Gasket for fuel cells

Abstract

A gasket for sealing two mating surfaces of a fuel cell is described. The gasket has a core layer comprising exfoliated vermiculite. The core layer is interposed between a first and second coating layer, the said coating layers each comprising glass, glass-ceramic and/or ceramic material. Methods for producing gaskets according to the invention are also described. A solid oxide cell or a solid oxide cell component comprising one or more of the gaskets; use of the gasket to improve sealing properties in a solid oxide cell; and a method of producing a solid oxide cell or of sealing a solid oxide cell comprising incorporating at least one of the gaskets into the solid oxide cell are also defined.

Claims

1. A gasket for sealing two mating surfaces of a fuel cell comprising a core layer comprising exfoliated vermiculite and first and second coating layers said core layer interposed between the said first and second coating layers, the said coating layers each comprising glass, glass-ceramic and/or ceramic material.

2. The gasket of claim 1, wherein the core layer has a density prior to use of 1.7-2.0 g/cm.sup.3.

3. The gasket of claim 1, wherein the coating layers each have a glass transition temperature, and wherein the core layer is more compressible than the coating layers when the gasket is at a temperature below the glass transition temperatures of the coating layers.

4. The gasket of claim 1, wherein the core layer is compressible in the direction perpendicular to its facing surfaces.

5. The gasket of claim 1, wherein the exfoliated vermiculite is chemically exfoliated vermiculite (CEV).

6. The gasket of claim 5, wherein the proportion of CEV is at least 30% w/w of the core layer.

7. The gasket of claim 1, wherein the exfoliated vermiculite core layer is in the range of 10 -2000 μm thickness.

8. The gasket of claim 1, wherein the coating layers are of an amorphous, crystalline or semi-crystalline character.

9. The gasket of claim 1, wherein the glass, glass-ceramic and/or ceramic material has a softening temperature in the range between 450 and 1000° C.

10. The gasket of claim 1, wherein each coating layer has a thickness of between 0.1 and 50 μm.

11. The gasket of claim 1, wherein the densities of the glass or glass-ceramic coatings are in the range 2-4 g/cm.sup.3.

12. A method for producing a gasket, comprising; a. applying a glass or glass-ceramic coating layer onto each of two opposed surfaces of an exfoliated vermiculite gasket core layer to form a coated gasket; b. locating the coated gasket in a fuel cell between mating surfaces to be sealed.

13. A method for producing a gasket, comprising; a. applying a glass or glass-ceramic coating layer onto each of two mating surfaces to be sealed to form coated mating surfaces; b. locating an exfoliated vermiculite gasket core layer between the coated mating surfaces; c. mating the coated surfaces and interposed gasket core layer together to form the gasket between the mating surfaces.

14. The method of claim 12, wherein the glass or glass-ceramic coating layer is applied in the form of a liquid suspension or paste-type formulation.

15. The method of claim 14, wherein the liquid suspension or paste-type formulation comprises an organic binder component, a glass or glass-ceramic powder, and a liquid carrier component.

16. The method of claim 15, wherein the liquid carrier component comprises a solvent and/or non-solvating liquid carrier.

17. The method of claim 16, wherein the solvent comprises an organic solvent selected from one or a combination of terpineols; ketones; alcohols; ether based alcohols unsaturated aliphatic alkyl monocarboxylates; lactates; and ether based ester.

18. The method of claim 17, wherein the terpineols comprise α-, β-, γ-, and/or 4-terpineol.

19. The method of claim 16 wherein the solvent comprises water.

20. The method of claim 16, wherein the non-solvating liquid carrier is water.

21. The method of claim 15, further comprising drying the liquid suspension or paste-type formulation, wherein after drying, the organic binder component comprises 1-60% w/w of the dried glass or glass-ceramic coating layer.

22. A solid oxide cell or a solid oxide cell component comprising a gasket of claim 1.

23. The method of claim 12, further comprising heating the gasket.

24. The method of claim 23, wherein the heating removes volatile components.

25. The method of claim 23, wherein the heating effects sintering of the coating layers.

26. The method of claim 23, wherein the heating effects wetting of the coating layers.

27. The method of claim 13, further comprising heating the gasket.

28. The method of claim 27, wherein the heating removes volatile components.

29. The method of claim 27, wherein the heating effects sintering of the coating layers.

30. The method of claim 27, wherein the heating effects wetting of the coating layers.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) FIG. 1 shows a schematic drawing of the testing apparatus according to example 1

(2) FIG. 2 shows pressure versus leak rates for embodiments of the invention

(3) FIG. 3 shows pressure versus leak rates for embodiments of the invention

(4) FIG. 4 shows pressure versus leak rates for embodiments of the invention

(5) FIG. 5 shows pressure versus leak rates for a comparative example

(6) FIG. 6 shows leak rate versus time for an embodiment of the invention

(7) FIG. 7 shows leak rate versus time for an embodiment of the invention

(8) FIG. 8 shows leak rate versus time for a comparative example

(9) FIG. 9 shows leak rate versus time for an embodiment of the invention

(10) FIG. 10 shows leak rate versus time for an embodiment to the invention

(11) FIG. 11 shows leak rate and voltage during thermal cycles

(12) FIG. 12 shows variable pressure difference and the effect on anode to cathode leak rates over time

DETAILED DESCRIPTION OF THE INVENTION

EXAMPLES

(13) In the following examples, embodiments of the invention described herein were prepared and tested as described below.

(14) Examples 1 to 6 are gaskets according to the present invention. All the materials of these examples were prepared for an average stack operating temperature of 700° C.

Example 1

(15) The coating carrier composition contained 80 wt % α-terpineol (from Merck), 15 wt % ethanol and 5 wt % ethyl cellulose (from Fisher Scientific) and glass powder (GM31107, available from Schott), with a glass to organic ratio of 2:1 w/w. The glass has a Tg of 532° C. and a softening temperature of 649° C. The exfoliated vermiculite core layer was (Thermiculite 866, available from Flexitallic). The Thermiculite was consolidated to a density of 1.9 g/cm.sup.3 before use in order to smooth the outer surfaces and therefore minimise the amount of leak channels formed between the core layer and the mating surfaces which normally arise due to the natural relative roughness of exfoliated vermiculite.

(16) The Ethyl cellulose was mixed with terpineol and ethanol at 35° C. with a magnetic stirrer for 24 h. After that the glass powder was added and the mixture was stirred for 1 h.

(17) The coating carrier composition was applied to the core layer by brush. Application by this method allowed for a thicker consistency and good coverage was easily achieved with a single layer.

(18) After application of the coating, the sheets were dried at 80° C. for 2 h and then cut to the required shape. Leak tests were conducted using ring-shaped seals having 40 mm outer diameter and 5 mm width. The gaskets were placed on top of a 20 mm Crofer 22 H steel available from ThyssenKrupp VDM GmbH mating plate and a 1 mm Crofer 22 H steel mating plate was placed on top of the gasket. Heat up procedure occurred as follows: 1. Heat up from room temperature to 700° C. at 60K/hr 2. Test run at 700° C. 3. Cool down to ambient temperature 1 K/min.

(19) The sample achieved sufficiently low viscosity and surface tension of the glass to achieve good wetting of adjacent surfaces and penetration to the vermiculite pores.

(20) To test the sample, gas was fed through the thick bottom plate. FIG. 1 presents the experimental setup for leak rate measurements. Samples were exposed to a 25 mbar overpressure using 50/50 mix of H2/N2 at 700° C. Periodical leak rate measurements were conducted by shutting off the valves (V1, V2) and measuring the pressure decay.

Example 2

(21) The coating composition contained 44 wt % α-terpineol, 53 wt % ethanol, 3 wt % ethyl cellulose and glass (GM31107, available from Schott), with a glass to organic ratio of 1:2 w/w. The exfoliated vermiculite core layer (Thermiculite 866) was prepared in the same manner as example 1. The coating carrier composition was also prepared using the method given for example 1 except the additional ethanol solvent was added and stirred into the mixture at the end.

(22) On this occasion, a wet spraying application was used to coat the core layer. The carrier had been thinned with ethanol to achieve suitable viscosity for the spray gun (U-POL Maximum HVLP mini with 1.0 mm nozzle). Several layers were sprayed from a distance of 10 to 20 cm. The viscosity of the resulting spraying suspension was 3.5 to 4.0×10.sup.−2Pa.Math.s.

(23) Heat-up and testing was conducted as for example 1.

(24) FIG. 2 presents the leak rates of examples 1, 2 and an uncoated Thermiculite 866 comparative example as a function of pressure at 0.1 MPa compressive stress.

(25) FIG. 3 presents the leak rates of example 2 and the uncoated Thermiculite as a function of pressure at 0.4 MPa compressive stress.

(26) The results of examples 1 and 2 show that the gaskets of the present invention provide substantially better leakage rates than a comparative Thermiculite only seal, especially at low compression stress levels. The gaskets according to the present invention show leaks rates of 0.1 to 0.03 ml(m min).sup.−1, which is a reduction of 60 to 90% compared to uncoated samples.

(27) Furthermore, the leak rate is shown to be almost independent of overpressure indicating that the primary leak mechanism is diffusion rather than advection. This was further tested by measuring leak rates with different gas compositions.

(28) FIG. 4 presents the leak rates of the coated gasket according to example 2 at different gas compositions and FIG. 5 also presents the leak rates of an uncoated Thermiculite gasket of the type used for example 2 with different gas compositions.

(29) Extrapolating the curves measured with air, one obtains more or less zero leak rate at zero pressure difference. However, with other gas compositions than air, there is clearly a diffusion component present. As such, leak rates can vary depending upon the gas combinations used.

(30) FIGS. 6, 7 and 8 present the leak rates of gaskets over time and show the effects of thermal cycling on leak rates. The figures are related to gaskets according to example 1, 2 and the comparative example respectively. The compressive stress used in these test runs was 0.1 MPa and the thermal cycling period was between 300 and 530 hours. FIGS. 6 and 7 show how the low leak rates of gaskets according to the present invention are maintained or even improved following a period of thermal cycling. In comparison, the uncoated Thermiculate gasket has a higher initial leak rate that worsens after thermal cycling.

Examples 3 and 4

(31) To further test the sealing properties of the coated seals with different temperatures and gas overpressures two coated seals were manufactured. The coating slurry formulation was manufactured by mixing the organic components a-terpineol, Elvacite 2045 and ethanol in a proportion of 80/11/9. Glass powder (Schott GM31107) was then added to the organic slurry with constant stirring using a magnetic stirrer. Doctor blade casting was used to apply the coating formulations to 0.7 mm thick consolidated Thermiculite 866 core layers (available from Flexitallic). The following samples were formed:

(32) Example 3. A coated seal with 5/10 w/w organic components to glass ratio

(33) Example 4. A coated seal with 5/13 w/w organic components to glass ratio

(34) After drying at ambient temperature for 72 h, the samples were cut into 40 mm OD, 30 mm ID sealing rings. The rings were assembled between two Crofer 22 APU plates available from ThyssenKrupp VDM GmbH and 0.870 mm thick spacers were inserted in the middle of the rings to correspond to the fuel cell in a stack. The test apparatus was assembled according to FIG. 1 and measurements were taken according to the methodology of example 1. Gas was fed inside the sealing ring through a hole in the middle of the bottom plate. A weight corresponding to 0.4 MPa of compressive stress was applied on top of the seals.

(35) The samples were heated up to 700° C. at a rate of 60 K/h with air at 2.5 mbar overpressure. After heat up, the gas mixture was changed to 50/50 H.sub.2/N.sub.2 and the overpressure was elevated to 25 mbar. FIGS. 9 and 10 show the leak rates of examples 3 and 4 respectively. It can be noticed that the leak levels remain at the very low level of ˜0.5 ml/m/min at 20 mbar overpressure. In addition, the leak rate is almost independent of the temperature or pressure and unaffected by the thermal cycle. Further leak rate measurements were taken for examples 3 and 4 after prolonged use. After 1300 hours the leak rates for the examples were approximately 0.49 ml/min/m and 0.32 ml/min/m respectively. As such, the leak rates at the end of the test were substantially the same as the leak rates at the start of the test, showing excellent long term leak rates.

Example 5

SEM Analysis

(36) A SEM analysis of a gasket according to the present invention was undertaken. The seal was prepared by placing a sample of the gasket according to example 1 and 2 between two 1 mm Crofer 22 H sheets. The sample underwent heat treatment, as described above, but with a 50 h dwell at 700° C. Thin glass layers around 2 to 10 μm are formed at the interfaces of the vermiculite and Crofer 22 H plates. The glass accommodated the surface roughness of the vermiculite and penetrated into its pores. This behaviour indicates self-healing of cracks that could develop in the vermiculite core or in the glass layer due to thermo-mechanical stresses.

Example 6

Stack Test

(37) To verify the suitability of the invention in a SOFC stack environment, a simple one cell stack was constructed. The stack consisted of anode and cathode endplates (20 mm Crofer 22 APU) into which gas channels were machined. A chromium barrier coating of MnCo.sub.1.8Fe.sub.0.2O.sub.4 was coated on the cathode endplate by a high velocity oxygen flame method (as described in Development and Application of HVOF Sprayed Spinel Protective Coating for SOFC Interconnects, O. Thomann, M. Pihlatie, M. Rautanen, O. Himanen, J. Lagerbom, M. Makinen, T. Varis, T. Suhonen, and J. Kiviaho, Journal of Thermal Spray Technology, 2013). The cell used in this test was Elcogen ASC-10B having an active area of 80 cm.sup.2. The stack had two seals: a seal between cell electrolyte and cathode end plate and a second seal between the end plates. The seals were formed according to the procedure of example 2. The compressive force on the stack was 120 kg corresponding to about 0.3 MPa on the gaskets.

(38) The stack was heated up according to the heat-up method given in example 1. After reaching 700° C. operating temperature, the anode was reduced using H.sub.2 in N.sub.2. Gas flows were then set to 2.011 NLPM air and 0.843 NLPM H.sub.2. With these nominal flows cathode inlet pressure was 10 mbar and anode inlet pressure 1 mbar. With 100% H.sub.2 at the anode the open circuit voltage was 1225 mV, indicating a water vapour content of less than 0.3% at the anode compartment. This means that the total oxygen leak from cathode and ambient to anode was around 1 ml.sub.N/min. Thermal cycles were conducted by reducing the temperature of the stack to 150° C. and then increasing it back to operating temperature at a rate of 120 K/h. After 1000 h dwell, the open circuit was measured again showing a value of 1230 mV indicating that the oxygen leak to anode had not increased.

(39) FIG. 11 presents the results of six thermal cycles. H.sub.2 leak to cathode remained steady after three thermal cycles and the OCV remained at a high level between thermal cycles, indicating a very low H.sub.2O concentration at the anode (<3%). This shows a low cathode to anode leak at ambient temperature.

(40) FIG. 12 presents the anode to cathode H.sub.2 leak versus pressure results. The anode pressure was increased and this can be seen by the increased pressure plots for the anode inlet and outlet. The cathode pressure was not increased. The cathode inlet pressure is shown as generally constant. The pressure at the cathode outlet was measured at zero throughout testing (not shown in FIG. 12) due to the presence of a large diameter outlet pipe. The rate of hydrogen leak during the test remained substantially constant, which shows that leak rates are independent of pressure difference between anode and cathode.

(41) Attention is directed to all papers and documents which are filed concurrently with or previous to this specification in connection with this application and which are open to public inspection with this specification, and the contents of all such papers and documents are incorporated herein by reference.

(42) All of the features disclosed in this specification (including any accompanying claims, abstract and drawings), and/or all of the steps of any method or process so disclosed, may be combined in any combination, except combinations where at least some of such features and/or steps are mutually exclusive.

(43) Each feature disclosed in this specification (including any accompanying claims, abstract and drawings) may be replaced by alternative features serving the same, equivalent or similar purpose, unless expressly stated otherwise. Thus, unless expressly stated otherwise, each feature disclosed is one example only of a generic series of equivalent or similar features.

(44) The invention is not restricted to the details of the foregoing embodiment(s). The invention extends to any novel one, or any novel combination, of the features disclosed in this specification (including any accompanying claims, abstract and drawings), or to any novel one, or any novel combination, of the steps of any method or process so disclosed.