TUBULAR REACTORS

Abstract

The invention relates to tubular reactors. In particular the invention provides for a reactor internal component for a fixed bed reactor which is axially receivable within a portion of an internal reaction cavity of a reactor tube. The reactor internal component includes a tubular insert, having a tubular wall with an outer surface shaped and dimensioned to fit into the internal reaction cavity of the reactor tube, the tubular insert having an inner passage of varied diameter which is operable to change a profile of the internal reaction cavity, in use to improve temperature distribution in a catalyst bed provided within the internal reaction cavity of the reactor tube.

Claims

1. A reactor internal component for a fixed bed reactor, axially receivable within a portion of an internal reaction cavity of a reactor tube, which includes a tubular insert, having a tubular wall with an outer surface shaped and dimensioned to fit into the internal reaction cavity of the reactor tube, the tubular insert having an inner passage of varied diameter which is operable to change a profile of the internal reaction cavity, the tubular insert has two ends, a first end operable to be positioned before a second end, relative to the direction of flow in the fixed bed tubular reactor, such that the flow is from the first end to the second end, the tubular insert having a neck portion positioned between the two ends, defined where an inner diameter of the inner passage is the smallest, the neck portion separates the tubular insert into a funnel portion and a functional tube portion, the funnel portion is defined by a section of the tubular insert between the first end and the neck portion and the functional tube portion is defined by a section of tubular insert between the neck portion and the second end, at the functional tube portion, the inner passage of the tubular insert gradually increases in diameter in an axial direction, from the neck portion to the second end.

2. The reactor internal component as claimed in claim 1, in which a diameter of the outer surface of the tubular insert is constant throughout the length of the tubular insert, with the tubular wall being of varying thickness to provide the varied diameter of the inner passage.

3. The reactor internal component as claimed in claim 1, in which the change in the profile of the internal reaction cavity includes decreasing an internal diameter of the internal reaction cavity in at least a portion of the reactor tube.

4. The reactor internal component as claimed in claim 1, in which an outer diameter of the tubular insert either: matches or is slightly less than an internal diameter of the reactor tube in which it is to be installed, such that the reactor internal component fits snugly into the reactor tube.

5. The reactor internal component as claimed in claim 1, in which, at the funnel portion, the inner passage of the tubular insert decreases in diameter in an axial direction from the first end to the neck portion, the funnel portion operable to function as a draft tube for gaseous reactants.

6. The reactor internal component as claimed in claim 1, in which the gradual increase in diameter of the inner passage in the functional tube portion is selected from any one or more: a linear increase, a stepped increase, a parabolic increase and a curved increase.

7. The reactor internal component as claimed in claim 1, in which the inner diameter at the neck portion is between 10% and 90% of the inner diameter of the reactor tube.

8. The reactor internal component as claimed in claim 1, in which the inner diameter at the neck portion is between 30% and 50% of the inner diameter of the reactor tube.

9. The reactor internal component as claimed in claim 1, in which the length of the reactor internal component is between 25% and 90% of the length of the reactor tube.

10. The reactor internal component as claimed in claim 1, in which the length of the functional tube portion is between 25% and 50% of the length of the reactor tube.

11. The reactor internal component as claimed in claim 1, in which the inner passage in the functional tube portion of the tubular insert is frustum shaped, which is operable to change the profile of the internal reaction cavity, which is normally cylindrical, to a frustum cavity.

12. The reactor internal component as claimed in claim 1, in which the reactor internal component is of a material with good thermal stability and high thermal conductivity, selected from any one of: metal, aluminium, steel, copper, an alloy, corundum, GH3044, metallic oxide, titanium, ceramic, silicon carbide, boron nitride, graphite and graphene.

13. A modified reactor tube for use in a fixed bed reactor, which includes a reactor tube having a cylindrical internal reaction cavity; and at least one reactor internal component, as claimed in claim 1, seated in the internal reaction cavity or forming part of a tubular wall of the reactor tube, which changes a profile of the internal reaction cavity, and decreases a diameter of the internal reaction cavity in at least a portion of the reactor tube, the at least one reactor internal component stabilizing the temperature distribution profile of the reactor tube when the fixed bed reactor is operational.

14. The modified reactor tube as claimed in claim 13, which includes catalyst particles in the internal reaction cavity providing a catalyst bed.

15. The modified reactor tube as claimed in claim 14, in which the at least one reactor internal component is located at an upstream section of the catalyst bed in the reactor tube.

16. The modified reactor tube as claimed in claim 15, in which the at least one reactor internal component is located such that the neck is positioned at the start of the catalyst bed.

17. A method of installing a reactor internal component to improve temperature distribution in a reactor tube of a fixed bed reactor, which includes providing a reactor tube with an internal reaction cavity; inserting at least one reactor internal component, as claimed in claim 1, into a portion of the reactor tube to change a profile of the internal reaction cavity, thereby providing a heat transfer improved internal reaction cavity; and filling the heat transfer improved internal reaction cavity with catalyst particles to provide a catalyst bed within the reactor tube.

18. The method as claimed in claim 17, which includes a prior step of removing a layer of ceramic balls on the upper side of the catalyst bed, and then removing a volume of catalyst particles to make space for the reactor internal component, before the reactor internal component is inserted.

19. The method as claimed in claim 17, in which the reactor internal component is inserted by axially aligning the internal component with the reactor tube, and sliding the internal component into the inner reaction cavity of the reactor tube.

20. The method as claimed in claim 17, in which the portion of the reactor tube into which the reactor internal component is inserted is proximate a top boundary of the catalyst bed.

21. The method as claimed in claim 18, which includes the later step of reloading the ceramic balls above the reactor internal component.

22. The method as claimed in claim 17, which includes increasing the length of the catalyst bed in the reactor tube, by reducing the volume of inert solid particles at ends of the reactor tube and replacing the volume with catalyst particles, to compensate for the volume of catalyst bed lost due to the volume taken up in the reactor tube by the reactor internal component.

Description

FIGURE(S)

[0059] In the figure(s):

[0060] FIG. 1 shows a design of one example of a reactor internal component;

[0061] FIG. 2 shows the reactor internal component shown in FIG. 1 installed in a reactor tube;

[0062] FIG. 3 shows three more examples of reactor internal components;

[0063] FIG. 4 shows a schematic diagram of axial-cross views of a tubular reactor without (a) and with (b) reactor internals installed;

[0064] FIG. 5 shows a temperature contour (a) in the tubular reactor and a comparison (b) of measured and predicted temperatures at different radial positions from the inlet of the tubular reactor;

[0065] FIG. 6 shows a comparison of CO consumption rates in the centre of the reactor for C1 and C3;

[0066] FIG. 7 shows a comparison of temperature contour in the tubular reactors of FIG. 4, without (a) and with (b) reactor internals installed under Fischer-Tropsch synthesis conditions; and

[0067] FIG. 8 shows a graph of axial temperature distribution in the tubular reactors of FIG. 4, without (blue) and with (red) reactor internals installed.

[0068] In the figures, like reference numerals denote like parts of the invention unless otherwise indicated.

EMBODIMENT OF THE INVENTION

[0069] In the figures, reference numeral (10) refers to an example reactor internal component for a tubular fixed bed reactor, in accordance with the invention. The reactor internal component (10) is insertable within a portion of an internal reaction cavity (52) of a reactor tube (50) (see FIG. 2). As shown in FIG. 1, the reactor internal component (10) includes a tubular insert (12), having a tubular wall (12.1) with an outer surface (12.2) shaped and dimensioned to fit into the internal reaction cavity (52) of the reactor tube (50). The tubular insert (12) has an inner passage (14) of varied diameter which is operable to change a profile of the internal reaction cavity (52), in use to improve temperature distribution in a catalyst bed (54) provided within the internal reaction cavity (52).

[0070] As best shown in FIG. 1.2 on the left-hand side, the diameter of the outer surface (12.2) of the tubular insert (12) is constant throughout the length of the tubular insert (12), is cylindrically shaped, and dimensioned to fit securely into a reactor tube (50). An outer diameter of the tubular insert (12) is therefore matched to the inner diameter (102) of the reactor tube (50) in which it is installed, such that the reactor internal component (12) fits snugly into the reactor tube (50). The tubular insert (12) has an inner surface (12.3) with a diameter smaller than the inner diameter (102) of the reactor tube (50). The tubular wall (12.1) is of varying thickness to provide the varied diameter of the inner passage (14). In the example shown in FIG. 2, the reactor internal component (10) changes the profile (shape and dimension) of a portion of the cylindrical internal reaction cavity (52) to a conical frustum cavity.

[0071] FIG. 3 shows three different examples of reactor internal components (10) in accordance with the invention. It is to be appreciated that the design of the reactor internal component (10) is not limited in these examples. As can be seen in these figures, the change in the profile of the internal reaction cavity (52) includes decreasing the internal diameter of the internal reaction cavity (52) in at least a portion of the reactor tube (50). The reactor internal components (10) are operable to narrow the passage (internal reaction cavity (52)) in the reactor tube (50), thereby decreasing heat build-up and resulting in more heat being removed when an exothermic reaction (e.g. Fischer-Tropsch Synthesis) takes place in the reactor. As shown in FIG. 2, the internal reaction cavity (52) receives catalyst particles which provides the catalyst bed (54). By decreasing the internal diameter of the internal reaction cavity (52), the amount of catalyst particles receivable in that section of the reactor tube (50) decreases and the heat transfer is intensified.

[0072] The reactor internal component (10) is axially receivable within the reactor tube (50), as shown in FIG. 2. In particular, the tubular insert (12) is placed at an upstream section (inlet, initial, entrance) of the catalyst bed in the reactor tube (50). Advantageously, this upstream section is the portion of the reactor tube (50) which is prone to hot spot formation, and in use, the tubular insert (12) reduces the temperature increase at that position.

[0073] As shown in FIGS. 1.2 and 2, the reactor internal component (10) has two ends, a first end (12.4) positioned before a second end (12.5), relative to the direction of flow in the tubular fixed bed tubular reactor, such that the flow is from the first end (12.4) to the second end (12.5). In use in a reactor, the first end (12.4) is positioned above the second end (12.5).

[0074] The tubular insert (12) has a neck portion (16) positioned between the two ends (12.4, 12.5). The neck portion (16) is defined where the inner diameter (104) of the inner passage (14) is the smallest and the tubular wall (12.1) is at its thickest. The inner diameter (104) at the neck portion (16) is between 10% and 90% of the diameter (102) of the reactor tube (50), preferably between 30% and 50%. In this example, the inner diameter (106) at the neck portion (16) is 50% of the diameter (102) of the reactor tube (50).

[0075] The neck portion (16) separates the tubular insert (12) into a funnel portion (18) and a functional tube portion (20). The funnel portion (18) is defined by a section of the tubular insert (12) between the first end (12.4) and the neck portion (16). The functional tube portion (20) is defined by a section of tubular insert (12) between the neck portion (16) and the second end (12.5).

[0076] The funnel portion (18) functions as a draft tube for gaseous reactants which affects the flow dynamics and pressure drop of the catalyst bed (54). The funnel portion (18) forms a constriction between the internal reaction cavity (52) proximate a first end (12.4) of the tubular insert (12) in which a ceramic ball layer is disposed, and the internal reaction cavity (52) at the functional portion (20) of the tubular insert (12) in which catalyst particles are provided, creating a venturi effect. At the funnel portion (18), the inner passage (14) of the tubular insert (12) decreases in diameter from the first end (12.4) to the neck portion (16). As best shown in FIG. 2, the diameter (106) of the inner passage (14) at the first end (12.4) is substantially matched to the inner diameter (102) of the reactor tube (50). To achieve this, the tubular wall (12.1) increases in thickness from the first end (12.4) to the neck portion (16).

[0077] At the functional tube portion (20), the diameter of the inner passage (14) of the tubular insert (12) gradually increases in the axial direction, from the neck portion (16) to the second end (12.5). The diameter (108) of the inner passage (14) at the second end (12.5) is substantially matched to the inner diameter (102) of the reactor tube (50). To achieve this, the tubular wall (12.1) decreases in thickness from the neck portion (16) to the second end (12.5). The axial length (110) of the functional tube portion (20) is between 25% and 100% of the length of the reactor tube (50). In this example, the axial length (110) of the functional tube portion (20) is 33% of the length of the reactor tube (50). It is to be appreciated that this reflects only one example of axial length (110) and other lengths of the functional tube portion (20) can be used depending on the reactor tube (50) length.

[0078] In the example shown in FIG. 2, the reactor tube (50) has a 50 mm diameter (102) and a 1000 mm height, and the tubular insert (12) has a length of 251 mm and the neck portion (16) of the tubular insert (12) has a diameter (104) of 25 mm. Again, this is only one example of the dimensions of the tubular insert (12) which can differ depending on the dimensions of the reactor tube (50).

[0079] FIG. 3 shows another three examples of the reactor internal component (10). The type or shape of the gradual increase in diameter of the inner passage (14) in these examples is a linear increase (FIG. 3.1), a stepped increase (FIG. 3.2), and a parabolic increase (FIG. 3.3), respectively.

[0080] The reactor internal component (10) is of a material with good thermal stability and high thermal conductivity. In these examples, the material is selected as copper. In use, the reactor internal component (10) improves temperature distribution in the catalyst bed (54) during an exothermic reaction.

[0081] An example of the design and evaluation which led to the present invention follows, with a condensed description of the outcome thereunder.

Full Study: Design and Evaluation

[0082] Design of ring and tube type internals:

[0083] Since the concentration of reactants decreases in the direction of flow in a fixed bed reactor, the reaction rate is relatively high at the entrance to the catalyst bed. Higher reaction rates result in an increased rate of release of reaction heat which in turn, increases the local temperature and consequently accelerates the reaction rate. Usually a hot spot firstly forms in the initial part of the catalyst bed, because the rate of reaction heat release exceeds the heat removal capacity of reactor..sup.2, 23

[0084] Reactor internal components were designed to be placed in the inlet section of the catalyst bed, so as to partially change the effective inner diameter of the bed along the axial direction. FIG. 1 is a drawing of the assembled reactor internal component, while FIG. 2 shows the axial section view. As can be seen in FIGS. 1 and 2, the proposed reactor internal component comprises a ring & tube type structure, with the outer diameter designed to fit perfectly into the inside of the reactor tube (shown best in FIG. 2), while the inner diameter varies in the axial direction. The neck position, where the inner diameter is the smallest, is at the top boundary of the catalyst bed and divides the internals into two parts, namely: the outer part (funnel portion) which works as a draft tube for the gaseous reactants; and the inner part (functional portion) which is the functional part. The configurations of the “draft tube” part, affects the flow dynamics and pressure drop. In the functional portion, the inner diameter increases linearly in the axial direction, which means that the effective reactor tube diameter is adjusted. Since the desired characteristics of the reactor internal component should include: good thermal stability, high thermal conductivity, and being cost-effective and easy to manufacture, the available material for manufacturing can be copper (which was used in the simulation), aluminum, steel, titanium, metallic oxide, steel, alloy, corundum, GH3044 alloy. Nonmetal materials can also be used, including ceramics, silicon carbide, boron nitride, graphite and graphene.

[0085] FIG. 2 also shows that it is easy to assemble the reactor internal component (or “internals”). The process is: firstly remove the layer of ceramic ball on the upper side of the catalyst bed (if applicable); remove a volume of catalyst particles equal to the volume of the cavity of the conical frustum of the internals; insert the internals inside the tube; lastly, fill the internal cavity with catalyst particles and re-load the ceramic balls (if used) above the insert. Obviously, if it is required to keep the total volume of catalyst in the reactor tube constant, the total length of the catalyst bed will be longer after the internals are installed. However, given that both ends of the catalyst bed are normally packed with inert solid particles, there is typically some flexibility to pack the reactor tube fully, and the required increase in catalyst bed height should be containable. Thus, no additional modification to the original reactor or additional operational procedures are needed to install the internals.

[0086] As shown in FIG. 2, the effective inner diameter of the reactor tube is directly influenced by the neck diameter (D.sub.neck); while its rate of increase is dependent on the length of the cavity of the conical frustum (h). In order to compare reaction performance with and without the internals, the total amount of catalyst loaded should be maintained, i.e. the volume of the cavity of the conical frustum should be equal to the volume of replaced cylindrical shaped catalyst bed. The equations describing the volume of the cavity of the conical frustum (Equation 1) and the volume of cylinder (Equations 2) are as follows:

[00001]$\begin{array}{cc}{V}_{\mathrm{con}}=\frac{1}{3}\pi h\left(D_{\mathrm{neck}}^2+R^2+D_{\mathrm{neck}}R\right)& \left(1\right)\end{array}$$\begin{array}{cc}{V}_{\mathrm{cyl}}=\pi R^{2}H& \left(2\right)\end{array}$

[0087] Where V.sub.con and V.sub.cyl represent the volume of the cavity of the conical frustum and the cylinder respectively, mm.sup.3; H is the length of the replaced cylindrically shaped catalyst bed, mm; and R is the inner diameter of the reactor tube, mm. The length of the cavity of the conical frustum cavity height h is greater than H, because the total amount of catalyst is kept as constant, namely V.sub.con=V.sub.cyl. It is more convenient to use the volumetric proportion of the replaced cylindrically shaped catalyst bed v as the configuration variable of the reactor internals. Since both the h or H are proportional to V.sub.cyl, the configuration of either the conical frustum cavity or the replaced cylindrically shaped catalyst bed can be determined in terms of the same variable. The proportion v was varied from 15% to 25% in this study for the design of the internals. The values of the variables used for the different simulations (C1 to C6) are summarized in Table 1 below. In the table, C1 indicates the blank case, in which no ring & tube type internals were used.

TABLE-US-00001 TABLE 1 Summary of specifications for the different simulations NO. C1 C2 C3 C4 C5 C6 v/% — 20 20 20 15 25 a/mm — 13 25 38 25 25 h/mm — 264 201 150 150 251 Note: v represents the conical frustum cavity volume proportion of the total catalyst bed.
Reactor model and validation:
Reactor model:

[0088] The reactor model was built based on a practical bench scale TFBR which was 50 mm in diameter and 1000 mm in height. Its geometry is shown in FIG. 1, and it can be seen that it consists of two parts: a catalyst bed on the tube side; and an annular oil bath on the shell side. When simulating the reactor with the reactor internal components installed, only the geometry of the model was changed. FIG. 2 shows a schematic of the reactor model with internals installed. The individual geometry is different for C2 to C6, since the specifications for the internals is different in each case, but the configuration of the reactor was kept constant. Assuming the mean bed void is constant, three solid porous zones were considered, namely a ceramic ball layer, the catalyst bed, and another layer of ceramic balls..sup.31 The physical properties of the ceramic ball layers and the catalyst bed are listed in Table 2 below.

[0089] Reynolds number for flow in the bed indicated that the flow is laminar, hence a laminar flow model was applied. The boundary between the reaction region and the oil bath region was set as a coupled wall, so that the corresponding heat transfer coefficients at different axial positions could be calculated based on the local fluid properties. The other walls adjacent to the atmosphere were set as adiabatic walls, since the experimental apparatus was covered by a layer of insulating material. The simulation software calculated the built-in governing equations, including the Navier-Stokes equation, energy balance, species balances, etc., in each individual cell of the model..sup.32 33 The SIMPLE algorithm was chosen for the Pressure-Velocity Couple scheme. The simulation results were regarded as convergent only when the calculated residuals were smaller than the absolute criterion of 10.sup.−6.

TABLE-US-00002 TABLE 2 Physical properties of the ceramic ball layer and the catalyst bed Ceramic ball Catalyst Diameter/mm 3.6 ± 0.3 1.8 ± 0.2 Thermal conductivity/(W .Math. m.sup.−1 .Math. K.sup.−1) 1.2 1.4 Pecking density/(g .Math. ml.sup.−1) 1.35 0.69 Bed void/% 46 62

[0090] For a Co-supported catalyst FTS system, side reactions (for example the water gas shift reaction), can be neglected. The FTS products were assumed to be only alkanes, and methane and ethane were used to represent the C.sub.1 and C.sub.2 product respectively. Because the formation rates of the C.sub.3+ products (hydrocarbons with carbon number bigger than 3) are dependent on the C.sub.2 product and chain growth probability, it is acceptable to use pentane to represent the C.sub.3+ products, for the purpose of simplifying the reaction kinetics. All the reactants and products are considered to be in the gas phase. The semi-empirical kinetics employed in this investigation were the same as that used in the previous study..sup.34 35 The reaction scheme is summarized in Table 3 below, and the equations for the CO consumption rate (Equation 4) and product formation rates are listed in Equations 5-7.

TABLE-US-00003 TABLE 3 FTS reaction scheme Reaction R.sub.1 CO + 3H.sub.2 .Math. CH.sub.4 + H.sub.2O R.sub.2 2CO + 5H.sub.2 .Math. C.sub.2H.sub.6 + 2H.sub.2O R.sub.3 5CO + 11H.sub.2 .Math. C.sub.5H.sub.12 + 5H.sub.2O

r.sub.FT=k.sub.1.Math.exp(−E.sub.1/RT).Math.C.sub.CO.Math.C.sub.H.sub.2/(1+k.sub.2.Math.exp(−E.sub.2/RT).Math.C.sub.CO).sup.2  (3)

r.sub.CO=−r.sub.FT  (4)

r.sub.C.sub.1=k.sub.3.Math.exp(−E.sub.3/RT).Math.r.sub.FT  (5)

r.sub.C.sub.2=k.sub.4.Math.exp(−E.sub.4/RT).Math.r.sub.FT  (6)

r.sub.C.sub.3+=[1−k.sub.3.Math.exp((−E.sub.3/RT)−2×k.sub.4.Math.exp(−E.sub.4/RT)].Math.r.sub.FT/5  (7)

[0091] Where r.sub.FT is the FTS reaction rate, kmol/(m.sup.3.Math.s); r.sub.CO is the CO consumption rate, kmol/(m.sup.3.Math.s); r.sub.C.sub.1, r.sub.C.sub.2 and r.sub.C.sub.3+ are the formation rates of methane, ethane and pentane, respectively; C.sub.CO and C.sub.H.sub.2 are the concentrations of CO and hydrogen. The mass balance of the model was checked, and the values of the eight constant coefficients used in the model are listed in Table 4 below. The four pre-exponential factors (k.sub.1-k.sub.4) were adjusted according to the FTS experimental results, while the activation energies (E.sub.1-E.sub.4) are taken from the literature.sup.36 37 38.

TABLE-US-00004 TABLE 4 List of kinetics parameters used in this study k.sub.1 k.sub.2 k.sub.3 k.sub.4 E.sub.1(kJ/mol) E.sub.2(kJ/mol) E.sub.3(kJ/mol) E.sub.4(kJ/mol) parameters 4.94 × 10.sup.9 4.68 8.58 × 10.sup.7 1.08 × 10.sup.3 100 20 81 49

[0092] All the simulations were conducted under the same conditions, namely: 1200 ml of combined catalyst and ceramic (made up of 300 ml 15% Co—SiO.sub.2 catalyst and the balance being ceramic balls); 458 K operating temperature; 20 bar operating pressure; flowrate of 1.5 Nl/min reactant mixture (H.sub.2/CO=2).

Model validation:

[0093] The model validation was done by comparing the simulation results obtained from the blank case (C1) to the experimental results. The comparison of the CO conversion and product selectivity are given in Table 5 below. As can be seen, the relative error in the CO conversion is only 8.5%, and the predicted selectivity is even more accurate, therefore it is concluded that the reaction kinetics are reliable and are suitable for describing the actual FTS reaction. More importantly, the heat transfer behaviour in the catalyst bed was also validated by comparing the predicted and measured temperature profiles. A specially designed temperature measurement system was implemented in the experiment set-up where axial temperature was measured at different radial position in the reactor, corresponding to radii of 8.5 mm, 17 mm and 21 mm respectively. FIG. 6(a) shows the predicted temperature contour in the reactor. The position of the catalyst bed is denoted by the region inside the dashed-lines, and the positions of the thermocouple are also indicated (P=8.5 mm; P=17 mm; P=21 mm). The comparison of the experimental and the predicted temperature profiles, at corresponding positions, are shown in FIG. 6(b). The maximum mean absolute error of all the temperature data is only 3.2 K, which is considered acceptable when compared to the operating temperature of 458 K. Thus, it is reasonable to assume that the heat transfer behaviour is well described. As mentioned above, only the geometry changes in the different simulation cases C2 to C6, and the inventors believe that the use of the reactor internals does not affect the simulation methodology. Therefore, the inventors claim that the simulations described in this study are reliable.

TABLE-US-00005 TABLE 5 Comparison of experimental results and simulation data for the blank case (C1) Experimental Simulation Relative result data error/% CO conversion/% 48.3 52.4 −8.5 S.sub.CH4/% 7.04 6.93 1.56 S.sub.C2/% 0.68 0.68 0 S.sub.C3+/% 92.3 91.7 0.61

[0094] Results and discussion:

[0095] Predicted performance of internals of different geometries:

[0096] Various neck diameter and frustum cavity height options were investigated ceteris paribus. The simulation results are summarized in Table 6 and Table 7 below. In order to evaluate the performance of different reactor internal components, the rate of change (R) is defined as:

R=(A.sub.int−A.sub.org)/A.sub.org×100%  (8)

[0097] Where: A can be maximum temperature increase ΔT.sub.MAX, CO conversion X.sub.CO, selectivity of methane S.sub.C1 or selectivity of C.sub.3+ products S.sub.C3+—depending on its use; the subscripts org and int indicate whether the parameter A refers to the original (org) tubular reactor (also known as the blank case C1) or the tubular reactor with the reactor internal component internals (int) installed (C2 to C6) respectively; A negative value of R indicates that the parameter decreases when reactor internal components are used.

[0098] A comprehensive comparison of the simulation results for the blank case (C1) to those with reactor internal components installed (C2 to C6) is given in Tables 6 and 7. These show that: the ΔT.sub.MAX in C2 to C6 dropped as the R is negative, reaching as low as −22.6% in C6; the CO conversion was almost constant as the rate of change is no more than 2.1%. Thus, we can conclude that: the temperature rise was inhibited by applying the reactor internal components; while the CO conversion was not significantly affected; the average temperature of the catalyst bed (T.sub.AVE) decreased in the case of C2-C6; the methane selectivity declined slightly; S.sub.C3+ increased compared to C1.

[0099] Table 6 shows that when increasing the neck diameter from 13 to 38mm, both ΔT.sub.MAX and T.sub.AVE showed a minimum. The lowest value of ΔT.sub.MAX was obtained in case C3 with D.sub.neck of 25mm. The reaction performance of FTS is directly related to the temperature of the catalyst bed, thus S.sub.C1 increased with increasing T.sub.AVE, while S.sub.C3+ showed the opposite trend. In addition, although the values of R.sub.XCO R.sub.SC1 and R.sub.SC3+ were quite small, the changes were obvious when decreasing D.sub.neck from 38 mm to 25 mm. This means that the FTS results were more sensitive in this range.

TABLE-US-00006 TABLE 6 The performance of internals with different neck diameters original internals NO. C1 C2 C3 C4 D.sub.neck/mm — 13 25 38 ΔT.sub.MAX/K 16.9 14.5 14.0 14.7 R.sub.ΔTmax/% — −14.1 −16.9 −12.8 T.sub.AVE/K 463.4  463.2 463.1 463.4 X.sub.CO/% 51.8 51.2 51.1 51.7 R.sub.Xco/% — −1.19 −1.34 −0.12 S.sub.C1/%  6.83 6.58 6.55 6.74 R.sub.SC1/% — −3.65 −4.09 −1.35 S.sub.C3+/% 92.5 92.8 92.7 92.6 R.sub.SC3+/% — 0.29 0.25 0.10

TABLE-US-00007 TABLE 7 The performance of internals with different proportions of catalyst being replaced original internals NO. C1 C5 C3 C6 v/% — 15 20 25 ΔT.sub.MAX/K 16.9 14.8 14.0 13.1 R.sub.ΔTmax/% — −12.3 −16.9 −22.6 Tave/K 463.4  463.2 463.1 462.9 X.sub.CO/% 51.8 51.3 51.1 50.7 R.sub.Xco/% — −0.99 −1.34 −2.13 S.sub.C1/%  6.83 6.63 6.55 6.46 R.sub.SC1/% — −3.03 −4.09 −5.41 S.sub.C3+/% 92.5 92.6 92.7 93.0 R.sub.SC3+/% — 0.20 0.25 0.48

[0100] As indicated in Table 7, T.sub.AVE declined gradually as the proportion of the original cylinder shape catalyst bed was changed from 15% to 25%, while the ΔT.sub.MAX dropped as low as 13.1 K with the change in rate of −22.6% for C6. The results demonstrate that a longer frustum cavity results in a lower peak temperature within the catalyst, as well as better product distribution for longer chain hydrocarbons. Furthermore, the methane selectivity decreased from 6.63% to 6.46%, while the C.sub.3+ products selectivity rose slightly from 92.6% to 93.0% (see Table 7). However, the CO conversion correspondingly dropped slightly from 51.3% to 50.7% which was caused by the lower average catalyst bed temperature; and the maximum changing rate was only −2.13%.

[0101] When the reactor internal component was applied to an existing TFBR, the packed catalyst volume was usually kept constant to maintain the reactor productivity at the same level. Therefore, there were slight increases in the catalyst bed height. The total height of the catalyst bed (H) in the different cases and their corresponding rate of change for each (R.sub.H) are listed in Table 8. We can see that H increased with decreasing neck diameter D.sub.neck or when enlarging the proportion of replaced original cylinder shape catalyst bed (v). Normally, there is extra space at both ends of the catalyst bed for the layers of inert solid supports. Thus, the fixed bed reactor may be designed to 1.5 times longer at most than its catalyst bed. However, actual TFBR design varies from case to case, and the available extra space in TFBR for the application of the reactor internal components cannot be determined in general. For example, in this experiment setup, inserting the reactor internal component increased the bed height by 20% of the total catalyst bed height at most, which is acceptable.

TABLE-US-00008 TABLE 8 Summary of the total catalyst height (H) in different cases and the corresponding rate of change (R.sub.H) in different cases NO. C1 C2 C3 C4 C5 C6 H/mm 585 732 669 618 647 690 R.sub.H/% — 25.1 14.4 5.6 10.6 17.9

[0102] The results suggest that, when designing reactor internal component for a new tubular reactor or when modifying an existing reactor, the neck diameter D.sub.neck should be optimized; and while a bigger proportion of replaced original cylinder catalyst bed is preferred, the actual value should be determined according to the available tube length.

Mechanism:

[0103] There are two mechanisms that reduce the maximum temperature when using the reactor internal components. The simulation results for C1 and C3 can be used as examples for comparison and the axial CO consumption rate along the centre of the catalyst bed in C1 and C3 is shown in FIG. 6. On the one hand, the reactor internal components increase the heat removal capacity by partially reducing the effective reactor diameter. Comparing the temperature increase in the thermal oil between the inlet and the outlet in C1 and C3 supports that there is increased heat removal (see Table 9). The initial inlet temperature for the thermal oil is the same in both cases, therefore, a higher temperature in the thermal oil at the outlet indicates that more reaction heat is removed. One can see that the temperature difference increased slightly from 0.0623 K (C1) to 0.0704 K (C3) and thus the internals resulted in more reaction heat being removed, even though the total amount of reaction heat released in C3 is lower than that in C1, since CO conversion in C3 decreased by 1.34%, as shown in Table 7.

TABLE-US-00009 TABLE 9 Temperature of thermal conductive oil at inlet and outlet, and temperature difference C1 C3 T.sub.inlSt/K 458.00 458.00 T.sub.outlSt/K 458.06 458.07 ΔT/K 0.0623 0.0704

[0104] On the other hand, the reaction intensity at the “critical” zone was dispersed over a longer axial distance (see FIG. 6) as the original cylindrical shaped catalyst bed (in C1) was replaced with a longer conical shaped bed in C3. It is inevitable that a large amount of heat is released in the FTS process as the reaction is extremely exothermic. An acceptable method to control the temperature of the hot spot is to reduce the rate of release of reaction heat in the initial part of catalyst bed by distributing the reaction heat over a longer axial distance. Although the volume of the cavity of the conical frustum was the same as that of the replaced cylindrical shaped catalyst bed, the catalyst bed packed in C3 had a longer and narrower shape, which means that the reaction rate and the reaction heat release rate per volume of reactor will be lower in the initial part of the catalyst bed, thereby reducing the undesirable temperature rise.

Conclusions:

[0105] In the present invention, a new reactor internal component was developed (ring & tube type internals), to inhibit the hot spot formation in a catalyst bed in FTS. A CFD model showed that modifying a reactor tube with this insert reduced the maximum temperature of the hot spot and improved the selectivity of C3+ products. The reactor model was based on an actual bench-scale TFBR with a 50 mm diameter and 1000 mm length. It was validated by choosing parameters so as to fit both the measured reaction conversions and selectivities. The measured temperature profiles from experiments conducted under typical low temperature FTS conditions with a cobalt catalyst where compared to the predicted profiles, and it was shown that the model predicted both the axial and radial temperature profiles which validated the simulations. By using a blank case (corresponding to no reactor internals) for comparison purposes, it was shown that the internals inhibited hot spot formation in the catalyst bed, while having little effect on the overall FTS reaction rate, i.e.: the maximum temperature in the catalyst bed dropped 22.6% when using internals (Case 6); while the rate of change in CO conversion was less than 2.13%. Furthermore, because of the reduced maximum temperature in the bed, the C.sub.3+ product selectivity increased slightly when using internals. The effect of the specifications of the internals, namely the diameter at the neck position D.sub.neck and conical frustum cavity height h was investigated. For the purposes of comparison, the inventors kept the amount of catalyst used constant in all the simulations and this resulted in the length of the catalyst bed varying for the different cases. The maximum temperature ΔT.sub.MAX in the bed showed a minimum with varying diameter of the neck of the insert D.sub.neck and an increasing trend with increasing length of cavity of conical frustum h. The overall reaction rate was not very sensitive to the presence of the reactor insert. The internals essentially reduced the effective inner diameter of the reactor tube, which enhanced the heat removal capacity and dispersed the heat release over the hot spot region over a longer axial distance. Given the design of ring & tube type internals, other benefits include ease of manufacturing, simple assembling and disassembling.

Condensed Results and Outcome

[0106] The reactor internal component (or ring and tube type internal), which has a linear increase in diameter in the functional tube portion (as shown in FIG. 3.1), was thus verified in a fixed bed tubular reactor by CFD simulation under typical Fischer-Tropsch synthesis conditions, which is strongly exothermic process. Before testing the internal component, a blank case study, which was without the reactor internal component installed in the reactor tube, was conducted. According to the blank case study, a 2D axisymmetric tubular reactor model with size of 50 mm in diameter and 1000 mm in length was developed by ANSYS Fluent 18.1. The catalyst bed, which was set as a porous zone in this model, was sandwiched by two ceramic ball layers. The Fischer-Tropsch synthesis included a series of reactions as described by a semi-empirical kinetics..sup.34 35 The SIMPLE algorithm was chosen for the Pressure-Velocity Couple scheme. Finally, this model in the blank case study was validated, with experiment results obtained under the conditions of: 20 bar, 458 K, Co-based catalyst, 300 ml catalyst, space velocity=300 h-1 and CO/H.sub.2 ratio is 2.

[0107] For the test of the reactor internal component, only the model geometry where an example internals was applied was changed while keeping the other parameters the same. FIG. 4 shows the schematic diagram of axial-cross views of a tubular reactor with (b) and without (a) reactor internals installed. As previously mentioned, the specifications of this example of the internals included: 25 mm in neck diameter and 330 mm in functional portion length, with the inner diameter of the internal reaction cavity changing in a linear manner. The simulation results are shown and compared in FIGS. 7 and 8, as well as Table 10 below. In particular, FIG. 7 shows a comparison of the temperature contours along the reactor tube with and without reactor internals. FIG. 8 compares the temperature plots of the reactor tube with and without internals. Table 10 below provides a comparison of the simulation results from tubular reactors with and without internals installed in a Fischer-Tropsch synthesis reaction, including the maximum temperature in the catalyst bed (T.sub.MAX/K), the maximum temperature rise (ΔT/K), the change rate of maximum temperature rise comparing to the blank case (change rate/%) and the CO conversion (X.sub.CO %).

TABLE-US-00010 TABLE 10 Comparison of simulation results from tubular reactors with and without internals installed in Fischer-Tropsch synthesis change T.sub.MAX/K ΔT/K rate/% X.sub.CO % Without internals 476.1 18.1 52.0 With internals 469.3 11.3 −22.6 50.7 300 ml Co-based catalyst, 20 bar, 458K, CO/H.sub.2 = 2, space velocity = 300 h.sup.−1

[0108] It is evident that the temperature rise in the catalyst bed can be improved. In particular, as shown, using the reactor internals can reduce the change rate of maximum temperature rise to as low as 22.6%, whilst simultaneously reducing the CO conversion slightly.

[0109] The inventor therefore believes that the present invention provides a novel tubular reactor internal component design which increases the heat transfer capacity across the tubular reactor wall, facilitating heat removal during highly exothermic reactions (e.g. FTS process) and reducing the temperature gradient in the catalyst bed. This allows for the maintenance of isothermal operation, preventing catalyst deactivation and increasing product selectivity. The reactor internal component design presents the further benefits of not substantially increasing the mass of the tubular reactor, and not causing a loss of volumetric efficiency of the tubular reactor, meaning that the reactor performance is not sacrificed. Advantageously, the present tubular reactor internal design can also be used directly and easily in existing TFBR applications. The invention further provides a reactor tube with such reactor internal components and a method of assembling same.