SEPARATION OF GASES VIA CARBONIZED VINYLIDENE CHLORIDE COPOLYMER GAS SEPARATION MEMBRANES AND PROCESSES THEREFOR

Abstract

A carbonized PVDC copolymer useful for the separation of an olefin from its corresponding paraffin may be made by heating a polyvinylidene chloride copolymer film or hollow fiber having a thickness of 1 micrometer to 20 micrometers to a pretreatment temperature of 100? C. to 180? C. to form a pretreated polyvinylidene chloride copolymer film and then heating the pretreated polyvinylidene chloride copolymer film to a maximum pyrolysis temperature from 350? C. to 750? C. A process for separating an olefin from its corresponding paraffin in a gas mixture is comprised of flowing the gas mixture through the aforementioned carbonized polyvinylidene chloride (PVDC) copolymer to produce a permeate first stream having an increased concentration of the olefin and a second retentate stream having an increased concentration of its corresponding paraffin.

Claims

1. A method of making a carbonized polyvinylidene chloride copolymer useful to separate an olefin from its corresponding paraffin comprising, (a) providing a polyvinylidene chloride copolymer film or hollow fiber having a thickness of 1 micrometer to 20 micrometers, (b) heating the polyvinylidene chloride copolymer film to a pretreatment temperature of 100? C. to 180? C. to form a pretreated polyvinylidene chloride copolymer film, and (c) heating the pretreated polyvinylidene chloride copolymer film to a maximum pyrolysis temperature from 350? C. to 750? C.

2. The method of claim 1, wherein the film or fiber is restrained in steps (b) and (c) by applying a force.

3. The method of claim 1, wherein the maximum pyrolysis temperature is at most 650? C.

4. The method of claim 1, wherein the polyvinylidene chloride copolymer film is a polyvinylidene chloride copolymer comprised of vinylidene chloride and at least one of the following: a vinyl monomer; a vinyl chloride monomer; an acrylate monomer; a methacrylate monomer; a styrenic monomer; acrylonitrile, methacrylonitrile; itaconic acid; chlorotrifluoroethylene that have been copolymerized.

5. The method of claim 1, wherein the thickness is from 1 micrometers to 10 micrometers.

6. The method of claim 1, wherein the polyvinylidene chloride copolymer film is formed by melt-extrusion at a stretch ratio from 1 to 8.

7. A process for separating an olefin from a gas mixture having the olefin's corresponding paraffin, the method comprising (i) providing the carbonized polyvinylidene chloride copolymer membrane of claim 1 and (ii) flowing the gas mixture through said carbonized polyvinylidene chloride copolymer membrane to produce a permeate first stream having an increased concentration of the olefin and a second retentate stream having an increased concentration of its corresponding paraffin.

8. The process of claim 7, wherein the olefin is ethylene or propylene.

9. The process of claim 8, wherein the paraffin is ethane or propane.

10. The process of claim 7, wherein the polyvinylidene chloride copolymer membrane has an average pore size greater than the olefin or paraffin as determined by gas permeation employing gas probe molecules of differing sizes.

11. The process of claim 10, wherein the average pore size is greater than 4 angstroms.

12. The process of claim 11, wherein the average pore size is greater than 5 angstroms.

Description

EXAMPLES

PVDC Copolymer Film Preparation:

Melt Extruded Films of MA 4.8 wt % Copolymer

[0040] Base PVDC copolymer with 4.8 wt % methyl acrylate (MA) comonomer (named as MA4.8 wt %, Mw=96,000, The Dow Chemical Company, Midland, Mich.) was blended with 2 wt % epoxidized soybean oil (based on total amount of blend), 4 wt % dibutyl sebacate, and 2 wt % PLASTISTRENGTH L-1000 an acrylic lubricant available from Arkema PLC, France. The blend was extruded through a 1.75 inch width film die (controlled at 174? C.) followed by water quench and stretch winding. The wind rate was controlled to obtain films of 2 mil (1 mil=25.4 micrometer). The films after winding were cut into approximately 12 inch wide and 2 feet length pieces and laid on flat desktop for about one week. Coupons of ? inch diameter were cut for carbonization as described below.

Melt Extruded Films of MA 8.5 wt % Copolymer

[0041] PVDC copolymer with 8.5 wt % methyl acrylate comonomer (named as MA8.5 wt %, Mw=85,000, The Dow Chemical Company) was blended with 2 wt % epoxidized soybean oil and 2 wt % PLASTISTRENGTH L-1000. The blend was extruded in the same manner as above. The wind rate was controlled to obtain films that are 2 mil (?50 micrometers) thick. The films after winding were cut into approximately 12 inch wide and 2 feet length pieces and laid on flat desktop for about one week.

Melt Bubble Extruded Films of MA 4.8 wt % Copolymer

[0042] Base PVDC copolymer with 4.8 wt % methyl acrylate (MA) comonomer (named as MA4.8 wt %, Mw=96,000, The Dow Chemical Company, Midland, Mich.) was blended with 2 wt % epoxidized soybean oil (based on total amount of blend), 4 wt % dibutyl sebacate, and 2 wt % PLASTISTRENGTH L-1000 an acrylic lubricant available from Arkema PLC, France. The resin was melt extruded using a commercial bubble extruder to form films having a thickness of 0.4 mil (?10 micrometers).

Melt Bubble Extruded Films of VC17.6 wt % Copolymer

[0043] PVDC copolymer with 17.6 wt % vinyl chloride (VC) comonomer (named as VC17.6 wt %, The Dow Chemical Company) was blended with 2 wt % epoxidized soybean oil and 2 wt % PLASTISTRENGTH L-1000. The resin was melt extruded using a commercial bubble extruder to form films having a thickness of 0.4 mil (?10 micrometers).

TABLE-US-00001 TABLE 1 Precursor Films Precursor Film thickness Film # Preparation method Base resin [mil] 1 Melt extrusion MA4.8 wt % 2 2 Melt extrusion MA8.5 wt % 2 3 Bubble Melt extrusion MA4.8 wt % 0.4 4 Bubble Melt extrusion VC17.6 wt % 0.4

Carbon Membrane Formation

[0044] A two-step pyrolysis approach was used. The precursor films were heated to a first temperature of 130-150? C. for 24 hours in a low temperature oven purged by 2 L/min of air (pretreated films), which was followed by further heating to pyrolyze the pretreated films to temperatures in the range of 350-950? C. in a 6 ID quartz tube furnace purged by 5 L/min of nitrogen.

[0045] For the initial low temperature pretreatment, 12 disks (? inch diameter) sandwiched between porous ceramic honeycomb plates, through which HCl generated should be transported out swiftly. A scrubber connected to this oven contained a 10 wt % sodium hydroxide aqueous solution. A loaded oven was heated at 1? C./min to the temperature shown in Table 2 (Pretreatment Temperature) and held for 24 hours under 2 L/min of air purge.

[0046] For the second heating step, the 12 pretreated disks were sandwiched between the porous ceramic honeycomb plates were loaded into a 6 ID quartz tube furnace. A scrubber connected to this furnace contained a 10 wt % sodium hydroxide aqueous solution. The furnace was raised to 500? C. at 3? C./min), and held for 30 minutes at the final temperature and then cooled down to room temperature (?25? C.). After cooling down, the carbon membranes were put into a storage box continuously purged with dry nitrogen at a flow rate of 5 Liter/min.

Carbon Membrane Permeation Test

[0047] The carbon membranes were masked onto a standard 25 mm filter holder (Millipore #4502500, EMD Millipore Corp., Germany) using an impermeable aluminum tape, leaving an open defined permeation area. A two-part epoxy (J-B Weld twin tube) was then applied along the interface of the tape and the carbon membranes. Mixed permeation tests of several gas species were conducted at 20? C. with a continuous upstream feed of either an equimolar mixture of H.sub.2/CO.sub.2/CH.sub.4 (total 75 sccm, 1 atm) or an equimolar mixture of C.sub.2H.sub.4/C.sub.2H.sub.6/C.sub.3H.sub.6/C.sub.3H.sub.8 (total 50 sccm, 1 atm), and downstream He purge (2.0 sccm, 1 atm). The permeate carried by the He purge gas was analyzed by a GC (gas chromatograph) with a TCD (thermal conductivity detector for H.sub.2 and CO.sub.2) and FID (flame ionization detector for all the hydrocarbons). The concentrations in all gases were lower than 5%, so the gas flow rate in downstream was considered the same as the He flow rate. The membrane permeate rate was calculated using the He purge flow rate times the permeate concentrations measured by GC. The tests were run for several hours to days until the permeate concentrations were steady. The parameters to make the carbon membranes are shown in Table 2. The resulting permeation results are shown in Table 3.

TABLE-US-00002 TABLE 2 CMS film Pre-treatment Temp. Final Pyrolysis Temp. Precursor Example [? C.] [? C.] * film # Ex. 1 130 500 3 Ex. 2 130 500 4 Comp. Ex 1 130 500 1 Comp. Ex. 2 130 500 2

TABLE-US-00003 TABLE 3 Ethylene/ Propylene/ Hydrogen Methane Ethylene Ethane Propylene Propane ethane propane CMS film Permeance Permeance H.sub.2/CH.sub.4 Permeance Permeance Permeance Permeance selectivity selectivity Ex. (GPU) (GPU) Selectivity (GPU) (GPU) (GPU) (GPU) [] [] Ex. 1 132 1.76 75 3.30 0.92 2.14 0.09 3.6 23.5 Ex. 2 133 1.47 90 2.66 0.70 1.43 0.03 3.8 51.1 Comp. Ex. 1 96 0.28 343 0.59 0.10 0.24 BDL 6.1 N/A Comp. Ex. 2 83 0.53 156 1.05 0.17 0.42 BDL 6.3 N/A

[0048] Comp. Ex.1 and Comp. Ex.2 are highly selective CMS membrane for H.sub.2/CH.sub.4 separation. However, the flux (permeance) of hydrocarbon molecules are too small, particularly ethylene and propylene which have a permeance lower than ?1GPU, making them not useful industrially.

[0049] Ex.1 and Ex.2 were made essentially the same as Comp. Ex. 1 and Ex. 2 respectively but were much thinner (?50 micrometers versus ?10 micrometers). The thinner CMS films of Ex.1 and Ex.2 had substantially worse H.sub.2/CH.sub.4 selectivities, making them unattractive for that application anymore. However, the fluxes of ethylene and propylene increased to industrially useful ranges (>1 GPU at 20? C.) for olefin/paraffin separations and displayed useful olefin/paraffin selectivities.