8–11. Here, it is proposed to develop a hybrid continuous process to prepare flat sheet composite membranes (FSCM). 3 ACS Paragon Plus Environment

Industrial & Engineering Chemistry Research

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 43

Various polymers have been used for the fabrication of composite membranes as a permselective layer. To remove hydrocarbons from other gasses, PDMS is the most commonly used polymer for the separation of higher hydrocarbons from other gases. Lately, several investigations were devoted to the separation of O2, N2, H2, CO2, CH4, and C2-C4 olefins and paraffins using PDMS membranes12–18. To the best of the authors knowledge, this paper is the first report disclosing the preparation of a FSCM by combining coating with melt-extrusion/salt leaching. In this work, a selective PDMS layer was coated on a LDPE microporous support to continuously produce a FSCM. Then, the permeability and solubility of C3H8, CO2, CH4, N2, and H2 through the composite PDMS/LDPE membranes were determined at different feed pressures. According to the solution-diffusion theory, the concentrationaveraged diffusion coefficient can also be determined. Finally, the ideal selectivity of C3H8 with respect to other gasses was determined to show the efficiency of these membranes to remove hydrocarbons from light components.

2. Theory Gas permeation in a PDMS film can be approximated by the solution-diffusion model via permeability (P) defined as:

P=

Nl p2 − p1

(1)

4 ACS Paragon Plus Environment

Page 5 of 43

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

Industrial & Engineering Chemistry Research

where N is gas flux through a film with thickness l and a partial pressure difference (p2p1), p1 is the downstream pressure and p2 is the upstream pressure (p2 > p1). However, the diffusion can be modeled by Fick’s law as19,20:

N =−

Dloc  dC    (1 − ω)  dx 

(2)

where Dloc is the local diffusion coefficient, ω is the gas mass fraction in the polymer and x is the diffusion direction. Integration of the expression yields: 1 P= p 2 − p1

C2

∫ Deff

dC

C1

(3)

where C is the concentration in the polymer and Deff is the effective diffusion coefficient

(Deff

=

Dloc/(1-ω)).

If

the

diffusion

coefficient

is

independent

of

concentration, the integration gives:

P=

C2 − C1 Deff p2 − p1

(4)

But if the diffusion coefficient is a function of concentration, Deff is substituted by the  ). Finally, when the downstream pressure (permeate side) is averaged diffusivity ( negligible compared to the upstream pressure (feed side), Equation (4) simplifies to: P=SD

(5)

 and the solubility (S) should be evaluated at the feed conditions as: where D is Deff or 

S =

C p

(6) 5 ACS Paragon Plus Environment

Industrial & Engineering Chemistry Research

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

series

resistance

model

can

explain

the

component

Page 6 of 43

permeation

through

a

composite membrane as21:

 l MS l SL  N A = ( p2 − p1 ) MS + SL  PA   PA

−1

(7)

where NA is the flux of component A at steady-state. The superscripts MS and SL are associated to the microporous substrate and permselective layer, respectively. The ideal selectivity may be written as the ratio of the component fluxes at a given a pressure as21:

α A/ B =

NA NB

(8)

Therefore, the ideal selectivity of a gas for a composite membrane can be calculated as:

α A/ B

l MS / PBMS + l SL / PBSL = MS l / PAMS + l SL / PASL

(9)

From Equation (9), it can be concluded that the ideal selectivity of a gas for a composite membrane is based on the properties of both the substrate and the permselective layers. When the main resistance of a FSCM to component transport is in the selective layer (as in the present work), the terms lMS/PAMS and lMS/PBMS can be neglected to give:

α A/ B =

l SL / PBSL PASL = l SL / PASL PBSL

(10)

6 ACS Paragon Plus Environment

Page 7 of 43

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

Industrial & Engineering Chemistry Research

Equation (10) can also be expressed using Equation (5) to get a product of solubility selectivity and diffusivity selectivity as: S  D 

α A / B =  A  ×  A   S B   DB 

(11)

In addition to parameters such as pressure, temperature, and composition, the solubility strongly depends on the gas condensability. In general, solubility increases with increasing condensability. Since gas condensability is related to the normal boiling point and critical temperature of a compound22, the solubility selectivity increases with increasing

condensability

difference

between

both

gases.

Larger

molecules

having

higher normal boiling point and higher condensability have higher solubility than smaller

molecules.

So

diffusivity

diffusivity selectivity increases

drops

when

when

molecular

size

increases

the size difference between

both

and

the

molecules

increases, with the smaller molecule having higher diffusivity22. Therefore, a trade-off generally occurs between diffusivity selectivity and solubility selectivity, with the overall selectivity being related to the order of magnitude of these two terms.

3. Experimental 3.1. Material The polymer used for the support was low density polyethylene (LDPE) Novapol LA 0219-A (Nova Chemicals, Canada) with a density of 919 kg/m3 (ASTM D792) and a

7 ACS Paragon Plus Environment

Industrial & Engineering Chemistry Research

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 43

melt index of 2.3 g/10 min (ASTM D1238). NaCl with particles sizes of 5-55 µm and a density of 1981 kg/m3 (Soda-LO Extra Fine) was supplied by Tate & Lyle (USA). The polydimethylsiloxane (PDMS, Dehesive 944, density 0.91 g/cm3 at 20°C) was supplied by Wacker Silicones Corporation (USA). This PDMS is reported to be suitable

for

manufacturer's

coating

polyethylene

technical

hydrogenpolysiloxane) and

datasheet. a catalyst

and

polypropylene

Wacker

also

(Catalyst

films

provided

OL). Toluene,

a

according

to

cross-linker as

a solvent,

the (V24, was

purchased from Anachemia (Canada) with a purity of 99.5%. All the gasses, with purity above 99.0%, were purchased from Praxair (Canada).

3.2. Membrane fabrication The composite membranes were produced by combining a coating and a melt extrusion/salt leaching method (Fig. 1). In this study, a LDPE flat support was made using a co-rotating twin-screw extruder (Leistritz ZSE 27) with a L/D ratio of 40. In the first step (melt extrusion), LDPE pellets were fed in the first zone of the extruder (main feeder) at a rate of 5 g/min, while NaCl particles were fed in the third zone of the extruder (side feeder) at a rate of 10.5 g/min to produce blends containing 68% wt. of salt23. The temperature profile in the extruder was controlled at 135 °C for zone 1, 140 °C for zones 2-8, and 145 °C for zone 9, 10 and the die (zone 11). The extruder was operated at a constant mass flow rate of 15.5 g/min and a screw speed of 60 rpm. A flat die with dimensions of 15 cm width and 250 µm gap opening was used to form the melt into a flat sheet. The material was then introduced in a calendar with a drawing 8 ACS Paragon Plus Environment

Page 9 of 43

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

Industrial & Engineering Chemistry Research

speed of the take-up unit set at 16 cm/min. In the second step (coating), the PDMS coating solutions were prepared by dissolving an appropriate amount of PDMS polymer resin, cross-linking agent and catalyst (weight ratio of 10:1:1 (w/w/w)) in toluene to obtain a homogeneous 50% PDMS solution. The coating solution was poured onto the LDPE/NaCl sheet and cast with a roller to make a uniform layer. In the third step (salt leaching/solvent evaporation), the membranes were put in contact with the surface of a hot water bath (50 °C) to leach out the salt and evaporate the solvent of the coated layer. In the fourth step (curing), the membranes were kept for 48 h at room temperature to cure. In the last step, the membranes were placed in a vacuum oven for 2 h at 80 °C to complete PDMS cross-linking and remove the remaining solvent. For comparison, the LDPE porous support layer without coating was also produced. The porosity of the substrate (PO) was determined as:  ρ  ΡΟ (% ) =  1 − 1  × 100 ρ2  

(12)

where ρ1 and ρ2 are the densities of the neat LDPE and porous LDPE flat sheet calculated by the ratio of volume over weight.

3.3. Membrane characterization 3.3.1. Scanning electron microcopy (SEM) The morphology of the membrane was examined via scanning electron microscopy (SEM) using a JEOL JSM-840A (JEOL, Japan). Samples were first immersed in liquid 9 ACS Paragon Plus Environment

Industrial & Engineering Chemistry Research

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

nitrogen (60 s) and then fractured to obtain well-defined cross-sections. Finally, pore size distributions were calculated by analyzing the SEM images using the Image-J V1.50i software (National Institutes of Health (NIH), USA).

3.3.2. Thermogravimetric analysis (TGA) TGA analysis was performed using a TA Instruments (USA) model Q5000 IR for a temperatures range of 50 to 900 °C in a nitrogen atmosphere at a flow rate of 25 ml/min and a heating rate of 10 °C/min.

3.3.3. Fourier transform infrared spectroscopy (FTIR) FTIR analysis was performed using a Nicolet model 730 (Nicolet Instruments, USA) Fourier transform infrared analyzer in the reflectance mode (Golden-Gate accessories) between 500 and 3500 cm-1 with a 0.5 cm-1 resolution of 128 scans.

3.3.4. Sorption and permeation measurement The pressure decay method was used to determine the gas sorption isotherms24. Pure gas permeation was measured using a constant volume/variable pressure method25. More details on the procedure are presented as supporting information.

10 ACS Paragon Plus Environment

Page 10 of 43

Page 11 of 43

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

Industrial & Engineering Chemistry Research

4. Results and discussion 4.1. Morphology 4.1.1. Morphology of the salt As seen in Figure 2, SODA-LO contains NaCl and a small amount of an organic material introduced to prevent agglomeration26. The hollow salt particles have a structure composed of individual NaCl crystals linked together in each particle. Figure 2(b) presents the initial particle size distributions of the salt where the average is 32 microns with a standard deviation of 20 microns.

4.1.2. Morphology of the LDPE substrate Typical SEM images of the LDPE microporous substrate before and after salt leaching are presented in Figure 3. Figure 3(b) shows that the support has a microporous structure and the resistance to gas permeation across this layer can be assumed to be negligible. Figure 3(c) reports the salt particle size distribution salt inside LDPE where the average is 5 microns with a standard deviation of 3 microns. It can be seen that the salt particles are smaller than before extrusion (Fig. 2(b)). This size reduction is associated to the stresses (shear and elongation) involved in the extruder leading to particle dispersion and break-up. Figure 3(d) reports the pore size distribution of the microporous LDPE substrate. The pore size is in the range of 8-11 µm giving a porosity of 54% as calculated by Equation (12), but the porosity calculated by the volume of salt removed from the polymer is 47%. This 7% difference can be associated

11 ACS Paragon Plus Environment

Industrial & Engineering Chemistry Research

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 43

to a large amount of void created around the salt particles during melt calendaring (Fig. 3a high magnification) due to different coefficients of thermal expansion which is around 39.8×10-6 °C-1 for the salt particles27, and 3.20×10-4 °C-1 for LDPE at room temperature28. As the LDPE/salt melt was cooled, the polymer matrix shrank more than the salt particles leading to interfacial voids. This is also the reason why the pore size after leaching (8-11 µm) is larger than the salt particle size in LDPE before leaching (34 µm).

4.1.3. Morphology of the FSCM Figure 4 presents a typical SEM image of the PDMS/LDPE FSCM. It can be seen that the LDPE microporous layer is not pore-penetrated and covered by a dense PDMS layer with a thickness of around 50 µm. A thick layer of the PDMS is formed due to the high concentration of the PDMS solution (50%). The high concentration was chosen for three reasons. Firstly, to be sure that the selective layer covers any holes on the support surface with a single coating step. Secondly, to avoid penetration of the coating solution into the substrate pores to reduce the resistance of the LDPE microporous layer to component transport. Thirdly, to improve the selectivity. 4.2. Formation of the porous structure In the formation of the LDPE porous support, the salt particles were removed by leaching in water. The salt content (68% wt.) was selected as this value was optimized in our previous study23. Since the structure of the porous substrate highly depends on

12 ACS Paragon Plus Environment

Page 13 of 43

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

Industrial & Engineering Chemistry Research

the amount of salt leached out from the polymer matrix, this value was determined by TGA. Figure 5 presents the TGA curve for neat LDPE before extrusion, initial salt particles (Soda-LO), the LDPE substrate before and after leaching, as well as the FSCM. In the range of 220-320 °C, nearly 5 and 3% weight losses occurred for the salt and LDPE substrate before leaching due to the thermal degradation of the organic material coating Soda-LO. Above 800° C, a second loss was observed related to NaCl degradation. In the range of 360-485 °C, about 33, 82, 98, and 100%, weight losses were observed for the LDPE substrate before leaching, FSCM, LDPE substrate after leaching and neat LDPE before extrusion, respectively. These degradations steps are associated to the thermal degradation of LDPE. It can be approximated from Figure 5 that the salt loading is around 67% which is close to the expected value (68%) within experimental uncertainty. The TGA results combined with SEM (Fig. 3(b) at high magnification) revealed that salt leaching was probably not complete, even after 160 min of continuous leaching and about 2% salt remains in the LDPE layer and FSCM after leaching. In the 485-580 °C range, nearly 16% weight loss occurred for FSCM due to the thermal degradation of PDMS29.

4.3. FTIR Figure 6 presents the ATR-IR spectra of the LDPE support without coating and PDMS/LDPE FSCM. The strong adsorption bands of the LDPE support at 2916 cm-1 and 2847 cm-1 are associated to C–H stretching vibrations of –CH3, –CH2 and –CH 13 ACS Paragon Plus Environment

Industrial & Engineering Chemistry Research

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 43

groups30. But these bands disappeared in the FSCM spectra indicating that the surface was completely covered by the PDMS layer31. A peak at 783 cm-1 is attributed to −CH3 rocking and −Si−C− stretching in −Si−CH3. The bands at 1004 cm-1, 1256 cm--1 and 2961 cm-1 are assigned to −Si−O stretching in −Si−OH, symmetric −CH3 deformation in −Si−CH3 and −CH2− stretching in −Si−CH2−, respectively. The multiple peaks between 650 cm-1 and 900 cm-1 are related to CH3 rocking and Si−C stretching. Finally, the peak at 1463 cm-1 is attributed to C−H bending deformation29.

4.4. Solubility The solubility of all tested gasses as a function of pressure at 27 °C is represented in Figure 7. It can be seen that the sorption isotherms for all gases are linear (H2, N2, CH4, and CO2) or nearly linear (C3H8). The results show that the dissolved C3H8 in the FSCM is 26 and 6 times higher than CH4 and CO2 at 70 psi, while the dissolved CO2 in is 42 and 21 times higher than H2 and N2 at 290 psi. The concentration of low solubility gases such as N2, H2, CH4 and CO2 in rubbery polymers can be approximated by Henry’s law22:

C = kd p

(13)

where kd [cm3(STP)/(cm3 psi)] is the Henry’s law constant and the values are reported in Table 1. The nonlinear behavior of C3H8 at high pressures is related to PDMS swelling due to the high concentration of dissolved C3H832–36. The concentration of

14 ACS Paragon Plus Environment

Page 15 of 43

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

Industrial & Engineering Chemistry Research

high solubility gases in uncrosslinked rubbery polymers obeys the Flory-Huggins (FH) expression as32:

ln a = ln φ2 + ( 1 − φ2 ) + χ( 1 − φ2 )2

(14)

where a is the activity which can associated to a relative pressure, p/psat where psat is the saturation vapor pressure of the gas, χ is the Flory-Huggins interaction parameter and φ2 is the volume fraction calculated as37:

 22,414 φ2 = 1 +  CV2  

−1

(15)

where  is the partial molar volume and C is the equilibrium penetrant concentration in the polymer. The Flory-Rehner (FR) model is a modified FH equation for crosslinked rubbery polymers giving32: ν ln a = ln φ 2 + (1 − φ 2 ) + χ (1 − φ 2 ) 2 + V 2  e V 0

   1 − φ 2   × (1 − φ 2 )1 / 3 −    2   

(16)

where V2 is the penetrant molar volume, νe is the effective number of crosslinks, and V0 is the volume of penetrant-free polymer. C3H8 saturation vapor pressure (145 psi), partial molar volume (80 cm3/mol) and molar volume (76 cm3/mol) were taken from the literature37. The average crosslink density (νe/V0) was calculated by the following steps: 1- C3H8 sorption in uncrosslinked silicone oil was measured (67.5 cm3 gas/cm3 oil at 80 psi and 27°C). The FH interaction parameter (χ) was calculated to be 0.387. 15 ACS Paragon Plus Environment

Industrial & Engineering Chemistry Research

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 43

2- The volume fraction of C3H8 (φ2) in the crosslinked polymer was then calculated at the same pressure and temperature to be 0.187. 3- Crosslink density (νe/V0) was calculated by Equation (16) to be 6.3×10-4 mol/cm3. The dash line in Figure 7 represents the model based on the FR theory using a concentration-averaged interaction parameter ̅ as37:

χ=

1

φ2,max

φ2, max

∫ χ (φ )dφ 2

(17)

2

0

which was found to be 0.315. Figure 7 shows that the FR model is better than Henry’s law at higher concentration. The solubility parameters as a function of pressure for any gas can be determined from the sorption isotherms via Equation (6). As shown in Figure 8(a), the solubility of C3H8 increases with increasing pressure, while the sorption isotherms of the other gases are nearly constant. So a linear relation can be used as:

S = S ∞ + np

(18)

where n represents the pressure dependence of the solubility and S∞ is the infinite dilution solubility given by: S ∞ = lim (C p )

(19)

p →0

The values for both parameters are reported in Table 1. Without any interactions between the components and the membrane, the sorption isotherms are commonly 16 ACS Paragon Plus Environment

Page 17 of 43

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

Industrial & Engineering Chemistry Research

related to the condensability of the component (i.e. critical temperature, Tc)24,37. To determine the solubility difference for each component, S∞ can be plotted as a function of Tc. As expected, Figure 8(b) shows that the logarithm of S∞ linearly increases with T c. For the relatively low sorption gases such as N2, H2, CH4, and CO2, n is close to zero indicating that the solubility of these components is fundamentally independent of pressure. On the other hand, for the highly soluble C3H8, a large value for n is obtained indicating a stronger solubility dependence on pressure38.

4.5. Permeability Pure gas studies showed that permeability of more soluble permeants increases with increasing pressure, while that of less soluble ones remains constant or can even decrease with increasing pressure35,36,39. Figure 9 presents the effect of transmembrane pressure on H2, N2, CH4, CO2, and C3H8 permeabilities in the FSCM. It can be seen that the permeability coefficient increases in the following order: N2 < H2 < CH4 < CO2 < C3H8 In general, a linear relation between permeability and pressure is observed which can be written as33:

P = P0 + m∆p

(20)

17 ACS Paragon Plus Environment

Industrial & Engineering Chemistry Research

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 43

The permeability of a component, determined at ∆p = 0, is referred to as P0, while m is the slope of the permeability vs. ∆p curve. All the parameters for the tested gases and their relative values with respect to C3H8 (ideal selectivity at ∆p = 0 (P0,C3H8/P0,gas)) are reported in Table 2. The value of the slope m can reveal three phenomena occurring in a membrane: plasticization, hydrostatic pressure and penetrant solubility32,34. Plasticization is associated to increased diffusion as a result of increasing polymer segmental motion due to the presence of dissolved molecules between

polymer

chains37,40. For strong sorbing permeants such as C3H8 and CO2, the solubility in rubbery polymers increases with pressure. As the upstream pressure and penetrant concentration in polymer increases, the tendency to plasticize the polymer matrix increases. In addition to these dual effects, which affect the diffusion coefficient, the permeability frequently increases with pressure. Also, an increase of applied pressure on the membrane can slightly compress the polymer matrix, thereby reducing the amount of free volume available for molecular transport leading to lower diffusion coefficient41. Therefore, membrane compaction and solubility reduction with increasing pressure are two factors leading to lower permeability for low-sorbing molecules such as hydrogen. This can also be associated to negative m values (Table 2), while the value is positive for C3H8 and CO2. Table 2 shows that the infinite dilution gas permeability and selectivity are in good agreement with the values reported by Merkel et al. for PDMS composite membranes37. This means that the support layer prepared by this new solvent free method did not induce

18 ACS Paragon Plus Environment

Page 19 of 43

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

Industrial & Engineering Chemistry Research

significant

transport

resistance

to

gas

permeation

through

the

PDMS

composite

membrane.

4.6. Diffusivity Diffusivity in the FSCM at 27 °C was determined using Equation (5). As shown in Figure 10(a), the pressure dependence of the diffusion coefficient for all gases can be represented by a linear expression as: D = D 0 + q∆p

(21)

The diffusivity of each component at ∆p = 0 is called D0, while q is the slope of the diffusivity vs. ∆p curve. The values reported in Table 3 for H2, N2, and CH4 show that the diffusivity term slightly decreases with pressure due to the compression effect leading to negative q values. Furthermore, the solubility of these components is independent of pressure. Therefore, the permeation of these gases decreases slightly with pressure, while for CO2 and C3H8, an increase in both permeability and diffusivity is observed with pressure. At low pressure, the diffusivity increases in the following order which is agreement with their decreasing kinetic diameter (δ) (Table 3): C3H8 < CO2 < CH4 < N2 < H2 Figure 10(b) presents the D0 values of each gas in the FSCM as a function of the critical volume (Vc) which varies over one order of magnitude. As observed, the diffusivity is a relatively weak function of size.

19 ACS Paragon Plus Environment

Industrial & Engineering Chemistry Research

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 43

The relation between diffusivity and Vc is commonly expressed by the following expression42: D = τ VC−η

(22)

where τ and η are adjustable parameters. η controls the decreasing level or rate of diffusivity

with

component

size.

Higher

η

are

associated

to

higher

diffusivity

selectivity in the FSCM, but membranes with high η have diffusion strongly related to component size. According to the results of Figure 10(b), the value of η for the FSCM is 2.073 which confirms that the components diffusivity is a relatively weak function of size.

4.7. Ideal selectivity For

C3H8/gas

pairs,

the

solubility

selectivity,

diffusivity

selectivity,

and

ideal

selectivity of the FSCM at ∆p = 0 were reported in Tables 1, 2 and 3, respectively. Since C3H8 is substantially more soluble than the other gases, the C3H8/gas solubility selectivity is higher than 1. As reported before, the solubility of C3H8 is higher at high pressure. Therefore, the C3H8 sorption increase is more significant than for the other gases and the C3H8/gas solubility selectivity increases with pressure (Fig. 11(b)). But the C3H8/gas diffusivity selectivity is below unity indicating that the C3H8 diffusivity is smaller than for the other components. As observed, the diffusivity selectivity, like the solubility selectivity, also increases with pressure.

20 ACS Paragon Plus Environment

Page 21 of 43

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

Industrial & Engineering Chemistry Research

The C3H8/gas ideal selectivity for the FSCM as a function of pressure is presented in Figure. 11(a). This increase was found to be related to the difference between the components condensability. At p = 120 psi, the FSCM is about 6, 16, 23, and 55 times more permeable to C3H8 than CO2, CH4, H2, and N2, respectively. According to Equation (11), the FSCM selectivity depends on the relative diffusion coefficients (DA, DB) of both components (A and B) in the FSCM and on the relative solubilities (SA, SB) of the components in the FSCM. Consequently, the FSCM favorably separates the larger,

more

condensable

component

(C3H8)

over

the

smaller,

less

condensable

components (H2, N2, CH4, and CO2). Therefore, it can be concluded that the FSCM can be a good candidate for the separation of higher hydrocarbons from light components.

5. Conclusion In this work, a continuous processing technology was presented to prepare a composite membrane made from a porous polymer layer produced by melt extrusion/salt leaching and coated with a selective layer. The proposed method is simple and cost effective since it is based on inexpensive materials (LDPE and PDMS) and uses a lower amount of an environmentally friendly solvent (water). The results showed that the LDPE layer has an open cell microporous structure while the PDMS layer was compact. From the membranes produced, gas permeability, solubility, diffusivity, as well as selectivity where determined at 27 oC and different pressures for: C3H8, CO2, CH4, N2, and H2. C3H8 solubility was found to be much higher than the other gases due to its higher condensability. On the other hand, C3H8 diffusivity was smaller because its molecules

21 ACS Paragon Plus Environment

Industrial & Engineering Chemistry Research

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

(being larger) interact

more strongly with

Page 22 of 43

the segments of the polymer chains

compared to the other gases. However, the permeation of each component through the membrane is mainly controlled by solubility. Finally, it can be concluded that the high C3H8 permeability (20595 Barrer) for an upstream pressure of 110 psi with high selectivities (55, 23, 16 and 6 for N2, H2, CH4, CO2) are interesting for the gas separation

of

higher

hydrocarbons

from

light

components,

especially

for

natural

gas/biogas purification and the petrochemical industry.

Acknowledgments The authors would like to acknowledge the financial support of the National Science and Engineering Research Council of Canada (NSERC) and the technical support of the Centre de recherche sur les matériaux avancés de l’Université Laval (CERMA). The technical help of Mr. Yann Giroux was also much appreciated.

22 ACS Paragon Plus Environment

Page 23 of 43

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

Industrial & Engineering Chemistry Research

References (1)

Kim, S. S.; Lloyd, D. R. Microporous Membrane Formation via Thermally-Induced Phase Separation. III. Effect of Thermodynamic Interactions on the Structure of Isotactic Polypropylene Membranes. J. Membr. Sci. 1991, 64 (1), 13–29.

(2)

Pinnau, I.; Koros, W. J. A Qualitative Skin Layer Formation Mechanism for Membranes Made by Dry/wet Phase Inversion. J. Polym. Sci. Part B Polym. Phys. 1993, 31 (4), 419– 427.

(3)

Krause, B.; Boerrigter, M. E.; van der Vegt, N. F. A.; Strathmann, H.; Wessling, M. Novel Open-Cellular Polysulfone Morphologies Produced with Trace Concentrations of Solvents as Pore Opener. J. Membr. Sci. 2001, 187 (1–2), 181–192.

(4)

Gore, R. W. Process for producing porous products. US Patent US3953566A, 1976.

(5)

Xi, Z.; Xu, Y.-Y.; Zhu, L.-P.; Du, C.-H.; Zhu, B. Effect of Stretching on Structure and Properties of Polyethylene Hollow Fiber Membranes Made by Melt-Spinning and Stretching Process. Polym. Adv. Technol. 2008, 19 (11), 1616–1622.

(6)

Lin, S.-H.; Tung, K.-L.; Chen, W.-J.; Chang, H.-W. Absorption of Carbon Dioxide by Mixed Piperazine–Alkanolamine Absorbent in a Plasma-Modified Polypropylene Hollow Fiber Contactor. J. Membr. Sci. 2009, 333 (1–2), 30–37.

(7)

Kim, J.-J.; Jang, T.-S.; Kwon, Y.-D.; Kim, U. Y.; Kim, S. S. Structural Study of Microporous Polypropylene Hollow Fiber Membranes Made by the Melt-Spinning and Cold-Stretching Method. J. Membr. Sci. 1994, 93 (3), 209–215.

(8)

Verdolotti, L.; Colini, S.; Porta, G.; Iannace, S. Effects of the Addition of LiCl, LiClO4, 23 ACS Paragon Plus Environment

Industrial & Engineering Chemistry Research

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 43

and LiCF3SO3 Salts on the Chemical Structure, Density, Electrical, and Mechanical Properties of Rigid Polyurethane Foam Composite. Polym. Eng. Sci. 2011, 51 (6), 1137– 1144. (9)

Chu, R. K. M.; Naguib, H. E.; Atalla, N. Synthesis and Characterization of Open-Cell Foams for Sound Absorption with Rotational Molding Method. Polym. Eng. Sci. 2009, 49 (9), 1744–1754.

(10)

Dangtungee, R.; Supaphol, P. Melt Rheology and Extrudate Swell of Sodium ChlorideFilled Low-Density Polyethylene: Effects of Content and Size of Salt Particles. Polym. Test. 2010, 29 (2), 188–195.

(11)

Narkis, M.; Joseph, E. Tensile Properties of Rigid Polymeric Foams Produced by Salt Extraction. J. Cell. Plast. 1978, 14 (1), 45–49.

(12)

Cao, J.; Lin, Y.; Li, J.; Chen, C. Separation of Propane from Propane/Nitrogen Mixtures using PDMS Composite Membranes by Vapour Permeation. Chinese J. Polym. Sci. 2009, 27 (5), 621-627.

(13)

Sadrzadeh, M.; Amirilargani, M.; Shahidi, K.; Mohammadi, T. Gas Permeation through a Synthesized Composite PDMS/PES Membrane. J. Membr. Sci. 2009, 342 (1–2), 236–250.

(14)

Li, P.; Chen, H. Z.; Chung, T.-S. The Effects of Substrate Characteristics and Pre-Wetting Agents on PAN–PDMS Composite Hollow Fiber Membranes for CO2/N2 and O2/N2 Separation. J. Membr. Sci. 2013, 434, 18–25.

(15)

Majumdar, S.; Bhaumik, D.; Sirkar, K. K. Performance of Commercial-Size Plasmapolymerized PDMS-Coated Hollow Fiber Modules in Removing VOCs from 24 ACS Paragon Plus Environment

Page 25 of 43

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

Industrial & Engineering Chemistry Research

N2/air. J. Membr. Sci. 2003, 214 (2), 323–330. (16)

Shi, E.; Huang, W.; Xiao, Z.; Li, D.; Tang, M. Influence of Binding Interface between Active and Support Layers in Composite PDMS Membranes on Permeation Performance. J. Appl. Polym. Sci. 2007, 104 (4), 2468–2477.

(17)

Yuan, F.; Wang, Z.; Li, S.; Wang, J.; Wang, S. Formation–structure–performance Correlation of Thin Film Composite Membranes Prepared by Interfacial Polymerization for Gas Separation. J. Membr. Sci. 2012, 421–422, 327–341.

(18)

Loh, C. H.; Zhang, Y.; Goh, S.; Wang, R.; Fane, A. G. Composite Hollow Fiber Membranes with Different Poly(dimethylsiloxane) Intrusions into Substrate for Phenol Removal via Extractive Membrane Bioreactor. J. Membr. Sci. 2016, 500, 236–244.

(19)

Prabhakar, R. S.; Raharjo, R.; Toy, L. G.; Lin, H.; Freeman, B. D. Self-Consistent Model of Concentration and Temperature Dependence of Permeability in Rubbery Polymers. Ind. Eng. Chem. Res. 2005, 44 (5), 1547–1556.

(20)

Ghosal, K.; Freeman, B. D. Gas Separation Using Polymer Membranes: An Overview. Polym. Adv. Technol. 1994, 5 (11), 673–697.

(21)

Pinnau, I.; Wijmans, J. G.; Blume, I.; Kuroda, T.; Peinemann, K. V. Gas Permeation through Composite Membranes. J. Membr. Sci. 1988, 37 (1), 81–88.

(22)

Ghosal, K.; Freeman, B. D. Gas Separation Using Polymer Membranes: An Overview. Polym. Adv. Technol. 1994, 5 (11), 673–697.

(23)

Mosadegh-Sedghi, S.; Rodrigue, D.; Brisson, J.; Iliuta, M. C. Highly Hydrophobic Microporous Low-Density Polyethylene Hollow Fiber Membranes by Melt-Extrusion 25 ACS Paragon Plus Environment

Industrial & Engineering Chemistry Research

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 43

Coupled with Salt-Leaching Technique. Polym. Adv. Technol. 2013, 24 (6), 584–592. (24)

Shah, V. M.; Hardy, B. J.; Stern, S. A. Solubility of Carbon Dioxide, Methane, and Propane in Silicone Polymers: Effect of Polymer Side Chains. J. Polym. Sci. Part B Polym. Phys. 1986, 24 (9), 2033–2047.

(25)

Chen, X.; Rodrigue, D.; Kaliaguine, S. Diamino-Organosilicone APTMDS: A New CrossLinking Agent for Polyimides Membranes. Sep. Purif. Technol. 2012, 86, 221–233.

(26)

Shen, S.; Hoffman, A. J.; Butler, S. E. Method of producing salt composition. WO Patent WO2015015151, 2015.

(27)

Sowbhagya Madhusudhan Rao, A.; Narender, K.; Gopal Kishan Rao, K.; Gopi Krishna, N. Thermophysical Properties of NaCl, NaBr and NaF by γ-Ray Attenuation Technique. J. Mod. Phys. 2013, 4, 208–214.

(28)

Tripathi, D.; Dey, T. K. Thermal Conductivity, Coefficient of Linear Thermal Expansion and Mechanical Properties of LDPE/Ni Composites. Indian J. Phys. 2013, 87 (5), 435– 445.

(29)

Beltran, A. B.; Nisola, G. M.; Cho, E.; Lee, E. E. D.; Chung, W.-J. Organosilane Modified Silica/Polydimethylsiloxane Mixed Matrix Membranes for Enhanced Propylene/Nitrogen Separation. Appl. Surf. Sci. 2011, 258 (1), 337–345.

(30)

Gulmine, J. V.; Janissek, P. R.; Heise, H. M.; Akcelrud, L. Polyethylene Characterization by FTIR. Polym. Test. 2002, 21 (5), 557–563.

(31)

Zhang, Y.; Zhang, H.; Li, Y.; Mao, H.; Yang, G.; Wang, J. Tuning the Performance of Composite Membranes by Optimizing PDMS Content and Cross-Linking Time for 26 ACS Paragon Plus Environment

Page 27 of 43

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

Industrial & Engineering Chemistry Research

Solvent Resistant Nanofiltration. Ind. Eng. Chem. Res. 2015, 54 (23), 6175–6186. (32)

Fleming, G. K.; Koros, W. J. Dilation of Polymers by Sorption of Carbon Dioxide at Elevated

Pressures.

1.

Silicone

Rubber

and

Unconditioned

Polycarbonate.

Macromolecules 1986, 19 (8), 2285–2291. (33)

Merkel, T. C.; Gupta, R. P.; Turk, B. S.; Freeman, B. D. Mixed-Gas Permeation of Syngas Components in Poly(dimethylsiloxane) and poly(1-Trimethylsilyl-1-Propyne) at Elevated Temperatures. J. Membr. Sci. 2001, 191 (1–2), 85–94.

(34)

Prabhakar, R. S.; Merkel, T. C.; Freeman, B. D.; Imizu, T.; Higuchi, A. Sorption and Transport Properties of Propane and Perfluoropropane in Poly(dimethylsiloxane) and Poly(1-Trimethylsilyl-1-Propyne). Macromolecules 2005, 38 (5), 1899–1910.

(35)

Raharjo, R. D.; Freeman, B. D.; Paul, D. R.; Sarti, G. C.; Sanders, E. S. Pure and Mixed Gas CH4 and n-C4H10 Permeability and Diffusivity in Poly(dimethylsiloxane). J. Membr. Sci. 2007, 306 (1–2), 75–92.

(36)

Raharjo, R. D.; Freeman, B. D.; Sanders, E. S. Pure and Mixed Gas CH4 and n-C4H10 Sorption and Dilation in Poly(dimethylsiloxane). J. Membr. Sci. 2007, 292 (1–2), 45–61.

(37)

Merkel, T. C.; Bondar, V. I.; Nagai, K.; Freeman, B. D.; Pinnau, I. Gas Sorption, Diffusion, and Permeation in Poly(dimethylsiloxane). J. Polym. Sci. Part B Polym. Phys. 2000, 38 (3), 415–434.

(38)

Choi, S.-H.; Kim, J.-H.; Lee, S.-B. Sorption and Permeation Behaviors of a Series of Olefins and Nitrogen through PDMS Membranes. J. Membr. Sci. 2007, 299, 54–62.

(39)

Sadrzadeh, M.; Amirilargani, M.; Shahidi, K.; Mohammadi, T. Pure and Mixed Gas 27 ACS Paragon Plus Environment

Industrial & Engineering Chemistry Research

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 43

Permeation through a Composite Polydimethylsiloxane Membrane. Polym. Adv. Technol. 2011, 22 (5), 586–597. (40)

Pandey, P.; Chauhan, R. S. Membranes for Gas Separation. Prog. Polym. Sci. 2001, 26 (6), 853–893.

(41)

Singh, A.; Freeman, B. D.; Pinnau, I. Pure and Mixed Gas Acetone/Nitrogen Permeation Properties of Polydimethylsiloxane [PDMS]. J. Polym. Sci. Part B Polym. Phys. 1998, 36 (2), 289–301.

(42)

Freeman, B. D.; Pinnau, I. Polymeric Materials for Gas Separations. ACS Symposium Series. 1999, 733, 1–27.

28 ACS Paragon Plus Environment

Page 29 of 43

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

Industrial & Engineering Chemistry Research

Tables

Table 1. Values of Tc, S∞, n, kd and infinite dilution solubility selectivity of C3H8 with respect to the other gases at 27 °C. S∞

n × 103

kd

(K)

(cm3(STP)/ cm3 psi)

(cm3(STP) /cm3 psi2)

(cm3(STP) /(cm3 psi))

H2

33

0.0037

0.0000

0.0037

92.5

N2

126

0.0055

0.0011

0.0058

61.2

CH4

191

0.0289

0.0293

0.0349

11.7

CO2

304

0.1304

0.0491

0.1445

2.61

C3H8

370

0. 3399

5.5877

0. 5757

1.00

Tc Gas

29 ACS Paragon Plus Environment

S∞C3H8/S∞gas

Industrial & Engineering Chemistry Research

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 43

Table 2. Infinite dilution gas permeability (P0), m and C3H8/gas ideal selectivity at ∆p = 0 and 27 °C for this study and the results of Merkel et al. obtained at 35 oC37. Merkel et al.37

This study P0

m

P0,C3H8/P0,gas

P0

m

P0,C3H8/P0,gas

(Barrer)

(Barrer/psi)

(-)

(Barrer)

(Barrer/psi)

(-)

H2

988

-0.70

3.75

890

-0.23

4.61

N2

421

-0.41

8.81

400

-0.09

10.3

CH4

1360

-0.34

2.72

1200

-0.02

3.42

CO2

3322

1.43

1.12

3800

1.03

1.08

C3H8

3711

142

1.00

4100

114

1.00

30 ACS Paragon Plus Environment

Page 31 of 43

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

Industrial & Engineering Chemistry Research

Table 3. Values of δ, Vc, D0, q and ideal diffusivity selectivity of C3H8/gas at ∆p = 0 and 27 °C. δ

Vc

D0 × 106

q × 108

D0,C3H8/D0,gas

(Å)

(cm3/mol)

(cm2/s)

(cm2/psi s)

(-)

H2

2.89

64

139

-9.86

0.07

N2

3.64

90

39

-4.22

0.24

CH4

3.80

99

24

-2.40

0.38

CO2

3.30

94

13

0.06

0.70

C3H8

4.30

203

9

1.71

1.00

Gas

31 ACS Paragon Plus Environment

Industrial & Engineering Chemistry Research

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

Figures

Figure 1. Schematic representation of the melt extrusion, sheet drawing, active layer coating and salt leaching steps for the production of FSCM.

32 ACS Paragon Plus Environment

Page 32 of 43

Page 33 of 43

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

Industrial & Engineering Chemistry Research

Figure 2. SEM images of: (a) the salt particles at different magnification and (b) the initial particle size distribution of the salt.

33 ACS Paragon Plus Environment

Industrial & Engineering Chemistry Research

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 43

Figure 3. Typical SEM images at different magnification of the LDPE support: (a) before and (b) after salt leaching, (c) the particle size distribution of the salt inside LDPE and (d) the pore size distribution of the microporous LDPE substrate.

34 ACS Paragon Plus Environment

Page 35 of 43

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

Industrial & Engineering Chemistry Research

Figure 4. Cross-sectional SEM images of a PDMS/LDPE FSCM.

35 ACS Paragon Plus Environment

Industrial & Engineering Chemistry Research

110 100 90

Weight (%)

80 70 60 50

Salt (Soda-LO)

40

LDPE substrate before leaching

30

PDMS-LDPE FSCM LDPE substrate after leaching

20

Pure LDPE before extrusion

10 0 0

100

200

300

3.0

400 500 600 Temperature (°°C)

700

800

900

Pure LDPE before extrusion LDPE substrate after leaching

2.5

Derv. Weight (%.°° C-1)

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 43

PDMS-LDPE FSCM LDPE substrate before leaching

2.0

Salt (Soda-LO)

1.5 1.0 0.5 0.0 0

100

200

300

400 500 600 Temperature (°° C)

700

800

900

Figure 5. TGA results for neat LDPE pellet before extrusion, initial salt particle (Soda-LO), LDPE substrate before and after leaching, and PDMS/LDPE composite membrane. 36 ACS Paragon Plus Environment

Page 37 of 43

783

PDMS/LDPE FSCM 1004

LDPE support 2916 2847 Transmittance

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

Industrial & Engineering Chemistry Research

719

1463 1256

862 701 2961

3500

3000

2500 2000 1500 Wave Number (cm-1)

Figure 6. FTIR-ATR spectra of the LDPE support and the FSCM.

37 ACS Paragon Plus Environment

1000

500

Industrial & Engineering Chemistry Research

80 FR

C (cm3(STP)/cm3 polymer)

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 38 of 43

70

Henry

C3H8

60 50

CO2

40 30 20 10

CH4

0 0

N2

H2

50 100 150 200 250 300 350 400 450 Pressure (psi)

Figure 7. Sorption isotherms of the different gases in FSCM at 27 °C.

38 ACS Paragon Plus Environment

Page 39 of 43

1.E+0

1.E+0

C3H8

C3H8 CO2

1.E-1 CH4 1.E-2

N2 H2

S∞ (cm3(STP)/cm3 psi)

S (cm3(STP)/cm3 psi)

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

Industrial & Engineering Chemistry Research

(a)

1.E-3 0

y = 0.0016e0.0143x 1.E-1

1.E-2 H2

1.E-3 50 100 150 200 250 300 350 400 450

CO2 CH4

N2

(b) 0

50

100 150 200 250 300 350 400 Tc (K)

Pressure (psi)

Figure 8. (a) Solubility of the gases in FSCM as a function of pressure at 27 °C and (b) infinite dilution solubility as a function of Tc.

39 ACS Paragon Plus Environment

Industrial & Engineering Chemistry Research

1.E+5

Permeability (Barrers)

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 40 of 43

C3H8 1.E+4 CO2 CH4 H2

1.E+3

N2 1.E+2 50

60

70

80

90

100 110 120 130

∆p (psi)

Figure 9. Permeability through the FSCM as a function of the pressure difference at 27 °C.

40 ACS Paragon Plus Environment

Page 41 of 43

1.E-3

1.E-4

H2

CH4 1.E-5

N2 CO2

C3H8

1.E-6

Diffusivity (cm2/s)

1.E-3

Diffusivity (cm2/s)

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

Industrial & Engineering Chemistry Research

(a) 50

y = 0.4015x-2.073

N2 CH4

C3H8

1.E-5

1.E-6 100 150 200 250 300 350 400 450

H2

1.E-4

CO2

(b) 60

90

∆p (psi)

120 VC

150

180

210

(cm3/mol)

Figure 10. (a) Diffusivity as a function of pressure difference at 27 °C and (b) infinite dilution diffusion coefficient D0 as a function of Vc.

41 ACS Paragon Plus Environment

Industrial & Engineering Chemistry Research

C3/N2

50 40 30

C3/H2

20

C3/CH4

10

C3/CO2

C3/CH4→

250

C3/N2→ 200 ←C3/H2

150

←C3/N2

50

←C3/CH4 ←C3/CO2

0 60

80

100

120

0.1

C3/H2→

100

0 40

1

C3/CO2→

(b)

20

Diffusivity selectivity

300

(a)

Solubility selectivity

60

Ideal 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

Page 42 of 43

0.01

40

60

80

100

120

140

Upstream pressure (psi)

Upstream pressure (psi)

Figure 11. C3H8/gas overall (a) solubility and diffusivity, and (b) selectivity of the FSCM as a function of the upstream pressure.

42 ACS Paragon Plus Environment

Page 43 of 43

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

Industrial & Engineering Chemistry Research

Graphical abstract

43 ACS Paragon Plus Environment