|Title of Invention||
"CHEMICAL MICROREACTORS AND METHOD FOR THEIR MANUFACTURE"
|Abstract||Chemical micro reactors for chemical synthesis and their methods of manufacture are known, but have disadvantages such for example as extremely high manufacturing costs or poor flexibility for adaptation to various cases of application. By means of the microreactors and manufacturing methods according to the invention these disadvantages are avoided. The microreactors are characterized in that the reactors contain fluid ducts in at least one plane and feed and return lines for fluids, the fluid ducts in at least one plane and feed and return lines for fluids, the fluid ducts being defined by side walls of metal lying one opposite the other and further side walls of metal or plastic extending between said first side walls, and in which the planes are connected together and/or with closure segments closing open fluid ducts by means of appropriate solder or adhesive layers. The manufacturing methods are characterized by process sequences in which the individual reactor planes produced by means of electrolytic methods, are connected together by soldering or gluing.|
|Full Text||The present invention relates to method for manufacturing chemical microreactors and an apparatus thereof.
The invention relates to chemical rnicroreactors which can be used in the -chemical industry for synthesising processes, to methods for their manufacture end to a preferred use of the microreactors.
There have been reports for a number of years in the literature relating to; chemical reactors which' have advantages in comparison to previous ,._ • production systoms for manufacturing chemical compounds. In converting chemical methods to a large industrial production scale there is the basic problem that tha dimensions of the production systems are larger by several orders of magnitude than the apparatus used on a laboratory scale for developing the processes. If for example a chemical synthesis is considered, then the relevant scale of siza of the chemical spscies reacting with cna another is determined by their molecular size, which generally is in . the range of beneath one nanometer up to a few nanometers. For diffusion and heat transfer phenomena lengths of a few millimetres down to the micrometre range are relevant. Due to the production volumes required in large-scale industry, chemical reactors usually have dimensions which lie in' the range between 'a few centimetres up to several metres. Therefore at least' for homogeneous chemica! reactions the experience gained on a laboratory seals with reaction volumes of a few litres up to about 100 litres relating tq the process management, cannot be directly adopted on an industrial scale, Even at the stage of mixing liquids, a stirring mechanism is primarily necessary in order to increase the transport of materials in such a way thet the distances between areas of differing concentration are reduced. The so-called scale-up problem also arises from the various dimensions of the reactor. A chemical reaction which has beon optimised' on the laboratory scale cannot thereafter immediately be transferred to the -
production system, but must be firstly transferred to a pilot system of dimensions between the laboratory and production scales (technical scale), before it is finally used in industrial production. A problem is that each stage of this process development requires its own optimisation cycle, each of these cycles being additively involved in the development time required for introduction of the process. In heterogeneous catalysis on the other hand, the catalyst particles are often applied to porous carriers, whose pore size lies in the range of order of magnitude (millimetre to micrometre range) relevant for the transport of materials.
When the process guidance is not at its optimum and based purely on experience from the laboratory scale, for example the yield of the chemical synthesis can be too small, as excessively large proportions of undesired secondary products are formed due to secondary reactions which are taking place.
In order to solve the above problems in transferring a process from the laboratory to the production scale, the concept of so-called microreactors was developed a few years ago. This involves a parallel arrangement of a plurality of reaction cells, whose dimensions extend from a few micrometres up to a few millimetres. These reaction cells are so formed that physical, chemical or electrochemical reactions can take place in them. In contrast to a previous porous system (heterogeneous catalysis), the dimensions of the cells in a microreactor are defined, i.e. produced according to plan in accordance with a technical process. The disposition of the individual reaction cells in the ensemble of the reactor is likewise ordered, particularly periodic in one or two dimensions. Counted among the reactors in the extended sense are also the necessary feed (inlet) and return (outlet) structures for the fluids (liquids and gases), and sensors and actors, for example valves, cooling and heating members, which influence or monitor the flow of material and heat in the individual cells.
One individual reactor cell has a lateral extension which lies in an order of magnitude favourable for optimum transport of material and heat. As the volume flow through one individual reactor cell is extremely small, the entire reactor is enlarged (scale-out) by parallel multiplication of these elementary cells to the industrially necessary scale. Due to the small dimensions, local differences in concentration and temperature in the fluid flows are reduced to a minimum. Thus the processes may be much more accurately adjusted to the optimum reaction conditions, so that the conversion rates in a chemical reaction can be increased for an identical duration time of the reaction medium in the reactor. In addition, the purity and yield of the synthesised materials can be optimised by setting almost the most favourable reaction conditions. In this way such chemical reactions can be realised, which were not manageable in the previous way, for generating for example intermediate products being trapped in a controlled manner.
There are a series of proposals for manufacturing the chemical microreactors.
On the one hand a microreactor can be produced for example by stacking a plurality of copper foils, in which grooves are machined by means of a diamond tool in order to form flow ducts. Such a microreactor is described by D. Honicke and G. Wiesmeier in the article "Heterogeneous Catalysed Reactions in a Microreactor" in DECHEMA Monographs, Volume 132, Papers of the Workshop on Microsystem Technology, Mainz, 20 to 21, February 1995, pages 93 to 107, which is used for partial oxidation of propene to form acrolein. The individual reactor layers are connected together by diffusion bonding and subsequent electron beam welding. It was necessary in order to carry out the chemical reaction to convert copper of the resultant ducts by partial oxidation into copper-(l)-oxide.
For precise and reproducible manufacture of the fine structures, a micro-positioning table suitable for such purposes is required. The individual reaction cells are basically produced serially and thus in a time- and-cost-intensive way.
By means of the LIGA process (Lithographie Galvano-Formung, Abformung), a plastic layer, usually polymethylmethacrylate (PMMA) is exposed by means of synchrotron radiation and is then subsequently developed. The structure produced in this way is electrolytically filled with a metal. The metal structure can then be again duplicated in further process steps by means of a plastic replication. Such a method is described by W. Ehrfeld and H. Lehr in Radiat. Phys. Chem., Volume 45 (1995), Pages 349 to 365, and W. Menz in Spektrum der Wissenschaft, February 1994, pages 92 to 99 and W. Menz in
Automatisierungstechnische Praxis, Volume 37, (1995), pages 12 to 22. According to the details in the article in Spektrum der Wissenschaft loc. cit., individual components or sub-systems, which are produced separately, are connected together by suitable jointing techniques.
A technique related to the LIGA process, which operates without the extremely expensive synchrotron radiation, is the so-called laser-LIGA method. In this case the plastic layer of PMMA is structured by a powerful UV laser and then electrolytically duplicated as in the LIGA process (W. Ehrfeld et al., "Potentials and Realization of Microreactors" in DECHEMA Monographs Volume 132, pages 1 to 29).
W. Menz in Automatisierungstechnische Praxis, loc. cit. also proposes a modified method according to which a microelectronic circuit has been formed on a silicon substrate in a previously known way, firstly a protective layer, thereupon an entire-surface metallising layer and thereon a plastic moulding compound are applied. Then by means of a metal matrix which
has been produced according to the LIGA process, the image of the fluid duct structures is impressed into the moulding compound. Thereafter the residual layers of the moulding compound covering the metal layer in the recesses formed are removed by plasma etching, and metal is deposited electrochemically in the recesses. The plastic structures are then removed and the exposed metal areas of the basic metallisation are removed by etching.
Both the previous LIGA process and also the laser-LIGA process are extremely expensive, as they require extremely expensive devices for structuring the plastic layer (synchrotron radiation source).
There is also known from the above named article by W. Ehrfeld et al., "Potentials and Realisation of Microreactors", a method of manufacturing chemical microreactors in which a photosensitive glass, for example FOTURAN® (Schott Glaswerke, Mainz) is used. For this purpose an image of the structure to be produced is transferred by UV light on to the glass member. By means of a subsequent heat treatment only the exposed areas in the glass crystallize. These can thereafter be preferably etched away in a hydrofluoric acid solution. This method has the advantage that the reaction ducts can be rapidly reproduced due to the parallel light exposure and the etching process. However, only certain glasses can be used, so that this manufacturing method is on the one hand expensive and on the other hand in particular is restricted to only a few cases of application.
Also the methods, which have been developed in the semiconductor industry for structuring silicon surfaces, have been taken over for manufacturing microreactors; for example a method has been described by J.J. Lerou et al. in the article "Microfabricated Minichemical Systems: Technical Feasibility", DECHEMA Monograph, Volume 102, pages 51 to 69, in which three etched silicon wafers and two end wafers at the outer
sides are connected together. Further, a heat exchanger filled with polycrystalline silver particles, which was likewise designed as a microreactor, was used. This method also may only be used to a restricted degree, as only silicon can be used.
A method of manufacturing panel heat exchangers is described in EP 0 21 2 878 A1. According to this, the duct structures required for the heat exchanger are formed by means of a mask (screen printing, photo printing) on plates of steel, stainless steel, brass, copper, bronze or aluminium, and the ducts themselves are produced in the surface areas not covered by the mask by a chemical etching process. Afterwards in a diffusion bonding process, a plurality of these plates are connected together. Such a heat exchanger, formed from plates welded together by diffusion bonding, is also disclosed in EP 0 292 245 A1.
The previously known methods for manufacturing microreactors therefore have many disadvantages, among which is the fact that only metal surfaces structured by means of a time-intensive and/or cost-intensive method can be produced in the reactor, or exclusively glass or silicon can be used, which are not well suited for specific applications.
The reactors according to EP 0 212 245 A1 and EP 0 292 245 A1 have the further disadvantage that with the configuration shown, only heat exchangers can be manufactured, so that many possible applications for chemical microreactors cannot be considered. In particular, complex reactors cannot be produced by this method, which have in addition to this ducts also electronic semiconductor circuits, fibre optic wave guides and other elements such as actors and sensors.
Therefore the object of the present invention are a chemical microreactor and a manufacturing method for chemical microreactors which are suitable
for a plurality of different applications and which are equipped with different and possibly complex elements such as electronic switching circuits, optical wave guides, actors and sensors as well as catalytic, corrosion-protective and other functional layers in the ducts. Further, the manufacturing process is intended to be cost-effective and capable of rapid execution. In particular, it is also intended to be possible to produce such microreactors in large numbers.
The present invention relates to method for manufacturing chemical microreactors, having substrate which has fluid ducts as well as feed and discharge lines for fluids, without using plastics moulding methods, said method having the following method steps:
a. forming fluid duct structures on the metal surfaces situated on
the substrate by means of a photoresist layer or a screen-printing lacquer layer,
such that the metal surfaces are partially covered by the layer;
b. at least partial currentless and/or electro-chemical etching away
of metal from the bare surfaces of the substrate;
c. completely removing the photoresist or screen-printing lacquer
d. forming solder layers;
e. superimposing the substrates and a closure segment closing the
fluid ducts and interconnecting the substrates and the closure segment by
The present invention also relates to method for manufacturing chemical microreactors, having one substrate which has fluid ducts as well as feed and discharge lines for fluids, without using plastics moulding methods, said method having the following method steps:
a. forming fluid duct structures on metal surfaces situated on the
substrate by means of a photoresist layer or a screen-printing lacquer layer,
such that the metal surfaces are partially covered by the layer;
b. currentless and/or electrochemical deposition of a metal layer on
the bare surfaces of the substrate;
c. completely removing the photoresist or screen-printing lacquer
d. at least partial currentless and/or electro-chemical etching away
of the metal of the substrate, forming fluid ducts;
e. forming adhesive and/or solder layers;
f. superimposing the substrates and a closure segment closing the
fluid ducts and interconnecting the substrates and the closure segment by
gluing and/or soldering.
The present invention also relates to method for manufacturing chemical micro reactors, having one substrate with fluid ducts as well as feed and discharge lines for fluids, without using plastics moulding methods, said method having the following method steps:
a. forming fluid duct structures on metal surfaces situated on the
substrate or directly on the substrate by means of a photoresist layer or a
screen-printing lacquer layer, such that the substrate surfaces or the metal
surfaces on the substrate are partially covered by the layer;
b. depositing a metal layer on the bare surfaces of the substrate or
on the metal surfaces;
c. completely removing the photoresist or screen-printing lacquer
d. forming adhesive and/or solder layers;
e. superimposing the substrates and a closure segment closing the
fluid ducts and interconnecting the substrates and the closure segment by
gluing and/or soldering.
The present invention also relates to chemical microreactor having fluid ducts in at least one plane as well as feed and discharge lines for fluids, in which microreactor the fluid ducts are delimited by metal side walls facing one another and by additional metal or plastics side walls extending between said side walls, wherein the planes are connected to each other and/or to a closure segment, which closes fluid ducts lying open, by means of suitable solder layers.
The chemical microreactors are advantageously suitable tor manufacturing toxic, unstable or explosive chemical products, particularly ofcyanogen chloride, phosgene, ethylene oxide, selenium compounds, mercaptanes, methyl chloride, methyl iodide, dimethyl sulphate, vinyl chloride and phosphines.
By moans of using industrial electrolytic methods for manufacturing the individual reactor planes, extremely flexible adaptation is possible to the respective case of application by means of the selection of appropriate comb nations of materials for manufacturing the planes.
In adcition, the opportunity is afforded of integrating the connection of the structured reactor planes into an overall process, in order to be able to produce stacked reactors. A diffusion welding process, such as is used when copper foils are in use and which represents a heavy temperature stress on the reactor components, or an anodic bonding process in the use of silicon wafers, are not used. Rather, the individual reactor planes are connected together by soldering or gluing in this way individual planes of the mrcroreactor can be combined into stacks with only medium temperature stress on the substrates, so that temperature-sensitive substrates and temperature-sensitive reactor elements, being integrated ' into the planes before their combination, for example semiconductor circuits or swellnble gels for forming actors, can be used. The soldering temoerature can be reduced to small values by the selection of specific solders, or the strength of the stack can be set at high values by the selection of specific hard soldors. By selecting solders which melt at low temperatures or by means of gluing it is possible to prepare even temperature-sensitive substrate surfaces for use in chemical synthesis
before combination of the reactor planes.
The inner surface of the reactor according to the invention can still be chemically or structurally altered even after combination, and thus can be optimised in accordance with the requirements of the specific chemical process. Furthermore it is possible in addition to the metal layers, to integrate any plastic layers into the reactor, as composite materials or metals with plastics are available in almost limitless numbers. Thus the materials used can be adapted to the specific requirements of the respective case of application.
The ducts to be produced can be manufactured extremely uniformly. The formation of burrs as occurs in the mechanical grinding of copper foils, and tool wear, do not occur. The dimensions of the fluid ducts preferably lie in a range of 1 millimetre or less. For example, fluid ducts with an approximately rectangular cross-section can be produced even with a width of 100 µm and a height of 40 µm. In particularly preferred embodiments of the invention, the fluid ducts have structural heights of 300 µm and less. Where the cross-section of the ducts is not rectangular, the width dimensions are intended to relate to width dimensions of the ducts measured at half the height. Further ducts with an approximately semicircled concave cross section can be produced.
A further substantial advantage resides in the fact that all the reactor planes may be produced simultaneously. It is not necessary to pass through the individual process stages in sequence. As the individual duct planes or modules can be substantially produced simultaneously, the entire reactor can be produced with closer tolerances. In addition, a high degree of reproducibility of the basic structures is enabled.
The reactors produced are inexpensive, as no excessively complex devices
are necessary for the manufacturing process. The resist structures formed in the LIGA process have in fact an extreme edge steepness and a very high aspect ratio. While these properties are essential for the production of micromechanical components for which this method was originally developed, they are not necessary for the manufacture of chemical microreactors. By avoiding the expensive synchrotron radiation or the expensive UV laser devices and the expensive masks necessary for this, structures can be produced photolithographically or even by means of screen-printing, by means of which the requirements of the average dimensions in microreactors are satisfied.
Compared to the heat exchangers or the manufacturing process described in EP 0 21 2 878 A1 and EP 0 292 245 A1, the reactors and the manufacturing process according to the invention have the advantage that temperature-sensitive materials can be used, as the diffusion bonding process is not used. In particular, semiconductor circuits, fibre optic wave guides, actors and sensors as well as temperature-sensitive coatings can be integrated into the reactor before its combination. This leads to a substantial expansion of the field of application and simplifies the design and manufacturing strategy for the reactors.
For the reasons mentioned above, the method according to the invention may be used with extraordinary flexibility. The individual components can be manufactured in large numbers, cost-effectively and with a high degree of dimensional accuracy.
Chemical microreactors are understood to be devices with fluid ducts from at least one reactor layer, which in addition to the actual reaction zones if necessary also have auxiliary zones which serve for mixing, metering, heating, cooling or analysis of the initial materials, of the intermediate products or of the end products. Each zone is characterised by a structure
adapted to the respective requirements. Whilst heating and cooling zones are designed either as heat exchangers or as reactor compartments equipped with electrical resistive heating systems or electrical cooling elements, and analysis zones have adapted sensors, metering zones contain for example microvalves and mixing zones, for example ducts with appropriately shaped inserts for swirling the combined fluids. The structure of the microreactors according to the invention can also be designed for specific cases of application in such a way that only heat is transported from or to the fluidic medium, for example in that heat is exchanged between the medium to be heated or cooled and another heating or cooling medium. The fluid ducts in the individual reactor layers are generally closed by stacking on top of one another a plurality of layers, and by closing the last layer with a closure segment.
Various substrates can be used to manufacture the microreactors: on the one hand, metal films are suitable for this, for example steel, stainless steel, copper, nickel or aluminium films. Their thickness should lie in a range between 5 /vm to 1 mm. Foils with a thickness of less than 5 //m are less suitable, as ducts with sufficient width cannot be formed therefrom. If a pure metal foil is used as a substrate, then in the case of such low metal layer thicknesses there arises the further problem that these foils are extremely difficult to handle. Foils with a thickness of more than 1 mm on the other hand would lead to an excessively thick reactor stack.
In addition, plastics, ceramics or glass films coated on one or both sides with metal may also be used. For example, epoxy resin or polyimide laminates lined with copper films are suitable. An opportunity of producing the plastic films coated with metal also resides in the fact of metallising these by known chemical methods. For this purpose the film must firstly be surface-treated by means of chemical or physical methods, being roughened for example in etching solutions or by means of a plasma
discharge using appropriate gases. Thereupon the plastic films, after appropriate further pre-treatment, for example cleaning, conditioning and activation, are metallised with an electroless and/or electrochemical method. The strength of the plastic layer, particularly of epoxy resin, is frequently increased by embedding glass fibre or aramide fabrics. Another possibility resides in pressing plastics and metal foils together under pressure and temperature effects (lamination).
Other chemically resistant materials are among others polytetrafluor-ethylene or other halogenised polyalkanes. Such chemically resistant materials can for example be activated by pjasma-enhanced chemical gas phase (vapor) deposition (PECVD). Securely-adhering nickel-phosphorus or copper layers can for example be formed on such activated surfaces by electroless metal deposition. The securely adhering coating of glass or ceramic materials is also possible directly according to known methods, for example by alkaline etching before activation and electroless metallization. By means of coating the chemically resistant plastics with metals, these materials can more simply be connected together in a securely adhering manner. Such a composite of polytetrafluorethylene films is not directly possible with metallized laminates.
Various methods can be used in order to form the fluid ducts. In one procedure the outset consists of a substrate coated over the entire surface with metal, for example copper. The methods suitable for this purpose for forming the ducts have been previously shown diagramatically. According to another variant method, the fluid ducts may also be generated by additive build-up of the metal layers exclusively in the areas on the substrates which do not correspond to the duct structures. The methods according to the invention are likewise available for this purpose.
In order to obtain sufficiently deep fluid duct structures, the thickness of
the metal layer to be etched off or deposited must be sufficiently thick. As there are frequently problems in uniformity of production of thick metal layers, particularly on large-area substrates, small substrate blanks, upon which the ducts are formed, are preferably used.
In order to form the fluid ducts by the etching method, the resist layer (screen-printed or photoresist layer) is applied to the substrate surfaces in such a way that the surface areas forming the fluid ducts are not covered by the resist layer.
For additive production of the fluid ducts, it is also possible to start with films not coated with metals. In this case firstly a screen printed layer or a photoresist layer is applied to the film surfaces in such a way that the surface areas corresponding to the fluid ducts are covered by the resist. The same also applies in the case of the reversal method. In order to enable electroless metal deposition when the additive technique is used, the film surfaces must first be pre-treated in an appropriate way. For this purpose the same methods are used as for the whole-surface metallization of the films. Thereupon the metal structures can be deposited in the exposed areas of the photoresist on the film surfaces. For example, the typical methods from printed circuit technology can be used. In this respect express reference is made to the details relevant to this matter in the "Handbuch der Leiterplattentechnik" Volume III, ed. G. Herrmann, pages 61 to 11 9, 1 993, Eugen G. Leuze Verlag, Saulgau, DE. The details on process technology contained therein are hereby incorporated. After metal deposition, the photoresist layer is totally removed.
Liquids and gases are processed as fluids in the finished microreactors.
According to the method according to the invention, firstly fluid duct structures are formed on the substrates, a screen-printing process or in
particular a photolithographic process being used. For this purpose a photoresist is applied to one or both surfaces of the film. In the diagram in Figure 1 there is shown by way of example a method for forming these structures (reversal method); the photoresist 2 can either be laminated as a film onto the substrate 1, or be applied as a liquid by spin-coating or curtain technique or by electrodeposition (process step A in Figure 1). Thereafter the photoresist layer is exposed with the image of the fluid duct structure to be produced, and the structure is then worked out in a development process (process step 2).
In addition to the fluid duct structures, other functions may also be provided on the substrates. On the one hand, so-called actors and sensors can be integrated into the microreactor. The actors involve switch members, particularly valves, controllable externally or automatically by measurement signals, but also electrical resistive heating systems or cooling elements operating according to the Peltier effect. Valves can for example be formed by swellable gel plugs. Microreactors in which actors and sensors are provided may be optimised to localised requirements with appropriate connection of actors and sensors in terms of the regulation technology. At the same time, sensor outputs may also be used for external monitoring of the reactor condition (such for example as ageing, toxification of catalyst and similar parameters).
If necessary, electrical connection lines for controlling or detecting measurement signals on the substrates are to be provided for the actors and sensors. Appropriate structuring elements must be taken into account during the photoprocess for these elements.
If substrates coated with metal are used, other elements may also be integrated in the interior of microreactors. For example, microchips for controlling actors and sensors can be integrated, a recess being provided in
a plastic laminate, into which the microchip is inserted. The electrical connections to appropriate control and signal lines can be produced by bonding or other known connection techniques, such for example as by soldering or gluing with electrical conductive adhesive.
Furthermore, in formation of the structures, at the same time also peripheral reactor components, such as feed lines, mixing zones, heating or cooling circuits can be formed additional to the reactor cells, so that the manufacturing costs are reduced. Therefore these elements are to be provided during the photostructuring stage. Moreover, the problems of sealing which normally occur are minimised.
The individual planes of the reactors can preferably be produced in multiple blanks. For this purpose the individual plane images are formed on a larger panel or film as fields lying next to one another. These elements are separated from one another after completion and can then be combined in a stack. Figure 2 shows the arrangement of a plurality of planes into a stack with identical fluid duct structures disposed crosswise to one another, the three lower planes being already connected together.
Thereupon the metal areas now exposed, which are not covered by the resist, can be further processed. For this purpose electroless or electrochemical methods are available. Preferably, either the metal of the metal foil is at least partially removed, in order to form the ducts, or further metal layers are built up on the exposed areas of the foil by electroless or electrochemical methods or by a combination of these methods (Figure 1). In the first-named process the fluid duct structures are formed in such a way that during development of the photoresist the areas corresponding to the ducts are produced. During the reversal process these areas on the other hand are kept covered by the photoresist, while the other areas not corresponding to the duct structures, are exposed.
According to a method variant of the reversal method, a metal resist layer 3, which is different from the basic metal layer, is applied on the exposed areas of the metal surface (process step C in Figure 1). Preferably, a tin, lead, lead/tin alloy, bismuth, tin/bismuth alloy, nickel or cobalt layer or an alloy of nickel and cobalt or an alloy of these elements with other elements such as boron or phosphorus are applied. By means of using these metals, the metal layers lying underneath the photoresist layers can be etched after removal of the photoresist layer, without the metal resist layers being attacked. This method offers the advantage that the photoresist layer needs no outstanding chemical resistance to the etching solution.
After the metal layer has been at least partly etched off in the exposed areas according to the etching method, or after the metal resist layer has been applied according to the reversal method, the photoresist layer is removed (process step D in Figure 1). According to the reversal technique the substrates are then treated in an etching bath which attacks the substrate material, but not the metal resist mask (process step E). According to the etching method, the substrate is etched after formation of the fluid duct structures in the photoresist layer. In both cases structures are worked out of the substrate in this way. The fluid ducts are formed by means of removal of the metal of the metal foil or metal coating.
Then a plurality of reactor planes produced in this manner are stacked one above the other and connected together (process step F).
If a solderable metal is used as a metal resist, in this method simultaneously a layer is obtained which simplifies the subsequent combination of the structured layers into a stack. A possible combination is the use of copper as a foil material or as a metal coating and tin or a tin/lead alloy as a solderable metal resist. The electrolytic application of solder containing silver, which is also used for hard soldering, is likewise
When using the reversal method, another method procedure for forming such solderable layers consists in applying to the exposed surfaces of the metal foil or the metal coating not covered by the photoresist, a tin layer, upon which in turn bismuth is deposited. If surfaces coated in this way are brought together during stacking of the individual reactor planes and the stack is heated, there forms at the boundary surface between bismuth and tin a eutectic mixture melting at low temperatures. About 58% by weight of bismuth is contained in this mixture. The mixture has a melting temperature of less than 140°C. The planes can be soldered together at the melting point of the eutectic alloy. Then the connection is heat treated, bismuth diffusing further into the tin. Thus the composition of the alloy changes so that the melting point of the metal layer increases. A soldered connection produced in this way for this reason remains resistant to a temperature far above the original soldering temperature after heat treatment. Therefore there is the additional advantage for later combination of the reactor stack that the application of a solderable and temperature-resistant layer can be simply integrated in the overall process. Moreover, tin may be applied to areas of one side of the planes and bismuth to the corresponding areas of the other side, which sides come to lie on one another during stacking, so that upon heating, the low temperature melting eutectic is formed upon heating. Naturally, metals other than tin and bismuth may be used, which form eutectics.
Furthermore, further metal layers can be deposited on the metal surfaces, where specific requirements of the particular case of application dictate this. Thus for example particularly wear-resistant layers against corrosion and abrasion, consisting for example of chrome, nickel/phosphorous alloy or palladium, or surfaces of catalytically active metal (e.g. platinum, palladium, rhodium, iridium, copper, silver, iron, nickel, cobalt, vanadium, chromium,
tungsten, molybdenum and the compounds thereof, e.g. complexes thereof), can be deposited with an electrolytic or electroless method. Also magnetic layers, for example of a ferromagnetic nickel/cobalt alloy can be necessary for specific applications, such for example as with the use of magnetic valves as actors. Furthermore, the surface structure can be roughened or smoothed also by etching techniques.
In order to produce complex three-dimensional fluid duct structures in the microreactor, if necessary connections must be formed between the various duct planes. For this purpose a continuous structuring of the substrate material is necessary. This can be brought about before assembly of the individual planes to form a stack either in a serial process such for example as by laser drilling or mechanical drilling. Alternatively it is also possible in a second photostructuring process after formation of the fluid ducts, to leave those points of the reactor planes unprotected, which are further removed in a second etching process, until the substrate material at this point has a continuous connection to the opposite side of the substrate.
Figure 3 shows diagramatically the formation of continuous perforations through the substrate material. According to this the substrate, for example with a metal core 1 and fluid duct walls 4 consisting of metal, are covered by a photoresist layer 2 (process step B in Figure 3). This layer is exposed by means of an appropriate layout and developed, so that at those points at which the continuous perforations are to be formed, exposed areas of the metal core result (process step C). Then the metal core is etched, forming continuous holes 5 (process step D) and thereafter the photoresist layer is again removed (process step E).
Such etching methods can also be used in order totally to remove the metal in specific layers of the microreactor consisting of a plurality of substrates,
by etching it off in defined areas. Thus, light-permeable windows can be formed, for example in order to pass a light beam from a light source through a fluid duct. By means of such windows, which are preferably formed by a transparent layer of plastic in order to prevent the passage of liquid, analyses can be carried out by means of measurement of light absorption or emission, if an appropriate additional light sensor is provided. Such sensors can also be produced by light-conductive layers within the planes.
When pure metal films are used, chemical and electrochemical methods are principally used as etching methods. For copper films for example a hydrochloric solution of copper (l)-chloride or iron (lll)-chloride can be used. For aluminium films an alkaline solution is suitable. When plastics carriers, for example of polyimide, are used, a chemical method with alkaline etching solutions, a plasma or a laser etching method can be used. The passage holes between the reactor planes can also be formed in the multiple blank, so that in turn the advantage of time, economy and uniformity of the etching method exists for all continuous perforations. Contrary to mechanical drilling devices, etching devices can be operated continuously without problems. Interruption in production due to a defective tool, as frequently occurs in drilling, can be eliminated.
In addition, photostructuring enables the production of connecting ducts which are considerably smaller than the ducts produced by mechanical drilling. This permits the build-up of a finely structured three-dimensional connecting network. This is necessary if a reactor stack consists not only of the layering of identical reactor chambers, but additional components, for example for chemical analysis or for monitored metering of further substances, are contained. In all it is therefore of advantage if the technique for interconnecting the individual planes enables the same structural simplicity and precision as that used in structuring the planes.
Preferably, larger multiple blanks of the substrate with a plurality of plane elements of the reactor are processed in a continuous system with horizontal transport. Such systems are known from the manufacture of printed circuit boards. In this way a uniform and rapid treatment of all the substrates passing through is achieved.
In order to provide adaptation to the desired throughput quantity of fluid, an appropriate number of planes must be integrated into a stack. Figure 4 shows the formation of a stack of three elements. Proceeding from individual layers 1 and a closure segment 6 closing the fluid ducts, a stack is formed which is subsequently soldered for example by heating by means of the solder layer 3 (process step B in Figure 4). Such stacks can then again be considered as reactor modules and if necessary can be combined to form larger blocks.
When the reactor planes are combined into stacks, two conditions must be fulfilled: on the one hand fixing of the planes together, and on the other sealing of the structures through which fluids flow. It is of advantage to close the stacks by front plates which are so designed that they are able to absorb the forces arising during operation of the reactor under high pressure. A plurality of planes are integrated to form a stack in such a way that the planes are secured and sealed with positional accuracy relative to one another. For this purpose appropriate registering elements are used. Particularly suitable are the techniques known from the manufacture of printed circuit boards and semiconductors, such for example as the application and use of so-called tooling holes or optical markers.
The planes can be combined according to DIN 8593 by soldering or by bonding together using adhesives (gluing). Selection of the technique depends on the operational parameters of the microreactor. Important operational parameters are the temperature, the pressure, mechanical
stresses and the chemical composition of the reaction components. Preferably, a soldering process is used which, when the reversal method is used, allows preparation of a soldering process by deposition of appropriate solder layers to be integrated into the manufacturing process.
Adhesive coatings can for example be applied by the screen-printing method. Among other things, adhesives on the basis of epoxy resins, acrylate resins, e.g. also cyanoacrylate resins, polyesters, polyurethanes, amino resins and phenolic resins are used. Due to their chemical resistance, epoxy resins are preferred. In the case of subsequent pre-treatment of the fluid duct surfaces and deposition of metal layers on the surfaces, the end surfaces of the adhesive layers exposed to the fluid can however be sealed, so that chemical compatibility with the fluids in this case is of no consequence.
As during operation of the microreactor at the end surfaces of the stack under pressure forces arise which would exceed the mechanical strength of the external microstructured planes, suitably dimensioned front plates which close off the stack must be provided. These absorb the forces. The front plates can be secured together by appropriate screwing techniques.
In a preferred embodiment, the chemical and morphological structure of the internal reactor surface after combination of the reactor can be modified. For this purpose appropriate fluid (liquid, gas) is passed through the reactor. Such a process sequence is necessary when the process parameters used during jointing are not compatible with the optimum surfaces for the reaction. For example, the solderable intermediate layer can be covered by electroless application of a chemically resistant nickel alloy layer 7 (process step C in Figure 4). If furthermore a temperature increase occurs during jointing, the surface morphology of the metal surface can alter disadvantageously due to recrystallisation. In such a case it is possible to
optimise the surface structure by passing through an appropriate etching solution, e.g. for enhancing the surface roughness, or by depositing a further metal layer. The chemical composition of the surface can also be disturbed by the jointing process. For example, a catalyst surface consisting of a plurality of phases can alter disadvantageously due to phase alteration. Carrier-mounted catalysts are likewise extremely sensitive to temperature alterations.
In particular, the inner surfaces of the microreactor can be subsequently covered by an additional metal layer if by means of the jointing technique only an insufficient seal is achieved against the fluid flows, which are usually passed through the ducts under pressure. In such a case, for example, the insufficiently sealed joints between the planes can be subsequently sealed by a layer of an electroless deposited metal layer, for example a copper or nickel layer.
As a consequence of utilising the opportunities arising from a three-dimensional connecting structure, multi-functional reactor types can be produced. In these, different functional areas are interlinked with one another in a complex manner. Thus it is possible to produce multi-stage synthesis reactors, consisting of cascade circuits of mixing, heat exchanger and reaction zones. In such a module the three-dimensional image of the duct structures is produced after the optimum arrangement of the various zones. A module may be defined similarly to a microelectronic component via its inputs and outputs. Such modules can further be combined to form new reactors.
The microreactors produced according to the method according to the invention can be used in particular for producing toxic, unstable or explosive chemical products according to known synthesis methods. In this way the separate manufacture of these compounds is avoided, so that
dangerous storage is eliminated. If in this case the further processing of specific intermediate products is involved, these compounds are produced on the production site for the end product and are passed directly thereafter into the reactor of the production system. As such compounds are frequently unstable, there is no risk that portions of the intermediate product produced will decompose again before being further processed. Furthermore, dangerous intermediate products need no longer be separately treated, such for example as in preparation or storage.
Examples of manufacture are given in the following in order further to explain the invention.
Example 1 (manufacture of a microreactor by using the etching method and connection of individual reactor planes by soldering)
The etching technique was used to manufacture a microreactor/heat exchanger in microstructuring technology.
Figure 5 schematically shows in a detail the arrangement of three successive copper plates.
The microreactor was made up of a stack of sixty copper plates 1. The plates were so structured that a gas or liquid flow can pass through in one direction. The layers were respectively stacked rotated through 90° to one another. Thus a plate heat exchanger for cross flow operation was formed.
As a substrate material for manufacturing the heat exchanger, copper plates 1 25µm thick with respective dimensions of 1 50 mm x 1 50 mm were used. Four respective plates for the heat exchanger were produced
from these plates.
The first manufacturing step consisted in the application of a photosensitive, negatively operating dry film resist (for example Riston® 4630 of Du Pont de Nemours Inc., USA). After exposure of the resist with the layout and development, copper was uniformly removed to a thickness of 60µm from the exposed areas in a copper etching solution.
There resulted a duct structure with a duct width of about 430 µm, which was predetermined by the layout, a duct depth of 60 µm and a web width of 70 /ym. Thereafter, the photoresist was removed from the entire surface again.
Then a tin/lead layer 4 /ym thick was applied to the structured copper plates.
Thereafter the copper plates were cut into four segments, which corresponded to the individual layers in the reactor packet. The segments were identically structured and were stacked in the prescribed way.
As end plates for the stacks, stainless steel plates 5 mm thick were used. The stack was screwed together by four M8 screws located at the corners. The holes 8 necessary for this, like the necessary gas or liquid passages 9, were drilled in the copper plates and stainless steel plates. In the final step the copper plates were soldered together, the stack being heated to 300°C, so that the tin/lead layer became liquid on the individual copper planes.
A stack produced in this way had respectively 2640 ducts for each flow direction with a cross-sectional area respectively of 0.7 cm2. The entire inner area came to 3000 cm2. The inner area was covered with a tin/lead layer 4 /ym thick.
For a special application for use in a chemical synthesis, the tin/lead layer was removed by passing through a tin/lead etching solution which did not attack copper. In a further method step an extremely thin palladium layer was applied by cementative metal deposition (e.g. 0.02 /ym thick). The microreactor could then be used for carrying out heterogeneously catalysed reactions. As the copper plates separating the individual planes had a very high heat transfer coefficient, such a microreactor was particularly suitable for highly endothermic or exothermic reactions.
Example 2 (manufacture of a microreactor by using the reversal method and connection of the reactor planes by soldering)
In a first manufacturing step, a photosensitive dry film resist (Laminar® HG 2.0 MIL of Morton International GmbH DE), was applied to a copper substrate (corresponding to method step A in Figure 1). After exposure of the resist with the layout (surface areas corresponding to the fluid ducts to be produced were not covered with resist) and subsequent development, a tin/lead layer 6 µm thick was applied by an electrolytic metallizing method in the exposed areas (corresponding to method steps B and C in Figure 1). This layer served both as an etching resist in the subsequent structuring and as a soldering layer for combining the individual layers.
The dry film resist was then again totally removed (corresponding to method step B in Figure 1). There followed a treatment with a copper etching solution (iron (lll)-chloride/hydrochloric acid) by means of which the areas not covered by the tin/lead layer were etched off within a thickness of 60µm (corresponding to method step E in Figure 1).
The copper plates carrying a quadruple copy blank were thereafter cut into four identically structured segments. Forty of these individual layers were
stacked as in manufacturing example 1 after the incorporation of registering, assembly and throughflow bores (corresponding to method step F in Figure 1), between two stainless steel plates coated with copper, of which one served as the closure segment closing the uppermost fluid ducts, the layers then being screwed together and finally soldered together by heating to about 300°C.
The inner surface of the ducts thus produced was not covered with a tin/lead layer and could therefore be covered, in order to carry out a catalyzed chemical reaction, directly by cementative metal deposition with a thin palladium layer (from an aqueous PdS04/H2S04 solution).
Example 3 (manufacture of a composite microreactor by using the
additive method and connecting the reactor planes by gluing)
In a first manufacturing step a dry film resist 100µm thick was applied to a polyimide film (Kapton® of DuPont de Nemours Inc., USA) coated on both sides with a copper film 25µm thick. After exposure of the resist with the layout according to Figure 5 and subsequent development of the resist, those points being exposed on the copper surfaces which were not intended to correspond to the fluid ducts to be formed, a copper layer 80/vm thick was electrolytically deposited in the exposed areas. The resist was then removed.
The copper substrates in turn produced in quadruple copy blanks, were cut into four identical structured segments. Twenty of these individual layers were stacked as in manufacturing example 1 after the incorporation of registering, assembly and throughflow bores.
In this stack a single layer of fluorinated polyethylene (FEP) was integrated
after the pre-treatment described in the following:
The FEP layer, about 1 mm thick (same size and bores as the copper layers) was coated in a radio frequency plasma (PECVD) using organo metallic additives (n-allyl-rr-cyclopentadienyl-palladium-(ll)), with palladium catalyst, the catalyst layer was metallized in an electroless nickel bath with sodium hypophosphite as a reduction agent with a nickel-phosphorous layer about 1 µm thick, and thereupon a copper layer about 30 µm thick was applied electrolytically from a sulphuric copper bath. The copper and nickel/phosphorous layers were structured with the layout of an appropriate electronic circuit according to known methods of printed circuit board technology. Thereafter electronic components such as semiconductor components (microchips) were mounted in the FEP layer and connected to control and signal lines by bonding and soldering methods.
In the lateral areas outwith the duct structures, both on twenty of the structured polyimide/copper layers and also on the FEP layer, a two-component adhesive on an epoxy resin base was thinly applied. The copper layers with the FEP layer in the middle were stacked together, the stack screwed between two stainless steel plates coated with copper and glued together.
All the disclosed features as well as combinations of the disclosed features are the subject-matter of this invention, insofar as they are not expressly referred to as known.
1. Method for manufacturing chemical microreactors, having one
substrate which has fluid ducts as well as feed and discharge lines for
fluids, without using plastic moulding methods, said method having the
following method steps:
a. forming fluid duct structures on metal surfaces situated on
the substrate by means of a photoresist layer or a screen-printing
lacquer layer, such that the metal surfaces are partially covered by the
b. at least partial currentless and/or electro-chemical etching
away of metal from the bare surfaces of the substrate;
c. completely removing the photoresist or screen-printing
d. forming solder layers;
e. superimposing the substrates and a closure segment closing
the fluid ducts and interconnecting the substrates and the closure
segment by soldering.
and optionally comprising between step (b) & (c) the step of depositing a metal layer on the bare surfaces of the substrate or on the metal surface.
2. Method as claimed in claim 1 wherein a metal layer is deposited on
the bare surfaces of the substrate by currentless and/or electrochemical
deposition and the step of etching takes place after removal of the
lacquer layer, thereby forming fluid ducts.
3. Method as claimed in claim 1 wherein the step of etching is
replaced by the said step of deposition of the metal layer on the bare
surfaces of the substrate and wherein the fluid duct structures are
alternatively formed directly on the substrate.
4. Method as claimed in one of claims 1 to 3, wherein the substrate is
coated in method step a. with a photoresist layer, the photoresist layer is
exposed with a positive or negative image of the fluid duct structure and
5. Method as claimed in one of the preceding claims, wherein a
substrate is used which has surfaces formed from at least one metal,
selected from the group comprising steel, stainless steel, copper, nickel
6. Method as claimed in claim 2, wherein in method step of
currentless and/or electrochemical deposition of a metal layer, at least
one metal, selected from the group comprising tin, lead, nickel, cobalt,
bismuth, silver, gold and an alloy of these metals, is applied by
currentless and/or electrochemical means.
7. Method as claimed in claim 6, wherein the substrates are
interconnected by soldering, by the layer of tin, lead, or bismuth or the
alloy layer being connected with the metal under the action of heat.
8. Method as claimed in one of claims 6 or 7, wherein in method step
of currentless and/or electrochemical deposition, a tin layer is applied to
one side of the substrates and a bismuth layer to the other side, and in
that the substrates are superimposed in such a way that the tin layers
and the bismuth layers lie the one above the other and are then soldered
to each other.
9. Method as claimed in one of the preceding claims, wherein in
individual layers of the microreactor which comprises a plurality of
substrates, the metal is completely removed by etching in defined areas
in order to form translucent windows.
10. Chemical microreactor constructed by the method as claimed in
claim 1, having fluid ducts in at least one plane as well as feed and
discharge lines for fluids, in which microreactor the fluid ducts are
delimited by metal side walls facing one another and by additional metal
or plastics side walls extending between said side walls, wherein the
planes are connected to each other and/or to a closure segment, which
closes fluid ducts lying open, by means of suitable solder layers.
11. Microreactor as claimed in claim 10, wherein the side walls are
formed from at least one metal selected from the group comprising steel,
stainless steel, copper, nickel and aluminium.
12. Microreactor as claimed in one of claims 10 and 11, wherein the
solder layer is formed by metals selected from the group comprising tin,
lead, bismuth, antimony, silver and alloys thereof.
13. Method for manufacturing chemical microreactor s substantially as
herein described with reference to the accompanying drawings.
14. Chemical microreactor substantially as herein described with
reference to the accompanying drawings.
|Indian Patent Application Number||419/DEL/1998|
|PG Journal Number||10/2008|
|Date of Filing||18-Feb-1998|
|Name of Patentee||ATOTECH DEUTSHLAND GMBH|
|Applicant Address||ERASMUSSTRASSE 20-24, D-10553 BERLIN, GERMANY.|
|PCT International Classification Number||C03F 1/20|
|PCT International Application Number||N/A|
|PCT International Filing date|