Hollow Fiber Membrane For Use in an Anesthetic Circuit

Abstract

Hydrophobic poly(4-methyl-1-pentene) hollow fiber membrane for retention of anesthetic agents with an inner and an outer surface and between inner and outer surface an essentially isotropic support layer with a sponge-like, open-pored, microporous structure free of macrovoids and adjacent to this support layer on the outer surface a dense separation layer with a thickness between 1.0 and 3.5 μm. The membrane has a porosity in the range of greater than 35% to less than 50% by volume and a permeance for CO.sub.2 of 20-60 mol/(h.Math.m.sup.2.Math.bar), a gas separation factor α(CO.sub.2/N.sub.2) of at least 5 and a selectivity CO.sub.2/anesthetic agents of at least 150.

The process for producing this membrane is based on a thermally induced phase separation process in which process a homogeneous solution of a poly(4-methyl-1-pentene) in a solvent system containing components A and B is formed, wherein component A is a strong solvent and component B a weak non-solvent for the polymer component. After formation of a hollow fiber the hollow fiber is cooled in a liquid cooling medium to form a hollow fiber membrane. The concentration of the polymer component in the solution may be in the range from 42.5 to 45.8 wt.-% and the hollow fiber leaving the die runs through a gap between die and cooling medium with a gap length in the range of 5-30 mm.

Claims

1. A membrane comprising: an integrally asymmetrical hydrophobic hollow fiber membrane for retention of anesthetic agents, which membrane is composed primarily of a poly(4-methyl-1-pentene) and has an inner surface facing towards its lumen, an outer surface facing outwards, a support layer with a sponge-like, open-pored, microporous structure between inner surface and outer surface and adjacent to this support layer on the outer surface a separation layer with denser structure, wherein the support layer is free of macrovoids and the pores in the support layer are on average essentially isotropic, wherein separation layer has a thickness in the range between 1.0 and 3.5 μm and, the membrane has a permeance for CO2 of 20-60 mol/(h.Math.m.sup.2.Math.bar), a gas separation factor α(CO2/N2) of at least 5 and a selectivity CO.sub.2/Sevoflurane of at least 150 the membrane has a porosity in the range of greater than 35% to less than 50% by volume.

2. The membrane according to claim 1, wherein the membrane structure changes abruptly in the transition from the separation layer to the support layer.

3. The membrane according to claim 1, wherein the separation layer has a thickness between 1.0 μm and 2.0 μm.

4. The membrane according to claim 1, wherein the membrane has the permeance for CO2 of 25-40 mol/(h.Math.m.sup.2.Math.bar).

5. The membrane according to claim 1, wherein the membrane has the permeance for CO2 of 25-40 mol/(h.Math.m.sup.2.Math.bar) and an O2 permeance of at least 10 mol/(h.Math.m.sup.2.Math.bar).

6. The membrane according to claim 1, wherein the membrane has the selectivity CO2/Sevoflurane of at least 220.

7. The membrane according to claim 1, wherein the membrane has the selectivity CO2/Sevoflurane of at least 400.

8. The membrane according to claim 1, wherein the membrane has the gas separation factor α(CO2/N2) of at least 8.

Description

DESCRIPTION OF THE FIGURES

[0071] FIG. 1 shows a schematic of the experimental system;

[0072] FIG. 2 graphs the Permeance results listed in Table 3;

[0073] FIG. 3 graphs the Selectivity results listed in Table 4;

[0074] FIG. 4 graphs the CO2 Permeance; and

[0075] FIG. 5 graphs the Selectivity results listed in Table 5.

[0076] The following methods were employed to characterize the membranes obtained:

Determination of the Volume Porosity:

[0077] A sample of at least 0.5 g of the membrane to be examined is weighed in a dry state. The membrane sample is then placed for 24 hours into a liquid that wets the membrane but does not cause it to swell, such that the liquid penetrates into all pores. This can be detected visually in that the membrane sample is transformed from an opaque to a glassy, transparent state. The membrane sample is then removed from the liquid, liquid adhering to the sample removed by centrifugation at about 1800 g, and the mass of the thus pretreated wet, i.e., liquid-filled, membrane, determined.

[0078] The volume porosity in % is determined according to the following formula 6:

[00005]$\begin{array}{cc}\mathrm{Volume}.Math..Math.\mathrm{porosity}.Math.\left[\%\right]=100.Math.\frac{\left(m_{\mathrm{wet}}-m_{\mathrm{dry}}\right).Math.\text{/}.Math.{\square }_{\mathrm{liq}.}}{\left(m_{\mathrm{wet}}-m_{\mathrm{dry}}\right).Math.\text{/}.Math.{\square }_{\mathrm{liq}.}.Math.+m_{\mathrm{dry}}.Math.\text{/}.Math.\square {\square }_{\mathrm{polymer}}}& \left(6\right)\end{array}$

where
m.sub.wet=weight of the wet, liquid-filled membrane sample
m.sub.dry=weight of the dry membrane sample
□.sub.liq.=density of the liquid used
□.sub.polymer=density of the membrane polymer

Determination of Permeance and Selectivity:

Gases and Gas Mixtures:

[0079] For determination of membrane permeance and selectivity the following gases and gas mixtures were employed:

[0080] A certified mixture of 5% CO.sub.2 balanced in O.sub.2 was used as a carrier gas in the ternary gas mixture experiments and for calibrating the mass spectrometer. A certified calibration gas mixture of He, O.sub.2 and N.sub.2 was also used to calibrate the mass spectrometer. The gases and gas mixtures used are listed in Table 1:

TABLE-US-00001 TABLE 1 Gas Specification Supplier CO.sub.2 USP grade, 99.5% pure, Praxair, Mississauga, DIN: 02014459 ON, CAN O.sub.2 USP grade, 99.99% pure, Praxair, Mississauga, DIN: 02014408 ON, CAN N.sub.2 Air Liquide, Dartmouth, NS, CAN 5% CO.sub.2 balanced USP grade, 99.5% pure, Praxair, Mississauga, in O.sub.2 DIN: 02014432 ON, CAN He, O.sub.2 and N.sub.2 SPG-3MX0021714 Air Liquide, Dartmouth, NS, CAN

[0081] Sevoflurane (Abbvie, North Chicago, Ill., United States), Desflurane and Isoflurane (both from Baxter Corporation, Deerfield, Ill., United States) were administered with anaesthesia vapourizers (Draeger, Lubeck, Germany). The permeation apparatus was assembled with plastic Legris ¼-inch vacuum rated PTFE fittings and ¼-inch polyethylene tubing, as well as TruWave ⅛-inch tubing and fittings for the permeate line and mass spectrometer sampling point.

Mass Spectrometer

[0082] Gas concentrations or gas compositions for determination of permeance and selectivity were measured using a mass Spectrometer. A quadrupole mass spectrometer (Omnistar Model PTM81217131, Pfeiffer Vacuum, ABlar, Germany) was used to measure the gas composition of the feed, retentate streams as well as the permeate streams for the mixed-gas tests. The mass spectrometer was controlled through Pfeiffer Vacuum's Quadera software (v4.50.004). Mass spectrometer operation, data analysis, display and storage were all controlled via Quadera. It was operated using Secondary Electron Multiplier (SEM) ion detection to achieve a higher resolution and to enable accurate detection at low concentrations. The same operating conditions were used for all of these experiments (SEM voltage of 910 V, Ion Current of 5.1.sup.−09 A, Spectral Resolution of 50 and a dwell of one second).

[0083] The mass spectrometer was calibrated for mass specific concentration determination per Pfeiffer Vacuum's recommended calibration procedure. An offset calibration was performed to eliminate any inherent offset of the measured SEM signal. A mass scale adjust calibration was then conducted to tune the mass scale to adjust the measured value of each mass of interest to an integer value. A background determination calibration was then performed to identify any mass peaks from residual gases that were not associated with the gases being measured. Background spectra from this calibration were subtracted from subsequent measurements. The final calibration was the gas specific calibration, where ion peaks are translated to concentration by the gas specific calibration, which entails the assignment of compounds to mass peaks. A calibration factor library was then constructed using certified calibration gas mixtures. The mass spectrometer calibration was maintained throughout the course of the experiment testing.

[0084] Anaesthesia calibration mixtures were mixed by manually injecting a defined amount of liquid Sevoflurane, Desflurane, or Isoflurane into a syringe filled with oxygen and agitating it to vapourise the liquid. The composition was calculated using the compound properties and the ideal gas law. The mixture was then used to calibrate the mass spectrometer. The concentrations supplied by the vaporizers were then verified with the calibrated device.

Determination of Permeation Characteristics:

[0085] Permeation characteristics of the membranes in the modules were measured using a variation of the constant-pressure variable-volume method.

[0086] FIG. 1 shows a schematic of the experimental system.

[0087] The feed gas flow rate was controlled using a needle valve and flow was measured using a glass tube flow meter (Scott Specialty Gases, Plumsteadville, Pa., USA) and the feed pressure using an analogue pressure gauge (Speidel Keller, Jungingen, Germany, accuracy: ±5 mmHg or ˜6.67 mbar) connected to the lumen outlet stream. The feed pressure on the lumen side was held constant at 1.2 bar for each experiment. The permeate pressure on the shell side was atmospheric. One port of the shell side of the module was plugged and the other port was fed through a 3-way valve either to a modified, horizontally placed glass pipette (Fischer Scientific, 5 mL total volume) for flow measurements or to the mass spectrometer for concentration determination.

[0088] To calculate the mean flow rate, the pipette was filled with water prior to each trial and time measurements were taken at 0.5 mL intervals as the permeated gas displaced the water. Immediately following the volumetric flow rate measurement, the permeate was then fed to the mass spectrometer sampling port and the gas composition only for the mixed gas experiments. Feed gas concentrations were measured when both shell-side ports on the module were plugged. The gas composition was measured from the retentate port of the module. The system was purged with oxygen between each trial.

Pure Gas Permeation Measurements

[0089] The pure gas permeance of the membranes was determined for pure carbon dioxide (CO.sub.2), oxygen (O.sub.2), and nitrogen (N.sub.2) supplied at 1.2 bar. Once the concentration of the gases stabilized, the permeate line was connected to the pipette setup. Four time readings were then measured every 0.5 mL as the permeate displaced the water. These readings were averaged and counted for one trial. A minimum of three trials were performed for each module at each set of operating conditions. A minimum of three trials were performed for each module at each set of operating conditions. The pure gas permeance was calculated according to Equation 7:

[00006]$\begin{array}{cc}{K}_i^P=\frac{Q}{A\left(p_f-p_p\right)}& \left(7\right)\end{array}$

[0090] Where Q is the permeate molar flow rate (mol/h), pf is the absolute pressure of the feed (bar), p.sub.p is the absolute pressure of the permeate (bar), and A is the inside active membrane area (m.sup.2). Permeance is reported in units of mol/h/m.sup.2/bar. The pure gas selectivity (α.sub.ij.sup.P) was calculated based on the ratio of the mean permeance of two pure gases i and j per Equation 8:

[00007]$\begin{array}{cc}{\alpha }_{\mathrm{ij}}^P=\frac{K_i^P}{K_j^P}& \left(8\right)\end{array}$

Where K.sub.i.sup.P and K.sub.j.sup.P are the pure gas permeances for components i and j, respectively.

Mixed Gas Permeation Measurements

[0091] The mixed gas permeation procedure was similar to that of the pure gas system but also required concentration measurements of the permeate. The feed and retentate compositions were determined as described above. In addition, following the flow measurements, the permeate was directed to the mass spectrometer for determination of the permeate composition.

[0092] Anaesthetic vapour concentrations for the experiments were chosen around their Minimum Alveolar Concentration (MAC). The MAC is a measure of potency of the anaesthesia gas and is defined as the minimum concentration of vapour required to immobilize 50% of patients. The MAC for Sevoflurane, Desflurane and Isoflurane is defined as 2.13%, 6.0% and 1.13% respectively. The vapour experiments were conducted with vapour concentrations at the MAC for each vapour. This set of experiments was performed near clinically relevant operating conditions, at room temperature and low feed pressure (1.2 bar) with either Sevoflurane, Isoflurane or Desflurane at their MAC mixed into a 5% carbon dioxide/oxygen gas mixture to mimic exhalation during anaesthesia administration. Three modules of each membrane type were used and three runs of each module were performed for a total of n=9 trials for Sevoflurane, Isoflurane and Desflurane.

[0093] The mixed gas tests entailed monitoring the feed, retentate and permeate gas composition with the mass spectrometer, in addition to the flow rate. The permeances of the mixed gas components were calculated using Equation 9:

[00008]$\begin{array}{cc}{K}_i^M=\frac{y_p.Math.Q}{A.Math..Math.\Delta .Math..Math.p_{i,\mathrm{ave}}}& \left(9\right)\end{array}$

[0094] The log-mean driving force (Equation 4) was used to compute ΔP.sub.i,ave. The mixed gas selectivity, alpha.sup.M, was calculated using the permeance of two components i and j in a gas mixture per Equation 10:

[00009]$\begin{array}{cc}{\alpha }_{\mathrm{ij}}^M=\frac{K_i^M}{K_j^M}& \left(10\right)\end{array}$

Examples 1 to 3 and Comparative Examples 1a, 1b, 2 and 3

[0095] In the Examples 1 to 3 and the Comparative Examples 1a, 1b, 2 and 3, the membranes were manufactured according to the following procedure:

[0096] Poly(4-methyl-1-pentene) (TPX DX845) as polymer component was dissolved in a nitrogen atmosphere in a container with stirrer at a temperature of 260° C. in a solvent system consisting of 70% by weight dioctyl adipate, 20% by weight glycerin triacetate, and 10% by weight castor oil. The concentrations of polymer and solvent system used in Examples 1 to 3 and in Comparative Examples 1a, 1b, 2 and 3 are listed in Table 1.

[0097] After degassing, the resulting clear and homogeneous solution was fed with a gear pump to a hollow-fiber die with an annular-gap outside diameter of 1.2 mm, which had been heated to 248° C., and extruded to form a hollow fiber. Nitrogen was metered into the lumen of the hollow fiber through the interior bore of the hollow-fiber die. After an air section of 24 mm, the hollow fiber passed through an approx. 1 m long spinning tube, through which glycerin triacetate, tempered to ambient temperature, flowed as a cooling medium. The hollow fiber, solidified as a result of the cooling process in the spinning tube, was drawn off from the spinning tube at a rate of 72 m/min, wound onto a spool, subsequently extracted with isopropanol, and then dried at approx. 120° C.

[0098] Hollow-fiber membranes were obtained with an inner diameter of about 200 μm and a wall thickness of about 90 μm, for which no pores were observable on its exterior surface in a scanning-electron-microscopic (SEM) image even at 60000× magnification. The interior surface facing the lumen had an open-pored, network-like structure with approximately circular openings. The sponge-like, open-pored, microporous support structure is covered by an approx. 1.5-2.0 μm thick separation layer.

TABLE-US-00002 TABLE 1 Comp. Comp. Exampl. Comp. Comp. Membrane Ex. 1a Ex. 1b Exampl. 1 Exampl. 2 3 Ex. 2 Ex.3 Polymer (wt. %) 46.0 46.0 45.8 45.5 45.0 42.5 41.0 concentr. Solvent (wt. %) 54.0 54.0 54.2 54.5 55.0 57.5 59.0 system concentr. Solvent 1.17 1.17 1.18 1.20 1.22 1.35 1.44 system/ polymer ratio Fiber wall (μm) 90 90 90 90 90 90 90 thickness Fiber internal (μm) 200 200 200 200 200 200 200 diameter Active inside (cm.sup.2) 24 26 26 26 26 26 26 area module Active (cm.sup.2) 45 49 49 49 49 49 49 outside area module Module % 20.00 20.06 20.06 20.06 20.06 20.06 20.06 filling degree

[0099] With the membranes of each example and comparative example test modules were prepared for testing the properties of the membranes. For this purpose, a minimum of three modules of each membrane type was produced by arranging the hollow fiber membranes in a cylindrical housing and potting the ends of the hollow fiber membranes in tube shells made from polyurethane. The lumina of the hollow fiber membranes were in fluid connection with a feed inlet and a feed outlet, respectively. The space in the module surrounding the hollow fiber membranes was in fluid connection with a permeate outlet.

[0100] Details of the membrane modules are listed in Table 2.

[0101] For the membranes of each example and comparative example the permeance for CO.sub.2, O.sub.2, and N.sub.2, the gas separation factor α(CO.sub.2/N.sub.2) and the selectivity CO.sub.2/anesthetic agents was measured according to the test methods described above.

[0102] The pure gas permeances for carbon dioxide (CO.sub.2), oxygen (O.sub.2) and nitrogen (N.sub.2) are listed in Table 3 and are shown in FIG. 2. The property of most interest for the intended application is the carbon dioxide permeance. In addition, the permeabilities of the three gases are presented in Table 2 for the examples and the comparative examples, wherein the permeabilities are calculated on the basis of the respective permeances by multiplying the permeances by the membrane thickness (fiber wall thickness=90 μm).

TABLE-US-00003 TABLE 3 K.sub.i P.sub.ip (mol/m.sup.2/bar/h) (10.sup.3 mol cm/m.sup.2/bar/h) Selectivity α.sub.ij CO.sub.2 N.sub.2 O.sub.2 CO.sub.2 N.sub.2 O.sub.2 CO.sub.2/O.sub.2 O.sub.2/N.sub.2 CO.sub.2/N.sub.2 Example 1 23.1 1.6 10.7 208 14 96 2.2 6.9 15.0 Example 2 32.3 3.0 12.3 291 27 111 2.6 4.2 10.9 Example 3 31.8 2.7 14.9 286 24 134 2.1 5.7 12.1 Comp. 16.7 2.3 7.1 150 21 64 2.3 3.1 7.2 Ex. 1a Comp. 19.8 1.7 7.8 178 15 71 2.5 4.6 11.5 Ex. 1b Comp. 88.6 10.1 36.4 797 91 328 2.4 3.6 8.8 Ex. 2 Comp. 110.9 10.6 42.2 998 95 380 2.6 4.0 10.5 Ex. 3

[0103] For the membranes of each example and comparative example mixed gas experiments were performed, i.e. experiments with gas mixtures containing either Sevoflurane (2.0 Vol.-%), Isoflurane (6.13 Vol.%), or Desflurane (1.13 Vol.%) with a balance of 5 Vol.% CO.sub.2 in O.sub.2 and the CO.sub.2 permeance, K.sub.CO2 (in (mol/m.sup.2/bar/h)), in these mixtures was compared to the permeance K.sub.CO2 of pure CO.sub.2 as determined in the pure gas experiments (see above). The results obtained are listed in Table 4 and shown in FIG. 3.

TABLE-US-00004 TABLE 4 K.sub.CO2 K.sub.CO2 K.sub.CO2 K.sub.CO2 (Pure (CCO.sub.2/ (CO.sub.2/ (CO.sub.2/ CO.sub.2) O.sub.2/Sevo) O.sub.2/Iso) O.sub.2/Des) Example 1 23.1 15.8 — — Example 2 32.3 33.1 11.4 25.2 Example 3 31.8 23.9 10 40.5 Comp. Example 1a 16.7 10.4 5.3 10.1 Comp. Example 1b 19.8 18.4 7.6 16.6 Comp. Example 2 88.6 63.3 — — Comp. Example 3 110.9 86 — —

[0104] In Table 5 the selectivity of CO.sub.2 over Servoflurane as obtained from the mixed gas experiments is listed together with the pure CO.sub.2 permeance for hollow fiber membranes of Examples 1 to 3 and Comparative Examples 1 to 3. Sufficiently high selectivities CO.sub.2/Servoflurane resulting in a good retention of anesthetic gases by the membranes with at the same time sufficiently high permeances for CO.sub.2 are obtained for the membranes of Examples 1 to 3. FIG. 4 shows a diagram of the selectivity CO.sub.2/Servoflurane in dependency of the pure CO.sub.2 permeance.

TABLE-US-00005 TABLE 5 Pure CO.sub.2 Solvent system/ Permeance Selectivity α.sub.ij polymer ratio (mol/m.sup.2/bar/h) CO.sub.2/Sevoflurane Example 1 1.18 23.1 1130 Example 2 1.20 32.3 550 Example 3 1.22 31.8 250 Comp. Example 1a 1.17 16.7 1450 Comp. Example 2 1.35 88.6 86 Comp. Example 3 1.44 110.9 93

[0105] FIG. 5 plots the CO.sub.2/Sevoflurane selectivity as a function of the CO.sub.2 permeance for the Examples and Comparative Examples listed in Table 5.