RU2588519C2 - Layered, leak-tight chemical reactors, methods of production and methods of use - Google Patents

Abstract

FIELD: chemistry.

SUBSTANCE: invention relates to laminar chemical rectors, particularly microchannel devices, methods for production and operation thereof. Chemical reactor for carrying out one or more processing steps includes a layered microchannel assembly including first and second adjacent sheets which are substantially flat, cross sectional area of sheet is greater than 100 cm², wherein first sheet comprises a system of parallel microchannels separated by a partition and including welds which extend along length of partitions and connected to first and second sheets.

EFFECT: invention provides reactors of better quality and lowers cost of manufacture.

31 cl, 23 dwg, 6 tbl, 10 ex

Description

Related Applications

This application claims priority over US patent applications with provisional serial numbers 61 / 394,328, filed October 18, 2010, and 61 / 441,276, filed February 9, 2011.

Introduction

The traditional idea of microchannel technology is that optimal heat transfer in a microchannel heat exchanger can only be achieved by diffusion welding and / or soldering. These methods are based on the formation of continuous metal contact surfaces between the layers. The solid contact surface eliminates the thermal contact resistance between the layers and is considered necessary for transferring high levels of heat from the exothermic reaction to the chambers to remove heat or to add heat during the endothermic reaction.

Soldering requires the addition of an intermediate layer of material that melts at a temperature below the melting temperature of the materials of the structure. During diffusion soldering or during brazing, the intermediate layer becomes liquid. The liquid intermediate layer flows, filling gaps or cavities and thus connecting materials. During the flow of the intermediate layer, its diffusion also occurs, as a result of which the materials of the intermediate layer can diffuse into the main material, and the material from the main layer can diffuse into the intermediate layer. As diffusion develops, the local composition of the material of the intermediate layer changes. Then, after the start of temperature reduction, when the maximum temperature is reached, the liquid intermediate layer hardens and fills the free space between the two main layers. Curing can be determined by temperature or composition. In the latter case, a melting point depressant, such as phosphorus or boron, is added to allow the intermediate layer to melt at a lower temperature with respect to the base material. In a similar example, a device made by diffusion welding also forms a tight thermal contact between the heat transfer layers.

Summary of the invention

In one aspect, the invention provides a method for manufacturing a layered device, comprising: providing a bottom sheet, a top sheet, and characterized in that the gap between the surface of the top sheet and the surface of the bottom sheet [note that the top and bottom sheets can be part of subnodes]; providing a heat-conducting finned insert, the height of which is at least 1% greater (preferably at least 2% greater, and in some embodiments, 1-10% greater) of the height of the gap, the location of the finned insert in the gap; and compressing the sheets so that the ribbed insert is deformed in accordance with the internal shape of the gap.

The insert may or may not have a catalytic coating. In some preferred embodiments of the invention, the resulting device has an irregular configuration in which the insert is deformed incorrectly, for example, not all crumpled walls are bent in one direction or in alternating directions. In some embodiments, the finned fins are advantageously (greater than 50%, preferably greater than 80%) curved in one direction. Compressed ribbed plates do not have the correct shape, such as that shown on the corrugated sheet.

In some preferred embodiments, the surface of the bottom and / or top sheet has fusion lines protruding from the surface; in some preferred embodiments, the sheet includes one or more internal fusion lines with a continuous length of at least 50% (preferably at least 80%) of the length or width of the sheet; and at least one side of the finned insert is at least partially (preferably completely) on one or more fusion lines. Preferably, the flow direction through the insert is perpendicular to the fusion lines. Surprisingly, we found that good heat transfer was achieved when these fusion lines were present, even when the fusion lines were located between the reaction chamber (i.e., the press fit chamber and the catalyst) and the heat exchanger. In some embodiments, the finned insert is tack-welded to the upper or lower surface, in a more preferred embodiment, the finned insert is held in place by press fit and is not connected to any of the surfaces by welding. Preferably, the finned plate is located inside the microchannel, and / or the adjacent heat exchanger includes microchannels.

In addition to all the methods described herein, the invention also includes a device manufactured in accordance with these methods. The invention also includes the operation of a device manufactured in accordance with the method for performing one or more technological operations. Preferably, the devices described herein are chemical reactors that are designed to perform one or more process operations.

In another aspect, the invention provides a layered device, comprising: a device including a fluid inlet connected to a working chamber and an outlet of a working medium connected to a working chamber; characterized in that the working chamber includes a gap and a heat-conducting finned insert, which is located inside the gap and touches the upper and lower parts of the gap; further characterized in that the finned insert is irregular in shape, which is caused by at least 1% compression of the finned insert within the gap; and a heat exchanger in thermal contact with the working chamber. In this device, the irregular shape is caused by compression of the sheets on both sides of the gap so that the height of the finned insert is deformed by at least 1% of the height of the finned plate (where the height is determined in the same direction as the height of the stack), preferably at least 2%, in some embodiments, at least 5%, and in some embodiments, in the range of 1 to 10%.

This device may contain any of the features described herein. For example, in some preferred embodiments, the insert is not coated catalytically; in some preferred embodiments, the implementation of the catalyst particles are present in the gap. The conductive finned insert is a finned plate that transfers heat from the process taking place inside the finned section to adjacent heat transfer layers. The finned insert is preferably made of a material having a higher thermal conductivity than the material of the heat transfer layer construction. In some preferred embodiments, the conductive fin insert has a thermal conductivity that is more than 10 times greater and more preferably more than 100 times the effective thermal conductivity of the catalyst located inside the conductive finned structure. For most materials, thermal conductivity is known, and if not, it can be measured using the standard method of the American Society of Testing and Materials.

In a further aspect, the invention provides a method for conducting a chemical reaction, comprising: placing at least one reagent in a working chamber of a device in accordance with claim 2; carrying out the process in the working chamber and at the same time heat exchange between the working chamber and the heat exchanger.

In another aspect, the invention provides a method for manufacturing a layered device, the method comprising: providing a first subnode or first sheet and a second subnode or second sheet; characterized in that the first subnode or first sheet includes a first parallel channel system, and characterized in that the second subnode or second sheet includes a second parallel channel system, and characterized in that there is no intersection between the channels in the first subnode or first sheet and the second subnode or second sheet; welding the edges of the first subassembly with the edge of the second subassembly to form a combined layer of the subassembly or welding the edges of the first and second sheets to form a common welded sheet; and laying a combined layer or common welded sheet with one or more layers or sheets, and joining stacked sheets to form a layered device.

Despite the lack of interchannel mixing for the channels in the first and second sheets, the first and second parallel channel systems can share a common rear and / or front receivers. In one example of this method, a sheet (or a subnode including a series of sheets) is cut into four parts, and these four parts are again welded together along the edges - this is a rather paradoxical process, since the cut sheet is welded along the edges of the previous cut. However, it was found that this process significantly reduces warpage and, accordingly, provides a higher quality laminated product. In some extended aspects of this method, the first and second sheets (or subnodes) are not parts of a cut common blank, but can be made separately. Preferably, the stacked layers have the same width and length (in some embodiments, within 5% of the length and width; in some embodiments, within 1% of the length and width). The cuts (and reconnecting welds) should be parallel to the length of the channels; with this orientation there is no risk of blocked or intermittent channels.

In some preferred embodiments, the width / height and length / height ratio for the first subnode is greater than 2, more preferably greater than 10. In some preferred embodiments, the width / length or length / length ratio for the first sheet or first subnode is 1.5 or more, more preferably 2 or more, in some embodiments, 4 or more. By height is meant the stacking height, the width and length are mutually perpendicular, and the length is located in the direction of the fluid flow along the sheet. Preferably, the first subnode or first sheet includes at least 5 first parallel channels (preferably microchannels); the second subnode or second sheet includes at least 5 second parallel channels (preferably microchannels), and the first and second subnodes or sheets are connected along the edge to form one sheet or subnode in which the first parallel channels are parallel to the second parallel channels. Preferably, the first and second subnodes or sheets are connected so that adjacent parallel channels in the first and second subnodes are located within 2 cm, more preferably within 1 cm in the width direction (the length is parallel to the general direction of flow through the device). The resulting device will contain sheets with welds that connect the segments of the sheets. In some preferred embodiments, in particular with a ratio of width / height or length / height of the first sheet or first subassembly greater than 1, before welding the edges, the step of aligning one or more (and preferably all) subassemblies is performed.

In addition to all the methods described herein, the invention also includes an article made in accordance with these methods.

The invention includes a layered device, comprising: a first layer having dimensions of width and length; characterized in that the first layer includes a first section including a first row of parallel channels, and a second section including a second row of parallel channels, in which the first and second rows of channels are parallel; in which the first and second sections are joined together by a welded joint and in which the welded joint is parallel to the parallel channels; and a second layer, characterized in that the second layer is welded to the first layer. In some preferred embodiments, the first layer is a subnode and the second layer is a subnode, and the layers are joined by a weld along the perimeter of the layers.

In a further aspect, the invention provides a preliminary bending method to obtain a more even subassembly. This method of manufacturing a layered device includes providing a metal sheet; deformation of the metal sheet and welding connection of the deformed metal sheet into a layered device.

In a preferred embodiment, the metal sheet is deformed in an arc with a curvature of 10 degrees from a flat surface, in some embodiments, with a curvature in the range of 10 to 80 degrees from a flat surface. The parts are preferably pre-bent from a flat state to ensure deformation in such a way that when the parts are subjected to stresses from welding (especially from laser welding), they return to an almost even state. The arc of curvature from an even state is less than 90 degrees and preferably from 10 to 80 degrees. A more preferred range is from 30 to 70 degrees. In some preferred embodiments, the metal sheet is made of stainless steel.

In another aspect, the invention provides a method for manufacturing a layered device, comprising: providing a subassembly; subnode alignment welding a subnode with a sheet or second subnode. Preferably, the subunits are welded around the perimeters, with the exception of areas with cutouts for the inlet and outlet openings. In some preferred embodiments, this method is combined with any of the methods described herein.

In another aspect, the invention provides a method for carrying out a process in a channel layer in which the compression stress is changed or vice versa, comprising providing a device comprising a first channel layer and a second channel layer located directly adjacent to the first channel; characterized in that in the first step, the first channel layer contains a first fluid at a first pressure, and the second channel layer contains a second fluid at a second pressure; characterized in that the first pressure is higher than the second pressure; carrying out a technological operation in the first channel layer at the first stage; characterized in that in the second stage, the first channel layer contains a third working medium at a third pressure, and the second channel layer contains a fourth working medium at a fourth pressure; carrying out a technological operation in the first channel layer in the second stage.

In tension, the pressure in the first channel layer is higher than in the second channel layer, and in compression, the pressure in the first channel layer is lower than in the second channel layer. In the direction of height, the layer is defined by the bottom and top of the layer, and a tensile or compressive force affects the bottom or top of the layer. In some preferred embodiments, the first and third fluids are the same, and the second and fourth fluids are the same; for example, the first and third fluids can be a Fischer-Tropsch process stream (“FT”), and the second and fourth fluids can be a heat carrier, usually water (or some other heat carrier), which is partially boiled. Examples of other processes include the synthesis of ethylene oxide, propylene oxide, methanol, ammonia, styrene and hydrogenation and hydrotreatment. Preferably, the method is carried out in a layered device with a first channel located in the first layer and a second channel located in the adjacent layer. Most preferably, the process is carried out in a layered device containing multiple (at least 10) alternating layers of the first and second channels. In some preferred embodiments, the first channel comprises a press fit insert, and in some preferred embodiments, the first channel comprises a catalyst that is preferably used in combination with a press fit. Surprisingly, we found that a device made without diffusion welding or soldering (such as a press-fit device) can function successfully under these conditions. Preferably, this method is defined as being observed during continuous operation, and not during shutdown or start-up. This process can occur, for example, as a residue forms in the process channel and / or a loss of catalyst activity, and the process conditions become equal to compensate for changes in the reactor.

In a further aspect, the invention provides devices in which a crown connects sections of the device (described below). Thus, the invention provides a device comprising a first layered unit comprising a series of stacked sheets; characterized in that the first side of the first laminate assembly comprises a series of inlet or outlet openings; characterized in that a receiver protruding from the first side and enclosing a series of inlet and outlet openings is attached to the first side. Typically, the receiver is made of metal and is welded to the first side of the first laminate assembly.

In some preferred embodiments, the apparatus includes a second laminate assembly comprising a series of stacked sheets; and characterized in that the first side of the second laminate assembly includes a second row of inlet and outlet openings; and characterized in that the receiver connects the first row of inlets or outlets to the second row of inlets or outlets.

In another aspect, a layered microchannel device having a length and a width, characterized in that the cross-sectional area, defined as the product of the sheet length by the sheet width, exceeds 100 cm ² (in some embodiments, exceeds 500 cm ² ), connected to the upper plate to form a welded assembly with a linear density of laser welded joints from 0.05 to 20 cm / cm ² (preferably from 0.1 to 10 cm / cm ² ) on a sheet surface section within the microchannel device; this section includes at least 50% (preferably at least 90%, in some embodiments 100%) of the continuous area of the main surface (sheets have two main surfaces). In most cases, the sheet forms the surface of the subnode in a larger device. Welds connect two or more sheets. Welds can also provide a connection between two adjacent channels of the internal stream. In this device, “100% continuous area” means the entire surface (not just 100% of the selected rectangular area); similarly, 50% and 90% mean 50% and 90% of the entire surface. Preferably, the ratio of length to width of the sheet is greater than 2. In addition to this or as an alternative to the indicated linear density, the device of the invention can be defined as having internal welds (i.e., welds on the inside of the sheet, and not along the perimeter), the length of which exceeds at least 10 times, preferably at least 100 times, the circumference (perimeter) of the device.

In another aspect, the invention provides a layered microchannel assembly including a first sheet and a second sheet; characterized in that each sheet has a length and a width, wherein the cross-sectional area, defined as the product of the sheet length by the sheet width, exceeds 100 cm ² (in some embodiments, exceeds 500 cm ² ); characterized in that the first and second sheets are essentially flat (the sheets may have some warpage, but are not corrugated); characterized in that the first sheet contains a system of parallel microchannels, the microchannels being separated from each other by a partition (microchannels can partially extend through the thickness of the first sheet (eg, etched channels) or extend through the entire thickness of the first sheet); characterized in that the first sheet and the second sheet are adjacent; and including welds that extend along the length of the partitions (welds may be continuous or intermittent) and connect the first sheet to the second sheet.

The phrase "pass along the length" means that the welds are directed in the same direction as the partitions and are in contact with the partitions.

In any of the methods described herein, the assembly can be formed by joining two or more welded sub-assemblies, the welding method of two or more sub-assemblies to form the assembly can be the same welding method or another welding method. In some embodiments, the welded subunits may also contain joints formed using techniques other than welding.

In a further aspect, the invention provides a method of forming a node, comprising welding the upper sheet with the lower sheet to form a series of channels located between the upper surface of the upper sheet and the lower surface of the lower sheet, characterized in that welding is used to form seams between the channels in a number of channels. The upper and lower surfaces are the upper and lower surfaces of the layered assembly. For example, the bottom sheet may contain etched channels, and the top sheet may be a flat sheet without etching. Preferably, the method of joining the sheets into nodes includes laser welding to create a seam between two adjacent channels of the internal stream. In any of the embodiments described herein, the channels may preferably be microchannels.

In another aspect, the invention provides a prefabricated welded assembly that has undergone a repair process to eliminate leaks or holes in the initial welding step, characterized in that the repair process may include the same primary welding technique (typically laser welding) or may use a secondary a process, such as gas-tungsten welding, pulsed laser, welding with cold transfer of metal or another, to reduce the number of leak points in the welded workpiece assembly.

In a further aspect, the invention provides a weldment blank that can withstand a pressure drop of more than 100 psi. inch hut at ambient temperature (more preferably more than 500 psi at ambient temperature). A welded assembly is a layered device in which, during operation, fluid flows are directed mainly perpendicular to the thickness of the sheet. In the welded assembly, the seals to withstand the differential pressure are made without diffusion welding or soldering.

In another aspect, the invention provides a method of forming a layered welded device, comprising: obtaining a welded workpiece assembly that has a curvature greater than 1 cm when mounted on a flat table; the implementation of the alignment process for the welded workpiece assembly with the creation of a substantially flat part with a lift of not more than 1 cm above the surface when mounted on a flat table; and welding the aligned workpiece assembly with a subassembly to form a layered weld device.

In a further aspect, the invention provides a prefabricated welded assembly comprising a series of channels that are sealed by welding (the seal is not caused by the use of polymer spacers, soldering, diffusion welding, or other conventional techniques) and having a leak rate of less than 0.5 psi. inch hut in 15 minutes; or a leak rate of less than 10 cm ³ / min of nitrogen (preferably less than 1 cm ³ / min of nitrogen) when nitrogen is injected at 100 psi. inch hut and ambient temperature.

In a further aspect, the invention provides a pressure-resistant layered device, comprising: a stack of sheets that are joined together (preferably by welding); characterized in that the stack of sheets includes a series of channels passing through the stack, and at least one inlet and one outlet openings connected to a series of channels; a continuous unconnected portion, which is the free volume in the stack of sheets; first and second end plates on opposite sides of the layered device; and further comprising a set of stiffeners that are held (for example, by welding) in tight contact with the main outer surfaces of the end plates and continuously protrude over the entire area bounded by a continuous unconnected portion. The stiffness of the elements of the set provides resistance to bending in the laying direction (i.e., the direction orthogonal to the plane of the layers). Preferably, the system further includes a process stream comprising hydrogen and / or hydrocarbon passing through a series of channels.

The invention also provides repair of welded devices. Devices can be restored by removing one or more welds. By removing the plate at one end of the channels, it is possible to block selected channels, thereby reducing lesions or otherwise avoiding channels containing defects. Alternatively, the device can be opened by removing the weld along the perimeter, after which it becomes possible to remove or replace subnodes. In some embodiments, the subnode may be removed and restored before reinserting into the device. After removing and / or replacing the subassembly, welding can be used again to close the device. To facilitate dismantling, one or more surfaces of the subassembly can be coated with a release layer, such as a coating of zirconium oxide or yttrium oxide, or an insert of graphoil or ceramic paper between the nodes can be used.

The invention also includes methods for maintaining a reactor by opening a weld and removing or replacing: a catalyst (such as a dispersed catalyst), a finned or wave-like plate, or subassembly. After this, re-welding of the system is possible. The invention further includes devices having weld metal structures associated with maintenance or repair.

In a further aspect, the invention provides a layered chemical reactor, comprising: a series of sheets in a stack; a stack having mutually perpendicular dimensions — height, width and length; characterized in that the height is the size of the stack; for open space inside the stack, the length is the largest of the dimensions, and the width is perpendicular to the length; characterized in that the stack includes at least one contact surface between the sheets, and during operation, internal pressure is created in the open space inside the stack and applied to the contact surface, the width of the open space being at least 0.07 m; and having such a resistance to leakage that when gaseous N _{2 is} supplied through the inlet to the free volume in the specified contact surface with the outlet closed, the pressure increases at a rate of 30-50 kPa / min, the pressure in the free volume increases to 790 kPa and the shutter speed is pressure for 15 minutes, followed by a return to ambient pressure by venting gaseous N _2, and further supply of water at a rate sufficient to increase the pressure through the inlet at a speed of 300-400 kPa / min With a closed outlet to increase the pressure in the free volume to a level of about 3000 kPa and continue to increase the pressure at a speed of about 100 kPa / min to 6000 kPa with a further decrease in pressure at a speed of 250-300 kPa to a level below 5300 kPa and continue to reduce the pressure to pressure environment, water draining and drying the reactor, and re-supplying N ₂ gas through the inlet at a rate of pressure increase of 30-50 kPa / min into the free volume in said contact surface at the closed outlet, increasing The pressure in the void volume of up to 790 kPa, and the overlap of the inlet opening to cut off the gas entering the free volume, the level of device leakage was not more than 100 kPa over the next 15 minutes.

The invention includes reactors with claimed leakage resistance, and not just reactors that have been tested by protocol. More preferably, the reactor has a leakage resistance of less than 30 kPa over the next 15 minutes, and in some embodiments, in the range of 1 to 50 kPa. In some preferred embodiments, the width of the open space is at least 0.1 m, in some embodiments, the width is at least 0.3 m. The reactor may be of any shape; and in some embodiments, the reactor consists of a stack of rectangular sheets, while in some other embodiments, the reactor consists of a stack of round sheets.

Preferably, the reactor has an end plate whose thickness is not more than 3 cm, preferably not more than 1 cm, and in some embodiments not more than 0.5 cm. Preferably, the reactor has a width and a length of at least 0.3 m; in some embodiments, at least 0.5 m. The invention includes any combination of features provided herein; for example, a reactor having a width and length of at least 0.3 m and an end plate whose thickness does not exceed 3 cm. The contact surface may be flat, but not necessarily flat.

In some embodiments, the device comprises more than 0.1 km of internal linear welds, in some embodiments, more than 1 km of internal linear welds and an exoskeleton. Preferably, the integrity of the layered chemical reactor is ensured by exoskeleton and welds. Preferably, the layered chemical reactor does not contain diffusion welding or soldering and does not have gaskets. No clamps are needed for bonding a layered chemical reactor, and due to the leakage resistance of the reactor (which is not connected with the clamps), it is not necessary to place it in a protective shell under pressure. In some preferred embodiments, at least 60% of the volume (in some embodiments of at least 80% of the volume) of the layered chemical reactor is made up of microchannels and other free volumes.

In the leakage test described above, the phrase “with the outlet closed” means that N ₂ has no other way out of the free volume, except for leakage through the contact surface between the sheets. In addition, it must be understood that this test is applied to a single interface or averaged sum of all the interfaces served by this inlet. The device passes the test, if at least one liquid circuit with the indicated parameters passes the test, preferred devices have at least 2 liquid circuits that pass the test, more preferably all liquid circuits pass the test (for example, if the device has two liquid circuits, each of which is served by one inlet and one outlet, both circuits withstand the test). The device is under the influence of environmental pressure, with the exception of sections of the device connected to the specified inlet.

The invention also includes a layered chemical reactor having an exoskeleton and having one or more (including any combination) of the features described herein. A continuous unconnected portion within the boundary of the internal pressure is the minimum distance between a given joint point between the layers within the contact surface under the influence of a given pressure in the layered device and an adjacent joint point between the same layers. In preferred embodiments, an exoskeleton is welded to the device, while in other embodiments, the exoskeleton is fixed by soldering, glue, or other means.

Exoskeleton is more effective than clamps. Clips can be easily removed (cutting or grinding is necessary to remove the exoskeleton). In addition, in a welded exoskeleton, the reinforcing elements may have a rectangular cross section oriented with the longer side parallel to the direction of load application with increased stiffness for resistance to bending stress. This allows the use of thinner sheathing sheets and reduces the weight and cost of the material necessary to resist loads of the same magnitude. Instead of an exoskeleton, clamps having thick plates with threaded joints could be used; however, the plates would have to be strong enough to resist bending stress, since threaded joints would not be loaded in this direction. The threaded joints would have to be strong enough to resist the full tensile stress caused by the force created by the pressure plate. The exoskeleton provides additional support plates in both cases. In addition, clamps are more likely to be subject to loosening and failure during cyclic operation.

The invention also includes a method of operating a layered chemical reactor, comprising: supplying gas to the reactor inlet to increase the pressure inside the reactor to a first pressure; additionally leak detection and additionally leak repair; gas bleeding; supplying fluid to the reactor to increase the pressure inside the reactor to a second pressure, the second pressure being higher than the first pressure; fluid removal; supplying gas to the reactor inlet to increase the pressure inside the reactor to a third pressure, the second pressure being higher than the third pressure; and measuring reactor leaks under a third pressure. In some preferred embodiments, the fluid is a liquid. This method is superior to the one-stage loading method of a device by pumping a fluid followed by a leak test.

In yet another aspect, the invention provides a layered device, comprising: a stack of sheets joined by welding; a stack of sheets containing the main package of sheets connected by welding around the perimeter of the sheets; and a connection for relieving stress in the stack of sheets; characterized in that the stress relieving connection includes two adjacent sheets that are stacked but not substantially connected to each other around the perimeter of two adjacent sheets.

In the present description, "adjacent" means directly adjacent without intermediate sheets.

The invention includes any method of using any device described herein; for example, a chemical process using any device described herein. Similarly, the invention includes any device for implementing any of the methods described in this application. The invention further includes any combination of the methods and / or designs mentioned herein. The sheets and inserts (if any) are preferably made entirely of metal. In the end devices, the metal may be coated with protective coatings and / or catalytic coatings, such as a layer of porous metal oxide with a catalyst deposited on the metal oxide.

The invention is intended to include, in various alternative embodiments, any general concepts and features that are described herein and may be determined by those skilled in the art. For example, the device of the invention may have any combination of features described in the present invention.

The invention is not limited solely to the options set forth above, and includes any methods, systems, and devices described herein. The invention includes all features or combinations of features described herein. The invention also includes chemical processing methods (including, for example, heat transfer, chemical reactions, Fischer-Tropsch synthesis reaction (FT)) and, for example, includes one or more (including any combination) of the conditions, transformations, etc. that are mentioned in this document. For cases of describing a process with reference to graphs or tables, the invention includes processes that have values within +/- 20%, more preferably about 10%, even more preferably about 5%, and in some embodiments within about 1% of the parameters of the conditions, ranges and / or values indicated herein. For example, the invention includes a FT synthesis method, with a contact time in the range of about 90 to about 278 ms, a CO conversion of about 58 to about 73%, and a methane selectivity of about 8 to about 34% (the term “about” should include the meanings within +/- 20%); the invention includes methods that are further determined by device designs; the invention may alternatively be defined as a system that includes both the features of the devices and the compositions of the fluids and / or conditions - for example, the system may be a device containing hydrogen gas and hydrogen oxide at a temperature of about 180 ° C.

Glossary

A “node” is two or more plates connected together to form a layered structure. The site, as a rule, consists of a number of "subnodes" and may be a functioning device or a blank for the device. A “subnode” is a “node” that is (or is intended to be) a component of a larger layered node. In some preferred embodiments, the assembly is completely sealed, with the exception of the inlet and outlet openings. The node is not required to be a fully functioning device; for example, it may be a preform or an intermediate product for a fully functioning device. For example, in some cases, a secondary pruning step is required to open the inlet and outlet channels of the stream. In some embodiments, the implementation of the node is made of plates having dimensions of the width and length of the end device; in some embodiments, the nodes may be cut into several subnodes or alternatively connected to form a larger node. In some preferred embodiments, the implementation of the node (or subnode) has a thickness of 1 cm or less; in some preferred embodiments, the implementation of the thickness is in the range from 0.1 to 1.0 cm; in some embodiments, the thickness is in the range of 0.2 to 0.4 cm. It is preferred that the plates in the assembly are substantially even and that the assembly has smooth upper and lower surfaces.

As used herein, the terms “plate”, “sheet”, “layers” and “gasket” are used interchangeably. The plates have a thickness of 1 cm or less; preferably 0.5 cm or less; more preferably 0.3 cm or less; and typically have a thickness of at least 0.02 cm.

An “exoskeleton” is a set of stiffeners that are held (for example, by welding) in tight contact with the main outer surfaces of the end plates of a layered chemical reactor and continuously protrude along the length of the end plate, which is located between the internal pressure boundary and the external pressure boundary of the stiffener set. The stiffness of the elements of the set provides resistance to bending in the laying direction (i.e., the direction orthogonal to the plane of the layers). The exoskeleton is not a clamp and does not require screws or bolts.

“Clearance” is the smallest microchannel size. Typically, in a layered device, the gap is in the stacking direction (i.e., height). When using the term “clearance” in preferred embodiments, it acts as a substitute for the height of the microchannel.

A "microchannel" is a channel having at least one internal dimension (wall to wall, excluding catalyst) of 10 mm or less, preferably 2 mm or less and more than 1 μm (preferably more than 10 μm) and, in some embodiments, from 50 up to 1500 microns, while from 500 to 1500 microns are particularly preferred when using a dispersed form of the catalyst; preferably the microchannel retains these dimensions along the length of at least 1 cm, preferably at least 20 cm. In some embodiments, in the range of 5 to 100 cm, and in some embodiments, in the range of 10 to 60 cm. Microchannels also determined by the presence of at least one inlet that does not coincide with at least one outlet. Microchannels are not just channels through zeolites or mesoporous materials. The length of the microchannel corresponds to the direction of flow through the microchannel. The height and width of the microchannel are substantially perpendicular to the direction of flow through the channel. In the case of a layered device in which the microchannel has two main surfaces (for example, surfaces formed by stacked and connected sheets), the height is the distance from the main surface to the main surface, and the width is perpendicular to the height. In some preferred embodiments of the present invention, the microchannels are straightforward or substantially straightforward — this means that a straight line can be freely drawn through the microchannel (“unhindered” means before feeding solid catalysts, sorbents, or other dispersed solid materials). Typically, devices contain a number of microchannels that have a common rear receiver and a common front receiver. Although some devices have one rear receiver and one front receiver, a microchannel device may have several rear receivers and several front receivers. Microchannels are also determined by the presence of at least one inlet that does not coincide with at least one outlet — microchannels are not just channels through zeolites or mesoporous materials. The height and / or width of the reaction microchannel is preferably 2 mm or less, and more preferably 1 mm or less. The walls of the microchannel are determined by the walls of the reaction channel. These walls are preferably made of a solid material, such as stainless steel or a heat-resistant alloy based on Ni, Co or Fe, such as FeCrAlY. Technological layers can be made of a material different from the material of the heat-exchange channels, and in one preferred embodiment, the technological layer is made of copper, aluminum or another material with a thermal conductivity above 30 W / (m · K). The choice of material for the walls of the reaction channel may depend on the reaction for which the reactor is intended. In some embodiments, the walls of the reaction chamber are made of stainless steel or Inconel, which are durable and have good thermal conductivity. As a rule, the walls of the reaction channel are made of material that provides the primary structural support for the microchannel device. The microchannel device can be made using known methods and, in some preferred embodiments, is made of alternating layered plates (also known as “gaskets”), and preferably gaskets intended for the reaction channels alternate with gaskets intended for heat exchange. Some microchannel devices include at least 10 layers (or at least 100 layers) constituting a layered structure in which each of these layers contains at least 10 channels (or at least 100 channels); the device may contain other layers with fewer channels.

In some devices, process channels contain catalyst particles. Preferably, the particles have a size (largest) of 5 mm or less, in some embodiments, 2 mm or less. Particle size can be measured using sieves, microscopy, or other appropriate techniques. For relatively large particles, screening through a sieve is used. Dispersed materials included in the process channels may be a catalyst, adsorbent or inert material.

The coolants can flow through heat transfer channels (preferably microchannels) adjacent to the process channels (preferably reaction microchannels), and can be gases or liquids, including steam, liquid metals or any other known coolants — the system can be optimized to change the phase in heat exchanger. In some preferred embodiments, the implementation of a number of heat transfer layers alternates with a number of reaction microchannels. For example, at least 10 heat exchangers alternate with at least 10 reaction microchannels, and preferably 10 layers of heat transfer microchannel systems have a contact surface with at least 10 layers of reaction microchannels. In some preferred embodiments, when there are “n” layers of process microchannels, “n + 1” heat transfer layers are present, the heat transfer layers contacting all process layers. Each of these layers may contain simple rectilinear channels or the channels inside the layer can have a more complex geometry. The invention includes systems that include both the device and the fluid present in the device. In "diffusion welding", the heating process is used to connect parts in which the diffusion of elements from one part to another occurs, resulting in a connected product due to diffusion of elements in the contact surface area (or next to the area representing the contact surface before diffusion welding).

When soldering, an intermediate layer is used, located in the gaps between the parts, the intermediate layer has a lower melting point than the parts.

When welding, heat is used to connect or seal parts. Unlike soldering, a fusible material is not required for welding, although it is recognized that a wire of the same material or similar material, which may have a slightly lower melting point, can be used for a weld, however, the joint is also referred to as a welded joint when sealing along the perimeter of a layered device with a certain penetration depth of the weld at the periphery, and not over the entire volume of the product. A “weld” in the finished part can be determined by a skilled worker — for example, a metallurgist can determine the weld using microscopic examination or other techniques known in the art.

“Joining” includes welding, diffusion welding, bonding and soldering. A join is any process that ties together two or more parts.

A “workpiece assembly” consists of a series of sheets that are connected to each other to form an integral laminated stack. The term “workpiece assembly” is sometimes used to designate a panel, and it may consist of upper and lower sheets defining a flow path, and, as a rule, contains a series of sheets in a stack defining numerous flow paths.

The invention also includes methods for performing technological operations within the device described herein. "Technological operation" means a chemical reaction, evaporation, compression, chemical separation, distillation, condensation, mixing, heating or cooling. “Technological operation” does not only mean fluid transfer, although technological operations are often accompanied by transfer. In some preferred embodiments, the process step is not solely mixing.

Microchannel reactors are characterized by having at least one reaction channel having at least one size (wall to wall excluding catalyst) of 1.0 cm or less, preferably 2 mm or less (in some embodiments, about 1 mm or less) and greater than 100 nm (preferably greater than 1 μm), and in some embodiments, from 50 to 500 μm. The channel containing the catalyst is a reaction channel. In general, a reaction channel is a channel in which a reaction occurs. The length of the reaction channel is usually higher. Preferably, the length of the reaction channel is greater than 1 cm, in some embodiments, greater than 50 cm, in some embodiments, greater than 20 cm, and in some embodiments, in the range of 1 to 100 cm.

“Press fit” describes a method for arranging a finned plate (preferably a copper wave plate) in the space inside the device. The press-fit fin plate is held in place by compression or tight fit in the free volume inside the device or subassembly. Although a small number of tack welds are possible, a press fit finned plate is not brazed or welded at all contact points in the installation position. Preferably, the press fit finned plate is held in place without the use of a binder or any other chemical bonding method.

Open space refers to the space inside the reactor in which there are no connected internal supports that resist tension. The open space may contain ribs or other structures that constitute compressive supports, but these structures are not fixed on both sides of the contact surface and therefore do not provide tensile resistance. The "open space" may be part of a larger space, however, in preferred embodiments, the layered device is welded or otherwise connected only at the periphery.

Free space is the space inside the device that is open for access to gaseous N ₂ entering one or more inlets of the device. The amount of free volume can be measured by evacuating gases from space for at least 10 seconds and supplying N ₂ gas to the space with measuring the volume of N ₂ gas filling the space.

Inner linear seam - welds that connect together two or more layers within the perimeter of the device with external welding around the perimeter.

In accordance with standard patent terminology, “comprising” means “including,” and none of these terms excludes the presence of additional or plural components. For example, when the device contains a layer, sheet, etc., it should be understood that the device of the invention may include a number of layers, sheets, etc. In alternative embodiments, the term “comprising” may be replaced by more stringent phrases “comprising essentially” or “consisting of”.

Brief Description of the Drawings

Figure 1 illustrates some of the elements from which the assembly of the technological layer is possible.

Figure 2 shows the stacking of components to form subnodes and stacked subnodes. Each illustrated process layer contains 3 finned inserts.

Figure 3 shows a layered stack in compression and a view of the angle of the device.

4A illustrates two plates forming a stress relieving connection.

Figv illustrates the nodes of the expansion joints welded to the upper and lower parts of the reactor chamber.

Figure 5 illustrates the "crowns" mounted on the outer part of the layered device.

Figure 6 shows the alignment of the surface of the coolant with the nodes of the channels of the coolant and chamfers of the end plate.

Figure 7 is a photograph illustrating a fillet weld added (illustration to the right) to create a more uniform platform for the catalyst holding assembly. The irregular shape of the fin plate shown on the right side of FIG. 7 is due to compression.

Figure 8 shows the process manifolds for the reactor of the invention. The figure shows the collectors (top and bottom) and the chamber of the layered reactor (in the center).

Figure 9 shows the collectors of the coolant for the device of the invention.

Figure 10 shows the transition to partial boiling and steady performance for an all-welded reactor.

Figure 11 shows a graph of thermal runaway in an all-welded reactor of the invention at a contact time of 70 ms.

12 illustrates a preliminary bend made parallel to the coolant channels to reduce curvature.

13 is a sectional view of a finned plate with a press fit adjacent to ridges formed during laser welding. The finned plate contacts the ridges, and a small open crack is observed between the finned plate and the heat transfer wall.

Fig. 14 shows fusion lines of laser seams that connect the top of the ribs between parallel and adjacent coolant channels formed in the bottom plate.

Figure 15 shows a chamber of a layered reactor (left) and a reactor with external supports (exoskeleton).

Fig.16 illustrates a node formed of 4 subnodes that were connected by spot welding.

Figure 17 illustrates the device of example 8, including external supports. The dimensions of the device’s chamber were approximately 0.6 × 0.6 × 0.08 m.

Figure 18 shows a graph of the pressure cycle used for hydrostatic testing of the process circuit of the device of example 8.

Figure 19 shows a graph of the pressure cycle used for hydrostatic testing of the coolant circuit of the device of example 8.

Figure 20 shows a graph of the temperature of the catalyst along the center line of the base layer. The height of the finned plate is 0.563 cm (0.225 ″).

Figure 21: Example 10: Catalyst temperature along the center line of the packed bed. The height of the finned plate is 0.5 ″.

Figure 22: Example 10: Catalyst temperature along the center line of the packed bed. The height of the finned plate is 2.5 cm (1.0 ″).

DESCRIPTION OF THE INVENTION

The described invention provides methods for manufacturing devices and devices that can be manufactured in accordance with these methods. The invention further includes methods for performing technological operations in devices. Technological operations may include chemical reactions, phase change, mixing, heat transfer and separation. The devices may be microchannel, or devices with larger characteristic sizes may be used. The characteristic microchannel size is defined as 10 mm or less in the range of 0.001 mm to 10 mm, in the preferred range of 0.01 mm to 2 mm, and in some embodiments, 0.1 to 2 mm.

In some embodiments, the method includes the first step of forming a subnode of at least two sheets — a bottom sheet and a top sheet (the bottom sheet may comprise etched channels); alternatively, a sheet with through channels may be located between the upper and lower sheets. In some embodiments, a subnode may be made up of three or more sheets. Small leaks are possible, but the fluid circulating through the first assembly essentially remains inside the first subassembly. In a second step, the first subassembly is laid adjacent to the process layer, and the subassembly and the process layer are connected to a press fit that provides thermal contact to form a assembly that includes at least two or more fluid channels.

One way to connect a subnode is to create a short stack of layers that includes two or more layers to create fluid channels. In alternative embodiments, more than two layers may be connected to form a system of parallel fluid channels or a subassembly that provides fluid channels for two or more fluids.

As an example of one embodiment, for forming the first assembly, a gasket or layer containing preformed channels (channels can be etched) is connected to the upper plate. The edges of the subassembly are substantially hermetically sealed along the edges to prevent leakage of fluid through the lateral edges and to maintain continuity of the bore so that more than 95% and preferably 99%, and more preferably 99.9% of the fluid that comes from the first inlet leaves the subnode through the first outlet, and does not flow out through the side edges or along other paths not provided for flow direction. In alternative embodiments, more than one inlet and / or outlet is possible, as determined by the layered geometry of the article.

Layered products are also preferably sealed along the upper or lower surface of the subassembly along the length of the corridors of a continuous metal or material; in some preferred embodiments, at least 50% of the corridors have continuous seals in the length direction on at least the upper and / or lower surface; seals are generally located along the walls of the channels that separate the channels. The connection can only be carried out in areas where, when laying the subassembly, metal contacts between the layers. It should be understood that areas containing a flow channel or a free volume for circulating fluids after manufacturing the device should not be closed. The seal is formed by the contact of two materials. It should also be understood that the connection of the subassembly along one or more surfaces of the subassembly can be continuous along the flow path or intermittent in accordance with the operational requirements for the design of the device. The fluid may flow or flow from one parallel channel to the next inside the first subassembly when testing the functionality of the device or when performing quality control checks before use. This small cross-flow volume is less than 20% of the flow per channel, and more preferably less than 10%, and even more preferably 25 or less; these percentages can be based on average cross-flow through all channels or cross-flow from any selected channel.

A subnode connection includes at least two layers, but may include three or more layers. In one embodiment, twenty or more layers are connected to a subnode. Methods for connecting the first subassembly include, but are not limited to, laser welding, resistance resistance welding, friction rotational welding, ultrasonic welding, diffusion welding, soldering, diffusion soldering or soldering with a transparent liquid phase, adhesive bonding, reactive bonding, mechanical bonding, etc. . The use of laser welding is the preferred embodiment when using laser welding of a certain type (including fiber lasers and ytterbium fiber lasers, in particular due to their low power consumption, which limits the level of metal deformation after joining).

The method of connecting the surface of the subassembly may be similar or different from the method of sealing the edges of the subassembly. In one embodiment, a fiber laser is used to densify along the perimeter, and in another embodiment, a pulsed laser is used. Other methods of welding or joining can also be used to seal the perimeter (with the exception of the areas of entry or exit from the layer of flow channels).

The connected or sealed subassembly is preferably subjected to a quality control (quality control) before being placed in the assembly. Evaluation can be performed for each subnode, quality control can be subjected to statistical sampling of subnodes, or quality assessment can be performed for a random selection of subnodes. Quality control may include a pressure test to check for leaks, a flow test to check the pressure drop, or a dye test to check the distribution of residence time, based on which a flow can be inferred between the proposed sealed internal parallel channels.

The connected or sealed sub-assemblies can then be connected to a node by alternating or repeating the connected sub-assemblies with a second sub-unit or a second system of fluid channels to create a device with two or more sets of fluid channels.

The fluid channel may include a wave or ribbed structure or an alternative structure suitable for chemical treatment, such as foam, felt, packing, cellular structures such as airgel, honeycombs, and the like. In some preferred embodiments, the implementation of a wave-like or finned structure creates channels or chambers having an aspect ratio (height to width) greater than one, where the height is the distance between two subnodes and the width is the distance between repeating finned plates or adjacent shoulders (wave-like surfaces) wavy plate. In an alternative embodiment, the second fluid channel may comprise any heat-conducting structure.

In a preferred embodiment, the second fluid channel includes process channels, and the first subassembly contains heat transfer channels, although in some embodiments this functional relationship may be inverse.

An example of elements that can be used to construct a fluid channel is shown in FIG. 1, where a wave plate is created from a flat foil. The outer portion of the fluid channel is sealed using butt planks (also defined as peripheral planks, or p-planks) or side planks, and may also include the use of support strips (o-planks).

The first subnode is placed or laid between the layer of the second fluid channel (shown as a wave-like layer). This is shown in FIG. One continuous second fluid channel is possible, or a series of continuous fluid channels (three are shown in FIG. 2) laid on each layer for the second fluid channel.

Technological wave plate can be connected to the first subassembly using welding on the peaks of the bulges or thermal bonding with binders or other materials that increase the thermal conductivity of the contact between the first subassembly and the second fluid channel. In one embodiment, the two layers are pressed into each other during installation and are welded without adding additional material to increase the thermal connection (this embodiment without soldering or welds for joining parts is called "press fit"). In another embodiment, additional material is added to reduce resistance between the first subassembly and the second fluid channel. In another embodiment, the thermal contact during chemical treatment is enhanced by the use of a multi-stage process in which a small clearance or gaps between the press fit technological structure and subassemblies are filled with liquid during processing under the action of capillary forces. The fluid may preferably fill the gaps and increase the conductivity of the composite structure during operation of a chemical plant.

After laying the combined stack, which includes the first subnode and the second fluid channel, the device of the invention is connected to form a stack using a method such as, but not limited to, external welding, binders and reactive bonding. Various methods of welding can be used to weld the stack, including, but not limited to, gas-tungsten welding, inert gas consumable electrode welding, laser welding, and electron beam welding. Also, soft soldering can be used to connect the perimeter if the operating temperature and pressure of the chemical reactor are favorable enough for the use of soft solder.

Before connecting the final node, the stack can be compressed to provide contact between the layers and reduce the cavities between the layers where the final connection of the device takes place. Compression can be performed, for example, using a clamping device applying a load by means of a bolted connection, or using an external press to apply a load to the stack. A press-fit wave plate may deform under compression and remain in a deformed state after compression has ceased.

Before laying, alignment of nodes may be required. One leveling method involves leveling by rolling a laser welded subassembly on a straightening machine. This method reduced deformation when using plates with dimensions of 15 × 60 cm (6 ″ × 24 ″). These panels had deformation in one direction - along the length of the fusion lines of the welds. Alignment by rolling gave less successful results for a subassembly with dimensions of 60 × 60 cm (24 ″ × 24 ″), when the part had deformations in two directions (bowl shape or three-dimensional parabolic shape). A standard leveling machine was used to align the deformed part, but this led to the destruction of laser welds. A non-standard soft-action manual roller was used for less aggressive bending of the deformed part to a more even state. A non-standard hand roller provided some reduction in deformation, but did not restore a substantially even state of the part, where a substantially even state is determined by a lag of no more than 1 cm behind any corner of the part laid on a flat surface. Thus, soft alignment can provide more high-quality device, in particular, for subnodes with different widths and lengths (i.e. non-square subnodes). In some preferred embodiments, alignment is performed for subnodes having a width of about 15 cm or less; in some embodiments, a width of from about 10 to 20 cm.

V-grooves are an advantage between sub-assemblies, since fillet welds can be used to fill V-grooves. A subnode may protrude slightly beyond a portion of the side slats or butt planks. In an alternative embodiment, the subassembly is installed substantially flush with the butt plate. Essentially flush means within 5 or less of the thickness of the subnode. As an example, if the thickness of the subassembly was 0.025 cm (0.01 ″), then the protrusion or recess of the edge of the subassembly would not be more than 0.125 cm (0.05 ″) from the edge of the butt strip. For a subassembly with a thickness of 0.15 cm (0.06 ″), the offset from the flush location is not more than 0.75 cm (0.3 ″) with a preferred offset of not more than 0.15 cm (0.06 ″) from the flush location, e.g. ., a protrusion of 0.15 cm (0.06 ″) or a recess of 0.15 cm (0.06 ″) as a preliminary embodiment.

A key advantage of this combined manufacturing method is to reduce the requirements for surface preparation for diffusion welding and / or soldering. For good diffusion welding and / or soldering, surfaces must be very clean, even and with tight tolerances for a precise fit. Eliminating the stage of diffusion soldering and / or welding also eliminates the need for high temperatures to affect the large device that is required for diffusion welding and / or soldering. The energy required for heating and cooling a large device is significant, as is the time required for heating and cooling a large device during welding or soldering to eliminate excessive temperature stresses and corresponding deformation. For a device made of stainless steel, consisting of predominantly even inner layers, the internal temperature gradient from the farthest corner from the center to the center point should be less than 30 ° C at temperatures above 500 ° C to prevent mechanical deformation of the layer. For a device with a cross-section of more than 0.5 × 0.5 m, it may take several days to heat up and several days to cool the device during soldering or welding during vacuum heat treatment. The necessary time for processing and preparing the surface of the parts increases the total cost of the reactor.

The inventive method for manufacturing a device eliminates the need for a diffusion welding and / or soldering step of the reactor. The method of the invention can provide the production of higher quality reactors at a reduced cost and in less time.

An unexpected result of the press-fit device of the invention is the effect of contact resistance between the layers. Press fit of the layers does not provide tight thermal contact, which is further aggravated as the size of the device increases and imperfect alignment of the initial part. Heat is transferred between the first subassembly and the second fluid channel through a low quality contact portion separating the fluid flows. In a device using diffusion welding or soldering, each layer is in close thermal contact, which is due to the nature of diffusion welding and / or soldering, while local surface roughness and / or roughness of the part or initial deformation reduce the heat transfer efficiency between the layers.

The significance of the thermal contact between the layers depends on the operational requirements of the process for the reactor or device. In some embodiments, the internal cavities between the two layers are filled during operation with a process fluid. In another embodiment, a thermally bonding material, such as a binder, putty, liquid, or a deformable solid, such as graphite or a similar intermediate layer, may be introduced into the press fit layer to reduce contact resistance between the two fluid layers (at least one of which is a press fit layer).

In some embodiments, an intermediate thermal contact layer is not required. In the reactor of the invention, a test was performed for the Fischer-Tropsch reaction without using an intermediate layer to increase thermal contact between the technological side of the wave-like plate and the coolant side of the laser welded subassembly. The performance indicators essentially corresponded to those measured for a solid brazed reactor of a similar design.

It is also contemplated that reactions or process steps involving hydrogen and / or liquid, including but not limited to hydrogenation, hydrocracking, or hydrotreatment reactions, do not require the use of a heat-conducting layer between the first subassembly and the second fluid channel. These fluids have a good coefficient of thermal conductivity, and when these cavities are filled with cavities, good thermal conductivity can be ensured. The fluids will provide sufficient capillary tension to penetrate into the cavity between the finned plate and the adjacent heat transfer surface. In addition, it should be noted that the surface tension of the oils is substantially lower on copper than on stainless steel, which further contributes to the capillary tension of the oil or wax in the case of the Fischer-Tropsch reactor for penetration into the cavity between copper and stainless steel (or another metal) in the course of the reaction. The increased surface tension of the oil or liquefied wax on stainless steel may not have the same effect or may require significantly smaller clearances for liquid to enter the cavity. It is also envisaged that copper finned plates will be less susceptible to manufacturing deviations.

It is contemplated that reactions involving oxidation reactions may require the use of an intermediate heat-conducting material. In one embodiment, a porous oxide coated catalyst or other retained fluid in a second fluid channel can also serve as an intermediate heat conducting material when filling cavities between two layers to provide heat transfer and control the reaction temperature.

Stress relief connection

A stress relieving joint made of two plates welded together can be added to the stack to reduce the stress present in the welds connecting adjacent layers in the chamber. The connection is designed to open (expand) as pressure is increased during operation. In this case, the sealing welds located on the outer surfaces of the reactor remain unstressed, which increases the service life of the device.

The expansion joint is made of two metal plates, usually of the same width and length with a stack. For example, in the device presented elsewhere in the present description, the plates have a width of ~ 60 cm (24 ″) and a length of 60 cm (24 ″), corresponding to the sizes of the other plates in the stack. Preferably, the bottom plate is thinner than the top plate in the stress relieving joint; for example, a base plate (i.e., a plate coplanar and in contact with the stack) may have a thickness of ~ 0.625 cm (0.25 ″), and the upper plate (the plate closer to the outer main surface) may have a thickness of 0.1 cm (0, 04 ″). In some preferred embodiments, the implementation of the base plate contains a number of holes, and the upper plate does not have separate features. The plates are arranged on top of each other and aligned along the edges. Next, the plates are welded using laser welding. The edges of the expansion joint plates do not have continuous welds; this ensures the movement of the plates during operation of the device. Preferably, the contour of the welds is designed so that the edges of the plates are connected only at the corners. This ensures separation of the edges during operation to compensate for any expansion of the reactor if necessary, without stressing the sealing welds on the reactor, and isolating the various internal seams from each other. The holes in the base plate provide an individual leak check of the welds before assembly into the reactor. After successfully passed qualification tests for expansion joints, the holes in the base plate can be filled using the standard gas tungsten welding process.

Preferably, two expansion joint assemblies are used in the finished reactor chamber: one at the top of the chamber, and one at the bottom. Preferably, one stress relieving plate is located opposite the reactor chamber and welded to the coolant subassembly along the entire circumference of the perimeter.

During operation, the reactor is under pressure. Pressure leads to elastic stretching of the outer supports. In the absence of stress relieving joints, this would lead to a corresponding extension of the reactor chamber itself and the development of stress-induced deformations in the sealing welds. In the presence of stress relieving joints, elastic tension causes the joints to open, reducing deformations in the welds.

An additional step of welding the crown (which is made of a solid hollow square or a metal ring of rectangular cross section or of two or more parts welded together with the formation of a solid ring that protrudes above the surface of the chamber). The crown creates an auxiliary design for the device and the final operational collectors (macro collectors) in such a way that during repair, the connection between the macro collector and the device can be cut and re-welded or connected for a subsequent period of operation. This use of the crown is particularly effective as a means for removing or reducing the catalyst contained in the reactor chamber. Wreaths, devices containing wreaths, methods for manufacturing devices containing wreaths, and methods of using devices containing wreaths are additional original aspects of the invention.

Examples:

Example 1. Welded subassembly - welded reactor - conductive finned plate with press fit

A welded reactor was manufactured and put into operation to confirm the same performance with a brazed reactor using the Fischer-Tropsch reaction as a test. The operation of the reactor was carried out continuously for more than 2000 hours and demonstrated that the contact of the technological plate containing the catalyst and the subassembly with the press fit proved to be sufficient for the effective operation of the reactor and ensured the performance of the soldered reactor of the same design.

Device description

The Fischer-Tropsch two-layer all-welded device was designed and manufactured to confirm the production process of the invention. The design of a multichannel microreactor includes two technological repeating units, alternating with three repeating coolant units. The coolant channels are oriented across the flow of technological channels.

Technological channels are made of wave-shaped copper plates with a length of 15.75 cm (6.2 inches), a width of 7.62 cm (3 inches) and a height of 0.635 cm (0.256 inches). The thickness of the wave plate is 0.015 cm (0.006 inches). The final device contains 274 technological channels in two layers. Each of these channels has the following average dimensions: width - 0.095 cm (0.0375 inches), height - 0.635 cm (0.25 inches) and length - 7.62 cm (3 inches). The finned plate exceeds 0.015 cm (0.006 m) of the nominal size of adjacent butt plates to provide excellent thermal contact.

The coolant channels of the device, including laser-welded sub-assemblies, are connected to the top plate by an ytterbium fiber laser (IPG production model YLR-600-SM: 600 Watt ytterbium fiber laser, 1.07 micron wavelength) with a weld thickness of 50 to 150 microns penetrating through the upper plate with a thickness of 500 microns into the lower channel plate with a thickness of 1000 microns, not melting it throughout the thickness. The subassemblies are made of two gaskets, which include an upper or cladding sheet or walls connected to a channel gasket that contains flow channels for heat transfer fluid. These coolant subassemblies are laid in a device with dimensions L × W × H-8 × 25 × 7 cm (~ 3 ″ × ~ 10 ″ × ~ 2.7 ″) and sealed around the perimeter using seams obtained by fusion welding and corner welded seams.

After welding the components of the chamber, the device was cleaned and loaded with 66.5 grams of highly active cobalt catalyst obtained from Oxford Catalysts, Limited (Oxford Catalists, Limited) and 120 grit silicon carbide supplied by Atlantic Equipment Engineers (Atlantic Equipments) Engineering).

The last stages of production included welding of the rear and front receivers (for the coolant and technological channels for external connections at the macro scale, i.e., large pipelines) and welding of supports. Welding of the supports was a structural need for this design, which also ensures the safe operation of the device without the need for a containment system under pressure.

Production of camera components

The all-welded reactor has two technological layers, each of which contains a copper wave-like plate and two stainless steel butt plates. Three coolant layers alternate with the technological layers and are made as subnodes using laser welding of the upper plate or stainless steel wall with a stainless steel duct gasket. The wall gaskets were cut to size from 0.05 cm (0.020 ″) thick stainless steel sheet. Heat carrier gaskets have non-linear and rectilinear channels, which are made using partial photochemical processing (PFC) in a stainless steel sheet 0.1 cm (0.040 ″) thick to a depth of 0.05 cm (0.020 ″) and 0.25 cm wide ( 0.100 ″) with ribs 0.1 cm (0.040 ″) in size for straight channels and to a depth of 0.0375 cm (0.015 ″) and a width of 0.0425 cm (0.017 ″) for channels in a non-straight section. The indirect section was made with a wavy or serpentine profile with 22 bends. Laser welding was performed for each rib along the entire length of the strip to create a seal between the channels, as well as between the channels farthest from the center and the perimeter to seal the channels from the outside. Further, the subnode was checked for leaks. The ability to check the subassemblies for leaks ensured the establishment and repair of leaks before assembling the device and avoided the production of defective devices. Another advantage of the preliminary assembly of coolant layers in the form of sub-assemblies is the reduction in the number of parts for laying in the assembly by at least about 20%.

The butt planks on the technological side were made of material of standard thickness (nominal thickness 0.625 cm (0.250 ″)) with minimal requirements for machining, which consisted only in cutting along the length and width and chamfering the edge. The wave-shaped copper plates were made in a standard molding process for finned plates in which thin coils bend at regular intervals to form a repeating finned structure. The finned fins were 0.64 cm (0.256 ″) high. The end plates required minimal machining with trimming along the length, width and chamfering. All parts do not require tight thickness tolerances associated with soldered devices, and therefore blanks can be used.

Assembly of the device for example 1

The device’s chamber (i.e., technological layers and coolant layers enclosed between two end plates) is created by stacking components with alternating technological layers and coolant layers. The number of technological layers is determined by the required yield of the FT reaction product, while the number of coolant layers per unit exceeds the number of technological layers so that each technological layer contacts both sides of the coolant. During the installation process, an installation device is necessary to align the parts, as well as maintain alignment during the assembly process and in the initial stages of welding. A clamping device was created, which is a platform for laying, as well as securing the stacked chamber for transportation to the welding stage. The clamping device consists of two plates made in the form of an elongated plus sign. Each contains four (4) grooves for installing threaded rods over the entire length with a diameter of 1.25 cm (1/2 ″). A base plate was located under the lower clamp to provide space for the threaded end and nut on the underside. To align the technological surfaces, four straight edges were fixed in place on each side of the clamping device using C-shaped clamps. The fifth straight edge was used to align one of the two surfaces of the coolant.

After installing and aligning the fixture, the first step is to place the end plate on the fixture with its centering between the straight edges of the technological surface. Next, the installation of a straight edge of the surface of the coolant before laying the first layer. The first laid layer was a coolant subassembly. Technological layers are located between two layers of the heat carrier. The subnode drops onto the end plate, moves into place with respect to the straight edge of the coolant surface and is centered between the chamfers on the end plate (figure 6). With satisfactory alignment, the first technological layer was placed on top of the coolant subassembly. At this stage, the butt edges were flush with the surface of the coolant. For fusion welding, this seals the contact surface between the joint strips and coolant sub-assemblies and the joint strips and end plates. The first end plate is flush with the butt plate of the coolant, followed by the location of the wave-like plate and then the second butt plate. The wave-like plate is installed in tight contact with the first butt strip and is centered on the coolant subassembly. The second butt strip is installed in tight contact with the wave-like plate, and its alignment is checked on another surface of the coolant. If the butt planks on both surfaces of the coolant are aligned within +/- 0.025 cm (0.010 ″) with the adjacent layer, alignment is acceptable. This process of laying the sub-nodes of the coolant and technological layers is repeated again with the laying of the next sub-node of the coolant. The resulting stack consisted of two (2) process layers and three (3) coolant layers. The last installable component of the chamber was the upper end plate. The upper end plate is installed flush with all four straight edges of the technological side and the straight edge of the coolant surface.

Welding of the chamber was carried out in two stages. At the first stage of welding, fillet welds were made with a width of 5 cm (2 ″) along the length of the butt planks. On each technological surface, six V-grooves were made, which were filled with fillet welds. Plots of curved cuts on the fixture provided access to these V-grooves to perform this welding step. The original fillet welds were designed to fix the end of the butt plate closest to the wave plates to prevent damage. Later fillet welds were added to the inner edge of the butt planks in accordance with the catalyst retention unit. The illustrations before and after performing these fillet welds are shown in Figure 7. After completing the fillet welds, the fixture was removed to provide access to the coolant surfaces for the next welding step. On each surface of the coolant there were three coolant sub-assemblies and, accordingly, six welds (one seam above each sub-assembly and under it), each of which represented a seam obtained by fusion welding, along the entire length of the surface. After that, the chamber was ready for cleaning and preparation for loading the catalyst.

Catalyst Cleaning and Loading

Before loading the catalyst, the technological side of the device was cleaned and the catalyst retention unit was welded into place.

Before loading, the catalyst is put in place and a catalyst retention unit is welded on one technological surface to hold the catalyst when the chamber is oriented in a vertical position. The catalyst retention unit consists of four parts: sieve, sieve fastener ring, foam and foam fastener ring. The sieve performs the function of retaining the catalyst in the device. The sieve fixing ring is a thin stainless steel frame that holds the sieve in place, ensuring close contact with the process surface. A small weld around the perimeter of the sieve fixing ring secures the sieve in place and ensures good catalyst retention in the device. Also, for better catalyst retention, the corner welds on the butt plates were raised to the inner edge of the butt planks. To protect the copper wave plates during welding along the perimeter of the mounting ring of the sieve, an aluminum protective partition can be used.

After installing the catalyst retention unit on one technological surface, the catalyst was loaded. The loading was a four-stage process. 4 end channels (1 at each end of each technological layer) have a partial finning and were blocked by a fixing ring of sieves, thus passing into the category of inoperative channels. These channels were completely filled with silicon carbide (SiC), an inert material with a particle size approximately corresponding to the catalyst, in the first loading step. The remaining channels were loaded with 3 layers: 1.6625 cm (~ 0.665 ″) SiC into the process inlet, 3.75 cm (~ 1.5 ″) with a catalyst bed and 1.875 cm (~ 0.75 ″) SiC into the process outlet. To ensure the required depth of each of these three layers, materials were loaded into the device in small portions, followed by tapping the side of the device (end plates) with a rubber hammer to seal the loaded material. After loading and compaction of each portion using measuring tips, depth measurements of all channels were performed. After completion of this process, for any given layer, the material was subjected to further compaction using ultrasound to approximate as much as possible to the SCPS (measured by external means of the average density of the compacted material layer). The loading was completed when, after compaction, the top layer remained flush with the butt planks. When the catalyst ceased to be densified by ultrasound, the catalyst bed was accepted as fully densified. After loading all three layers (two SiC layers, one catalyst layer), a catalyst retention unit was installed and welded on the open technological surface of the device using the same procedure that was previously described for a different technological surface.

Final welding for example 1

The manufacturing of the device was completed by the three final stages of welding, namely, the fastening of technological collectors, heat-transfer manifolds and base plates. Unlike soldered devices, the only internal components for an all-welded FT device with a through seal are coolant sub-assemblies. The remaining components (wave-shaped plates, butt plates and end plates) are connected to adjacent components only along the perimeter. Support plates provide the necessary structural support to maintain the integrity of the device under the action of high pressure drops during operation. The base plates also serve to replace the pressure protective sheath (ZOD) used for soldered devices. For the coolant, two outlet openings are used, which are designed for separate removal of water vapor and water in the liquid phase.

The process manifolds are made of 304L stainless steel and have the following approximate dimensions: length - 22.75 cm (9.1 ″), width - 6.75 cm (2.7 ″), depth - 4.75 cm (1.9 ″) ) An inner pocket with approximate dimensions of 20.25 × 4.25 × 3.00 cm (8.1 ″ × 1.7 ″ × 1.2 ″) is hermetically connected to the technological hole of the wave plate and provides support for the catalyst retention mechanism. The collectors are welded around the perimeter of the chamber using the traditional gas tungsten welding process. Pipes of one inch in diameter extend from the central part of each of the technological collectors, providing entry and exit from the process gas chamber. Manifolds are designed with a minimum wall thickness of approximately 1.25 cm (0.5 ″) to support process pressure during operation. See Fig. 8.

The coolant intake manifold is made of 304L stainless steel and has the following approximate dimensions: length - 14 cm (5.6 ″), width - 6.75 cm (2.7 ″), depth - 4.25 cm (1.7 ″) . An inner pocket with an approximate size of 12 × 4.75 × 3.25 cm (4.8 ″ × 1.9 ″ × 1.3 ″) is hermetically connected to the coolant inlet channels and is designed to evenly distribute the coolant on the coolant inlet surface. The collector is welded around the perimeter using the traditional gas tungsten welding process. Pipes with a diameter of one inch extend from the collector, providing entry into the coolant into the chamber. The manifold is designed with a minimum wall thickness of approximately 0.95 cm (0.38 ″) to support fluid pressure during operation.

The coolant exhaust manifold is made of 304L stainless steel and has the following approximate dimensions: length - 14 cm (5.6 ″), width - 6.75 cm (2.7 ″), depth - 11 cm (4.4 ″). An inner pocket with approximate dimensions of 12 × 4.75 × 10 cm (4.8 ″ × 1.9 ″ × 4.0 ″) is hermetically connected to the outlet channels of the coolant and is designed to provide unhindered exit of the coolant flow from the reactor chamber. The collector is welded around the perimeter using the traditional gas tungsten welding process. Pipes with a diameter of 2.5 cm (1 ″) extend from opposite sides of the collector. The upper pipe provides steam output, and the lower pipe provides water output in the liquid phase. The manifold is designed with a minimum wall thickness of approximately 0.95 cm (0.38 ″) to support fluid pressure during operation.

Two coolant collectors are welded directly to the all-metal end plates of the chamber and process collectors. In this case, the welds of the chamber are completely sealed in accordance with the parameters of the process and the coolant collectors and in this regard do not go directly to any outer surface of the reactor. See FIG. 9.

Next, supports (exoskeleton) are added using the traditional gas tungsten welding process. Only four sets of supports surround the reactor in a vertical direction (process flow) and one additional set is located in a horizontal direction above the exhaust manifold of the coolant. Each set of vertical supports is made of two identical parts made of stainless steel 304L, which have the following approximate dimensions: length - 22 cm (8.8 ″), height - 8.25 cm (3.3 ″), thickness - 0.625 cm (0 , 25 ″). Each set of supports is welded at the contact points of their ends, as well as using intermittent seam welding around the perimeter. Four sets of supports are spaced about 5 cm (2 ″) apart, and the first set is approximately 6.75 cm (2.7 ″) from the edge of the coolant intake manifold. While four vertical sets of supports provide support for the reactor chamber and process manifolds, a horizontal set provides additional support for a large-sized coolant exhaust manifold. The two horizontal supports have the following approximate dimensions: length - 13.5 cm (5.4 ″), height - 60 cm (2 ″), thickness - 0.625 cm (0.25 ″). They are located in the center on each side of the coolant exhaust manifold and are welded to this collector, as well as to the end plates of the chamber and the vertical supports furthest from the center.

Experimental setup for example 1

Technological side

The flow and composition of the synthetic gas (syngas) supplied to the fixed-bed microchannel reactor for Fischer-Tropsch synthesis was controlled by adjusting the flow rates of individual gases (carbon monoxide, hydrogen and nitrogen) using Brookes mass flow controllers. Gases were supplied through an activated carbon layer and a 13X molecular sieve to eliminate any impurities. The feed gases were preheated in a stainless steel microchannel heat exchanger before entering the reactor. A 3,000-watt Watlow tape heater was installed on the reactor and insulated to minimize heat loss. Performance data were measured using pressure transducers and type K thermocouples in a 316SS stainless steel jacket.

The product flow was directed through three collectors at elevated pressure with gradual cooling to ensure coarse separation of light and heavy hydrocarbon products and the aqueous phase. Most products were collected in the first tank of the product (with a temperature of ~ 100 ° C) and the second tank (with an ambient temperature of ~ 25 ° C). The aqueous phase and heavy hydrocarbon products (wax) entered the 1st tank, while the aqueous phase and the light liquid hydrocarbon phase entered the 2nd tank. Exhaust gas was removed from the 3rd tank. Samples of the produced gas were taken through a sampling channel located along the flow from the reactor and against the flow from the first reservoir for collecting products, and analyzed using an Agilent M200H gas column microchromatograph with two columns, 5A molecular sieve and PlotQ program.

Coolant side

A 76-liter (20 gal.) Carbon steel tank was used to store cooling water. The chemical composition of the water was maintained by the addition of Cortrol OS5300 and Optisperse AP302 reagents. Nitrogen was injected into the reservoir under pressure to maintain pressure in the steam circuit. A high pressure pump manufactured by Cat Pumps (model 231.3000) was used to supply water to the coolant circuit. An Appleton FLSC-62A flowmeter was used to measure and control the flow rate of the coolant. Before entering the reactor, the feed water passed through a 25-cm block of 5-micron filters for particle capture and a 60-micron Swagelok filter. The steam outlet of the reactor is connected to a nitrogen source to control the coolant pressure, while water (separated in the rear coolant collector) flows into the 2-liter stainless steel Swagelok tank, which was used to maintain the required water level in the system.

Operational data

The all-welded reactor of the invention was operated without a containment under pressure for ~ 2150 hours. The reactor had an exoskeleton welded to the outside to resist pressure when carrying out this reaction under high pressure.

The reactor was operated with a gradual tightening of operating conditions to the point of thermal runaway (about 70 milliseconds of contact time on the technological side). After thermal runaway, the reactor was regenerated to assess the degree of damage to the catalyst. After regeneration, the catalyst recovered about 50% of its original activity.

Test run and evaluation

The reactor was launched as follows: after the activation of the catalyst was completed, the reactor was cooled to ambient temperature and subjected to a pressure of 2413 kPa (350 psig). Cooling water was supplied to the coolant circuit at the target flow rate, and the reactor was heated to a syngas injection temperature of ~ 170 ° C. Then, a stepwise supply of syngas and heating of the reactor to the target operating temperature were started.

Upon completion of the launch, the reactor entered the following mode: H ₂ : СО = 2.0, p = 2413 kPa (350 psi), diluent ~ 16.8%, contact time ~ 290 ms. On the outer surface of the reactor, two rows of thermocouples were welded by tack welding at a distance of approximately 1.17 cm (0.46 in) and 3.2 cm (1.26 in) from the beginning of the catalyst layer (3.20 cm and 5.23 cm from reactor inlet). The all-welded microchannel reactor of the invention did not provide fully isothermal operation of the reactor; however, the measured temperature gradients were less than about 5 ° C. In addition, the internal gradients of the FT catalyst have not been measured and are expected to be higher than the measured temperature gradients in the walls of the reactor. The temperature profile of the surface of the reactor was controlled within ± 2 ° C of the average temperature.

A direct comparison of the parameters of the (all-welded) reactor of the invention with a brazed reactor is presented below in table 1 for one Fischer-Tropsch catalyst:

Table 1 Comparison of the (all-welded) reactor of the invention with a brazed reactor Reactor Soldered reactor Reactor of the invention H ₂ : CO in the feed syngas 2.0 2.0 Syngas dilution 16.5% 16.8% Pressure 25 bar 25 bar Contact time, milliseconds 290 ms 289 ms Temperature* 210 ° C 206.7 ° C Continuous operation time 231 h 209 h Operational parameters CO conversion 71.8% 74.0% Selectivity for CH ₄ 8.9% 8.7% Selectivity for CO ₂ 0.4% 0.3% Selectivity to C2 0.8% 0.8% Selectivity for C3 2.3% 2.3% Alpha 0.91 0.91 * Note: the position of the temperature measurement is slightly different for the two reactors. The temperature for the brazed reactor is measured on the surface of the coolant gasket, while for the (all-welded) reactor of the invention, it is measured on the outer surface of the reactor wall.

The continuous operation time of the all-welded reactor of the invention is comparable with other soldered and single-channel reactors under the same conditions. The wax product collected during this test period was analyzed to determine the distribution of hydrocarbon products. The results showed excellent consistency with wax from earlier tests under similar conditions using short and long single-channel reactors and pilot tests of soldered reactors.

Resistance to violations of technological parameters

After approximately 211 hours of continuous use, a failure of the flow meter has led to blockage. The failure of the flow meter triggered a zero flow alarm (despite the normal functioning of the pump), which caused the backup pump to start, which entailed a significant increase in the flow rate of the coolant, which caused a blockage. The system was restarted within 5 minutes. During this period, the reactor cooled to ~ 197 ° C. The consumption of CO and H ₂ was immediately restored after a system reset. Consumption N ₂ was restored after 2 minutes. The increase in the temperature of the reactor began immediately, and within 9 minutes (from the moment the system was reset), the maximum temperature recorded on the outer surface of the reactor reached ~ 240 ° C due to the absence of coolant flow (since the coolant channels were not pumped out, the water supply from the previous operation was on the coolant side and could start to evaporate). After reaching a temperature of 240 ° C (9 minutes after the system was reset), the heat transfer pumps were started manually. The temperature of the reactor began to drop to normal levels. After 28 minutes (37 minutes after the system was reset), the situation was under control and the reactor cooled to ~ 192 ° C. Further, the reactor temperature was gradually raised to a value before blocking (~ 206.6 ° C). After the reset, a higher value of H ₂ flow rate was established compared to the target, providing a change in the ratio of H ₂ : CO to 2.17 (instead of 2.00).

This experimental result was unexpected in that the thermal runaway did not occur seconds after the supply of the coolant failed and that 9 minutes elapsed between the loss of coolant and the increase in the temperature of the reactor by more than 40 ° C. The efficiency of the catalyst was returned to the expected levels after restarting the coolant and reaching the target temperature. A high ratio of the volume of the metal reactor block to the volume of the catalyst creates a thermostock, causing the absorption of exothermic heat of the reaction for several minutes until the violation of the system parameters. This provides a significant advantage over traditional FT fixed-bed tubular reactors in which temporary loss of coolant leads to thermal runaway and loss of catalyst efficiency. The FT reactor of the invention provides a temporary buffer for unwanted violations of thermal parameters, as shown by a return to the expected efficiency after 9 minutes without supplying a coolant. The temperature rise of the reactor was carried out as expected, but the high heat capacity of the metal structure prevented irreversible sintering of the catalyst.

For the all-welded experimental reactor, the catalyst volume was ~ 7% of the total reactor volume (63.1 ml of catalyst in the 0.934 L reactor block was 25 × 7.5 × 4.75 cm (10 ″ × 3 ″ × 1.9 ″)) . In this case, the ratio of the reactor volume to the catalyst volume of ~ 14: 1, an acceptable thermostat time without a coolant of 9 minutes was demonstrated.

For larger devices with a ratio of reactor volume to catalyst volume of less than 14: 1 and typically less than 10: 1, and more preferably less than 3: 1 or 2: 1, an acceptable time without supply of coolant should be less than 9 minutes and is expected to be within 5 minutes or less, in some embodiments, 30 seconds or less. In some preferred embodiments, the ratio of catalyst volume to reactor volume is from 2 to 60%; in some embodiments, reactors in accordance with the present invention are from 5 to 40%; where the total volume of the reactor includes the volume of channels, channel walls, built-in manifolds and external walls, but does not include the volume of external pipelines or protective shells under pressure.

In addition, after ~ 346 hours of continuous operation, the carbon monoxide mass flow controller (PMR) was replaced due to the low dry gas hours (RHF) for the effluent. The new RMR of CO was adjusted to a value lower than the target, which led to an increase in the ratio of H ₂ : CO. Over a subsequent period of 17 hours (up to ~ 363 hours of continuous operation), the ratio of H ₂ : CO was ~ 2.36, and the conversion of CO was increased to> 85%.

Surprisingly, we found that even with a high conversion of CO, the deactivation rate did not increase and the efficiency was restored to previous levels after adjusting the feed ratio. The reactor is stable in operation under changing conditions, including an increased ratio of H ₂ to CO. Conventional fixed-bed FT tube reactors do not provide a good response to rapid changes in heat transfer, including a sharp increase in the heat output from the reaction. The reactor of the invention continued to operate stably as the heat from the reaction associated with an increase in the ratio of H ₂ to CO increased. Further, the efficiency was returned to the expected parameters after the restoration of the target value of the syngas ratio.

Partial Boil Assessment

At the next stage of the demonstration of the reactor of the invention, the partial boiling of the coolant was tested and the results of temperature control and stability of the all-welded reactor of the invention were evaluated, in which the elements of the control of the boiling flow (22 wave-like bends of the small cross-section channel on the inlet side in front of the main coolant channels) are sealed only along the linear seam between parallel wave-like elements, and not along the contour of wave-like elements. One section of the wave-like element falls on the heat-transfer channel, and a simple linear seal between parallel coolant channels was sufficient to ensure stable operation of the controlled partial boiling reaction at high heat flux, for example, the flow did not go beyond the bends along the sections adjacent to the serpentine elements, which would then lead to a preliminary pressure drop and could potentially cause a destabilization of the coolant flow rate during boiling.

Operating conditions were maintained that corresponded to the initial conditions (H ₂ : СО = 2.0, p = 2413 kPa (350 psi), diluent ~ 16.8%, contact time ~ 290 ms). Starting from ~ 634 hours of continuous operation, the flow rate of the coolant was reduced from ~ 2 l / min in such a way that a steam output of ~ 1-3% boiling was ensured. The reactor temperature was also adjusted to maintain ~ 70% CO conversion. At ~ 679 hours of continuous operation, the flow rate of cooling water was reduced to 0.4 l / min and an output steam content of 1.5% was achieved. The temperature of the reactor was reduced to 204.3% to maintain the target conversion of CO. Efficiency is essentially the same as a single-phase operation. In addition, the use of partial boiling ensures the operation of a reactor with an increased net heat flow or heat production by a FT reactor, the operation of which is carried out with a contact time below the initial 290 ms.

The partial boiling conditions were maintained for ~ 300 hours, and the efficiency over this period is shown in figure 10.

The reactor of the invention steadily worked for> 250 hours of continuous operation with partial boiling of water and an output steam content of 1.5% under the initial condition of 290 ms, essentially corresponding to the efficiency of a single-phase coolant.

Performing a thermal stability assessment

At the next test stage, the ability of the reactor of the invention to efficiently remove heat was tested by increasing the heat output of the reaction (processing a greater syngas volume with a reduced contact time while maintaining CO conversion by controlling the temperature). It should be noted that the technological finned plates were only pressed for contact with the coolant subassembly. The contact resistance of the technological finned plate with the press fit and the coolant subassembly did not essentially change the efficiency of the FT exothermic reactor. In addition, the press fit of the finned plates in relation to one wall was hindered by a raised rib or thickening of a height of the order of 0.013-0.13 mm (0.5-5 mils) and a width of the order of 0.025-0.508 mm (1-20 mils) associated with a method of manufacturing a subassembly using laser welding.

While maintaining 16.5% dilution, Н ₂ : СО = 2 and an operating pressure of 2413 kPa (350 psig), the contact time of the reaction of the process decreased stepwise from 290 ms to ~ 70 ms. For example, taking into account 66.5 grams of catalyst and a catalyst bed volume of 63.1 cm ^{3, a} change from 290 ms (flow rate 13.1 tbsp / min) to 70 ms increases the consumption of syngas in the reactor to 54.1 tbsp / min . The composition of the syngas and transition details are presented in table 2 below. The temperature was raised from ~ 206.6 ° C to ~ 263 ° C to maintain ~ 70% CO conversion. As a result of this increased heat flux, the heat transfer ability of the heat carrier was tested. Key data for this phase of operation is presented below in table 2.

table 2 All-welded reactor process data for various contact times Contact time
[ms] Consumption N ₂
[Art. l / min] CO consumption
[Art. l / min] Consumption
N ₂
[Art. l / min] Temperature
[° C] CO conversion
[%] Selectivity for CH ₄
[%] Approx. heat flow per channel
[W] Average steam output 290 7.3 3.6 2.2 207 74.1 8.7 1,2 1.5% 210 10.0 5,0 3.0 215 72,2 9.5 1.7 2.9% 150 14.1 7.0 4.2 222 71.0 14.3 2,3 3.1% one hundred 21.1 10.5 6.3 241 70,2 26.5 3.6 5.4% 85 24.8 12,4 7.4 253 71.0 34.8 4.4 7.3% 70 30.1 15.1 8.9 263 69.6 39.9 5.3 10.5%

Based on the above data, the welded reactor of the invention can cope with a heating capacity that is more than four times the value provided that the contact time is 290 milliseconds (with an increase in the average vaporization from ~ 1.5% to ~ 10%).

The following are the individual stages of operation and comparison with similar test conditions for other devices (soldered and single-channel reactors):

The welded reactor of the invention was tested at a contact time of 210 ms from 945 to 1131 hours of continuous operation. Other process parameters were maintained at a constant level of H ₂ : CO = 2.0, p = 2413 kPa (350 psi), diluent ~ 16.5%. The temperature of the reactor was increased to ~ 214.6 ° C to maintain the conversion of CO.

The welded reactor of the invention was tested at a contact time of 150 ms from 1132 to 1182 hours of continuous operation. Other process parameters were maintained at a constant level of H ₂ : CO = 2.0, p = 2413 kPa (350 psi), diluent ~ 16.5%. The temperature of the reactor was raised to ~ 221.7 ° C to maintain the conversion of CO.

The welded reactor of the invention was tested at a contact time of 100 ms from 1205 to 1350 hours of continuous operation. Other process parameters were maintained at a constant level of H ₂ : CO = 2.0, p = 2413 kPa (350 psi), diluent ~ 16.5%. The temperature of the reactor was increased to ~ 241.2 ° C to maintain the conversion of CO. Over a period of ~ 1221–1228 hours of continuous operation, the CO mass flow rate regulator failed due to water ingress, which caused blockage and the need for replacement.

Further, the contact time was gradually reduced in 5 ms steps from 100 ms to 70 ms. Other process parameters were maintained at a constant level of H ₂ : CO = 2.0, p = 2413 kPa (350 psi), diluent ~ 16.5%. At a contact time of 70 ms and a reactor temperature of ~ 263.1 ° C (~ 1542 hours of continuous operation), an unstable runaway behavior was noted for a number of thermocouple positions on the reactor. Under constant conditions, there was a sharp rapid increase in temperature (before runaway control due to lower reactor temperature), as shown in the graph in figure 11.

The temperature shown in FIG. 22 is measured on the outer metal wall of the reactor, which is the opposite surface of the support plate 1.27 cm (0.5 in) thick, which is in contact with the coolant channel farthest from the center. The temperature of the metal rises by less than 10 ° C, but indicates an increase in the temperature of the catalyst layer, estimated above 50 ° C.

During this period of operation, the increased productivity of the inventive welded reactor was also demonstrated. Wax material obtained during operation with contact times of 290 ms, 210 ms and 150 ms was analyzed to determine the alpha number. The technological parameters of the device are presented below in table 3.

Table 3 Demonstrated high performance all-welded reactor of the invention with an alpha value of 0.89 or higher for the wax product. For contact times above 210 milliseconds, the alpha value was 0.91 or higher. Alpha is determined by the classical method known to those skilled in the chemistry of Fischer-Tropsch. Contact time 290 ms 210 ms 150 ms Temperature 206.5 ° C 214.6 ° C 221.7 ° C Operational parameters CO conversion 74.1% 72.2% 71.1% Selectivity for CH ₄ 8.7% 9.5% 14.4% C5 + productivity ~ 0.7 gal / day ~ 0.95 gal / day ~ 1.1 gal / day kg C5 + / lkat / h 1.24 1.75 1.99 Alpha 0.91 0.91 0.89

Steam content / partial boiling stability assessment

In this part of the study, from 1662 hours to 1783 hours of continuous operation, the coolant flow rate was reduced from 0.5 l / min (while maintaining other operating parameters at a constant level of Н ₂ : СО = 2, diluent 16.5 ° technological pressure 2413 kPa (350 psi)) to increase the degree of boiling at the same heat output and to provide increased output steam content. At ~ 1712 hours of continuous operation and a flow rate of 0.2 l / min, the flow meter reached the lower limit of measurement, and a further decrease in flow rate became impossible. The average steam content during this test phase was increased to ~ 15%, as shown in table 4 below.

Table 4 Operational performance of all-welded reactor at different output steam content Contact time
[ms] Coolant flow rate
[l / min] CO conversion
[%] Selectivity for CH ₄
[%] Average steam output
[%] 278 0.41 73.3 8.2 1,5 150 0.53 71.0 14.3 2.9 70 0.50 69.6 39.9 10.5 90 0.37 57.8 34.1 8.9 90 0.22 57.8 34.1 14.8

Pressure on laser welded subassembly

During the operation of the FT reactor, the laser-welded subnode gradually switches from compression to tensile work as the pressure inside the coolant subassembly increases with continuous operation. In particular, the section of the coolant subassembly, which is adjacent to the process layers, is the smallest sub-section of the coolant. For the all-welded reactor described in this example, approximately 60% of the subassembly is adjacent to the finned process bed and undergoes compression and tensile changes. For a larger reactor with a 60 × 60 cm (24 ″ × 24 ″) subassembly, more than 80% of the subassembly’s surface area transitions between compression and tension as the boiling point changes depending on pressure. The reaction temperature, as a rule, for the Fischer-Tropsch reaction increases as the catalyst is deactivated with the accumulation of wax. The characteristic initial temperature is in the range of 200 to 210 ° C. when the boiling point of the steam curve is in the range of 1448 to 1793 kPa (210 to 260 psi). The process feed pressure is typically between 1724 and 3101 kPa (250 to 450 psi). With increasing pressure on the side of the coolant, an increase in temperature occurs. At 220 ° C, the steam pressure during the boiling process (the preferred method for removing exothermic heat from the Fischer-Tropsch reaction) is approximately 2206 kPa (320 psi). At 230 ° C, the vapor pressure is approximately 2620 kPa (380 psi). For the experiments described in this example, the process temperature was raised to above 250 ° C at a vapor pressure of approximately 3861 kPa (560 psi), which significantly exceeds the process pressure of the reaction. At the beginning of the cycle, the laser welds were under compression under high pressure on the ribbed plate or the technological side. At maximum temperature, the laser welds worked in compression, and the pressure on the coolant side was 2289 kPa (332 psi) higher than the pressure on the process side. Before and after regeneration, the reaction temperature decreased to below 220 ° C, and the laser welds again made the transition from tensile work to compression work. The laser-welded subnode demonstrated resistance to operation under conditions of variable compression and tension for more than 1000 hours of continuous operation. In addition, it was revealed not only the resistance to operation under conditions of variable compression and tension for laser seams, but also the absence of the effect of changes in pressure inside the device on the thermal contact of the finned plate and the catalyst seal. An important parameter to maintain the good operational efficiency of the reactor of the invention is a well-loaded catalyst bed, preferably when the catalyst loading density is within 2% and even more preferably 1% of the theoretical SDES (average packed density) determined externally in accordance with the methods of the American Society of Specialists on tests and materials for a specific material. To compare the actual pressure drop with the calculated value from the Ergun equation, it is possible to use the cold flow pressure drop test. If the pressure drop goes beyond 5% (with a preferred level of 2%) of the calculated value from the Ergun equation, then the layer is not densely packed. With insufficient compaction of the layer, negative effects are possible when the flow is fed into the reactor, and changes between compression and tension in relation to the technological channels and coolant channels can lead to an increase in undesirable forces acting on the catalyst particles, which entails abrasion or grinding. During the destruction of the catalyst particles, the formed small particles are highly likely to cause a significant decrease in pressure in some channels relative to others, causing a violation of the flow distribution, local overheating, or preliminary thermal runaway.

Summing up, the operational efficiency of the all-welded reactor was confirmed by the test results and confirmed the absence of the need for soldering and / or diffusion welding. The thermal contact provided by the technology of all-welded production provided the desired results.

The initial excess of the coolant pressure over the process pressure was limited to 345 kPa (~ 50 psi) to prevent the welds from diverging between the coolant channels and deforming the wavy plates based on the dimensions of the laser welds for the reactor of this invention and tensile testing of similar laser welded parts . Throughout the cycle, this pressure increase was increased to 2289 kPa (~ 332 psi), and no visible visible deformations were noted. After testing, the opening of the reactor confirmed that the internal parts of the reactor were not compressed or deformed, and also did not undergo other undesirable changes. This astounding result suggests that ribbed plates with a loaded catalyst provided structural support for small laser welds that were less than 0.005 cm (0.002 in) wide. It is possible to perform laser welds with a wider width, for example, 0.0150 cm (0.006 in) or more, for operation at elevated coolant pressure without the necessary support from the process channel loaded with the catalyst. Thus, in preferred embodiments, laser welds of a width of 0.015 cm or more are used in the apparatus and methods of the invention.

EXAMPLE 3. Laser welds of large sheets

To demonstrate the effect of the non-uniform thickness of the parts on laser welded subassemblies for the reactor of the invention, three types of parts were tested. According to the initial theoretical assessment, the upper sheets should have been less prone to deformation due to lower input power to form a weld. However, surprisingly, it was found that thinner top sheets are more susceptible to deformation. Preference was given to thicker top sheets.

- The top sheet welded to the coolant channel gasket had a wall thickness of 0.05 cm (0.020 ″), 0.025 cm (0.010 ″) and 0.025 cm (0.005 ″), 0.125 cm.

- Initial welding was performed using intermittent laser welds to fix the parts (e.g. for every fifth row). Before performing long welds in every fifth row, intermittent laser welds were added in the direction along the length of the weld. The length of intermittent welds was approximately 2-4 cm at intervals of 5-20 cm without welding.

- After that, the upper plate was removed and welding was performed along the entire length.

- For two scenarios with a thinner wall, it was necessary to adjust the power setting and focus

- 60% power setting for the wall 0.025 cm (0.010 ″)

- 50% power setting for the wall 0.0125 cm (0.005 ″)

Initially, it was suggested that the deformation will be lower with decreasing wall thickness and power, but the opposite result was obtained. When measuring the wall subassembly of 0.05 cm (0.020 ″), a deformation of 6.875 cm (~ 2.750 ″) was determined.

Subsequent tests of the walls with a thickness of 0.025 cm (0.010 ″) and 0.0125 cm (0.005 ″) showed the following stress strain values:

- for sub-node 0.025 cm (0.010 ″) - 7.265 cm (2.906 ″)

- for the subnode 0.0125 cm (0.005 ″) - 7.4025 cm (2.961 ″)

More than 161 welds 60 cm (24 ″) long over a 60 cm (24 ″) node width were made in the test units, and the stresses in these components were not surprising, representing longitudinal and transverse stresses that are standardly observed when performing long welds. The task is to reduce stresses during the welding process or to reduce / eliminate stresses after processing the subassembly.

Preferably, the individual topsheets have a thickness of 0.04 cm or more, in some embodiments, in the range of 0.04 cm to 0.2 cm; more preferably 0.05 cm to 0.1 cm. "Top sheet" means a sheet that is located on top of a sheet or a series of sheets that contain channels or other cavities; the top sheet seals the channels or cavities in the height direction and completes the subnode.

Pre-bending tests for large parts

One of the ways to reduce welding stresses was to use preliminary bending of the components, followed by the operation being tested. The following process was used:

- process

- Preliminary bending of the coolant wall and channel coolant gasket in the kit

- Installing the fixture on the laser welding table and setting up and aligning the laser

- Implementation of the laser seam welding program

- Re-installing the fixture on the laser welding table

- Complete welding program

The components with preliminary bending were made using a device including clamps and connecting plates, which ensured a controlled and stable bending of the components (see Fig. 12). The bending of the sheets was performed in a direction perpendicular to the length of the channels. Using this technique, two sheets were bent, for which welding was further performed. The final subnode was subjected to visual comparison with the subnode, made in a similar way, but without preliminary bending.

- Stress relieving at temperatures below 400 ° C is acceptable practice, but provides only moderate stress relieve. One hour at 870 ° C typically relieves about 85% of residual stresses. However, stress relief in this temperature range can also lead to the deposition of carbides at grain boundaries, causing sensitization, which significantly impairs corrosion resistance in a number of media. To avoid these effects, it is recommended to use stabilized stainless steel (grades 321 or 347) or ultra-low carbon steel (304L or 316L), especially when stress relief is required over a considerable length. The stress relieving for a large part was tested at 400 ° C and an elevated temperature of 1000 ° C. In both cases, limited success was noted. A moderate decrease in deformation was observed, amounting to ~ 40% of the initial warpage, measured by the distance of one edge on a flat surface.

- Normalization (often referred to as annealing to a solid solution) not only recrystallizes riveted grains, but also returns chromium carbides (deposited at grain boundaries in sensitized steel) back to the austenite solution. Annealing also homogenizes the dendritic structures of the weld metal and relieves all residual stresses from cold working. Normalization temperatures typically exceed 1040 ° C, although some types can be annealed at precisely controlled temperatures up to 1010 ° C, when the fine grain size is important. The exposure time at temperature is often shortened to minimize surface scaly or to control grain growth, which can lead to an “orange peel” during molding.

Example 4. The effect of finned plates in welded reactors

The wave-like structures, including finned plates with a press fit and high aspect ratio, most often do not maintain a rectilinear shape and contain a bend or curvature of finned plates. After compression in contact with two solid planks on both sides of the finned plates, the plates are even more bent or curved. In the welded reactor of the invention, finned plates are pressurized to adjacent surfaces rather than soldered or welded, causing greater contact resistance or thermal resistance at the contact point of the finned plates with adjacent surfaces. In exothermic or endothermic reactions, heat from the reaction is transferred between the finned plates and adjacent surfaces. To improve thermal contact with the press fit, finned plates are made above the butt joints. When the fin plates are compressed in contact with an external load, they bend or bulge. As the finned plates bend, they are less likely to maintain an internal margin of safety.

Finned plates with a press fit made of Cu110 0.015 cm thick (0.006 ″) and 0.64 cm high (0.256 ″) are pressed against the butt plates 0.63 cm high (0.25 ″). Ribbed plates become more eccentric when pressed against adjacent heat-conducting walls. In the photograph in FIG. 13, horizontal lines are observed at the landing sites of the finned plates on the fusion lines of the laser weld, which seals the cross-flow heat exchange channels.

The press-fit reactor described in Example 1 showed amazingly good operational efficiency comparable to that of a brazed reactor, which had no ridges from laser welding, separating the finned plate from the heat transferring wall due to the combination of high thermal conductivity of the copper finning plate transferring more heat to the heat-conducting wall - including due to axial conductivity to improve thermal contact points, and fill a small gap (estimated from 5 to 150 microns) of water hydrogen and / or Fischer-Tropsch liquids produced during the reaction. The thermal conductivity of hydrogen and liquid oil is substantially higher than for most gases, and therefore reduces the effect of poor contact resistance between the press fit finned plate and the heat transfer wall. In addition, the advantage of using copper with a very high thermal conductivity is the effective axial conductivity for transferring heat to the heat transferring contact surface without the formation of local overheating and the creation or exacerbation of unwanted side reactions.

A load of 17,926 kPa (2,600 psi) was applied to the stack to ensure contact. A preferred load range may be from 3447 kPa (500 psi) to 3447000 (500000 psi) depending on the height of the finned plate, the material of the finned plate, the thickness of the finned plate and the eccentricity of the starting plate. After compression and the first gas-tungsten welding along the contour of the side slats or butt planks, the contact of the finned plate with the heat transfer unit is ensured.

Example 5. Linear density of welds per unit area of a subassembly for a large laser-welded subassembly

This example describes a laser-welded subassembly with dimensions of 60 × 60 cm (24 ″ × 24 ″) with fusion lines of laser welds 0.6 m long along the coolant flow direction. This panel included 161 coolant channels and 162 laser weld fusion lines for sealing the device between the coolant channels, and a structural support was presented to prevent deformation of the equipment during pressure operation.

This sample is characterized by a linear length of welds of 97.2 m in a section of 0.6 × 0.6 m or a linear density of welds of 270 m per m ² of surface area. Alternatively, the linear density can be represented as 2.7 cm / cm ² for the plot with an area of 3600 cm ² in this embodiment. For other embodiments, the linear density of the welds may be higher or lower than 2.7 cm / cm ² , and in preferred embodiments, the implementation may range from 0.1 cm / cm ² to 10 cm / cm ² .

Example 6. Registration of laser welds

Fig. 14 shows fusion lines of laser welds that connect the upper part of the ribs between parallel and adjacent coolant channels formed in the lower plate. Laser welds connect the bottom channel plate to the top plate. The laser weld can be made along the tops of the ribs, which are formed in the bottom, or channel plate. In this example, the rib width is 0.093 cm (0.037 ″) and the width of the laser seam can vary from 0.005 cm (0.002 ″) to 0.025 cm (0.01 ″). The laser weld can be made in the middle part of the rib or on either side along the width of the rib.

Example 7. Production of a large FT reactor

The all-welded chamber of the Fischer-Tropsch reactor is manufactured primarily as a two-layer unit with alternating sections of the coolant and technological sections. A feature of this design is that the coolant subassembly is welded to ensure mechanical integrity as an independent unit. One process for creating such a subunit is the use of laser welding to connect the upper one-piece gasket to the machined gasket by arranging the welds between each channel parallel to the ribs. It has been demonstrated that using this approach, the required mechanical integrity can be achieved. The second parameter to consider is the straightness of the part. Welding two thin sheets in this way can lead to significant deformation of the subassembly due to shrinkage of the material connected to the stainless steel weld. Deformation of the subassembly can lead to increased labor intensity and the need for efforts to re-align parts or modify the styling process to handle deformed parts.

One way to minimize deformations associated with welding is to limit the overall size of the subassembly. It has been found that the use of a subassembly that contains 60 cm (24 ″) welds between the channels, but is limited to a width of approximately 15 cm (6 ″), provides an acceptable level of straightness, while a similar assembly with a width of 60 cm (24 ″) does not perform this task (e.g., the total area is 15 × 60 compared to 60 × 60 cm (6 ″ × 24 ″ compared to 24 ″ × 24 ″)). In addition, several of these sub-assemblies with a width of 15 cm (6 ″) can be joined together by intermittent seam welding to ensure sufficient straightness. Thus, an even 60 × 60 cm (24 ″ × 24 ″) subunit can be made of four 15 × 60 cm (6 ″ × 24 ″) interconnected intermittent seam welds, providing substantially improved straightness than the 60 × 60 subunit cm (24 ″ × 24 ″) of the initial parts with dimensions of 60 × 60 cm (24 ″ × 24 ″).

Another useful feature that can be provided for a welded subassembly is the execution of parts that can be accurately checked for leaks and mechanical integrity before using them to form a complete welded stack of the FT reactor. This can be achieved by slightly increasing the size of the original subassemblies, adding a hole through which pressure can be applied, and initially sealing the ends of the subassemblies using a laser welding process. This pressure test may be hydrostatic or pneumatic. After welding and certification of individual parts, the units are cut to the correct size 15 × 60 cm (6 ″ × 24 ″), at which the flowing ends of the channels are opened, followed by intermittent seam welding of four 15 cm (6 ″) subassemblies together with the formation of a single coolant subunit 60 × 60 cm (24 ″ × 24 ″).

Further, the final subnodes alternate with the technological layers with the formation of the main chamber of the reactor. The laying process includes the initial installation of a clamping plate with a thickness of 5 cm (2 ″), onto which an end plate with a thickness of 2.5 cm (1 ″) is laid, and then alternating coolant layers and technological layers. The installation is completed by placing the last coolant subassembly with the upper end plate and the upper clamping plate laying on it. Pressure is applied to the stacking to pre-compress the wave-shaped copper plates to ensure metal-metal contact for all components. Applied pressure ranges from 138 kPa (20 psi) to 3,447,000 kPa (500,000 psi), with a preferred range of 138 kPa to 138,000 kPa (20 to 20,000 psi) or more. a preferred range of 138 kPa to 34,474 kPa (20 to 5,000 psi). The stack is then secured in place using a clamping system before relieving the applied pressure. The clamping system keeps the stack compressed to weld the chamber.

Welding of the chamber is carried out mainly in three stages: the implementation of load-bearing welds on both technological surfaces, the implementation of sealing welds on both surfaces of the coolant and the performance of two sealing welds with an end plate on each technological surface to prevent bypass flow. The reactor remains in a fixed state during each of these welding steps to ensure the best possible thermal contact in the chamber. It should be noted that each of these three stages of welding the chamber serves a separate purpose. Bearing welds on technological surfaces are carried out first and give the stack as a whole sufficient mechanical strength to simplify further operations with the camera (lifting, rotation or other manipulations) during the other two stages of welding, as well as all subsequent manufacturing steps. Sealing welds on two surfaces of the coolant are primary welds that protect the reactor from internal cross-leakage (process medium is coolant or vice versa). The sealing welds of the end plate are used to seal the most distant coolant sub-centers with the upper and lower end plates. Although they are sealed on the surfaces of the coolant in order to avoid cross-leakage, it is also necessary to seal them on the technological surfaces in order to prevent the transfer of the process gas flow between these parts, thus bypassing the catalyst layer. It should be noted that the catalyst is loaded into the reactor and between the technological finned plates after the final assembly.

Before welding the chamber into one assembly, the reactor should be substantially free of leaks. The welded reactor has not yet been prepared to pump significant internal pressure, so a bolt on the clamping mechanism can be used for support. After certification of the reactor chamber, the front coolant receiver is welded into place. After completing this step, the reactor can be tested at the flow rate of the cooler side, if desired or necessary. Next, the corresponding rear coolant receiver is welded into place. Both heat collectors are prepared for operation and form part of the base of the external support system of the reactor.

Since, with the exception of coolant subassemblies, all welding is performed around the perimeter of the chamber, the reactor cannot withstand significant internal process pressure without deformation of the upper and lower end plates due to the pressure caused by the pressure, and accordingly, the reactor cannot enter the operating mode in its current state. To ensure the mechanical integrity of the reactor, a system of external supports (exoskeleton) is welded along the perimeter of the reactor chamber. These supports are designed to balance the internal process pressure and thus to control any deformation of the upper and lower plates within an acceptable framework. External supports act as stiffeners that are welded to the upper and lower end plates of the reactor, after which the upper and lower parts are connected. They balance any load caused by internal pressure build-up and, accordingly, prevent any other possible deformation. Their thickness is less than the height, and they surround the reactor in several layers (see figure 15). Preferably, each stiffener has a height of at least 3 times, more preferably at least 5 times the thickness (height is implied in the laying direction). In some embodiments, the distance between the sets of supports, as well as their thickness and height, are determined based on the balanced loads of the process. In one example, the supports are made of a 0.75 inch thick stainless steel plate and project approximately 8 ″ on the upper and lower end plates with a spacing of about 3 ″ between them. As the last stage of certification, temporary collectors can be welded to the chamber on technological surfaces, which are subjected to high pressure tests to confirm that the reactor meets design criteria. After this stage, the technological collectors are removed, and the chamber can be prepared for loading the catalyst, after which the final technological collectors are welded into place.

FIG. 16 illustrates an example in which four 15 × 60 cm (6 ″ × 24 ″) subassemblies are arranged side by side and welded at different sites for connection to a 60 × 60 cm (24 ″ × 24 ″) assembly. Spot welding is preferred for joining subassemblies since it has been found that continuous welds cause increased deformation.

Example 8

The device of Example 8 is a welded reactor or device that provides cross-flow heat transfer between two fluid streams. Alternative flow patterns may be used, but the cross-talk is used in the specific example described. External supports were welded as a part of the kit (see figure 17) to the outer part of the device chamber in order to ensure pressure build-up in the internal channels of the device with respect to the low pressure of the external environment without compromising the integrity of the device or loss of tightness. An “exoskeleton” or a set of external supports allows the device to withstand high pressure drops with the external environment. The device was made of stainless steel 304L, including external supports. The use of an exoskeleton allows the use of a welded reactor for Fischer-Tropsch reactions and other reactions or heat transfer tasks or operations under high pressure without using an external pressure vessel. The devices operate in tension rather than compression in the presence of an external pressure vessel with a high pressure fluid surrounding the reactor or device of the invention.

The camera of the device of example 8 with dimensions 61 × 61 × ~ 6.5 cm was made of layers that are welded around the perimeter, as described in the attached application. The outer supports were made with a thickness of 1.9 cm, a length of 105 cm and a width of about 14 cm at the ends and about 17 cm in the area adjacent to the surface 61 × 61 cm. The supports were located at a distance of 10.2 cm from each other (between centers) with an arrangement between the supports of transverse elements 1.9 cm thick along each end of the device (so that two rows of cross elements are spaced 60 cm apart). The welds between the outer supports and the cross members were welds with grooves and full penetration.

Hydrostatic tests were originally performed for the process flow loop. The procedure used (shown graphically in figure 18) was as follows:

1. Determination of the initial leakage rate during a pressure test with nitrogen at 690 kPa (psi).

2. Filling the device with water using a pump (in this example, the Prep All HPLC pump Lab Alliance (Lab Elyans) was used).

3. Using a pump to increase pressure from external pressure (ie <450 kPa) to ~ 3300 kPa (464 psi) at a rate of ~ 300-400 kPa / min.

4. Relieve pressure at a rate of ~ 50-100 kPa / min to a level below ~ 3000 kPa (420 psi).

5. Using a pump to increase pressure from ~ 3000 kPa (420 psi) to ~ 3700 kPa (522 psi) at a speed of ~ 50-100 kPa / min.

6. Using a pump to increase pressure from ~ 3700 kPa to> 6000 kPa (855 psi) at a speed of ~ 100-150 kPa / min.

7. Relieve pressure at a rate of ~ 250-300 kPa / min to a level below 5300 kPa (754 psi).

8. Continue bleeding pressure until environmental conditions are reached and draining water from the device.

9. Repeat step 1.

Next, a second hydrostatic test was performed for the process flow loop. The procedure used (shown graphically in figure 19) was as follows:

10. Filling the device with water using a pump (the Prep All HPLC pump manufactured by Lab Alliance was used in this example).

11. Using a pump to increase pressure from external pressure (ie <250 kPa) to ~ 3500 kPa (495 psi) at a speed of ~ 2000-2500 kPa / min.

12. Using a pump to increase pressure from ~ 3500 kPa to> 6000 kPa (855 psi) at a speed of ~ 800-900 kPa / min.

13. Relieve pressure at a rate of ~ 400 kPa / min to below 5200 kPa (740 psi).

14. Continue bleeding pressure until environmental conditions are reached and draining water from the device.

15. Recheck the leak rate at a pressure of about 690 kPa (100 psi) and compare with the original leak rate for the pressure test.

Hydrostatic tests of the coolant and process circuits were performed using the above protocols. The device showed no signs of mechanical failure during hydrostatic tests. The leakage rate from the coolant circuit to the process circuit before and after these hydrostatic tests, measured as the pressure drop over a 15-minute period with an initial pressure of 690 kPa (100 psi) in the coolant circuit, was 0.6 kPa (0.09 psi) and 21 kPa (3.05 psi). Further, the device was prepared using welding, and the leakage rate was measured from the coolant circuit to the process circuit, which was defined as a pressure drop of 2.2 kPa (0.32 psi) over a 15-minute period with an initial pressure of 690 kPa (100 psi) Alternatively, gas tungsten welding, inert gas consumable welding, or other conventional welding methods may be used.

The welded chamber of the reactor was assembled from laser-welded sub-assemblies of the coolant, after which welding to the final reactor was performed. Further, the reactor was equipped with an exoskeleton, which ensures functioning at elevated pressure. For the Fischer-Tropsch reactor, hydrostatic pressure tests are performed at 5895 kPa (855 psi). Alternatively, hydrostatic tests with higher or lower pressure may be used for the Fischer-Tropsch operating conditions, depending on the desired operating conditions. The hydrostatic tests described in this example certify the reactor for operation at a peak temperature of 250 ° C and a peak design pressure of 3875 kPa (562 psi). The operating temperature and pressure should be lower than the peak design pressure to ensure an operational safety factor. For the desired increased working pressure for the Fischer-Tropsch reactor, the gaps between the external supports, as shown in FIG. 17, should be reduced, and additional supports added to certify the reactor for increased working pressure.

Other reactions in the case of operation at elevated temperature or pressure should be provided using the exoskeleton of the invention, but may require additional supports, which are located with a higher density. Alternatively, if a reduced design process pressure or temperature is required, a rarer arrangement of welded support elements should be used. The welded reactor of the invention can be operated at elevated temperature and elevated pressure without being placed inside the containment under pressure or without a soldered or diffusion-welded reactor chamber.

The exoskeleton of the invention allows the reactor or device to successfully withstand hydrostatic tests at high internal pressure and low external pressure while maintaining mechanical integrity and usability of the reactor.

Example 9. Leak Testing

All-weld leak test

Summary

All-welded FT devices were subjected to pneumatic leak tests to a pressure level of 690 kPa (100 psi) independently for the process side and the coolant side. In some embodiments, the leakage test pressure may be up to 1380 kPa (200 psi) or 3450 kPa (500 psi), and in one embodiment up to 6900 kPa (1000 psi). The pressure drop was recorded over time to determine the presence of leaks in the device, as well as leak tests of the contact surfaces to determine the location of the leaks. To ensure pressure build-up on either side, front and rear receivers sealed with gaskets are attached to the device using a clamping device with a thread along the entire length. Pressure is injected into the device in increments of 90-180 kPa (10-20 psi) with a stop after each step to check the rate of pressure drop, check gaskets and all fittings for leaks, and to identify any leaks inside the device.

Example 10. Ultra-high finned plates used in the reactor of the invention for the Fischer-Tropsch reaction and other chemical reactions

The all-welded reactor, in which the technological layers (including finned structures) are adjacent to the coolant sub-assemblies with the formation of the final assembly, allows the use of non-traditional technological layers, including ultra-high finned plates. The use of ultrahigh finned fins can be difficult to solder or diffuse welding, because they provide (if they provide at all) a weak support structure during soldering or diffusion welding and are able to bend, deform, or otherwise make it necessary to fix or straighten the final structure before operation in case operation is possible. Excessive deformation of ultra-high finned fins (such as about 1.25 cm or 0.5 inches or higher) after soldering or diffusion welding makes the device unsuitable for use.

Case A: Ribbed plate height 0.5625 cm (0.225 ″)

The Fischer-Tropsch reaction is carried out in a microchannel reactor. A microchannel reactor contains a number of parallel process channels with a compact FT catalyst. Heat from the reaction is removed through the coolant channels between the technological channels, where water acts as the coolant. The heat removal from the reaction generated inside the catalyst bed is improved due to the internal finned structure. A continuous finned plate inside the process channel forms a series of parallel flow paths. In this example, the finned plate is made of copper 110. The height of the process channel is 0.5625 cm (0.225 ″) and the length is 57.5 cm (23 ″). The fin plate has the same length, the distance between the ribs of the plate is 0.1 cm (0.04 ″), and the thickness of the fin plate is 0.015 cm (0.006 ″). It is assumed that there is no contact resistance between the ribbed plate and the walls of the process channel. The wall thickness of the process channel is 0.05 cm (0.02 ″), and it is made of stainless steel.

The process channel contains a certain amount of cobalt catalyst and is described in the first example. The porosity of the catalyst layer is approximately 0.4, and the effective thermal conductivity is approximately 0.3 W / (m · K). The loading density of the catalyst is 1060 kg / m ³ .

Complex FT reactions are modeled as a simplified reaction scheme with 6 volumetric reactions (see table 5). The parameters (table 6) in terms of speed are established using the test data of the catalyst in the FT laboratory reactor.

Table 5 FT reactions and kinetics No. Reactions Expressions of speed one 3H ₂ + CO → H ₂ O + CH ₄

2 5H ₂ + 2CO → 2H ₂ O + C ₂ H ₆

3 7H ₂ + 3CO → 3H ₂ O + C ₃ H ₈

four 9H ₂ + 4CO → 4H ₂ O + C ₄ H ₁₀

5 H ₂ O + CO → H ₂ + CO ₂ R _CO2 = k ₅ exp (-E ₅ / RT) C _CO C _H2O 6 29H ₂ + 14CO → 14H ₂ O + C ₁₄ H ₃₀ $$R_{ft}=\frac{k_{6}exR(-{\mathrm{<figure-callout\; id="6"\; label="E"\; filenames="00000020.png,00000022.png"\; state="\{\{state\}\}">E</figure-callout>}}_6/RT)C_{H2}{C}_{C0}}{{[\mathrm{one}+k_{a\cdot d}\exp (-E_{a\cdot d}/RT)C_{C0}]}^2}$$

Table 6 Parameters in Speed Equations Reaction k _i E _i speeds in kmol / (kg [cat.] s) J / kmol one 2,509 × 10 ⁹ 1.30 × 10 ⁸ 2 3,469 × 10 ⁷ 1.25 × 10 ⁸ 3 1,480 × 10 ⁷ 1.20 × 10 ⁸ four 1,264 × 10 ⁷ 1.20 × 10 ⁸ 5 2,470 × 10 ⁷ 1.20 × 10 ⁸ 6 3,165 × 10 ⁴ 8.0 × 10 ⁷ k _ad = 63.5; E _ad = 8.0 × 10 ⁷ J / kmol

Here, in the models of FT reactors, the following operating conditions were used:

- The temperature on the walls of the technological channels - 230 ° C

- The pressure at the beginning of the catalyst layer is 2840 kPa (412 psi)

- The ratio of H ₂ / WITH in the feed mixture is 2.1

- Nitrogen dilution of the feed mixture - 31.3% (v / v)

- Technological feed mixture is preheated to the temperature of the walls of the channels

- The flow rate of the feed mixture per catalyst package with dimensions of 0.1 × 0.563 × 57.5 cm (0.04 ″ × 0.225 ″ × 23 ″) is 1197 cm ³ / min. The contact time calculated based on the volume of the catalyst is 0.17 seconds.

The calculated temperature of the catalyst bed as a function of reactor length is shown in Figure 20. A temperature sample was obtained along the center of the catalyst bed so that the peak of the curve represents the maximum temperature of the catalyst bed. In this case, it is 239 ° C and is located at a small distance from the beginning of the catalyst layer. CO conversion is achieved at 76.0%, and methane selectivity at 15.3%.

Case B: ribbed plate height 1.25 cm (0.5 ″)

The Fischer-Tropsch reaction is carried out in a microchannel reactor. The microchannel reactor is similar to the reactor in case A. The only difference is that the height of the process channel and the copper finned plate is 1.25 cm (0.5 ″). It is expected that at a nominal ribbed height of 1.25 cm (0.5 ″), the initial height of the finned plate should be in the range of 1.253 cm (0.501 ″) to 0.13 cm (0.052 ″) with a preferred range of 1.26 cm (0.504 ″) to 1.275 cm (0.510 ″), close to the nominal size of 1.25 cm (0.5 ″) of the side strips, butt strips or p-strips.

In this example, the operating conditions are similar, except for the flow rate of the feed mixture, which is increased in accordance with the total catalyst loading volume to maintain the same reaction contact time at 0.17 seconds. The flow rate is 2661 cm ³ / min.

In this example, the same catalyst and reaction kinetics are used. The characteristics of the catalyst layer are also similar to those used in case A.

The calculated temperature of the catalyst layer as a function of reactor length is shown in Figure 21. A temperature sample was obtained along the center of the catalyst layer so that the peak of the curve represents the maximum temperature of the catalyst layer. In this case, it is 246 ° C and is located at a small distance from the beginning of the catalyst layer. CO conversion is achieved at 81.9%, and methane selectivity at 17.0%. During operation, it should be expected that the operating temperature due to the pressure from the steam side will be slightly reduced to 226-229 ° C, so that the CO conversion will be less than 76%, and the corresponding selectivity will be slightly lower than the calculated 17%. In general, according to calculations, with a ribbed plate height of 1.25 cm (0.5 ″), temperature-controlled operation will be provided for the reactor.

It is expected that a 1.25 cm (0.5 ″) high copper fin plate for a Fischer-Tropsch reactor can be used in an all-welded reactor of the invention. This transition to an ultra-high finned fin (0.625 cm (0.25 ″) or higher) is not possible in the case of the finned stainless steel finned plate or in the case of a brazed reactor. The ultrahigh finned fin plates of a brazed reactor undergo significant compression and crumple at high temperatures and during brazing at high loads. The use of an all-welded reactor platform provides more efficient Fischer-Tropsch reactors with high and ultra-high finned plates.

Case C: Ribbed Plate Height 2.5 cm (1 ″)

The Fischer-Tropsch reaction is carried out in a microchannel reactor. The reactor configuration is similar to the reactor in case A. The only difference is that the height of the process channel and the copper finned plate is 2.5 cm (1.0 ″).

In this example, the operating conditions are similar, except for the flow rate of the feed mixture, which is increased in accordance with the total catalyst loading volume to maintain the same reaction contact time at 0.17 seconds. The flow rate is 5321 cm ³ / min.

In this example, the same catalyst and reaction kinetics are used. The characteristics of the catalyst layer are also similar to those used in case A.

The calculated temperature of the catalyst bed as a function of reactor length is shown in Figure 22. A temperature sample was obtained along the center of the catalyst bed so that the peak of the curve represents the maximum temperature of the catalyst bed. In this case, it exceeds 600 ° C. Due to the fact that the volume of the catalyst is approximately 4 times the volume in the FT reactor in case A, the total amount of heat passing through each edge of the plate increases with multiplication by similar scale factors. At this level of heat from the reaction, even a low resistance to heat transfer inside the copper fins becomes critical. This can be observed by significant changes in temperature from the center to the edge. Due to the high temperature in most of the catalyst layer, the CO conversion is about 90% and the methane selectivity is also significantly higher than for lower finned plate heights. It is expected that for an exothermic effect of this type, the FT catalyst is quickly deactivated and, accordingly, the exothermic effect moves along the stream in the axial direction much like a burning cigar. After the FT catalyst burns out, the total conversion should be low (less than 40% per passage for similar conditions of flow, temperature and pressure), and the corresponding methane level should be high (above 10%).

The above examples imply excellent thermal contact of the finned plate with the wall and excellent flow distribution on the coolant side without blocked channels or low-flow coolant channels. It is expected that in a real industrial large-scale reactor using a dispersed FT catalyst, the stable height of the finned plate should preferably be less than 1.3 cm (0.5 in) to compensate for manufacturing defects or operational disturbances associated with potential contamination on the coolant side during operation .

Claims (31)

1. A chemical reactor for carrying out one or more technological operations, including a layered microchannel unit, which in turn includes a first sheet and a second sheet; each sheet has a length and a width, wherein the cross-sectional area, defined as the product of the sheet length by the sheet width, exceeds 100 cm ² ;
the first and second sheets are substantially flat;
the first sheet contains systems of parallel microchannels, the microchannels being separated from each other by a partition;
the first sheet and the second sheet are adjacent; and
including welds that extend along the length of the partitions and connect the first sheet to the second sheet. 2. The reactor according to claim 1, characterized in that the cross-sectional area, defined as the product of the sheet length by the sheet width, exceeds 500 cm ² . 3. The reactor according to claim 1, characterized in that the welds are continuous. 4. The reactor according to claim 1, characterized in that the welds on the second sheet have a linear density of at least 2.7 cm / cm ² . 5. The reactor according to claim 1, characterized in that the seams are laser welds and have a width of 0.015 cm or more. 6. A method of manufacturing a layered chemical reactor for one or more technological operations, including:
providing a first subnode or first sheet and a second subnode or second sheet;
wherein the first subnode or first sheet includes a first parallel channel system, the second subnode or second sheet includes a second parallel channel system, while there is no intersection between the channels in the first subnode or first sheet and the second subnode or second sheet;
welding the edges of the first subassembly with the edge of the second subassembly to form a combined subassembly layer or welding the edges of the first and second sheets to form a common welded sheet;
and laying a combined layer or common welded sheet with one or more layers or sheets, and joining stacked layers or sheets to form a layered device. 7. The method according to p. 6, characterized in that the first and second systems of parallel channels can have a common collector or lining. 8. The method according to p. 6, characterized in that the first subnode is welded to the second subnode by spot welding. 9. The method according to p. 6, characterized in that the common sheet or subnode is cut into many parts and subsequently welded to form a node. 10. The method according to p. 6, characterized in that the first subnode or first sheet has the same size as the second subnode or second sheet. 11. The method according to p. 6, further comprising the step of aligning the first subnode or first sheet before welding the edges. 12. A layered device manufactured according to the method of claim 6. 13. A method of carrying out a technological operation in a channel layer in a layered chemical reactor, in which tensile compression changes or vice versa, including:
providing a device comprising a first channel layer and a second channel layer located directly adjacent to the first channel;
wherein in the first step, the first channel layer contains a first fluid at a first pressure, and the second channel layer contains a second fluid at a second pressure; wherein the first pressure is higher than the second pressure;
carrying out a technological operation in the first channel layer at the first stage;
wherein in the second step, the first channel layer contains a third fluid at a third pressure, and the second channel layer contains a fourth fluid at a fourth pressure; wherein the fourth pressure is higher than the third pressure;
and carrying out the technological operation in the first channel layer in the second stage. 14. The method according to p. 13, characterized in that the first and third fluids are a Fischer-Tropsch process stream, and the second and fourth fluids are a heat transfer fluid. 15. The method according to p. 13, carried out in a welded layered device, characterized in that the first channel includes a heat-conducting insert with a press fit and a catalyst. 16. The method according to p. 13, characterized in that, since solid particles are formed in the first channel, the process conditions are regulated to adapt to changes in the reactor. 17. A layered microchannel device for carrying out a technological operation having a length and a width, the cross-sectional area of which, defined as the product of the sheet length by the sheet width, exceeds 100 cm ² and connected to the upper plate to form a welded assembly with a linear density of laser welded joint 0 , 05-20 cm / cm ² on the sheet surface section within the microchannel device; this section includes at least 50% (preferably at least 90%, in some embodiments, 100%) of the continuous area of the main surface. 18. The device according to p. 17, characterized in that the cross-sectional area, defined as the product of the sheet length by the sheet width, exceeds 500 cm ² and the device is connected to the upper plate to form a welded assembly with a linear density of laser welded joint of 0.05-20 cm / cm ² on the surface section of the sheet within the microchannel device; moreover, this section includes 100% of the main surface. 19. A method of forming a chemical reactor with a layered node for carrying out one or more technological operations, including: welding the upper sheet with the lower sheet with the formation of a number of channels located between the upper surface of the upper sheet and the lower surface of the lower sheet; in this case, welding is used to form seams between the channels in a number of channels. 20. The method according to p. 19, wherein the method of joining sheets into nodes includes laser welding for fastening two adjacent microchannels of the internal stream. 21. A layered chemical reactor for carrying out one or more technological operations, including: a stack of sheets connected by welding;
moreover, a stack of sheets includes:
the inner stack of sheets connected by welding around the perimeter of the sheets; and a connection for relieving stress in the stack of sheets;
wherein the stress relieving connection includes two adjacent sheets that are held in a stack but are not substantially connected to each other along the perimeter of two adjacent sheets. 22. The reactor according to claim 1, characterized in that the microchannels include hydrogen gas and carbon dioxide. 23. The reactor according to claim 1, characterized in that between the first sheet and the second sheet there is no intermediate layer of material. 24. The reactor according to claim 1, comprising a top sheet with a thickness ranging from 0.04 to 0.2 cm. 25. The device according to p. 17, characterized in that between the sheet in the device and the upper plate there is no intermediate layer of material. 26. The device according to p. 17, characterized in that the upper plate has a thickness in the range from 0.04 to 0.2 cm 27. The device according to p. 17, which is connected to the upper plate with the formation of a welded unit with a linear density of laser welded joints of 0.05-20 cm / cm ² on a section of the sheet surface within the microchannel device; moreover, this section includes at least 90% of the continuous area of the main surface. 28. The reactor according to p. 21, characterized in that the temperature seam is made of a lower plate and an upper plate, the upper plate closer to the outer main surface of the stack, the lower plate containing many holes, and the upper plate is devoid of characteristic features. 29. The reactor according to p. 21, characterized in that the two adjacent sheets are not connected, with the exception of their corners. 30. The reactor according to p. 21, characterized in that the chemical reactor is a reactor having an inner stack of sheets, and where the connection for stress relieving is located in the upper part of such a stack, and further comprises a second connection for stress relieving, including two adjacent sheets, which held in a stack, but essentially not connected to each other around the perimeter of two adjacent sheets, the second stress relieving connection located at the bottom of the inner stack of sheets. 31. The reactor according to p. 21, characterized in that the chemical reactor is a reactor having an inner stack of sheets, with one sheet of the connection to relieve stress is located opposite such a stack, and the connection to relieve stress is welded to the coolant subassembly.

Images