IN-SITU VAPORIZER AND RECUPERATOR FOR ALTERNATING FLOW DEVICE

Abstract

The present invention relates to a device for converting a liquid feed stream to a gaseous vapor stream comprising: (a) channel means having a first and second end, said channel means having a plurality of channels connecting said first and second end, said channel means having a substantially solid region and a void region, (b) inlet means for directing the liquid feed stream to the first end of the plurality of channels, and (c) outlet means for directing the gaseous vapor stream from said plurality of channels, where said channels have, at any distance d between the inlet and outlet, a geometric configuration, perpendicular to the feed flow direction, wherein the average void fraction ranges from about 0.3 to about 0.95.

Description

CROSS-REFERENCE TO RELATED APPLICATIONS

This application claims the benefit of provisional U.S. Application No. 61/411,512, filed Nov. 9, 2010, the contents of which are incorporated herein by reference.

FIELD OF THE INVENTION

The present invention relates to an in situ vaporizer and heat recovery apparatus for an alternating flow system that can be located on-board a transportation vehicle.

BACKGROUND OF THE INVENTION

Many industrial processes, including steam reforming, often require the conversion of liquid feed streams (such as gasoline or other liquid hydrocarbons, or water) to vapor streams prior to chemical conversion. Traditional vaporizers and boilers require significant thermal mass and generally perform poorly in transient operation due to fluctuating pressure and temperature changes. For mobile applications, the transient requirements to operate over a wide dynamic operating range (known as turndown) makes vapor stream handling difficult and provides challenges for equivalence ratio control between fuel and oxidizer. Many applications need this vaporization and mixing process to occur with precise mass flow control, operate over a large dynamic range, reside in-situ within the reactor volume to minimize heat transfer losses, incur low pressure drop losses, and be resistant to corrosion by the process fluid streams. Therefore, a need exists in the art for technology that is capable of providing precise mass flow control, operation over a large dynamic range of temperature and pressure, minimize heat transfer losses, incur low pressure drop losses, and be resistant to corrosion by the process fluid streams.

The current invention describes a novel in-situ vaporizer and heat recovery device particularly suitable for alternating flow reactor systems where the flow through the device alternates between a liquid vaporization mode and a reheating mode. In a preferred embodiment, it is part of a pressure-swing reformer (“PSR”) system as described in U.S. Pat. No. 7,491,250 for example. It enables the design of compact syngas generation systems, which may be deployed, for example, in transportation vehicles.

The present inventors have discovered particular design criteria of such an in-situ vaporizer and heat recovery device that are required for it to achieve the above described advantages. These features include specifications on geometry, heat capacity and heat transfer, flow channel dimensions, and void fraction and void fraction gradient to minimize the size and weight of the device. A unique combination of these design criteria produce a high efficiency vaporizer/recuperator as demonstrated in an application to pressure swing reforming as taught in the above referenced patent.

SUMMARY OF THE INVENTION

The present invention relates to a device for converting a liquid feed stream to a gaseous vapor stream comprising: (a) channel means having a first and second end, said channel means having a plurality of channels connecting said first and second end, said channel means having a substantially solid region and a void region, (b) inlet means for directing the liquid feed stream to the first end of the plurality of channels, and (c) outlet means for directing the gaseous vapor stream from said plurality of channels, where said channels have, at any distance x between the inlet and outlet, a geometric configuration, perpendicular to the feed flow direction characterized by (1) a void area A′(x), and (2) a total cross-sectional area A(x), where the void area A′(x) as a fraction of the total area A(x) is

${\phi }_x=\mathrm{void}\mathrm{fraction}\mathrm{at}\mathrm{distance}x=\frac{A^{\prime }\left(x\right)}{A\left(x\right)}$

and an average void fraction along a length of device L is

${\phi }_a=\mathrm{average}\mathrm{void}\mathrm{fraction}=\frac{{\int }_O^LA^{\prime }\left(x\right)x}{{\int }_O^LA\left(x\right)x}$

where the average void fraction ranges from about 0.3 to about 0.95.

In one embodiment, the void fraction varies along the length of the device.

In another embodiment, the average variation in void volume, as projected over the total length of the device, ranges from about 0.01 to about 0.5.

In yet another embodiment of the present invention the device has sequential constant void volume regions, which regions vary along the device length.

BRIEF DESCRIPTION OF THE FIGURE

FIG. 1 hereof is a schematic representation of an embodiment of the present invention.

DETAILED DESCRIPTION OF THE INVENTION

The present invention is directed to a device for converting a supplied liquid stream(s) to a gaseous vapor stream. More particularly, the invention relates to a device operating in a cyclic process whereby a liquid hydrocarbon and, optionally, a water mixture is converted into a mixture of hydrocarbon vapor and optionally steam. One particular application is for vaporization of a liquid hydrocarbon and water stream for use in a pressure swing steam reforming process.

The device is designed for continuous operation through a cyclic two-stage process. The first stage is a vaporization mode, and the second stage is a reheating mode. We have discovered particular design criteria or features that produce an in-situ vaporizer and heat recovery device capable of large dynamic ranges of operation, minimal pressure drop, corrosion resistance, and be sufficiently compact and light weight for mobile applications. These features include specifications on geometry, heat capacity and heat transfer capability, flow channel dimensions, and the void fraction and gradient of the void fraction within the vaporizer.

The device of the present invention is comprised of a partially porous media with void passages comprising a set of characteristic sizes and shapes. The void regions within the device form a set of connected pathways by which fluid (in either the liquid or gaseous state) can pass through the device. A simplified conceptual device geometry is shown in FIG. 1 for illustration. The void fraction, or fractional porosity, is the open fraction of the volume of the device at a given point in space. Fluid may enter either end of the device, and exit at the opposite end. A first purpose of the device is to vaporize a supplied liquid stream to vapor. As illustrated in FIG. 1, the surface by which the liquid stream to be vaporized enters the device (11) will be defined as the inlet surface here identified as (10). The surface through which the vaporized stream exits is defined as the outlet surface (12). In the re-heat stage of the device, a high temperature re-heat stream may enter through either the inlet or outlet surface, and exit the corresponding opposite end. In the schematic shown in FIG. 1, the reheat stream is shown flowing counterflow to the vaporization flow (16). The reheat stream (14) is illustrated in FIG. 1 as entering the outlet surface stream (12) and exiting at the inlet surface (10) as stream (15). In FIG. 1, the device is illustrated as a cylindrical device having constant cross-sectional area. In practice, the device cross-sectional shape is not limited to cylindrical, but may be rectangular, square, triangular, or otherwise. The cross-sectional area may also vary as a function of axial position.

Generally stated, there are two periods or stages of operation of the device, and these constrain the design characteristics of the invention. In the first period which operates for a time t_vap, a liquid or mixture of liquids or liquid-containing stream (16) enters the device at inlet (10) at a given volumetric liquid flow rate Q_liq. This liquid flow is vaporized to a gas-phase stream that can be characterized in terms of a Normal gaseous volumetric flow rate, Q_vap, with knowledge of the density and molecular weight of its components. The liquid phase may alternatively be supplied through a pre-atomized liquid stream forming a liquid droplet spray. The liquid or liquid spray enters, impacts the pre-heated surfaces within (11), and changes from the liquid to vapor phase, which exits the device as a vaporized stream (17). In other embodiments, a gaseous stream, either as vapor of liquid stream (16) or as diluent, may enter the device along with the liquid stream (16) and exit alongside vaporized stream (17). Any material that is already gaseous before entering the device during the vaporization period is not considered to be part of Q_liq or Q_vap, although it must be considered in calculation of dew points. In the second period of the device operation, potentially beginning after a delay period of t_d, a heated stream (14) is passed through the device for a period of time t_regen with a volumetric flow rate Q_regen, entering one end of the device and exiting the opposite end as a cooled stream (15). This is the re-heating or regeneration step. In the example of FIG. 1, this stream enters the device (11) at the outlet end (12) and exits at the inlet end (10). In other embodiments (not shown), the stream may enter the device (11) at the inlet end (10) and exits at the outlet end (12).

An element of the invention is that both vaporization and reheat periods of operation of the device share the same flow paths. The regenerative flow heats the device from an initial temperature (in one embodiment approximately around the dew point of the liquid feeds) to a final higher temperature which is higher than the dew point of the liquid streams at the prescribed operating pressure. The dew point, identified herein as T_DEW is known in the art as the temperature at which a vapor just begins to condense, and is dependent on stream composition and pressure. The flows are varied in a periodic manner, alternating between the endothermic vaporization phase (energy transfer to the fluid phase from the solid material) and exothermic regenerative heat transfer phase (energy transfer from the fluid flow to the solid material). The initial and final temperature of the device are determined from the coupling of the flow conditions (such as flow rate and velocity), geometric considerations such as device size and design, and the thermophysical properties of both the fluid and the device.

The present invention provides the specification for the geometric channels that conduct flow through the vaporizer. Such channels are defined as open or “void” spaces separated by solid walls. In some embodiments of the present invention, the walls themselves may include some porosity, typically with pore diameters well below 0.1 mm. Such porosity is not counted as “void” as used in the present specification, and is counted only as a reduction of the apparent density of the solid walls.

The geometric features of the device of the present invention are related to the magnitude of void fraction, the spatial variation of the void fraction, and the size and shapes of the small channel features which comprise the void fraction. Along a single imaginary spatial plane (19) as shown in FIG. 1, the local void fraction is defined as:

${\phi }_x=\frac{A^{\prime }\left(x\right)}{A\left(x\right)}.$

Here, ø_x is the void fraction at a given spatial location is denoted by the subscript x, the total cross-sectional area of the device at this plane is A(x), and the “open” or “void area” which contains no solid material in that plane is given by A′(x). For the purposes of the present invention, we adopt the convention that there is an axis of flow from inlet surface (10) to outlet surface (12), and “x” is the distance along this axis. Thus, x follows the direction of vaporizing flow (16), and has a value of zero at the device inlet (10) and a value of L_total at the device outlet (12), where L_total is the length of the device from inlet to outlet along the axis of fluid flow. We also adopt the convention that plane (19) is perpendicular to this axis of flow. Such planes are also known in the art as “axial planes”. The average porosity, or average void fraction, along a certain length of the device L is defined by:

${\phi }_a=\frac{{\int }_0^LA^{\prime }\left(x\right)x}{{\int }_0^LA\left(x\right)x}.$

If the averaging length L is taken to be the overall length of the device L_total, ø_a represents the average void fraction of the overall device noted as ø_avg. If the overall device occupies a total volume V_total, (solid+porous volume), the total open or void volume within the device is then V_void=ø_avgV_total, and the total volume of solid material in the device is then V_solid=(1−ø_avg)V_total=V_total−V_void.

We have discovered that the average void fraction ø_a of the device is a parameter that leads to successful operation of the device. We have discovered that acceptable average void fraction for the device of the present invention ranges between 0.3 and 0.95. Preferably, the device average void fraction ranges from 0.4 to 0.7.

In preferred embodiments, the void fraction varies axially along the length of the device. By length of the device, we mean the dimension along which the vapor flow predominantly occurs. In FIG. 1, this dimension or direction is axial, here indicated by dimension or direction “x.” The surface void fraction at the inlet surface (10) of the device has been found to preferably be between 0.5 and 0.995. A most preferred range at this location ranges from 0.65 and 0.995. The void fraction at the outlet surface (12) of the device is preferred to range from 0.2 and 0.7. A most preferred range of outlet void fractions ranges from 0.35 and 0.6.

For the preferred embodiment where the void fraction varies axially from the inlet to outlet surface, the variation may be generated either through a continuous change of void fraction or through a sequential series of constant void fraction regions. However the void fraction is varied, that variation of void fraction can be characterized as an average variation over the length of the device. For example, the set of ø_x values from inlet (10) to outlet (12) can be analyzed by least squares methods that are known in the art to compute a least-squares linear slope of ø_x versus x. This average variation can be expressed as a slope or gradient (i.e. change in void fraction per length), or as average total void fraction change, the latter being computed as the least squares slope multiplied by the total device length (L_total). In many embodiments of the present invention, the average total void fraction change is between 0.01 and 0.5. The preferred average total void fraction change is 0.15 to 0.35. We have discovered that the acceptable ranges of the axially device-averaged void fraction gradient vary between 0.01 and 0.5 void fraction decrease per linear inch of length. The preferred variation of the average void fraction gradient is 0.15 to 0.35 void fraction decrease per linear inch length. We have also discovered that in the embodiment where a series of sequential constant porosity regions are utilized, the number of preferred regions is greater than one and less than twenty. The more preferred number of regions is between two and ten, and a most preferred number between two and five.

The void volume is comprised of a large number of structured, small scale void regions denoted in the following as channels or channel regions. These channels are made of simple shapes and in a range of sizes. Preferred shapes for the cross-sections of the channel size void regions are highly structured, regular shapes, such as circular, semi-circular, annular, periodic wavy-walled corrugation, or rectangular channels and slots.

One embodiment of the device is to have the channels geometries nearly identical in size and shape throughout the device volume. A preferred embodiment of the device is to have nearly identical channel shapes within the device, with varying sizes at different axial locations (i.e. sets of circular channels with different diameters). An even more preferred embodiment of the device is to have an axial variation of both the channel shapes and sizes. Stated otherwise, changing the shape and/or size may increase the surface area available for heat transfer and thereby vaporization.

An alternative embodiment utilizes chaotic channels of wormhole networks comprised of millions of irregular channel shapes, such as a material characteristic of ceramic or metal foams. An additional embodiment utilizes the structures passages created by the interwoven networks of wire materials, such as made from stacking or weaving of metallic wires.

The channel shapes that make up the void regions of the device can be characterized by a set of spatial dimensions. One dimension is referred to as hydraulic diameter. For flow channels consisting of simple closed connected surfaces (i.e. circular cylinders, square, rectangular, triangular, or curved channels), the hydraulic diameter is defined as d_h=4 A/P, where A is the cross sectional area of the channel's flow-carrying void and P is perimeter around the closed surface. For complex channel configurations which are not comprised of simply connected shapes, a channel hydraulic diameter can be defined in terms of the axial plane (19) described in connection with FIG. 1. Within any axial plane, d_h′(x)=4 A′(x)/P′(x), where A′(x) represents the total void area of the device at a given axial location, and P′(x) is the total length of the intersecting surface between the solid and void region.

A characteristic horizontal size of the device h_char is approximated by taking the square root of the device volume to the total axial length, or

$h_{\mathrm{char}}=\sqrt{\frac{V_{\mathrm{total}}}{L_{\mathrm{total}}}}$

We have also discovered that certain channel properties are preferential within an inlet region of the device. The extent of this inlet region is preferably between about 5% and about 40% of the device length. In other words, the inlet region may extend from x=0 (the inlet surface (10)) to as little as x=0.05 L_total or to as much as x=0.4 L_total.) Preferably, the inlet region is between 10% and 30% of the device length. The orientation of the channels within the inlet region are preferably arranged such that movement of the incoming inlet flow stream is possible perpendicular to its average direction of flow. This allows a dispersive mixing component and a distribution component to the flow. A preferred design has a flow pathway perpendicular to flow axis that is in proportion to the characteristic horizontal size(h_char) concurrent with the flow in the axial direction. A preferred design has, within the inlet region, a perpendicular to flow axis continuous flow pathway that is at least 10%, to as much as 50% of the characteristic horizontal size(h_char). of the device. Other preferred designs have a continuous flow pathway of at least L_total in length within all axial planes within the inlet region of the device.

We have discovered that the acceptable characteristic channel hydraulic diameter sizes, d_h(x), in the inlet region of the device range from 0.1 to 10 mm, with a preferred range of 0.3 to 5 mm, and an even more preferred range of 0.7 to 2 mm. The characteristic channel sizes at the outlet of the device range from 0.2 to 5 mm, with a preferred range of 0.4 to 2 mm, and an even more preferred range of 0.5 to 1.5 mm.

In preferred embodiments, the ratio of the channel hydraulic diameter to channel length is between 0.5 and 10,000, with a preferred ratio between 10 and 5000, and an even more preferred ratio between 40 and 200.

An additional characteristic parameter of the device is the internal surface area available for vaporization of the incoming stream, and for reheating of the recuperator volume. We define a surface area per unit volume measure locally within the device to be S_v=S/V, or the surface area per unit volume, where S is the interior surface area contained within a prescribed total volume V (void+solid volume). One embodiment of the invention utilizes a uniform value of S_v. Alternative embodiments of the invention will employ a non-uniform value of S_v throughout different regions of the device. The average S_v for the device is simply the total device surface area divided by the total device volume, or S_v,avg=S_total/V_total. We have discovered that the acceptable average range of internal surface area per unit volume or the device to be between 10 in²/in³ and 2000 in²/in³. A preferred value of the average range of internal surface area per unit volume is between 20 in²/in³ and 1000 in²/in³. An even more preferred value of the average range of internal surface area per unit volume is between 50 in²/in³ and 250 in²/in³.

The available heat transfer area and device volume describe only the surface contact between the fluid and the solids. The composition and physical properties of the solid material of the device have been discovered to be instrumental to its successful operation, and its ability to function efficiently in terms of reheat capability, energy storage, and energy transfer. Accordingly, we have discovered that the composition of the materials which comprise the device are such that the thermal heat capacity have a value at least 100 J/kg-K, with a preferred value greater than 500 J/kg, and an even more preferred heat capacity value greater than 1000 J/kg.

From the aspect of heat management within the device, thermal contact between various regions of the device should be maximized. We have discovered that acceptable values of the thermal conductivity of the materials be at least 10 W/m-K, with a preferred value greater than 50 W/m-K, and an even preferred value greater than 200 W/m-K. We have also discovered that the composition of the materials which comprise the device are such that the density of the solid materials should be at least 2500 kg/m³, with a preferred value greater than 5000 kg/m³, and an even more preferred value greater than 7000 kg/m³.

The specific design of the device will depend on the liquid injection conditions and timing desired to accomplish the vaporization process. The amount of energy H required to vaporize a liquid stream flowing for a certain period of time is given by:

$\begin{array}{cc}H=\begin{array}{ccc}{\stackrel{.}{m}}_{\mathrm{liq}}& {\lambda }_{\mathrm{liq}}& {\tau }_{\mathrm{vap}}=\begin{array}{ccc}{Q}_{\mathrm{vap}}& {\lambda }_{\mathrm{vap}}& {\tau }_{\mathrm{vap}}\end{array}\end{array}& \\ {\stackrel{.}{m}}_{\mathrm{liq}}=Q_{\mathrm{vap}}\frac{\left(MW\frac{g}{\mathrm{mol}}\right)}{\left(\frac{22.4N}{\mathrm{mol}}\right)}{\lambda }_{\mathrm{liq}}={\lambda }_{\mathrm{vap}}\frac{\left(\frac{22.4N}{\mathrm{mol}}\right)}{\left(MW\frac{g}{\mathrm{mol}}\right)}\mathrm{units}\text{:}{\stackrel{.}{m}}_{\mathrm{liq}}=\left[g\text{/}\mathrm{hr}\right]{\lambda }_{\mathrm{liq}}=\left[J\text{/}g\right]{\lambda }_{\mathrm{vap}}=\left[J\text{/}N\right]Q_{\mathrm{vap}}=\left[N\text{/}\mathrm{hr}\right]{\tau }_{\mathrm{vap}}=\left[\mathrm{hr}\right]& \end{array}$

In these expressions, {dot over (m)}_liq is the mass flow rate of the liquid, λ_liq is the latent heat of vaporization of the liquid in mass units, τ_vap is the time period of injection, Q_vap is the Normal volumetric gaseous flow rate of vaporized liquid, and λ_vapis the latent heat of vaporization of the liquid in Normal gas volume units. The amount of energy required is proportional to the rate of liquid supply, the vaporization energy per unit mass (or volume) of the fluid, and the length of time of the process. Here, both the liquid basis and converted gaseous basis expressions are shown including their conversion factors. Normal gas conditions are known in the art and are typically taken as 0° C. and 1 atm absolute.

The vaporization capacity of the device is directly proportional to its available energy storage. The highest temperature of the device will be at the time of the beginning of the liquid injection step. The average temperature of the device at this time is T_DVI. As liquid is injected and vaporized, the device temperature will fall to a final average temperature T_DVF. at which point the vaporization portion of the operating cycle is complete. Here, we define ΔT_DEVICE to be the average temperature difference in the device from the beginning to the end of the liquid injection process (T_DVI−T_DVF). The absolute values of the high and low temperatures will depend on the device properties, and the heat balance and operating conditions of the process. The maximum vaporization capacity of the device H′ is given by

H′=ρ_deviceV_total(1−φ)C_p,deviceΔT_DEVICE=V_total(1−φ)C_p,deviceⁿΔT_DEVICE

units: ρ_device=g/l]V_total=[l]C_p,device=[J/g° C.]ΔT_DEVICE=[K]C_p,deviceⁿ=[J/l° C.]

Here ρ_device is the average density of the solid material of the device, φ is the average porosity of the device, C_p,device is the average specific heat capacity of the device is mass units, and C_p,deviceⁿ is the average specific heat capacity of the device is volume units ρ_deviceC_p,device=C_p,deviceⁿ. The vaporization capacity is proportional to the specific heat capacity, solid material volume, and material density of the device, and also directly proportional to the temperature differential during the process.

The space velocity of a system can be expressed as the Normal volumetric hourly gas flow rate of feed divided by the volume of the device, called the gaseous hourly space velocity, or GHSV. The gaseous feed rate is calculated as a molar rate of feed, and the Normal volume rate calculated as if the substances are gaseous species. As an example, a liquid water feed flowing at rate of 1 g/sec entering a 0.5 liter device has a gaseous hourly space velocity for the liquid injection step given by

$\mathrm{GHSV}=\frac{Q_{\mathrm{vap}}}{V_{\mathrm{total}}}=\frac{\left(1-\frac{g}{\sec }\right)\left(\frac{3600\sec }{\mathrm{hr}}\right)\left(\frac{\mathrm{mol}}{18g}\right)\left(\frac{22.4\mathrm{NL}}{\mathrm{mol}}\right)}{0.5L}=8960{\mathrm{hr}}^{-1}$

Here, Q_vap is the Normal volumetric gas flow rate (in units of NL/hr) and V_total is the total device volume. In general, the compactness and resulting efficiency of the device in terms of rate of vaporization capacity per unit volume is directly proportional to the space velocity. For integrated systems utilizing hydrocarbon feeds which undergo subsequent chemical or catalytic reactions, the overall space velocity of the system is proportional to the productivity of the system. It is desirable for the space velocity to have as high a value as possible.

In preferred embodiments of the invention, the space velocity GHSV is preferably greater than 500, and even more preferably greater than 1000.

For the device to operate in a cyclic fashion, the amount of heat consumed in the vaporization step is balanced by the amount of heat deposited in the device during the reheat (regeneration) portion of the cycle. If the liquid feed is fed at a high rate (high GHSV), then the heat is used up rapidly and the cycle time must be short. If the liquid feed is fed at a low rate (low GHSV), then the heat is used up slowly and the cycle time is longer. Combining the expressions above for H and H′ and setting them equal gives the expression

Q_vap λ_vap τ_vap=V_total(1−φ)C_p,deviceⁿΔT_DEVICE

Rewriting the terms to be non-dimensional on both sides of the equation and substituting the expression for GHSV from above gives the relationship

$\left(\mathrm{GHSV}\right)\left({\tau }_{\mathrm{vap}}\right)=\frac{\left(1-\phi \right)C_{p,\mathrm{device}}^{″}\left(\Delta T_{\mathrm{DEVICE}}\right)}{{\lambda }_{\mathrm{vap}}}$

Here, all of the variables assume their previous definitions.

The heat transfer requirements of the device can be expressed as a product of the volumetric heat of vaporization of the liquid feeds and the GHSV of the feed streams. The volumetric heat transfer requirement for vaporization is:

{tilde over (H)}_v=(GHSV)(λ_vap) units: H_v=[J/lhr] λ_vap=[J/l] GHSV=[hr^−1]

This is the required energy transfer per unit volume per unit time for liquid vaporization. The expression for {tilde over (H)}_v is noted because of the time dependence of hours rather than seconds. These expressions define the energy balance that is necessary for sufficient energy to be present for the vaporization of the liquid streams.

After the liquid enters the device for the time period τ_vap, a hot reheat stream is passed through the device to raise the temperature back to the initial high temperature at the start of the liquid injection cycle. The ability to raise the temperature of the device during this reheat stage is closely connected to geometric features of the device, as noted earlier. Variables include the porosity, hydraulic passage sizes, and the thermal properties of which the device is constructed. It is known in the art that porous media comprised of solid materials with characteristic channel passage shapes can be characterized by a heat transfer coefficient (h) and a characteristic heat transfer surface area (A). Preferred values for the surface area per unit volume characteristics were defined above. Correlations for the heat transfer coefficient based on gas and solid properties are also known in the art. These heat transfer coefficients are a function of flow rate and gas phase composition. Coefficients typically increase as the characteristic channel size of the porous material is decreased. The volumetric heat transfer coefficient can be defined and given in units of

$h_v=\frac{J}{\left(\mathrm{device}\right)\left(°C.\right)\left(\sec \right)}$

The volumetric heat transfer requirement for vaporization rewritten in consistent time units can be written as

$H_v=\frac{\left(\mathrm{GHSV}\right)\left({\lambda }_{\mathrm{sp}}\right)}{3600\sec \text{/}\mathrm{hr}}=\frac{J}{\left(\mathrm{device}\right)\left(\sec \right)}$

The present invention has a characteristic heat transfer temperature change as the ratio of the volumetric vaporization requirement to volumetric heat transfer coefficient for regeneration of the device. This characteristic temperature differential is expressed as

ΔT_HT=H_v/h_v

This temperature differential describes the balance between the heat transfer supply and demand during the cyclic operation of the device. As used here, this is based on heat transfer coefficients used in the reheat portion of the cycle, which is typically the lower heat transfer coefficient portion of the cycle and serves as a limiting design condition. This temperature difference is a basic design parameter for the device. The device design and material properties chosen to satisfy the requirements of the invention.

In the practice of the present invention, the characteristic ΔT_HT is preferably between about 0.1° C. and 600° C. More preferably, the characteristic ΔT_HT should be between 0.5° C. and 300° C.

A characteristic energy availability ratio for the device is defined to be

$R=\frac{\left(T_{\mathrm{DVI}}-T_{\mathrm{DVF}}\right)}{\left(T_{\mathrm{DVI}}-T_{\mathrm{DEW}}\right)}$

This is a preferred parameter for evaluating the effectiveness of the device. A value closer to one means ideal use of the energy available in the vaporizer, while values lower than 1.0 reflect the realities of temperature gradients that are needed to move heat rapidly. We have discovered that the required range for the energy availability R is from 0.05 to 0.7. A preferred range for the ratio R is found to be between 0.1 and 0.5. A most preferred range is between 0.2 and 0.4.

An additional feature of the device is a low axial resistance to flow in order to minimize pressure drop during both vapor production and reheat regeneration processes. An axial flow resistance can be defined for orthotropic resistance such as laminar flow through small channels:

$\frac{\Delta p}{L}{\int }_0^L\left[\frac{8\pi \mu {\rho }_c}{{\phi }_x^2}\left(\frac{\mathrm{GHSV}}{3600\sec \text{/}\mathrm{hr}}\right)\right]x={\int }_0^L\left[\frac{128\mu }{\pi {\rho }_cd_c^4}\left(\frac{\mathrm{GHSV}}{3600\sec \text{/}\mathrm{hr}}\right)\right]x$ $\mathrm{units}\text{:}\Delta p=\left[N/m^2\left(\mathrm{Pa}\right)\right]$ ${\rho }_{c}=\left[\mathrm{cells}\text{/}m^2\right]$ $\mu =\left[N\sec \text{/}m^2\right]$ $\mathrm{GHSV}=\left[{\mathrm{hr}}^{-1}\right]$ $L=\left[m\right]$ $d_c=\left[m\right]$ ${\phi }_x=\left[\mathrm{none}\right]$

Here, Δp is the pressure drop from frictional resistance, L is the averaging spatial length, dx is the local incremental axial distance, φ_x is the local void fraction, GHSV is the gaseous hourly space velocity (in units of hr⁻¹), λ is the fluid viscosity, and ρ_c is the cell number density per unit area (number of channels per unit cross sectional area of the device). All values are taken to be functions of the local coordinate, such that the overall pressure drop is the integral contribution from all axial sections of the device.

The pressure drop is constrained by the channel size, the number density of cells per unit area, the porosity, and the GHSV (flow per unit volume) of the device. The porosity is related to the channel size by the expression

$d_c^2=\frac{4{\phi }_x}{\pi {\rho }_{c}}$

The physical parameters of the device are selected in a manner to meet the design constraints of the system for operation. For those skilled in the art, for a device designed for a wide dynamic range of operating conditions, the design condition is selected based on maximum flow conditions in the regions of the device with the smallest flow passageways.

A feature of the device for successful operation is that it operate with a low axial resistance to flow, such that the overall average Δp per unit length of the device is less that 5 psi/inch. A range of parameters which is acceptable for the device allow a pressure drop between 0.01 and 5 psi/inch, with a preferred range between 0.03 and 1 psi/inch pressure drop.

In one embodiment, the device may be constructed using an arrangement of thin, corrugated sheets of various metallic composition rolled into a tight concentric rings or stacked into closely spaced layers. The corrugation geometry generates a series of small annular cells. The diameter of the cells may be varied by changing the thickness of the rolled material sheets and the density of the corrugated concentric rings or sheets, and by varying the tightness of packing along its axial length.

In one example of this embodiment, the material is comprised of sheets of Fecralloy® metal. The corrugation and sheet thicknesses for this design are selected to result in an overall porosity (open volume) of approximately 40% for the low void volume outlet region cross-sections. The low design porosity is used to meet the design requirements for maximum metal mass (for heat capacity and energy storage at the highest flow rate conditions). A high porosity inlet region has a porosity of approximately 80%. This high porosity enables a significantly higher degree of liquid penetration into the interior volume of the device. The interior of the device comprises a monolith having an intermediate value of porosity of about 60%, which is utilized as a transition between the low porosity inlet and high porosity outlet regions of the device. As noted earlier, the sheets are made of continuous pieces of material in the axial direction, but of varying lengths.

The cellular design of the device is such as to provide minimal pressure drop from the supplied fluid streams in either direction. Low pressure drop operation is particularly useful for applications involving reheating of the device by high velocity, high temperature gaseous streams which would incur substantial pressure drop losses.

This device is particularly useful with a cyclic endothermic steam reforming process with an exothermic regeneration process to produce synthesis gas. In the reactor bed configuration, a mixture of liquid hydrocarbon fuel and liquid water are injected using electronic fuel injectors onto the top (inlet) surface and into the internal volume of the vaporizer. The vaporized mixed stream then flows downward through a Gas Mixer and subsequently into the Reaction Zone where the feed is steam reformed into synthesis gas, using energy in the bed previously deposited from the reheat portion of the cycle. This synthesis gas passes out of the device at the bottom and can be utilized externally.

In the reheat portion of the cycle, a carbon monoxide, hydrogen, and possibly also fuel mixture is combusted. This high temperature stream is then used to reheat the catalyst bed of the Reaction Zone. At the end of the regenerative phase of the process, the cycle returns to liquid injection mode. These two cycle processes in this configuration are equivalent to the injection phase and reheat phase of the invention noted earlier.

Claims

1. A device for converting a liquid feed stream to a gaseous vapor stream comprising: φ x = void   fraction   at   distance   d = A ′  ( x ) A  ( x ) and a average void fraction along a length of device L φ a = average   void   fraction = ∫ O L  A ′  ( x )   x ∫ O L  A  ( x )   x said average void fraction ranging from about 0.3 to about 0.95.

(a) channel means having a first and second end, said channel means having a plurality of channels connecting said first and second end, said channel means having a substantially solid region and a void region,

(b) inlet means for directing the liquid feed stream to the first end of the plurality of channels, and

(c) outlet means for directing the gaseous vapor stream from said plurality of channels, where said channels have, at any distance d between the inlet and outlet, a geometric configuration, perpendicular to the feed flow direction characterized by 1. a void area A′(x) and, 2-channel total cross-sectional area A(x), where the void area A′(x) as a fraction of the total area A(x) is

2. The device of claim 1 where the void fraction varies along the length of the device from a void fraction at the inlet means ranging from about 0.5 to about 0.995, to a void fraction at the outlet means ranging from about 0.2 to about 0.7.

3. The device of claim 2 wherein the void fraction variation along the length of the device ranging from about 0.01 to about 0.5 void fraction decrease per linear inch of length.

4. The device of claim 3 wherein the variation ranges from about 0.15 to about 0.35 void fraction decrease per linear inch of length.

5. The device of claim 1 where the void fraction decreases from the inlet means to the outlet means in sequential constant void fraction regions numbering greater than one and less than twenty.

6. The device of claim 5 where the number of constant void fraction regions ranges from three to ten.

7. The device of claim 1 wherein the channels are further characterized as having a channel hydraulic diameter, dH, that ranges from about 0.1 to about 10.0 millimeters at the inlet, to about 0.2 to about 0.5 at the outlet, and a channel length.

8. The device of claim 7 wherein the channel hydraulic diameter ranges from about 0.3 to about 5.0 millimeters at the inlet, to about 0.4 to about 2.0 millimeters at the outlet.

9. The device of claim 7 wherein the ratio of the channel hydraulic diameter to channel length is between about 0.5 and about 10,000.

10. The device of claim 9 where the ratio is between about 10 and about 5000.

11. The device of claim 10 where the ratio is between about 40 and about 200.

12. The device of claim 1 where the channels are further characterized as having an average surface area per unit volume Sv,avg ranging from about 10 in2/in3 to about 2000 in2/in3.

13. The device of claim 12 where Sv,avg ranges from about 20 in2/in3 to about 1000 in2/in3.

14. The device of claim 13 where Sv,avg ranges from about 50 in2/in3 to about 250 in2/in3.

15. The device of claim 1 wherein the channels are made of materials having a thermal heat capacity of at least 100 J/Kg-K, therm and conductivity of at least about 10 W/m-K, and density of at least about 2500 Kg/m3.

16. The device of claim 15 wherein the thermal heat capacity is at least about 500 J/Kg, thermal conductivity of at least about 50 W/m-K, and density of at least about 5000 Kg/m3.

17. The device of claim 16 wherein the thermal heat capacity is at least about 100 J/kg, thermal conductivity of at least about 200 W/m-K, and density of at least about 7000 Kg/m3.

18. The device of claim 1 wherein the device operates at a gaseous hourly space velocity greater than about 500.

19. The device of claim 18 wherein the device operates at a gaseous space velocity above about 1000.

20. The device of claim 1 wherein the channels have a pressure drop, Δp, that is less than about 5 psi/inch.

21. The device of claim 20 wherein Δp is less than about 1 psi/inch.