Method of fractionating mixtures of low molecular weight hydrocarbons

Abstract

The invention relates to membrane gas separation, in particular to a method of fractionating mixtures of low molecular weight hydrocarbons based on the capillary condensation of the mixture components in the pores of microporous membranes having uniform porosity and a pore diameter of 5 to 250 nm, wherein, for capillary condensation, the temperature of the membrane and the pressure on the permeate side are kept below the temperature and the pressure of the feed mixture. The method provides significantly increasing membrane permeability with respect to condensable components, and also component separation factors, while also allowing to avoid deep cooling of the gas stream fed to a membrane module, and to carry out gas separation under insignificant cooling of the membrane on the permeate side (down to -50? C.). The invention provides for energy-efficient fractionation of hydrocarbon mixtures, including separation and drying of natural and associated petroleum gases.

Claims

1. A method of fractionating mixtures of low molecular weight hydrocarbons characterized in that the separation of a feed mixture into permeate and retentate is carried out using a microporous membrane having uniform porosity and a pore diameter ranging from 5 to 250 nm, wherein the temperature of the membrane on the permeate and also the pressure on the permeate side are kept below the temperature and the pressure of the feed mixture to provide capillary condensation of mixture components in the micropores of the membrane.

2. The method of claim 1 characterized in that the temperature of the membrane and the pressure on the permeate side are selected such that the equilibrium pressure of the saturated vapors of the separated components on the permeate side is lower than the partial pressure of the components in the feed stream.

3. The method of claim 1 characterized in that the temperature of the retentate is above the temperature of the membrane and the permeate.

4. The method of claim 1 characterized in that the liquid phase of hydrocarbons is withdrawn from the permeate.

5. The method of claim 1 characterized in that the pore size dispersion of the membrane material is not above 25%.

6. The method of claim 1 characterized in that microporous inorganic anodic aluminum oxide membranes are used as the membrane material.

7. The method of claim 6 characterized in that the structure of the anodic aluminum oxide membrane comprises splitting of larger diameter into several pores having smaller diameter, wherein all the pores have the same diameter at the same depth from the membrane surface.

8. The method of claim 1 characterized in that track membranes are used as the membrane material.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) The essence of the invention is explained by the drawings wherein

(2) FIG. 1 outlines a common scheme of fractionating mixtures of hydrocarbons using capillary condensation in the pores of microporous membranes;

(3) FIG. 2 demonstrates micrographs of the surfaces and pore size distribution in the membranes: (a) microporous asymmetric inorganic membrane made from anodic aluminum oxide with selective layer pore diameter of 6 nm (embodiment example 1); (b) microporous inorganic membrane made from anodic aluminum oxide with pore diameter of 43 nm (embodiment examples 2 and 3); (c) microporous inorganic membrane made from anodic aluminum oxide with pore diameter of 130 nm (embodiment example 4); (d) microporous inorganic membrane made from anodic aluminum oxide with pore diameter of 230 nm (embodiment example 5); (e) Whatman Nuclepore track polycarbonate membrane with pore diameter of 110 nm (embodiment example 6);

(4) FIG. 3 demonstrates a plot of permeability in the asymmetric microporous inorganic membrane made from anodic aluminum oxide vs. average pressure on the membrane for He and i-C.sub.4H.sub.10 at the membrane holder temperature of 10? C. and temperature of the fed gas of 25? C.

IMPLEMENTATION OF THE INVENTION

(5) The present invention is explained by specific embodiment examples, which, however, are not the only ones possible and do not limit the scope of the invention.

Examples 1 to 5. Separation of Test Hydrocarbon Mixture on Microporous Inorganic Membranes Made from Anodic Aluminum Oxide

(6) For illustrating the method of fractionating low molecular weight hydrocarbons using the approach of capillary condensation on microporous inorganic membranes anodic aluminum oxide membranes were formed 100 microns thick and with pore diameter of 5 to 250 nm.

(7) In view of low permeability of the membranes with small diameter pores and significant thickness (required for providing a sufficient mechanical strength) a membrane with 6 nm pore diameter (Example 1) was formed as a layered asymmetric membrane having a selective layer (with layer thicknesses: 90 microns at D.sub.pore=120 nm; 7 microns at D.sub.pore=40 nm; 3 microns at D.sub.pore=6 nm). The membrane was formed by anodic oxidation in 0.3M oxalic acid solution by stepwise reduction of voltage. The layer with 120 nm pore diameter was formed at 120V voltage, the layer with 40 nm diameterat 40V, the layer with 6 nm diameterat the voltage of up to 10 V.

(8) Anodic aluminum oxide membranes with 43 and 130 nm pore diameter (Examples 2, 3 and 4) were made by anodic oxidation in 0.3 M oxalic acid solution at the voltages of 40 V and 120V correspondingly. Membrane thickness control was carried out according to charge density for the charge passed during anodization, assuming oxidation efficiency of 0.451 (?m/cm.sup.2)/C for the 40 V voltage to 0.55 (?m/cm.sup.2)/C for the 120 V voltage.

(9) The anodic aluminum oxide membrane with 230 nm pore diameter (Example 5) was made by anodic oxidation in 0.3 M phosphoric acid solution at 190 V voltage. Membrane thickness control was carried out according to charge density for the charge passed during anodization, assuming oxidation efficiency of 0.45 (?m/cm.sup.2)/C.

(10) Removal of the metal barrier layer for all membranes was carried out by chemical etching in the acid solution with electrochemical detection of the pore opening moment. This approach provides reproducibility of barrier layer removal. Microphotographs of the membrane surface and pore size distribution are shown in FIG. 2. Data on membrane pore diameters and pore size distribution are also outlined in Table 1.

(11) Fractionation of low molecular weight hydrocarbons using capillary condensation on microporous inorganic anodic aluminum oxide membranes was carried out in a membrane module equipped with a cooled membrane holder. Also, thermostatic control of permeate section at lowered temperature was carried out.

(12) In order to determine the membrane permeability with respect to condensed gas, measurement of permeability of asymmetric microporous inorganic membrane using pure butane was carried out (FIG. 3). The obtained dependencies (plots) reflect a significant increase in membrane permeability of more than 500 m.sup.3/(m.sup.2.Math.atm.Math.h) upon approaching the condensation pressure in membrane pores. Therefore, capillary condensation of gases in membrane channels leads to significant increase in membrane permeability, which allows for significantly increasing membrane efficiency at industrial application.

(13) In order to test the method using gas mixtures a test mixture was made simulating contents of the associated petroleum gas, of the following contents: 67.0 vol. % CH.sub.4, 7.1 vol. % C.sub.2H.sub.6, 10.1 vol. % C.sub.3H.sub.8, 2.6 vol. % i-C.sub.4H.sub.10, 5.2 vol. % n-C.sub.4H.sub.10, 1.4 vol. % i-C.sub.5H.sub.12, 1.4 vol. % n-C.sub.5H.sub.12, 3.9 vol. % C.sub.6H.sub.14, 1.6 vol. % N.sub.2. The contents of the feed gas mixture and retentate gas mixture were determined by means of gas chromatography using the Perkin Elmer Clarus 600 gas chromatograph.

(14) Experiments with membranes having various pore diameters were carried out. During the experiments the membrane module was fed with gas mixture under the pressure P.sub.1 (6-7 bar) and the temperature T.sub.1=50? C. High temperature of the feeding gas mixture was used in order to avoid gas condensation prior to feeding it to the membrane module. The membrane and the permeate section were cooled to the required temperature. The permeate side pressure, P.sub.2 was maintained equal to the equilibrium pressure come to stay at cooling the gas mixture (introduced to the permeate section at P.sub.1 and T.sub.1=25? C.) or 1 to 3 bar lower than the equilibrium pressure. The input gas stream was controlled in such a way so to minimize release of C1 and C2 components to permeate at maximum stream. Full parameters of experiments are outlined in Table 1. Table 1 also outlines contents of retentates after membrane fractionating of the mixture. Averaged contents and separation coefficients with respect to components were calculated from feed mixture and retentate streams and compositions.

(15) According to experiments performed the capillary condensation of gases in membrane channels with pore diameter in the range of 5 to 250 nm leads to significant increase in efficiency of the membrane and to efficient separation of heavy hydrocarbons. The maximum separation coefficients achieved for the outlined conditions were ?(C.sub.3H.sub.8/CH.sub.4)=39.7; ?(i-C.sub.4H.sub.10/CH.sub.4)=61.6; ?(n-C.sub.4H.sub.10/CH.sub.4)=63.5; ?(i-C.sub.5H.sub.12/CH.sub.4)=69.5; ?(i-C.sub.5H.sub.12/CH.sub.4)=70.6; ?(i-C.sub.5H.sub.12/CH.sub.4)=71.4 for the membrane with 43 nm pore diameter and membrane holder and permeate section temperature of ?46.5? C. The minimum stage cut of light component (C1 and C2) was less than 2.5%, at feed stream of more than 340 m.sup.3/(m.sup.2.Math.atm.Math.h) and the permeate stream of up to 79.2 m.sup.3/(m.sup.2.Math.atm.Math.h), thus providing for possible practical applicability of the method. The measured retentate dew point with respect to hydrocarbons at experimental pressure was less than ?38? C., the calculated value was ?45.9? C. The calculated values of the dew points with respect to hydrocarbons for examples 1 to 5 are also outlined in Table 1.

(16) The results obtained for other membranes and process conditions, though inferior to the results outlined in Example 3, also illustrate high separation efficiency for the components of the mixture. Thereby, increasing in the membrane pore diameter results in increasing in the maximum permeate stream through the membrane and also to increasing stage cut of light components. It should also be noted that retentate composition changes insignificantly when experimental conditions are changed (see Examples 1 to 4), thereby making it possible to use this method for producing mixtures of constant composition. Moreover, sequential separation of fractions with various boiling points and equilibrium partial volumes by sequential use of membranes cooled to various temperatures is possible (Examples 2, 3).

Example 6. Separation of a Test Hydrocarbon Mixture Using a Track Polyethylene Terephthalate Track Membrane

(17) A commercially available sample of Whatman Nuclepore polycarbonate membrane with pore diameter of 100 nm was used as membrane material in this example. Microphotograph of the membrane surface is shown in FIG. 2. The pore diameter determined according to scanning electron microscopy was 110?27 nm (Table 1). The experiment on fractionating the low molecular weight hydrocarbons using capillary condensation on track membrane was carried out similarly to the experiments described in Examples 1 to 5. The experimental parameters are outlined in Table 1.

(18) The maximum separation coefficients achieved for the outlined conditions were: ?(C.sub.3H.sub.8/CH.sub.4)=3.5; ?(i-C.sub.4H.sub.13/CH.sub.4)=4.4; ?(n-C.sub.4H.sub.10/CH.sub.4)=4.1; ?(i-C.sub.5H.sub.12/CH.sub.4)=5.6; ?(i-C.sub.5H.sub.12/CH.sub.4)=6.8; ?(i-C.sub.5H.sub.12/CH.sub.4)=7.8. The light component (C1 and C2) stage cut was ?14% at the feed stream of 230 m.sup.3/(m.sup.2.Math.atm.Math.h). A relatively high light component stage cut in this case is associated with non-uniformity of the pores, as well as due to the presence of pores with a size twice as large in the membranes, resulted from combining of tracks. Despite the high light component stage cut, high permeability of the membrane in the capillary condensation mode in combination with commercial availability determines its practical applicability in the method.

(19) Therefore, according to the obtained data the claimed method allows for efficiently separating hydrocarbon mixture including natural and associated petroleum gases using capillary condensation on microporous membranes.

(20) TABLE-US-00001 TABLE 1 Process parameters for the fractionating of low molecular weight hydrocarbons using capillary condensation on microporous membranes for a mixture of test contents (67.0 vol. % CH.sub.4, 7.1 vol. % C.sub.2H.sub.6, 10.1 vol. % C.sub.3H.sub.8, 2.6 vol. % i-C.sub.4H.sub.10, 5.2 vol. % n-C.sub.4H.sub.10, 1.4 vol. % i-C.sub.5H.sub.12, 1.4 vol. % n-C.sub.5H.sub.12, 3.9 vol. % C.sub.6H.sub.14, 1.6 vol. % N.sub.2) for various membrane types Example No. 1 2 3 4 5 6 Membrane parameters Membrane type Anodic aluminum oxide Track polycarbonate Average pore 6 43 43 130 230 110 diameter, nm Pore size 1.5 (25) 6 (14) 6 (14) 20 (15) 40 (17) 27 (25) distribution, nm, (%) Porosity, % 14 12 12 8 8 3.8 Process parameters Feed mixture 50 50 50 50 50 50 temperature, T.sub.1, ? C. Membrane holder ?46.5 ?20 ?46.5 ?46.5 ?46.5 ?46.5 and permeate temperature, T.sub.2, ? C. Feed stream, 150.2 322.4 343.3 336.5 275.2 237.1 Nm.sup.3/(m.sup.2 .Math. h) Retentate 114.6 248.4 267.5 252.2 229.3 174.0 stream, Nm.sup.3/(m.sup.2 .Math. h) Feed stream/ 6.6 6.6 6.1 6.3 6.3 6.3 retentate pressure, P.sub.1, bar Permeate side 3.0 5.3 4.7 4.7 5.1 4.6 pressure, P.sub.2, bar Membrane 3.6 1.3 1.4 1.6 1.2 1.7 pressure drop, bar Permeate gas 0.3 3.3 0.5 1.3 5.6 11.2 phase stream, Nm.sup.3/(m.sup.2 .Math. h) Permeate liquid 151.1 310.0 307.6 342.3 146.0 228.8 phase stream, l/(m.sup.2 .Math. h) Calculated 34.4 70.6 70.1 77.9 40.2 52.1 liquid permeate phase stream, Nm.sup.3/(m.sup.2 .Math. h) Component contents in the retentate, % CH.sub.4 84.87 82.24 85.06 85.69 75.56 79.9 C.sub.2H.sub.6 8.52 6.74 7.91 7.03 7.16 6.86 C.sub.3H.sub.8 5.5 6.67 5.75 5.89 10.5 7.55 i-C.sub.4H.sub.10 0.45 1.33 0.46 0.52 2.36 1.53 n-C.sub.4H.sub.10 0.63 2.74 0.74 0.84 3.78 3.36 i-C.sub.5H.sub.12 0.021 0.215 0.048 0.02 0.248 0.551 n-C.sub.5H.sub.12 0.011 0.064 0.021 0.019 0.393 0.254 C.sub.6H.sub.14 <0.001 <0.001 0.004 <0.001 <0.001 0.001 Dew point with ?46.6 ?26 ?45.9 ?44.3 ?24.3 ?17.8 respect to C.sub.nH.sub.2n+2 (calcul., McWilson) Stage cut of components, % CH.sub.4 3.6 5.7 1.4 4.4 6.4 12.8 C.sub.2H.sub.6 8.4 23.6 9.4 22.5 12.3 26.1 C.sub.3H.sub.8 58.5 49.1 55.6 56.3 13.4 45.2 i-C.sub.4H.sub.10 86.8 60.6 86.2 85.0 24.4 56.8 n-C.sub.4H.sub.10 90.8 59.4 88.9 87.9 39.5 52.6 i-C.sub.5H.sub.12 98.9 88.2 97.3 98.9 85.2 71.1 n-C.sub.5H.sub.12 99.4 96.5 98.8 99.0 76.6 86.7 C.sub.6H.sub.14 >99.98 >99.98 99.95 >99.98 >99.98 99.98 C1 + C2 stage cut 3.84 7.48 2.23 6.22 6.99 14.08 C3+ stage cut 79.49 65.48 77.77 77.79 41.46 60.49 Total 23.70 22.95 22.08 25.05 16.68 26.61 Separation factor for C.sub.n/C.sub.1 C.sub.2H.sub.6/CH.sub.4 2.3 4.1 6.7 5.1 1.9 2.0 C.sub.3H.sub.8/CH.sub.4 16.3 8.6 39.7 12.8 2.1 3.5 i-C.sub.4H.sub.10/CH.sub.4 24.1 10.6 61.6 19.3 3.8 4.4 n-C.sub.4H.sub.10/CH.sub.4 25.2 10.4 63.5 20.0 6.2 4.1 i-C.sub.5H.sub.12/CH.sub.4 27.5 15.5 69.5 22.5 13.3 5.6 n-C.sub.5H.sub.12/CH.sub.4 27.6 16.9 70.6 22.5 12.0 6.8 C.sub.6H.sub.14/CH.sub.4 27.8 17.5 71.4 22.7 15.6 7.8