DE10009530A1 - Process for the isothermal and isobaric chemical gas infiltration or refractory materials into a porous structure comprises sealing a part of the surface of a porous structure - Google Patents

Abstract

Isothermal and isobaric chemical gas infiltration or refractory materials, especially carbon, into a porous structure comprises sealing at least one part of the surface of a porous structure subjected to educt gas, and inserting the porous structure into the reactor so the diffusion path for the gas is lengthened in the structure. Preferred Features: Sealing is carried out using a graphite foil which is adhered to the relevant part of the surface of the porous structure using an organic adhesive.

Description

The invention relates to a method for chemical gas phase infiltration of a por structure with refractory materials, especially carbon, as well as a preferred te use of the method.

Chemical gas phase infiltration is the most important process for infiltrating refractory materials into a porous structure and thus also the most important process for the production of fiber-reinforced composites with a matrix of silicon carbide or carbon. The problems of chemical gas phase infiltration are generally known and are described in detail in the scientific and patent literature [WV Kotlenski in Chemistry and Physics of Carbon (Edited by PL Walker and PA Thrower), p. 173, Marcel Dekker, New York, 1973; DE 41 42 261 A1 ( 1993 )]. The main problem with isobaric, isothermal chemical gas phase infiltration is that the reactant gas must diffuse into the porous structure before it is decomposed. If this does not succeed, there is an increased deposition in the surface areas of the porous structure. The result is deposition gradients that drop from the surface into the interior of the structure and ultimately lead to the closure of the surface, so that adequate infiltration of the interior is not achieved.

The patent [DE 196 46 094 C2 ( 1998 )] therefore proposed conditions for isothermal, isobaric chemical gas phase infiltration with which this problem should be largely manageable. Essential features are flow control with the help of internals, so that the gas flows through the reactor at high speed in narrow gaps. Due to the proposed components, infiltration takes place "on all sides", but in any case from opposite sides.

The invention has for its object a method for chemical gas phase to create seninfiltration, in comparison to the state of knowledge and the Technology both a much faster infiltration and to a special degree the porous structure is better filled.

According to the invention, the object is achieved with a method with the features according to Claim 1 solved. This procedure is based on a paradox, which means that the faster infiltration and especially the better filling of the porous structure be achieved when the diffusion path for the educt gas by shielding Sealing, coating or sealing selected surface areas of the porö This structure is lengthened and possibly maximized.

The easiest way to illustrate the procedure is the example of an infinitely extended porous plate with a thickness of 2 L. Usually you will infiltrate such a plate from both sides; in this case the diffusion path would be L. According to the method according to the invention, the diffusion path is 2 L. For this purpose, one side is shielded or sealed in such a way that the gas cannot diffuse into the porous plate from this side, or only to a negligible extent compared to the opposite side . The latter restriction only applies in the event that a complete sealing of a page is technically extremely difficult or complex.

In principle, it is also possible that despite shielding or sealing one side of the infinitely extended plate, the diffusion path 2 L is not sufficiently long. In this case, in order to extend the diffusion path, a further plate of the porous structure or a plate made of an auxiliary can be attached or fixed in front of the uncoated or sealed side of the plate. The latter has a high open porosity of preferably over 80% and consists of a very inexpensive material such as a felt or fleece made of carbon, stone or glass or similar fibers.

The essence and the advantages of the method according to the invention will follow the example of carbon infiltration with reference to the Figures shown. These show:

Fig. 1. a plate with the dimensions W × D × H to explain the sealing treatment of the surface.

Fig. 2. is a schematic sketch of the reactors used with inner diameters of 22 mm and 40 mm.

Fig. 3. Deposition rates of carbon in 17 mm long, one above the other, one-sided closed capillaries depending on the depth / length of the capillaries, obtained under the following reaction conditions: temperature 1100 ° C, methane pressure 20 kPa, residence time of the gas 0.08 s. (.) Result for capillaries open on both sides.

Fig. 4. Deposition rates of carbon in 17 mm long, one on top of the other, closed capillaries depending on the depth / length of the capillaries, obtained under the following reaction conditions: temperature 1125 ° C, methane pressure 20 kPa, residence time of the gas 0.08 s. (.) Result for capillaries open on both sides.

Fig. 5. Deposition rates of carbon in 17 mm and 34 mm long, one above the other, one-sided closed capillaries depending on the depth / length of the capillaries, obtained under the following reaction conditions: temperature 1100 ° C, methane pressure 20 kPa, residence time of Gases 0.2 s. (▱) Result for capillaries open on both sides.

Fig. 6a. Deposition rates of carbon in 17 mm and 34 mm long, one above the other, one-sided closed capillaries depending on the depth / length of the capillaries, obtained under the following reaction conditions: temperature 1100 ° C, methane pressure 20 kPa, residence time of the gas 0.08 s. (▱) Result for capillaries open on both sides.

Fig. 6b. Ratio of the deposition rates at the end and at the entrance of the capillaries closed at one side of 17 (x) and 34 (+) mm depth depending on the height in the reactor.

Fig. 7. Schematic representation of a "real" porous structure with bottleneck pores (a), and the infiltration result with homogeneous infiltration, ie without a gradient of the deposition rate (b), and with an increasing gradient of the deposition rate (c).

In Fig. 1 a plate of thickness D, the height H and the width B is shown, in which the principle of encasing or sealing a porous structure will be explained. The plate has the areas D × H, D × W and W × H. Assuming that H »D and W» D, this plate will be infiltrated from the areas W × H on both sides according to the prior art. According to the method according to the invention, the infiltration preferably takes place on one side of only one surface W × H under the assumed proportions of the dimensions, in which all other surfaces are sealed.

The condition H »D and B» D implies a very small thickness D, for example of less than 20 or 10 mm, the infiltration takes place in such a way that two plates used side by side so that they are facing the opposite inner Touch surfaces W × H and one of the outer surfaces W × H is sealed. A preferred solution under the assumed conditions of a very small thickness  D and H »D and B» D consist in the use of a second porous plate with the same height H and the same width B from an inexpensive auxiliary material, such as. B. from a felt or with a very high porosity fleece, which in an analogous way like described above is used with the porous plate to be infiltrated. If, on the other hand, D, H <D and B »D apply to a very small thickness, the infiltration takes place preferably one side of only one surface D × B, with all other surfaces verse be considered.

The procedure described above applies to more complex geometries analogous.

From the results of the examples below it follows that for the Infil tration depths from about 20 mm are attractive and the attractiveness with increasing infiltra depth increases. The upper limit of preferred infiltration depths is between 50 and 100 mm. Accordingly, there are clear boundary conditions for the procedure wise when sealing or when inserting a porous structure into the re actuator before. From practical experience, the case is highly unlikely to be everyone three dimensions of a porous structure to be infiltrated larger than 50 to 100 mm are. The method can therefore be used universally.

Process examples example 1

A schematic diagram of the reactor used for the infiltration with a usable diameter of 22 mm is shown in FIG. 2. It consists of an externally heated ceramic tube (not shown); In this there are ceramic internals, so that a cylindrical separation chamber with a conical inflow and outflow part is created.

Inside is a sample holder or holder, which is designed for flow reasons so that an annular gap of approximately 2 mm width is created in both the conical and the cylindrical part, through which the gas flows. The structure to be infiltrated consists of a plate made of impermeable pyrolytic carbon. It has the dimensions 18 mm wide, 20 mm high and 3 mm thick. In these platelets there are capillaries 1 to 7 , one above the other, with a diameter of 1 mm. The middle capillary 4 is open on both sides and has a length of 18 mm; the remaining capillaries are closed on one side and have a length of 17 mm. After infiltration has ended, the layer thickness of the deposited carbon is determined as a function of the depth of the capillary using an X-ray microscope. The average deposition rate r in µm / h is calculated from the layer thickness in µm and the deposition time in hours. In Fig. 3 deposition rates are shown in the capil 1 , 3 , 5 and 7 closed capil depending on the depth of the capillaries z. They were obtained at a temperature of 1100 ° C, a methane pressure of 20 kPa and a residence time of the gas in relation to the cylindrical deposition space (annular gap) of 0.08 s after an infiltration period of 152 h. In all capillaries closed on one side, the deposition rate increases significantly with increasing depth of the capillaries. It can also be seen that the increase in the deposition rate is still very pronounced even at the closed end of the capillaries; this means that at a greater depth, even higher deposition speeds can be expected. Such increasing deposition rates imply that the capillary, as a model of a porous structure, is infiltrated from "inside out".

The outstanding advantage of the method according to the invention results from a comparison of the deposition gradients in the capillaries closed on one side with that in the capillaries 4 open on both sides (closed symbols in FIG. 3), in which the diffusion path is only about half as large. Their gradient is vanishingly small, ie infiltration takes place much more slowly and, above all, only to a very limited extent "from the inside out".

Example 2

To accelerate the infiltration, infiltration was carried out under the same conditions as in Example 1, but at a temperature of 1125 ° C. Experience has shown that a temperature increase of 25 ° C increases the deposition rate by a factor of 2.5. The results reproduced in FIG. 4 confirm this statement. It also shows that the rate of deposition increases even more than at the lower temperature of 1100 ° C. In contrast, the deposition rate in the bilaterally open capillary 4 (closed symbols), ie with only half the diffusion away, shows almost no gradient.

Similar or analogous results were also obtained with a Ver received for a while.

Example 3

Based on the results of Example 1 were in a Fig. 2 speaking reactor, but with a useful diameter of 40 mm and a height of the cylindrical deposition space of 50 mm, using platelets made of pyrolytic carbon with the dimensions 35 × 50 × 3 mm a capillaries of 34 and 17 mm depth sealed on one side infiltrated simultaneously. Here, viewed from bottom to top, capillaries 1 , 3 , 5 and 7 had a depth of 34 mm, capillaries 2 and 6 had a depth of 17 mm (as in Examples 1 and 2), while capillary 4 was open on both sides was. The infiltration was carried out at a temperature of 1100 ° C., a methane pressure of 20 kPa and a residence time of the gas with respect to the cylindrical annular gap with a width of 2 mm and a height of 50 mm of 0.2 s.

The results shown in FIG. 5 directly show the comparison of the deposition rates and their gradients at capillary depths that differ from one another by a factor of two. At the entrance to all capillaries, the deposition rate increases continuously from capillary 1 to capillary 7 ; this means that there is no influence of the depth of the capillaries or their one-sided closure. The increase in the rate of deposition at the entrance to the capillaries results from the methane decomposing as it flows through the ring gap. If we now consider the gradients of the deposition speed at a depth of 17 mm, i.e. at the end of the capillaries with a depth of 17 mm and in the middle of the capillaries with a depth of 34 mm, values result for the capillaries 1 , 3 , 5 , 7 of 34 mm depth which are many times higher than those in capillaries 2 and 6 of 17 mm depth. The same conclusion results from a comparison of the gradients with respect to the total depth of the capillaries of 17 and 34 mm. These results prove beyond any doubt that the longer the capillary, ie the diffusion path, the greater the deposition speed and the gradient.

Also noteworthy is the profile of the deposition rate in the capillaries 4, which are open on both sides and are 35 mm long (▱). The gradient is smallest here; the fact that it is not completely symmetrical results from the fact that the sample holder with sample may not have been arranged completely centrally in the deposition space. In any case, the comparison of the capillaries closed on one side with the capillaries open on both sides is impressive proof that the one-sided closure leads to drastically increased deposition rates and gradients of the deposition rates in the sense of the method according to the invention.

The almost horizontal course of the deposition rates at the end of the 34 mm long closed capillaries suggests that with further extension The depth of the capillaries diffusion control starts with the consequence that the Deposition speeds do not continue to increase, but above a certain level Begin to decrease in depth. For this reason, were under the same conditions conditions, but with reduced dwell time, further infiltrations were carried out.

Example 4. Fig. 6 (a) shows deposition rates analogous to example 3, but with a residence time reduced from 0.2 s to 0.08 s. The differences in the deposition gradients between the capillaries of 17 mm in length and those of 34 mm in length are even more dramatic than in the previous example 3 (same symbols as there). To illustrate this, the ratios of the deposition velocities at the end and at the entrance of the capillaries as a function of the height are shown in FIG. 6 (b). In the case of capillaries 1 , 3 , 5 , 7 of 34 mm depth, values of more than 12 result, ie the deposition rate at the end of the capillaries is more than 12 times as high as at the entrance. For capillaries 2 and 6 of 17 mm depth, however, this ratio is only about 2.

Also noteworthy is the still strong increase in the deposition rate at the end of the 34 mm deep capillaries. In contrast to the vori no example of any influence of diffusion limitation. That means,  that the deposition rates increase with even longer diffusion paths next continue to climb.

Example 5. Capillaries are idealized models for porous structures and are therefore ideal for documenting infiltration processes. The following shows how increasing gradients of the deposition rate affect the filling of a real porous structure. Typical of such structures are pores which are connected to one another by so-called bottle necks ( FIG. 7 (a)). In Fig. 7 (b) the result is shown which results from a homogeneous infiltration of such a porous structure. Homogeneous is understood to mean that the deposition speed has no gradient, which is the ideal case in view of the state of knowledge and technology. In Fig. 7 (c) a gradient of 5 is assumed as an example for the infiltration. In this case, compared to homogeneous infiltration, there is a remarkable reduction in residual porosity by about 25%. In example 4, gradients of over 12 were documented ( FIG. 5 (b)), so that if the infiltration is carried out optimally, as shown in the examples, a further dramatic reduction in the residual porosity is possible. It succeeds because the pores are filled "from the inside out" due to the strong gradients of the deposition rate. This allows pores to be filled before the bottle neck is closed.

Infiltration "from the inside out", as documented using the example of FIG. 7 (c), also means that progressive infiltration leads to a decrease in the depth of infiltration.

If the method is carried out optimally, a po red plate made of an inexpensive auxiliary material in front of the side to be infiltrated from which the infiltration takes place.

Claims (25)

1. A method for an isothermal and isobaric chemical gas phase infiltration based on diffusion in the porous structure of refractory substances, in particular carbon, characterized in that at least part of the surface of the porous structure flowed through by the educt gas seals or the porous structure so in the reactor is used that the diffusion path for the educt gas in the porous structure is extended. 2. The method according to claim 1, characterized in that the sealing with Foil is made with the relevant part of the surface of the porous structure With the help of an adhesive. 3. The method according to claims 1 or 2, characterized in that Gra phit film is used. 4. The method according to claims 1 or 2, characterized in that a Foil made of an organic material is used, which when not heated oxidizing atmosphere forms a carbon film. 5. The method according to claims 1 or 2, characterized in that a Foil made of metal or another inorganic material is used. 6. The method according to any one of claims 1 to 5, characterized in that a organic glue is used, which when heated in non-oxidizing atmosphere sphere forms a coke or carbon residue. 7. The method according to any one of claims 1 to 5, characterized in that a inorganic glue is used. 8. The method according to claim 1, characterized in that the sealing with a slurry (slurry), which is an inorganic or organic  Contains binder that egg when heated in a non-oxidizing atmosphere forms a coke or carbon residue. 9. The method according to any one of claims 1 to 8, characterized in that before the unsealed part of the surface additionally a porous structure made of egg an inexpensive auxiliary with a high open porosity is attached. 10. The method according to any one of claims 1 to 9, characterized in that as inexpensive auxiliary material of high open porosity a felt or fleece with an open Porosity of more than 60%, preferably more than 80% is used. 11. The method according to any one of claims 1 to 10, characterized in that as Educt gas methane is used. 12. The method according to any one of claims 1 to 10, characterized in that as Educt gas Natural gas is used. 13. The method according to any one of claims 1 to 10, characterized in that as Educt gas a mixture of methane or natural gas with hydrogen is used. 14. The method according to any one of claims 1 to 13, characterized in that the temperature is 1000 ° C to 1200 ° C and the pressure is 1 to 40 kPa. 15. The method according to any one of claims 1 to 10, characterized in that a C ₂ hydrocarbon such as ethane, ethene or ethyne is used as the starting gas. 16. The method according to any one of claims 1 to 10, characterized in that a mixture of a C ₂ hydrocarbon and hydrogen is used as the starting gas. 17. The method according to any one of claims 1 to 10, 15 or 16, characterized in that the temperature is 800 ° C to 1000 ° C and the pressure 1 to 40 kPa be. 18. The method according to any one of claims 1 to 10, characterized in that a C3 or C ₄ hydrocarbon such as propane, propene, butane and butadiene is used. 19. The method according to any one of claims 1 to 10, characterized in that a mixture of a C ₃ - or C ₄ hydrocarbon and hydrogen are used as the starting gas. 20. The method according to any one of claims 1 to 10, 18 or 19, characterized in that the temperature is 700 ° C to 1000 ° C and the pressure is 1 to 40 kPa. 21. The method according to any one of claims 1 to 10, characterized in that as Educt gas a mixture of methyltrichlorosilane and hydrogen is used. 22. The method according to any one of claims 1 to 10, characterized in that as Educt gas a mixture of tetrachlorosilane, methane and hydrogen or of Si Lan, ethene and hydrogen is used. 23. The method according to any one of claims 1 to 10, 21 or 22, characterized in that the temperature is 600 ° C to 1100 ° C and the pressure is 1 to 40 kPa. 24. The method according to any one of claims 1 to 20, characterized in that carbon fiber reinforced carbon is produced. 25. The method according to any one of claims 1 to 10, 21 or 22, characterized in net that carbon fiber or silicon carbide fiber reinforced silicon carbide be put.