Login| Sign Up| Help| Contact|

Patent Searching and Data


Title:
A CARBON MOLECULAR SIEVE MEMBRANE, METHOD OF PREPARATION AND USES THEREOF
Document Type and Number:
WIPO Patent Application WO/2017/068517
Kind Code:
A1
Abstract:
The present disclosure relates to a novel carbon molecular sieve membranes - CMSM. More particularly, the disclosure relates to the polymeric precursor and carbonization conditions needed for producing a high-selective, age-free CMSM. More specifically, the present subject-matter discloses a process for obtaining from a cellophane film and in a single carbonization step. The carbon molecular sieve membrane obtainable be the process describe displays no pore blockage in the presence of water vapor and displaying very high ideal permselectivities and permeabilities for industrially relevant gas mixtures.

Inventors:
MAGALHÃES MENDES ADÉLIO MIGUEL (PT)
SILVA DE ANDRADE MÁRCIA RAFAELA (PT)
FERREIRA DA SILVA BOAVENTURA MARTA (PT)
VALE RODRIGUES SANDRA CRISTINA (PT)
Application Number:
PCT/IB2016/056295
Publication Date:
April 27, 2017
Filing Date:
October 19, 2016
Export Citation:
Click for automatic bibliography generation   Help
Assignee:
UNIV DO PORTO (PT)
International Classes:
B01D65/08; B01D53/22; B01D67/00; B01D69/10; B01D69/14; B01D71/02
Domestic Patent References:
WO2000053833A12000-09-14
WO2009129984A12009-10-29
Foreign References:
EP0671202A21995-09-13
US5288304A1994-02-22
EP0671202A21995-09-13
US5575963A1996-11-19
US5695818A1997-12-09
Other References:
CAMPO M C ET AL: "Carbon molecular sieve membranes from cellophane paper", JOURNAL OF MEMBRANE SCIENCE, ELSEVIER BV, NL, vol. 350, no. 1-2, 15 March 2010 (2010-03-15), pages 180 - 188, XP026915560, ISSN: 0376-7388, [retrieved on 20100104], DOI: 10.1016/J.MEMSCI.2009.12.026
SADIPAL: "Perforated Rolls / Roleaux perforés / Perforierte Rollen", 15 March 2010 (2010-03-15), XP055344683, Retrieved from the Internet [retrieved on 20170210]
246: "21 CFR 177.1200 - Cellophane", 11 March 1989 (1989-03-11), XP055344697, Retrieved from the Internet [retrieved on 20170210]
C.W. JONES; W.J. KOROS: "Characterization of Ultramicroporous Carbon Membranes with Humidified Feeds", IND. ENG. CHEM. RES., vol. 34, 1995, pages 158 - 163
S. LAGORSSE; F.D. MAGALHAES; A. MENDES: "Aging study of carbon molecular sieve membranes", J. MEMBR. SCI., vol. 310, 2008, pages 494 - 502, XP022440407, DOI: doi:10.1016/j.memsci.2007.11.025
C. NGUYEN; D.D. DO: "Adsorption of supercritical gases in porous media: determination of micropore size distribution", J. PHYS. CHEM., vol. 103, 1999, pages 6900 - 6908
C. NGUYEN; D.D. DO; K. HARAYA; K. WANG: "The structural characterization of carbon molecular sieve membrane (CMSM) via gas adsorption", J. MEMBR. SCI., vol. 220, 2003, pages 177 - 182, XP004443547, DOI: doi:10.1016/S0376-7388(03)00219-9
L.M. ROBESON: "The upper bound revisited", J. MEMBR. SCI., vol. 320, 2008, pages 390 - 400, XP022761147, DOI: doi:10.1016/j.memsci.2008.04.030
M. TEIXEIRA; M.C. CAMPO; D.A. TANAKA; M.A. TANCO; C. MAGEN; A. MENDES: "Composite phenolic resin-based carbon molecular sieve membranes for gas separation", CARBON, vol. 49, 2011, pages 4348 - 4358, XP028255243, DOI: doi:10.1016/j.carbon.2011.06.012
TANAKA; M.A. TANCO; C. MAGEN; A. MENDES: "Carbon-A120 -Ag composite molecular sieve membranes for gas separation", CHEM. ENG. RES. DES, vol. 90, 2012, pages 2338 - 2345
X. MA; B.K. LIN; X.WEI; J. KNIEP; Y.S. LIN: "Gamma-Alumina Supported Carbon Molecular Sieve for Propylene/Propane Separation", IND. ENG. CHEM. RES., vol. 52, 2013, pages 4297 - 4305
K-I. OKAMOTO; S. KAWAMURA; M. YOSHIRO; H. KITA: "Olefin/Paraffin Separation through Carbonized Membranes Derived from an Asymmetric Polyimide Hollow Fiber Membrane", IND. ENG. CHEM. RES., vol. 38, 1999, pages 4424 - 4432, XP002750617, DOI: doi:10.1021/ie990209p
Attorney, Agent or Firm:
TEIXEIRA DE CARVALHO, Anabela (PT)
Download PDF:
Claims:
CLAIMS

Carbon molecular sieve membrane obtained by carbonization of a carbonaceous polymer precursor having hydrophilic surface elements with a predetermined heating protocol up to a predetermined end carbonization temperature.

Membrane according to the previous claim wherein the polymer precursor is cellophane.

Membrane according to the previous claims wherein the hydrophilic surface elements are metallic and semi-metallic elements and oxygen functional groups.

Membrane according to the previous claims wherein the metallic elements are select of a list consisting of: lithium, sodium, potassium, magnesium, cupper, silicon, nickel, iron, calcium, barium and their oxides.

Membrane according to the previous claims wherein the oxygen functional groups comprise hydroxyl, carbonyl, ether, esther and carboxylic.

Membrane according to the previous claims wherein the said carbon molecular sieve membrane is age free after exposed to a humidified stream comprising at least 50 % relative humidity (RH) - at 25 °C and 1 bar.

Membrane according to the previous claim wherein the said carbon molecular sieve membrane the said carbon molecular sieve membrane is age free after exposed to a humidified stream comprising at least 80 % relative humidity (RH) - at 25 °C and 1 bar.

Membrane according to the previous claims comprising large pores between 0.5- 0.8 nm interconnected to pores between 0.3-0.5 nm.

9. Membrane according to the previous claims wherein the end carbonization temperature is between 500-625 °C.

10. Membrane according to the previous claim wherein the end carbonization temperature is between 525-600 °C.

11. Membrane according to the previous claim wherein the end carbonization temperature is between 500-575 °C.

12. Membrane according to any of the previous claims wherein the carbonization of the cellophane film has a heating rate between 0.1-10 C-min-1.

13. Membrane according to any of the previous claims wherein the carbonization of the cellophane film has a heating rate between 0.5-5°C-min-l, preferably 1 °C-min- 1.

14. Membrane according to any of the previous claims wherein the carbonization of the cellophane film has one or more dwells, in particular comprising a first dwell at 110 °C.

15. Membrane according to the previous claims wherein the cellophane film is obtained by the viscose process.

16. Membrane according to the previous claims wherein the membrane thickness varies between 1-10 μιη.

17. Membrane according to the previous claims further comprising a support reinforced with inorganic particles, wherein the support comprises the same shrinkage ratio during the carbonization step compared with the carbon molecular sieve membrane.

18. Membrane according to the previous claim wherein inorganic particles of the support are boehmite and/or glass micro fibers.

19. Hollow fiber carbon comprising the carbon membrane described in any of the previous claims, wherein the membrane comprises a thickness between 1 - 10 μιη.

20. Use of the carbon molecular sieve membrane described in any of the previous claims in a humidified gas streams comprising a constant gas permeation up to 80 % of relative humidity.

21. Use of the carbon molecular sieve membrane according to the previous claim for the gas separation of the following mixtures: air dehumidification, nitrogen and oxygen from air, natural gas dehydration, separation of hydrogen from syngas, removal of C02 from the natural gas wells, recover of hydrogen from natural gas, C02 separation from flue gas, helium recovery/separation, ethene/ethane or xenon separation from gas mixtures.

22. Method to obtain a carbon molecular sieve membrane described in any of the previous claims comprising:

providing a cellophane precursor produced by the viscose process, comprising hydrophilic surface elements wherein the hydrophilic surface elements are metallic and semi-metallic and oxygen functional groups;

carbonizing the precursor with a predetermined heating protocol up to a predetermined end carbonization temperature, wherein the end carbonization temperature is between 500-625 °C.

23. Method according to the previous claim wherein the end temperature is between 525-575 °C, preferably 550 °C.

24. Method according to any of claims 22-23, wherein the carbonization of the cellophane film has a heating rate between heating rate 0.1-10 C-min-1, preferably 0.5-5, more preferably 1 °C-min-l.

25. Method according to any of claims 22-24, wherein the carbonization of the cellophane precursor is in a controlled atmosphere, preferably of nitrogen or vacuum atmosphere.

26. Method according to any of claims 22-25, wherein the carbonization of the cellophane precursor is in an atmosphere comprising a nitrogen flow rate ranging between 100-200 ml min Λ.

27. Method according to any of claims 22-26 ,wherein the cellophane precursor obtained by the viscose process comprises 5-10 wt.% of moisture, 0-15 wt.% of plasticizers, 0.5-2 wt.% of sodium sulphate and 80-90 wt.% of cellulose.

Method according to claims 22-27, wherein the plasticizer is between 5-15 wt.%.

29. Method according to claims 23-29, wherein the plasticizers is select from a list consisting of: glycerol, PEG, monopropylene glycol, urea, and mixtures thereof.

Description:
D E S C R I P T I O N

A CARBON MOLECULAR SIEVE MEMBRANE, METHOD OF PREPARATION AND USES

THEREOF

Technical field

[0001] The present disclosure relates to a novel carbon molecular sieve membranes (CMSM). More particularly, the disclosure relates to the polymeric precursor and carbonization conditions needed for producing a high-selective, age-free CMSM. More specifically, the present subject-matter discloses a process for obtaining from a cellophane precursor and in a single carbonization step a CMSM that displays no pore blockage in the presence of water vapor and displaying very high ideal permselectivities and permeabilities for industrially relevant gas mixtures.

Background

[0002] Carbon molecular sieve membranes (CMSM) were proposed first by Koresh and Soffer in 1980 and studied since them. Though the very promising high permselectivities and permeabilities displayed by these membranes, they still not commercial.

[0003] CMSM are prepared from the controlled carbonization of a polymeric precursor; pre- and post-treatments are normally applied to improve the separation performance and stability of the prepared membranes. This precursor plays a key role in the production of carbon molecular sieve membranes since different precursors carbonized in the same conditions bring carbon membranes with different properties. The carbonization conditions (carbonization end temperature, heating rate, atmosphere and soaking time) are also of extreme importance for the resulting CMSM performance since they tailor the pore size and structure of carbon membranes.

[0004] CMSM are known to have high permeability of gases, high selectivity and to be thermal and chemically stable. However, they present significant complications related to their performance stability. All reported and disclosed CMSM suffer pore blockage in the presence of water vapor, normally above 30 % of RH, and display oxygen chemisorption that in some cases is most severe than in others. The chemisorbed oxygen reduces the pore size and affects the separation performance of the membrane. On the other hand when CMSM are exposed to humidity, water initially adsorbs onto hydrophilic sites and once the first water molecule is adsorbed, adsorbate-adsorbate interactions will promote the adsorption of further molecules through hydrogen bonds. When the resulting water cluster reaches the dispersive energy for rolling over the hydrophobic path, it eventually encounters a gas selective pore constriction blocking it. In consequence of this pore blockage, the membrane permeability decreases, normally abruptly, making it useless.

[0005] Jones and Koros [1] evaluated the performance of single-fiber test modules at various water activity levels (23 %-85 % of relative humidity) and observed that the membranes were unfavorably affected by exposure to water vapor. The membrane performance loss increased with the increase of the relative humidity in the stream. The same authors suggested, in 1995, a coating of the carbon membranes with a very thin film of highly porous and hydrophobic material such as Teflon [2]. This coating provided a protective barrier that reduced the adverse effects on the performance of the membrane caused by the humidity.

[0006] Carbon Membranes Ltd. (Israel) produced mechanically stable hollow fiber dense membranes from cellulose cupra-ammonia precursors [3]. A technique for repairing cracked hollow fibers was developed using a sealant [4] and a method for improving membrane selectivity was also carried out using chemical vapor deposition followed by activation [5]. However, these membranes presented aging problems when exposed to pure oxygen or air and exhibited pore blockage in the presence of humidity [6]. Eventually this company closed in 2001.

[0007] Various regeneration techniques are known in the art. These include thermal regeneration, ultrasonic regeneration, chemical regeneration, electrothermal regeneration and microwave regeneration. However, regeneration techniques are not favored since they are time consuming, expensive and often unsuccessful. [0008] The prior art had never reported a CMSM stable in the presence of a humidified stream above 40 % RH. The present patent discloses the process and applications of a CMSM stable to oxygen and humidified streams, produced in a single carbonization step and displaying extremely high gas separation performances.

[0009] Cellophane is a thin, transparent sheet made of regenerated cellulose. "Cellophane" is in many countries a registered trademark of Innovia Films Ltd. Cellophane remains a trademark in many countries in Europe and elsewhere, in the United States it is, by court decision, a generic name.

[0010] Cellophane emerged from a series of efforts conducted during the late 19th century to produce artificial materials by the chemical alteration of cellulose, a natural polymer obtained in large quantities from wood pulp or cottonlinters. In 1892 English chemists Charles F. Cross and Edward J. Bevan patented viscose, a solution of cellulose treated with caustic soda and carbon disulfide. Viscose is best known as the basis for the man-made fiber rayon, but in 1898 Charles H. Steam was granted a British patent for producing films from the substance.

[0011] These facts are disclosed in order to illustrate the technical problem addressed by the present disclosure.

General Description

[0012] It was surprisingly found , that contrary to what it was expected the membranes showed no noticeable pore blockage when treating humidified stream up to ca. 80 % RH, since the state of the art indicates otherwise. Preferably, the permeation of some species as well as separation factors was improved. Moreover, these membranes showed no noticeable oxygen chemisorption after a conventional treatment with propylene. This unprecedented result was obtained with a single carbonization step. This fact opens the door to a number of novel applications of these membranes. These new CMSM are advantageous for several industrial applications such as separation of nitrogen and oxygen from air, separation of hydrogen from syngas, removal of C0 2 from the natural gas wells, recover of hydrogen from natural gas, C0 2 separation from flue gas, air dehumidification, natural gas dehydration, helium recovery/separation, ethene/ethane or xenon separation from gas mixtures. CMS membranes can be flat sheet or hollow fibers and supported or unsupported.

[0013] The present disclosure relates to a novel carbon molecular sieve membranes. More particularly, the disclosure relates to the polymeric precursor and carbonization conditions needed for producing a high-selective, age-free CMSM.

[0014] More specifically, the present subject-matter discloses a process for obtaining CMSM from a cellophane film and in a single carbonization step.

[0015] The carbon molecular sieve membrane obtainable by the process described displays no pore blockage in the presence of water vapor and displaying very high ideal permselectivities and permeabilities for industrially relevant gas mixtures.

[0016] The cellophane film/precursor used in the present disclosure was obtained by the viscose process.

[0017] An aspect of the present disclosure relates to a carbon molecular sieve membrane obtained by carbonization of a carbonaceous polymer precursor comprising hydrophilic surface elements, in particular homogeneously distributed, with a predetermined heating protocol up to a predetermined end carbonization temperature. The hydrophilic sites allow water molecules to jump along the surface from hydrophilic group in hydrophilic group avoiding the formation of water clusters and displaying a quasi-linear water vapor adsorption isotherm.

[0018] The carbon molecular sieve membrane pore size distribution is controlled independently by the carbonization end temperature.

[0019] In an embodiment for better results, the polymer precursor may be cellophane.

[0020] In an embodiment for better results, the hydrophilic surface elements can be metallic and semi-metallic elements and oxygen functional groups.

[0021] In an embodiment for better results, the metallic elements may be select of a list consisting of: lithium, sodium, potassium, magnesium, cupper, silicon, nickel, iron, calcium, barium and their oxides. [0022] In an embodiment for better results, the hydrophilic surface elements are oxygen functional groups such as: hydroxyl, carbonyl, ether, esther and carboxylic.

[0023] In an embodiment for better results, the carbon molecular sieve membrane is age free - i.e. that displays no pore blockage - when submitted to a humidified stream comprising 50 % relative humidity (RH) - at 25 °C and 1 bar. Preferably, the carbon molecular sieve membrane is age free when submitted to a humidified stream comprising 80 % relative humidity (RH) - at 25 ° C and 1 bar. The measurement of the pore blockage can be performed by several methods, in this disclosure the pore blockage was measure as described in example 1.

[0024] In an embodiment for better results, the metallic elements are clusters of atoms, in particular clusters size in the range 1-20 atoms.

[0025] In an embodiment for better results, the carbon molecular sieve membrane of the present disclosure comprises large pores between 0.5-0.8 nm interconnected to pores between 0.3-0.5 nm. The measurement of the membrane pore size distribution can be performed by several methods, in this disclosure the pore size distribution was measured by the method proposed by Do et al. [7, 8] for carbonaceous materials.

[0026] In an embodiment for better results, the end carbonization temperature is between 500-625 °C, preferably between 525-600 °C, preferably between 500-575 °C.

[0027] In an embodiment for better results, the carbonization of the cellophane film has a heating rate between 0.1-10 C-min-1, preferably the cellophane film has a heating rate between 0.5-5°C-min-l, preferably 1 °C-min-l.

[0028] In an embodiment for better results, the carbonization of the cellophane film has one or more dwells, in particular comprising a first dwell at 110 °C.

[0029] In an embodiment for better results, the cellophane film is obtained by the viscose process.

[0030] In an embodiment for better results, the membrane thickness varies between 1-10 μιη. [0031] In an embodiment for better results, carbon molecular sieve membrane of the present disclosure may further comprising a support reinforced with inorganic particles, wherein the support comprises the same shrinkage ratio during the carbonization step compared with the carbon molecular sieve membrane, preferably inorganic particles of the support are boehmite and/or glass micro fibers.

[0032] In an embodiment for better results, the shrinkage ratio is controlled by adding further comprising phosphoric acid catalyst to the cellophane precursor.

[0033] In an embodiment for better results, the addition plasticizers improve the mechanical flexibility of the carbon molecular sieve membrane of the present disclosure. For better results the plasticizer concentration is between 5-15 wt.%, improving the control of the pore size distribution. Preferably the plasticizers may be select from a list consisting of: glycerol, PEG, monopropylene glycol, urea, and mixtures thereof.

[0034] Another aspect of the present disclosure relates to hollow fiber carbon comprising the carbon molecular sieve membrane of the present disclosure, wherein the membrane comprises a thickness between 1 - 10 μιη.

[0035] Another aspect of the present disclosure relates to the use of the carbon molecular sieve membrane described in the present disclosure in a humidified gas streams comprising a constant gas permeation up to 80 % of relative humidity.

[0036] In carbon molecular sieve membrane described in the present disclosure may be use in gas separation, in particular for several applications as: air dehumidification, separation of nitrogen and oxygen from air, natural gas dehydration, separation of hydrogen from syngas, removal of C02 from the natural gas wells, recover of hydrogen from natural gas, C02 separation from flue gas, helium recovery/separation, ethene/ethane or xenon separation from gas mixtures.

[0037] Another aspect of the present disclosure relates to a method to obtain the carbon molecular sieve membrane described in the present disclosure comprising: providing a cellophane precursor produced by the viscose process, comprising hydrophilic surface elements wherein the hydrophilic surface elements are metallic and semi-metallic and oxygen functional groups; carbonizing the precursor with a predetermined heating protocol up to a predetermined end carbonization temperature, wherein the end carbonization temperature is between 500-625 °C, preferably the end temperature is between 525-575 °C, more preferably 550 °C.

[0038] In an embodiment for better results, the carbonization of the cellophane film has a heating rate between heating rate 0.1-10 C-min-1, preferably 0.5-5, more preferably 1 °C-min-l.

[0039] In an embodiment for better results, the carbonization of the cellophane precursor/film is in a controlled atmosphere, preferably of nitrogen or vacuum atmosphere.

[0040] In an embodiment for better results, the carbonization of the cellophane precursor film is in an atmosphere comprising a nitrogen flow rate ranging between 100-200 ml min-1.

[0041] In an embodiment for better results, the plasticizer has a concentration between 5-15 wt.%.

[0042] These and other features of the present invention will become readily apparent upon further review of the following specification and figures.

Brief Description of the Drawings

[0043] The following figures provide preferred embodiments for illustrating description and should not be seen as limiting the scope of invention.

[0044] FIG.l Temperature history to prepare carbon molecular sieve membranes. [0045] FIG.2 Scanning electron images of a CMSM carbonized at 550 °C (A) cross- section; (B) surface view.

[0046] FIG. 3 Water vapor adsorption and desorption equilibrium data at 25 °C on a carbon molecular sieve membrane carbonized at 550 °C.

[0047] FIG. 4 Micropore size distribution for carbon molecular sieve membrane carbonized at 550 °C.

[0048] FIG.5 Gas permeation results for O2/N2 in carbon membranes carbonized at different temperatures and comparison with the respective upper bound limit.

[0049] FIG.6 HRTEM pictures of a (A) cellophane-based CMSM and (B) carbon molecular sieve membrane from Fraunhofer.

[0050] FIG. 7 TPD spectra of carbon molecular sieve membranes prepared at 400 °C and 550 °C and a comparison with the spectra of hollow fiber membranes by Carbon Membranes Ltd: (a) CO evolution; (b) CO2 evolution.

[0051] FIG. 8 Illustrative figure of the high flexibility of the cellophane based-CMSM. Figure proving the high flexibility of a CMSM produced at 550 °C.

Detailed Description

[0052] Carbon molecular sieve membranes contain pores larger than the ultramicropores required for the molecular sieving process. These larger pores connect ultramicropores that perform the molecular sieving mechanism and allow for high permeabilities. Generally, an increase in the carbonization end temperature results in a decrease in the gas permeability and an increase in the ideal selectivity, since higher carbonization temperatures originate CMSM with smaller interplanar spacing between the graphite-like layers of carbon.

[0053] The carbonization process is a very important step during the production of carbon molecular sieve membranes and can be considered the heart of the carbon membrane fabrication process. The polymer precursor is heated to the desired end temperature in a controlled atmosphere at a specific heating rate, originating an amorphous carbon membrane with a very narrow porosity that is responsible for the molecular sieve properties of the carbon membrane. Carbonization temperature lies between the decomposition temperature of the carbonaceous precursor and its graphitization temperature. The optimum carbonization temperature depends strongly on the type of the precursor. The choice of a suitable precursor is fundamental to guarantee the production of defect-free CMSM and it should withstand high temperature treatment without much shrinkage, should be thermosetting to avoid melting or softening during any stage of the carbonization process besides having a high carbon yield. Appropriate polymeric precursors include cellulose and derivatives thereof, polyfurfuryl alcohol, polyamides, polyimides, phenolic resins, acrylics, and the like. Other precursor materials may be useful for the preparation of ca rbon membranes and therefore the mentioned suitable precursors are not limiting.

[0054] CMSM suffer often quite fast oxidation in contact with ambient air and easily loses permeability (it may decrease orders of magnitude) due to water or other vapors adsorption as molecular clusters. Normally, the water vapor adsorption isotherm displays a S-shape behavior. This shape is related to the hydrophobic nature of the CMSM; for RH lower than ca. 30 %, the water molecules adsorb in hydrophilic sites and desorb permeating the membrane. However, above this humidity a catastrophic adsorption behavior is normally observed. Water molecules still adsorb in hydrophilic sites, but since the vapor concentration is high enough, other water molecules successively establish H-bonding until a m size cluster is reached. This behavior ends with a steep increase in the adsorbed water molecules and in pore blockage. Indeed, the resulting water clusters gain enough dispersion energy to detach from the hydrophilic site and move over the hydrophobic path towards pore constrictions, blocking them. Also, when CMSM are exposed to air, even at room temperature, oxygen atoms from air combine with some active sites forming oxygen surface groups that reduce the open porosity and increase the surface hydrophilicity.

[0055] To overcome the water vapor blockage problem and age-free behavior, the carbonized precursor shall have a hydrophilic character. Cellophane material was found to originate CMSM that meet these requirements. Cellophane is a natural polymer of glucose, 100 % biodegradable and inexpensive, produced from wood cellulose by the viscose process. CMSM made from cellophane paper show hydrophilic sites homogeneously distributed in its inner surfaces that allow water molecules to jump smoothly between sites avoiding the formation of water clusters. These hydrophilic sites are made of metallic and semi-metallic elements that incorporated during the polymeric film production, namely ionic sodium and silica nanoparticles, and oxygen functional groups such as hydroxyl, carbonyl, ether, esther and carboxylic. CMS membranes carbonized from cellophane under optimized conditions show a linear or quasi-linear water vapor adsorption isotherm. This linear isotherm is characteristic of an adsorbent displaying hydrophilic sites homogeneously distributed. In the present subject-matter, permeation studies demonstrated that up saturation, humidity does not affect membrane ability to permeate and separate gases. Moreover, these membranes show no noticeable oxygen chemisorption after a conventional treatment with propylene.

[0056] Carbonization operating conditions have a significant impact on the final properties of carbon membrane. Generally, an increase in the carbonization end temperature results in a decrease in the gas permeability and an increase in the ideal selectivity; the same is valid for higher soaking times but its effect is much more moderate. Carbonization under vacuum commonly originates more selective and less permeable CMSM when compared to a carbonization under an inert atmosphere. The carbonization history controls then the pore size distribution but also the surface chemistry. For controlling these two characteristics independently it is necessary to have two independent factors. In the present disclosure, CMSM with different pore size distributions are obtained even if produced with the same carbonization history by varying the cellulose concentration in the cellophane precursor film. Moreover, plasticizers, which are incorporated during the cellophane film production to avoid the precursor embrittlement, similarly display a relevant role in the CMSM performance: they are also able to tailor the pore size distribution of the carbon membranes independently of the carbonization end temperature besides improving the mechanical flexibility of the final carbon film/hollow fiber. [0057] In the present disclosure, surface chemistry studies demonstrate that the final carbonization temperature controls also the membrane surface chemistry. Carbonization is preferably carried out under an inert atmosphere, e.g. of nitrogen. The end carbonization temperature ranges between 400 °C and 900°C, preferably between 500 °C and 600 °C. The heating rate should be between 0.1 °C-min 1 and 10°C-min _1 , preferably between 0.5 °C and 5 °C-min 1 and more preferably 1 "C-min -1 . During the heating protocol various dwells are applied; these dwells are chosen according to the obtained thermogravimetric results, i.e. where the first derivative of mass loss shows a significant negative peak a dwell is applied. The first dwell, at ca. 110 °C, is very important for drying the membrane precursor and avoid the formation of cracks or defects in the carbon matrix and the subsequent dwells are needed to fine the pore size distribution. Dwells may be avoidable if heating rates are kept low. Once the end temperature is reached, the membranes are allowed to cool until room temperature before being removed from the furnace.

[0058] Carbon membranes can be divided in supported and unsupported. Unsupported membranes can be hollow fibers, flat membranes or capillary and supported membranes can be flat or tubular. The present carbon membranes can take any of these forms but are preferably unsupported or supported hollow fibers.

[0059] The disclosed CMSM are advantageous for industrial applications such as separation of nitrogen and oxygen from air, separation of hydrogen from syngas, removal of C0 2 from the natural gas wells, recover of hydrogen from natural gas, C0 2 separation from flue gas, air dehumidification, natural gas dehydration, helium recovery/separation, ethene/ethane and xenon separation from gas mixtures.

[0060] The invention will now be further described with reference to the following Examples, which are considered to be illustrative only, and non-limiting. EXAMPLE 1

Extremely high performance CMSM for humidified O2/N2 separation

[0061] Cellophane paper with c.a. 20 μιη thickness produced by the viscose process and containing 6.2 wt.% of moisture, 13.2 wt.% of plasticizers (glycerol and urea), 1.05 wt.% of sodium sulphate and 80.6 wt.% of cellulose was used as precursor.

[0062] Previous to the carbonization step, the precursor film was cut in disks with 48 mm in diameter. The carbonization of the cellophane films was accomplished in a quartz tube (80 mm in diameter and 1.5 m in length) inside a tubular horizontal Termolab TH furnace.

[0063] The basic protocol had an end temperature ranging between 500 °C and 900 °C, preferably between 525 °C and 575 °C, and more preferably 550 °C, without soaking time, a nitrogen flow rate ranging between 100-200 ml min 1 and a slow heating rate with some dwells to avoid a quick release of residual solvents and volatile matter that could damage the carbon matrix. The protocol is pictured in FIG.l. After the end temperature was reached, the system was allowed to cool naturally until room temperature and the carbon membranes were removed from the tubular furnace.

[0064] Micrographs of the produced carbon molecular sieve membranes were taken by scanning electron microscopy (SEM). FIG.2 presents the surface and cross-sectional views of an obtained CMSM; clusters of microspheres are observed. These microspheres are related to cellulose hydrothermal carbonization that happens at around 220 °C and originates more stable oxygen groups in the core (ether, quinone and pyrrone) and more hydrophilic oxygen groups in the shell (hydroxyl, carbonyl, carboxylic and esther).

To perform the permeation experiments, the carbon membranes were glued to steel o-rings. Epoxy glue (Araldite 8 Standard) was also applied along the interface of the steel o-ring and the carbon membrane. A sintered metal disc covered with a filter paper was used as support for the film in the test cell. Single gases were tested at 25 °C, feed pressure of 1 bar and vacuum at the permeate side. The tests were performed in a standard pressure-rise setup with LabView ® data logging. The system included the membrane module connected to a vessel with a calibrated volume at the permeate side and connected also to a gas cylinder at the feed side. The feed gas could either be used dry or passed through a bubbler with distilled water prior to the membrane module. The relative humidity was checked with a RH meter at an exit port. The permeability,?,, of the CMSM towards to pure component i was determined as described by the following equation:

where F! X j is the flux of the speciesi, AP t the partial pressure difference of species ί across the membrane and 8 the membrane thickness.

[0065] The ratio of two gases permeability coefficients is termed permselectivity (often designated as ideal selectivity):

[0066] Table 1 shows the permeation results for the produced CMSM when exposed to different levels of relative humidity for several hours.

TABLE 1

[0067] After CMSM contacting with a humidified oxygen stream (RH humidity relative- of 75-77 %), the overall permeability increased ca. 2.5 times (the permeability to oxygen stays roughly constant) and O2/N2 ideal selectivity increase from 19.0 to 41.2. [0068] Carbon materials are generally poorly wetted by water presenting large contact angles. Contact angles measurements were performed in a surface energy evaluation system (DataPhysics OCA-Series) using the sessile drop method and water, ethylene glycol and n-hexadecane as probe liquids. A needle connected to a microsyringe was used to place the liquid drops on the surfaces. For each drop 150 points were collected. Table 2 shows the measured contact angles for a carbon molecular sieve membrane produced at 550 °C.

TABLE 2

Measured Contact Angles (deg)

Sample

Water Ethylene glycol n-Hexadecane

CMSM 550 17.7 ± 0.9 13.9 ± 3.3 0.0 ± 0.0

[0069] The prepared CMSM 550 sample shows the highest hydrophilic character displaying a small contact angle.

[0070] The adsorption and desorption equilibrium isotherms of water vapor were obtained for the produced carbon molecular sieve membranes by the gravimetric method using a suspension magnetic balance from RubotherrrT at 25 °C. The sample was fragmented into flakes in order to fill the weighting basket. FIG. 3 shows the experimental values as a function of relative pressure, considering 32 mbar the water vapor at 25 °C. The experimental adsorption/desorption branches are well described by a quasi-linear isotherm. The significant amount at very low relative pressures confirms the hydrophilic character of the developed carbon molecular sieve membranes.

[0071] The pore size distribution and the porosity volume were obtained for carbon molecular sieve membranes based on the adsorption equilibrium isotherm of C0 2 at 0 °C using a suspension magnetic balance from Rubotherm ® . To perform the experiments the sample was also fragmented into flakes. FIG. 4 shows the obtained pore size distribution and Table 3 the micropore volume, the mean pore width and the characteristic energy for adsorption.

TABLE 3

Parameter CMSM 550

Micropore volume (cm 3 kg -1 ) 274.2

Mean pore width (nm) 0.435

Characteristic energy (kJ mol "1 ) 10.92

[0072] In an embodiment, CMSM 550 presents larger pores of 0.5-0.8 nm known as micropores (responsible for surface diffusion mechanism) interconnected to ultramicropores of 0.3-0.5 nm responsible for the molecular sieving properties of carbon membranes.

EXAMPLE 2

Extremely high performance CMSM for dry O2/N2 separation

[0073] Carbon molecular sieve membranes are prepared from cellophane according to Example 1, varying the carbonization end temperature between 575 °C and 625 °C, more preferably 600 °C. Table 4 shows the oxygen and nitrogen permeation results for the produced CMSM.

TABLE 4

Sample Nitrogen Permeability Oxygen Permeability O2/N2 ideal

(Barrer) (Barrer) selectivity

CMSM 600 <0.001 0.78 >800

[0074] The produced CMSM showed to be very selective to O2/N2 (ideal selectivity > 800) separation. EXAMPLE 3

Other gas separations

[0075] Carbon molecular sieve membranes are prepared from cellophane precursor according to Example 1, varying the carbonization end temperature between 400 °C and 600 °C. Table 5 and 6 show the obtained permeabilities and ideal selectivities for several dry gases using CMSM produced at different carbonization end temperatures, respectively.

TABLE 5

Permeability (Barrer)

CMSM

N 2 02 He H 2 CO2 CH 4 H2O CaHe CaHg

CMSM 400 0.07 0.73 5.43 8.35 3.39 0.008 12.1 n.d* n.d*

CMSM 500 0.06 0.95 10.24 18.94 8.21 n.d* 16.0 n.d* n.d*

CMSM 550 0.07 1.33 17.26 32.59 13.0 0.01 28.5 0.056 0.035

CMSM 600 <0.001 0.78 11.78 24.90 2.57 «0.001 25.1 0.065 0.025 n.d* = not determined

TABLE 6

Ideal Selectivity

CMSM

O2/N2 H2/N2 H2/CH4 CO2/N2 CO2/CH4 He/N 2 H2/O2 CaHe/Cal

CMSM 400 10.4 119.3 1044 48.4 423.8 77.57 11.4 n.d*

CMSM 500 15.8 315.7 n.d* 136.8 n.d* 170.7 19.9 n.d*

CMSM 550 19.0 465.6 3259 185.7 1300 246.6 24.5 1.6

CMSM 600 >800 >25 000 »25 000 >2600 »2600 >1200 44.5 2.6 n.d* = not determined

[0076] The results show that increasing the carbonization end temperature results in an increase in the carbon molecular sieve membrane selectivity. [0077] The upper bound relationship for membrane gas separations correlates the log of the selectivity versus the log of the permeability of the more permeable gas for the best performing polymer membranes and it was devised by Robeson [9]. FIG.5 shows the permeation results of 0 2 and N 2 in carbon membranes carbonized at 500 °C, 550 °C and 600 °C end temperatures and the upper bound limit. This figure illustrates that the CMSM prepared have a separation performance well above the referred upper bound limit.

[0078] The separation performance is extremely high, particularly for the carbon molecular sieve membrane produced at 600 °C. However, the C3H6/C3H8 ideal selectivity is extremely low when compared to other values reported in literature. After a brief research in the literature, it was perceived that when carbon molecular sieves shows high 0 2 /N 2 ideal selectivities, low C3H6/C3H8 ideal selectivities are then obtained and vice-versa (Table 7). This indicated the existence of two different sieving mechanisms for carbon molecular sieves: a gate sieving (for CMS with rounded shape pores) and a tunnel sieving (for CMS with needle shape pores). Figure 6 shows HRTEM pictures of a carbon molecular sieve membrane presenting a pore morphology of a gate sieving mechanism (Figure 6a) and a carbon molecular sieve membrane with a tunnel sieving mechanism (Figure 6b).

TABLE 7

Sample

ideal selectivity ideal selectivity

CMSM 550 1.6 19

CMSM 600 2.6 >800

CMSM [ as described in reference 10] 14.6 5.4

CMSM [ as described in reference 11] 5.8 37.8

CMSM [ as described in reference 12] 5.1 38

CMSM [as described in reference 13] 13 3.3 [0079] Contact angles measurements were performed in the different carbon molecular sieve membranes according to Example 1. Table 8 shows the measured contact angles for the different membranes.

TABLE 8

Measured Contact Angles (deg)

Sample

Water Ethylene glycol n-Hexadecane

CMSM 400 57.5 ± 2.3 28.5 ± 5.9 0.0 ± 0.0

CMSM 500 32.3 ± 4.6 24.9 ± 3.1 0.0 ± 0.0

CMSM 550 17.7 ± 0.9 13.9 ± 3.3 0.0 ± 0.0

CMSM 600 20.2 ± 0.8 13.9 ± 3.3 0.0 ± 0.0

[0080] The results show that with increasing the carbonization end temperature the produced carbon molecular sieve membranes become more hydrophilic until reaching a maximum at ca. 550 °C.

[0081] Temperature-programmed desorption (TPD) provides information on the surface functional groups on carbon materials. Surface complexes on carbon materials decompose upon heating by releasing CO and C0 2 at different temperatures. The nature of the groups can be assessed by the decomposition temperature and the gas released.

[0082] TPD experiments were carried out in a U-shaped quartz tube located inside an electrical furnace and connected to a Dycor Dymaxion Mass Spectrometer (Ametek Process Instruments). The samples (0.1 g) were heated to 1100 °C at 5 °C min 1 using a constant He flow rate of 25 ml min -1 . FIG. 7 shows the TPD profiles of the carbon molecular sieve membranes produced at 400 °C and 550 °C comparing with a hollow fiber membrane produced by Carbon Membranes Ltd. [0083] The results show that the TPD spectra of CMSM 400 and CMSM 550 prepared from the same precursor film are quite distinct. For CMSM 400 sample, both CO and C0 2 peaks originate from the decomposition of carboxylic anhydrides. For CMSM 550 sample, the first CO2 peak may be attributed to lactone groups and the higher temperature CO2 peak may be result from carboxylic anhydrides. Considering the CO profile, this sample produced at 550 °C shows a first maximum around 650 °C related to carboxylic anhydrides. The second maximum appears at around 780 °C, which originates from quinone groups. Regarding the hollow fibers produced by Carbon Membranes Ltd, some different oxygen surface functional groups were found when compared to the obtained CMSM 550, namely phenol and carboxylic groups. Considering the CO2 profile, lactone and carboxylic groups were identified. Phenol and quinone groups yield CO. Both CO and CO2 originate a peak related to decomposition of carboxylic anhydrides.

[0084] One of the major limitations of carbon molecular sieve membranes is their brittleness which makes their manipulation and assembly into the functional modules very difficult. Although the flexibility of the produced cellophane-based CMSM of the present disclosure tended to decrease as the carbonization end temperature increase, all the prepared membranes are very flexible and could be bent as shown in Figure 8. When a cellophane-based CMSM of the present disclosure is bent, the minimal distance until fracture is c.a. 3 mm (obtained value for a carbon molecular sieve membrane prepared at 550 °C).

[0085] The term "comprising" whenever used in this document is intended to indicate the presence of stated features, integers, steps, components, but not to preclude the presence or addition of one or more other features, integers, steps, components or groups thereof.

[0086] All references recited in this document are incorporated herein in their entirety by reference, as if each and every reference had been incorporated by reference individually. [0087] Where ranges are given, endpoints are included. Furthermore, it is to be understood that unless otherwise indicated or otherwise evident from the context and/or the understanding of one of ordinary skill in the art, values that are expressed as ranges can assume any specific value within the stated ranges in different embodiments of the invention, to the tenth of the unit of the lower limit of the range, unless the context clearly dictates otherwise. It is also to be understood that unless otherwise indicated or otherwise evident from the context and/or the understanding of one of ordinary skill in the art, values expressed as ranges can assume any subrange within the given range, wherein the endpoints of the subrange are expressed to the same degree of accuracy as the tenth of the unit of the lower limit of the range.

[0088] The disclosure should not be seen in any way restricted to the embodiments described and a person with ordinary skill in the art will foresee many possibilities to modifications thereof.

[0089] Where ranges are given, any value within the range may explicitly be excluded from any one or more of the claims. Any embodiment, element, feature, application, or aspect of the compositions and/or methods of the invention, can be excluded from any one or more claims.

[0090] The present disclosure should not be seen in any way restricted to the embodiments described and a person with ordinary skill in the art will foresee many possibilities to modifications thereof.

[0091] The above described embodiments are combinable. The following claims further set out particular embodiments of the disclosure.

[0092] All references recited in this document are incorporated herein in their entirety by reference, as if each and every reference had been incorporated by reference individually.

References

[1] C.W. Jones, W.J. Koros. Characterization of Ultramicroporous Carbon Membranes with Humidified Feeds. Ind. Eng. Chem. Res. 34 (1995) 158-163. [2] W. J. Koros, C.W. Jones, Composite carbon fluid separation membranes, US Patent 5,288,304, (1994)

[3] A. Soffer, J. Gilron, S. Saguee, R. Hed-Ofek, H. Cohen, Process for the production of hollow carbon fibers membranes, Eur. Patent 671,202 (1995).

[4] A. Soffer, S. Saguee, D. Golub, H. Cohen, M. Azariah, Selective clogging of failed fibers, US Patent 5,575,963 (1996).

[5] A. Soffer, A. Moshe, A. Abraham, C. Haim, G. Dan, S. Shlomo, Method of improving the selectivity of carbon membranes by chemical carbon vapour deposition, US Patent 5,695,818 (1997).

[6] S. Lagorsse, F.D. Magalhaes, A. Mendes, Aging study of carbon molecular sieve membranes, J. Membr. Sci. 310 (2008) 494-502.

[7] C. Nguyen, D.D. Do, Adsorption of supercritical gases in porous media: determination of micropore size distribution. . Phys. Chem. 103 (1999) 6900- 6908.

[8] C. Nguyen, D.D. Do, K. Haraya, K. Wang, The structural characterization of carbon molecular sieve membrane (CMSM) via gas adsorption. . Membr. Sci. 220 (2003) 177-182.

[9] L.M. Robeson, The upper bound revisited, J. Membr. Sci. 320 (2008) 390- 400.

[10] M. Teixeira, M.C. Ca mpo, D.A. Tanaka, M.A. Tanco, C. Magen, A. Mendes, Composite phenolic resin-based carbon molecular sieve membranes for gas separation, Carbon 49 (2011) 4348-4358. [9] M. Teixeira, M. Campo, D.A.

[ll]Tanaka, M.A. Tanco, C. Magen, A. Mendes, Carbon-AI 2 03-Ag composite molecular sieve membranes for gas separation. Chem. Eng. Res. Des. 90 (2012) 2338-2345.

[12] X. Ma, B.K. Lin, X.Wei, J. Kniep, Y.S. Lin, Gamma-Alumina Supported Carbon Molecular Sieve for Propylene/Propane Separation. Ind. Eng. Chem. Res.52 (2013) 4297-4305.

[13] K-l. Okamoto, S. Kawamura, M. Yoshiro, H. Kita, Olefin/Paraffin Separation through Carbonized Membranes Derived from an Asymmetric Polyimide Hollow Fiber Membrane. Ind. Eng. Chem. Res.38 (1999) 4424-4432.