DEVICE FOR CARRYING OUT MATERIAL EXCHANGE PROCESSES

Abstract

The invention relates to an apparatus for carrying out mass transfer processes, comprising a column having at least two inlet pipes for introducing a gaseous phase, where separation-active internals are accommodated in the column and a column section extends from the at least two inlet pipes to the separation-active internals, in which section a coverage of a cross-sectional area of the column is less than 25%, based on the total cross-sectional area, and where the at least two inlet pipes have a height offset which corresponds to not more than three times an inlet pipe diameter and the at least two inlet pipes are at an angle (α) of from 60° to 150° to one another and have asymmetry with respect to one another. The invention further relates to a use of the apparatus and also a method for designing the apparatus.

Claims

1-38. (canceled)

39. An apparatus (1) for carrying out mass transfer processes, comprising a column (2) having at least two inlet pipes (3, 5) for introducing a gaseous phase, where separation-active internals (9) are accommodated in the column (2) and a column section (29) extends from the at least two inlet pipes (3, 5) to the separation-active internals (9), in which section a coverage of a cross-sectional area of the column (2) is less than 25%, based on the total cross-sectional area, and where the at least two inlet pipes (3, 5) have a height offset which corresponds to not more than three times the largest inlet pipe diameter (6) and the at least two inlet pipes (3, 5) are at an angle (α) of from 60° to 150° to one another and have asymmetry with respect to one another, wherein the asymmetry is given by the at least two inlet pipes (3, 5) each having different inlet pipe diameters (6, 17).

40. The apparatus (1) according to claim 39, wherein the asymmetry is given by the at least two inlet pipes (3, 5) being distributed asymmetrically around the circumference (19) of the column (2).

41. The apparatus (1) according to claim 39, wherein the angle (α) differs by at least 10° from a further angle (β) between two of the at least two inlet pipes (3, 5).

42. The apparatus (1) according to claim 39, wherein the at least two inlet pipes (3, 5) are arranged at the same height (8) on the column (2).

43. The apparatus (1) according to claim 39, wherein the at least two inlet pipes (3, 5) are arranged at the bottom of the column (2) or as side inlet on the column (2).

44. The apparatus according to claim 39, wherein the apparatus (1) comprises precisely two inlet pipes (3, 5) for introducing a gaseous phase, where the two inlet pipes (3, 5) have a height offset which corresponds to not more than three times the largest inlet pipe diameter (6).

45. The apparatus (1) according to claim 39, wherein the at least two inlet pipes (3, 5) open radially into the column (2).

46. The apparatus (1) according to claim 39, wherein the separation-active internals (9) comprise a structured packing and/or packing elements.

47. The apparatus (1) according to claim 39, wherein the separation-active internals (9) comprise trays without guided flow.

48. The apparatus (1) according to claim 39, wherein the separation-active internals (9) comprise crossflow trays.

49. The apparatus (1) according to claim 39, wherein vaporizers (20, 30) are attached via the at least two inlet pipes (3, 5) to the column (2).

50. A method comprising carrying out a mass transfer process in the apparatus (1) according to claim 39, for producing isocyanates, styrene or an alkyl acrylate.

51. The method according to claim 50, wherein the asymmetry is given by the flow velocities through the at least two inlet pipes (3, 5) being different.

52. The method according to claim 50, wherein the apparatus (1) for carrying out mass transfer processes is used as rectification column (40) in a process for the continuous production of an alkyl acrylate (H.sub.2C═CH—C(═O)OR, where R = n-butyl or isobutyl), where aqueous 3-hydroxypropionic acid is reacted under dehydrating and esterifying conditions in the presence of the appropriate butanol (R—OH) in a reactor comprising the rectification column (40) and butyl acrylate formed, unreacted butanol and also water used and water formed are distilled off as ternary azeotrope at the top and, after separation into a respectively liquid aqueous phase and liquid organic phase, the aqueous phase and the organic phase are each at least partly discharged and the organic phase comprising the butyl acrylate and the butanol is fractionally distilled.

53. The method according to claim 50, wherein a liquid phase (12) is taken off from the rectification column (40), at least partially vaporized and at least partly recirculated via the at least two inlet pipes (3, 5) to the rectification column (40).

54. The method according to claim 53, wherein the liquid phase (12) is at least partially vaporized in at least two vaporizers (20, 30).

55. The method according to claim 50, wherein the pressure at the top of the rectification column (40) is in the range from 0.2 bar to 5.0 bar absolute.

56. The method according to claim 52, wherein the fractional distillation of the organic phase comprising the butyl acrylate and the butanol is carried out by distilling off the butanol in an additional rectification column (40) and distilling off the butyl acrylate from the resulting bottoms in a further additional rectification column (40).

57. A method for designing the apparatus (1) for mass transfer according to claim 50, comprising the following steps: (a) specification of the position and orientation of the at least two inlet pipes (3, 5) on the column (2); (b) calculation of the gas flow in the column (2) using a flow simulation; (c) repetition of the steps (a) and (b) with different positions and orientations of the at least two inlet pipes (3, 5) and (d) selection of the position and orientation of the at least two inlet pipes (3, 5) at the flow which displays the most uniform flow pattern.

Description

[0102] Working examples of the invention are illustrated in the figures and are explained further in the following description.

[0103] The figures show:

[0104] FIGS. 1 to 5 cross-sectional views of a column having various inlet pipe arrangements,

[0105] FIG. 6 a three-dimensional depiction of a column having two inlet pipes,

[0106] FIG. 7 a three-dimensional depiction of a column having two inlet pipes of different diameter,

[0107] FIG. 8 a longitudinal sectional view of the column with inlet pipe,

[0108] FIG. 9 a histogram depicting the relative velocity on entry into separation-active internals in the form of a packing,

[0109] FIG. 10 frequency functions of the relative velocity on entry into separation-active internals in the form of a packing for different inlet pipe arrangements,

[0110] FIG. 11 frequency functions of the relative velocity on entry into separation-active internals in the form of trays for different inlet pipe arrangements,

[0111] FIG. 12 frequency functions of the relative velocity on entry into separation-active internals in the form of trays for different inlet pipe arrangements and unequal flow velocities in the inlet pipes,

[0112] FIG. 13 frequency functions of the relative velocity on entry into separation-active internals in the form of trays for different inlet pipe arrangements and unequal inlet pipe diameters,

[0113] FIG. 14 frequency functions of the relative velocities for one and two inlet pipes and

[0114] FIG. 15 a schematic depiction of an apparatus for carrying out mass transfer processes.

[0115] FIGS. 1 to 5 show cross-sectional views of a column 2 having various arrangements of inlet pipes 3, 5.

[0116] In FIG. 1, in accordance with the prior art, a first inlet pipe 3 and a second inlet pipe 5 are arranged at an angle α of 180° on the column 2. Furthermore, the inlet pipes 3, 5 are oriented in a radial direction 4 on the column 2.

[0117] The columns 2 depicted in FIGS. 2 and 3 also have precisely two inlet pipes 3, 5. Here, the angle α between the two inlet pipes 3, 5 is less than 180° . In FIG. 2, the angle α is 90° and in FIG. 3 the angle α is 120° . A further angle β between the two inlet pipes 3, 5 is 270° in FIG. 2 and 240° in FIG. 3. Accordingly, the two inlet pipes 3, 5 have inequality with one another owing to their nonuniform distribution around the circumference 19 of the column 2, since circumference sections having different lengths are formed between the inlet pipes 3, 5.

[0118] Three inlet pipes 3, 5, 25 for introduction of a gaseous phase are arranged on the column 2 in FIG. 4. The angle α between the inlet pipes 3, 5, 25 is 120° in each case and the inlet pipes 3, 5 are uniformly distributed over the circumference 19. This is an embodiment according to the prior art in so far as equal inlet pipe diameters 6, 17 and equal velocities in the inlet pipes 3, 5, 25 are present. In the case of at least two different inlet pipe diameters 6, 17 and/or at least two different velocities through the inlet pipes 3, 5, 25 arranged as depicted, an embodiment according to the invention is present.

[0119] FIG. 5 shows a column 2 having four inlet pipes 3, 5, 25, 27. These are uniformly distributed around the circumference 19 of the column 2 and the angle α between the inlet pipes 3, 5, 25, 27 is 90° in each case. This is likewise an embodiment according to the prior art, in so far as equal inlet pipe diameters 6, 17 and equal velocities in the inlet pipes 3, 5, 25, 27 are present. In the case of at least two different inlet pipe diameters 6, 17 and/or at least two different velocities through the inlet pipes 3, 5, 25, 27 arranged as depicted, an embodiment of the invention is present.

[0120] FIG. 6 shows a three-dimensional depiction of a column 2 having two inlet pipes 3, 5. To give a better overview, only the region of the column 2 in which the inlet pipes 3, 5 are arranged is shown.

[0121] Two inlet pipes 3, 5 for introduction of a gaseous phase are arranged on the column 2. The inlet pipes 3, 5 are at an angle α in the range from 60° to 150° to one another. In the embodiment depicted here, the angle α is 90° . In particular, no inlet pies 3, 5 are arranged directly opposite one another in the arrangement of the inlet pipes 3, 5. Furthermore, it is advantageous when more than two inlet pipes 3, 5 are provided that the angles between the inlet pipes 3, 5 are different. As a result of this, direct impingement of the gaseous phases introduced via the inlet pipes 3, 5 is avoided and a more uniform flow distribution is achieved in this way.

[0122] Above the inlet pipes 3, 5 for introducing the gaseous phase, there are separation-active internals 9 in the form of a packing having an entry 11 in the column 2.

[0123] FIG. 7 shows a three-dimensional depiction of a column 2 having two inlet pipes 3, 5, which essentially corresponds to the column 2 according to FIG. 6, the difference being that in this case inlet pipes 3, 5 have different inlet pipe diameters 6, 17. The first inlet pipe 3 has an inlet pipe diameter 6 larger than a further inlet pipe diameter 17 of the second inlet pipe 5.

[0124] FIG. 8 shows a longitudinal sectional view of the column 2 according to FIG. 6, which has a column section 29 and inlet pipes 3, 5. The column section 29 has a free cross-sectional area and has a section height 28. The inlet pipes 3, 5 have the inlet pipe diameter 6 and are each arranged at a height 8 on the column 2 which has a column diameter 7.

EXAMPLES AND COMPARATIVE EXAMPLES

Example 1

[0125] The distribution of the relative velocity at the entry 11 into a packing 9 of a column 2 was determined. The calculation is based on an arrangement of two inlet pipes 3, 5 arranged at the same height at an angle α of 120° from one another.

[0126] For the calculation of the vapor flow presented here, a packing having a height of 1 m and a pressure drop of 1 mbar was assumed as separation-active internals 9. The column diameter 7 of the column 2 of 3200 mm and the inlet pipe diameter 6 of the two inlet 3, 5 of 1000 mm was assumed for the calculation. A pressure in the column 2 of 5.5 bar, a gas density of 16.6 kg/m.sup.3, a gas viscosity of 1.3.Math.10.sup.-5 Pa.Math.s, a velocity in the inlet pipes 3, 5 of 1.07 m/s with an F factor of 4.34 and a velocity in the column 2 of 0.21 m/s with an F factor of 0.85 were prescribed as boundary conditions for the calculation of the velocities. The F factor refers to the steam loading in the column 2 and is the product of the average velocity of the gaseous phase in m/s multiplied by the square root of the gas density in kg/m.sup.3.

[0127] In the interior of the column 2, a system of a plurality of eddy structures which are not shown here and in which flow lines, likewise not shown here, move upward in the direction to the separation-active internals 9, i.e. the packing, is established.

[0128] The vertical velocity component at the entry 11 of the separation-active internals 9 is a measure of the incorrect distribution established in the column 2.

[0129] In order to be able to employ the incorrect distribution appropriately as a measure of the flow uniformity, it is useful firstly to depict the calculated vertical velocities at the entry 11 of the column 2 in a histogram. Such a histogram is shown by way of example in FIG. 9.

[0130] To produce the histogram, it is possible, for example, firstly to depict the vertical velocities at the entry 11 into the separation-active internals 9 calculated using a suitable simulation program for flow calculations graphically by means of a grayscale and generate the histogram from the shades of gray. The histogram shows, for each velocity, the proportion of the cross-sectional area in which this velocity occurs. Here, the velocity is plotted on the abscissa 21 and the cross-sectional area is plotted on the ordinate 23.

Example 2

[0131] A cumulated frequency function as shown in each case in FIGS. 10 to 14 for different inlet pipe arrangements, inlet pipe configurations and modes of operation was calculated from histogram data as depicted in FIG. 9. The difference between the velocity which is of such a magnitude that the velocity is greater on only 5% of the cross-sectional area and the velocity which is of such a magnitude that the velocity is lower on only 5% of the cross-sectional area is calculated as measure for the nonuniform distribution. The smaller this difference, the more uniform is the flow distribution.

[0132] FIG. 10 shows frequency functions of the relative velocity at the entry 11 into the separation-active internals 9, i.e. the packing, for different arrangements of the inlet pipes 3, 5. Here, the velocity is plotted on the abscissa 31 and the cumulated proportion by area from 0 (no proportion at all) to 1 (the total area) is plotted on the ordinate 33.

[0133] In a first arrangement, the angle α between the inlet pipes 3, 5 is 90° . The corresponding first curve of the frequency function is designated by the reference symbol 35. A second curve 37 shows the frequency function for an arrangement of the inlet pipes 3, 5 at an angle α of 120° and a third curve 39 shows the frequency function for an arrangement of the inlet pipes 3, 5 at an angle α of 180°.

[0134] In contrast to the histogram in FIG. 9, the velocities in FIG. 10 were calculated for a column 2 having a column diameter 7 of 6400 mm. An inlet pipe diameter 6, 17 of in each case 3000 mm, a pressure of 0.025 bar, a gas density of 0.118 kg/m.sup.3, a gas viscosity of 7.8.Math.10.sup.-6 Pas, a velocity in the inlet pipes 3, 5 of in each case 11.7 m/s with an F factor of 4 and a velocity in the column 2 of 5.46 m/s with an F factor of 1.87 were prescribed as further boundary conditions.

[0135] The intersection of the curves 35, 37, 39 with a cumulated proportion by area 41 of 95% is the velocity which is of such a magnitude that the velocity is greater on only 5% of the cross-sectional area and the intersection of the curves 35, 37, 39 with the cumulated proportion by area 43 of 5% is the velocity which is of such a magnitude that the velocity is lower on only 5% of the cross-sectional area. The difference can then be determined in a simple manner from the graphs. When all curves 35, 37, 39 as depicted here are shown in a graph, the nonuniform distribution can be read off directly. The greater the distance between the intersections of in each case one curve 35, 37, 39 with the straight line 41 or 43, the greater is the nonuniform distribution. In an arrangement of 2 inlet pipes 3, 5, it can thus be seen that the greatest nonuniform distribution occurs at an angle α of the inlet pipes of 180°, so that a smaller angle α should be selected. The difference of the nonuniform distribution for an arrangement of the inlet pipes at 90° or 120° is so much smaller compared to the nonuniform distribution at 180° that the exact angle can, for example, be matched to the circumstances of the piping around the column.

Example 3

[0136] FIG. 11 shows frequency distributions for a column 2 having a column diameter 7 of 2900 mm and an inlet pipe diameter 6, 17 of the inlet pipes 3, 5 of in each case 900 mm. Trays having a pressure drop of 3 mbar at the entry 11 into the separation-active internals 9, i.e. the lowermost tray, were assumed as separation-active internals 9 in the column 2. The calculation was based on an arrangement of two inlet pipes 3, 5 arranged at the same height and at an angle α of 60°, 90°, 120° or 180° from one another in each case.

[0137] A pressure in the column 2 of 1.2 bar, a gas density of 1.63 kg/m.sup.3, a gas viscosity of 1.2.Math.10.sup.-5 Pa.Math.s, a velocity in the inlet pipes 3, 5 of in each case 7.4 m/s with an F factor of 9.4 and a velocity in the column 2 of 1.82 m/s with an F factor of 1.43 were prescribed as boundary conditions for the calculation of the velocities.

[0138] A system of a plurality of eddy structures which are not depicted here and in which flow lines, likewise not depicted here, move upward in the direction of the trays is established in the interior of the column 2. The vertical velocity component at the entry 11 into the lowermost tray is a measure of the incorrect distribution which is established in the column 2.

[0139] Cumulated frequency functions were calculated in each case for the different inlet pipe arrangements from the histogram data not shown here in a manner analogous to FIG. 9 and are shown in FIG. 11.

[0140] The difference between the velocity which is of such a magnitude that the velocity is greater on only 5% of the cross-sectional area and the velocity which is of such a magnitude that the velocity is lower on only 5% of the cross-sectional area is calculated as measure for the nonuniform distribution. The smaller this difference, the more uniform is the flow distribution.

[0141] FIG. 11 shows frequency functions of the relative velocity at the entry 11 into the separation-active internals 9, i.e. into the trays, for the different inlet pipe arrangements. Here, the velocity in m/s is plotted on the abscissa 31 and the cumulated proportion by area from 0 (no proportion at all) to 1 (the total area) is plotted on the ordinate 33.

[0142] In a first arrangement, the angle α between the inlet pipes 3, 5 is 60°. The associated fourth curve (broken line) of the frequency function is denoted by the reference symbol 44. A second curve 45 (solid line) shows the frequency function for an arrangement of the inlet pipes 3, 5 at an angle α of 90° from one another, a third curve 46 (dotted line) shows the frequency function for an arrangement of the inlet pipe 3, 5 at an angle α of 120° and a fourth curve 47 (broken line) shows the frequency function for an arrangement of the inlet pipes 3, 5 at an angle α of 180°.

[0143] The intersection of the curves 44, 45, 46 and 47 with a cumulated proportion by area 41 of 0.95 is the velocity which is of such a magnitude that the velocity is greater on only 5 out of 100 parts of the cross-sectional area and the intersection of the curves 44, 45, 46 and 47 with the cumulated proportion by area 43 of 0.05 is the velocity which is of such a magnitude that the velocity is lower on only 5 out of 100 parts of the cross-sectional area. The difference can then be determined in a simple manner from the graphs. When all curves 44, 45, 46 and 47 are depicted in a graph, as shown here, the nonuniform distribution can be read off directly. The greater the distance between the intersections of in each case one of the curves 44, 45, 46 and 47 with the straight line 41 or 43, the greater is the nonuniform distribution. In an arrangement of two inlet pipes 3, 5, it can thus be seen that the greatest nonuniform distribution is at the angle α of the inlet pipes of 180° here. The results depicted in FIG. 11 are summarized in table 1.

TABLE-US-00001 Angle α Nonuniform distribution at a pressure drop of 3 mbar [%] 60° 19.3 90° 18.6 120° 17.9 180° 20.0

Example 4

[0144] FIG. 12 shows frequency distributions for a column 2 which correspond essentially to the column 2 of example 3. Here, different velocities prevail in the two inlet pipes 3, 5, with the inlet pipe diameters 6, 17 each being 900 mm.

[0145] A velocity of 8.9 m/s with an F factor of 11.3 in the first inlet pipe 3 and a velocity of 5.9 m/s with an F factor of 7.5 in the second inlet pipe 5 were used as a basis for the calculation.

[0146] FIG. 12 shows, in a manner corresponding to FIG. 11, frequency functions of the relative velocity at the entry 11 into the separation-active internals 9 for different inlet pipe arrangements, with different gas velocities prevailing in the inlet pipes 3, 5 here.

[0147] In a first arrangement, the angle α between the inlet pipes 3, 5 is 60° . The associated eighth curve (dash-dot line) of the frequency function is denoted by reference numeral 60. A ninth curve 62 (solid line) shows the frequency function for an arrangement of the inlet pipes 3, 5 at an angle α of 90° from one another, while the tenth curve 64 (dotted line) shows the frequency function for an arrangement of the inlet pipes 3, 5 at an angle α of 120° and an eleventh curve 66 (broken line) shows the frequency function for an arrangement of the inlet pipes 3, 5 at an angle α of 180° .

[0148] The greatest nonuniform distribution prevails at the angle α of the inlet pipes of 180°, and the uniformity of the distribution is improved further for the remaining arrangements compared to an embodiment with equal velocities (cf. table 1). The results depicted in FIG. 12 are summarized in table 2.

TABLE-US-00002 Angle α Nonuniform distribution at a pressure drop of 3 mbar [%] 60° 15.6 90° 18.6 120° 15.1 180° 22.0

Example 5

[0149] FIG. 13 shows frequency distributions for a column 2 which corresponds essentially to the column 2 of example 3. Here, two inlet pipes 3, 5 with different inlet pipes diameters 6, 17 are present, with the gas velocities in the inlet pipes 3, 5 each being 7.4 m/s with an F factor of 9.4.

[0150] An inlet pipe diameter 6 of the first inlet pipe 3 of 794 mm and a further inlet pipe diameter 17 of the second inlet pipe 5 of 995 mm were used as a basis for the calculation.

[0151] FIG. 13 shows, in a manner corresponding to FIG. 11, frequency functions of the relative velocity at the entry 11 into the separation-active internals 9 for different inlet pipe arrangements, with the inlet pipes 3, 5 here having different inlet pipe diameters 6, 17.

[0152] In a first arrangement, the angle α between the inlet pipes 3, 5 is 60° . The associated twelfth curve (dash-dot line) of the frequency function is denoted by reference numeral 68. A thirteenth curve 70 (solid line) shows the frequency function for an arrangement of the inlet pipes 3, 5 at an angle α of 90° from one another, while a fourteenth curve 72 (dotted line) shows the frequency function for an arrangement of the inlet pipes 3, 5 at an angle α of 120° and a fifteenth curve 74 (broken line) shows the frequency function for an arrangement of the inlet pipes 3, 5 at an angle α of 180° .

[0153] The greatest nonuniform distribution prevails at an angle α of the inlet pipes of 180°, with the nonuniform distribution being improved further compared to an embodiment with equal inlet pipe diameters (cf. table 1). The results depicted in FIG. 13 are summarized in table 3.

TABLE-US-00003 Angle α Nonuniform distribution at a pressure drop of 3 mbar [%] 60° 12.1 90° 9.3 120° 13.7 180° 18.2

Example 6

[0154] As boundary conditions for FIG. 14, the same conditions as for the depiction in FIG. 9 were selected, except that the velocity in the inlet pipes 3, 5 was increased by a factor of 1.5 and the inlet pipe diameters 6, 17 were correspondingly decreased in order to obtain the same velocity in the column.

[0155] For a curve 55 which shows the frequency distribution for two inlet pipes 3, 5, an angle α between the inlet pipes 3, 5 of 120° was prescribed. For the calculation of a curve 59 using only one inlet pipe 3, an inlet pipe 3 enlarged by a factor 2.sup.0.5 in diameter was used. As a result, the F factor and the velocity in the column 2 and thus also the pressure drop in the separation-active internals 9, i.e. the packing, remain constant compared to the introduction of the gaseous phase via two inlet pipes 3, 5.

[0156] The normalized velocity is plotted on the abscissa 51 and the proportion by area is plotted on the ordinate 53 here.

[0157] The first curve 55 shows the frequency distribution for two inlet pipes 3, 5. The second curve 59 shows the frequency distribution for one inlet pipe 3.

[0158] It can clearly be seen from the comparison in FIG. 14 that a better uniform distribution is achieved in the case of a column 2 having two inlet pipes 3, 5.

Example 7

[0159] The effects of a nonuniform distribution of a vapor phase exiting from two vaporizers, based on the cross-sectional area of a column, were examined with the aid of a thermodynamic simulation of an overall plant for producing n-butyl acrylate.

[0160] The thermodynamic simulation was carried out using the software Aspen Plus®. Model data banks for modeling unit operations and also materials data banks were imported in respect of specific materials properties which are implemented in the software. Mixing properties were calculated by means of the software based on various thermodynamic models of materials data of the pure substances.

Example 7a

[0161] To determine the energy consumption in the case of a uniformly distributed gaseous phase, three vapor streams 15a, 15b, 15c having an identical size, as are depicted in FIG. 15, were used as a basis.

[0162] A column 2 as rectification column 40 was represented with three identical subcolumns 2A, 2B, 2C in the simulation and the three subcolumns 2A, 2B, 2C were each simulated with 13 theoretical plates.

[0163] At the top of the subcolumns 2A, 2B, 2C, a runback 16 in the form of a liquid phase was divided into three equal-sized liquid streams 16a, 16b, 16c and distributed over the three subcolumns 2A, 2B, 2C.

[0164] At the bottom of the column, bottom offtake streams 12a, 12b, 12c and liquid streams 14c, 14d exiting from vaporizers 20, 30 were combined to form a total stream 12, i.e. a liquid phase from the column 2, and mixed with a feed stream 10. A small substream 18 of the total stream 12 was discharged from the plant and the main stream 13 of the total stream 12 was divided into two identically sized streams 13a, 13b and fed to the two vaporizers 20, 30.

[0165] Vapor streams 14a, 14b which exit from the vaporizers 20, 30 and were fed, for example, through inlet pipes 3, 5 to the column 2 were combined to form a vapor feed stream 15 and then divided into three identically sized streams 15a, 15b, 15c, each in the form of a gaseous phase, and introduced into the three subcolumns 2A, 2B, 2C.

[0166] The division of the vapor feed stream 15 was effected equally:

TABLE-US-00004 Vapor stream (15a) 33.33% by weight, Vapor stream (15b) 33.33% by weight, Vapor stream (15c) 33.33% by weight, in each case based on the vapor feed stream 15.

The thermodynamic simulation of the total plant indicated the following quantities of heat required in the vaporizers 20, 30:

TABLE-US-00005 Vaporizer 20: 8922 kW, Vaporizer 30: 8922 kW.

Example 7b

[0167] To determine the energy consumption in the case of an unequally divided gaseous phase, three unequal vapor streams 15a, 15b, 15c as per FIG. 15 were used as a basis. The procedure was otherwise as in example 7a.

[0168] A nonuniform distribution of the vapor streams 14a, 14b exiting from the vaporizers 20, 30 over the cross-sectional area of the column 2, which is caused by an unfavorable arrangement of the inlet pipes 3, 5 at the circumference of the column, was simulated by the different-sized vapor streams 15a, 15b, 15c which were fed to the subcolumns 2A, 2B, 2C.

[0169] The division of the vapor feed stream 15 was effected unequally:

[0170] All other conditions remained unchanged compared to example 7a.

[0171] The thermodynamic simulation of the total plant indicated the following quantities of heat required in the vaporizers 20, 30:

TABLE-US-00006 Vaporizer (20): 9237 kW, Vaporizer (30): 9237 kW.

[0172] Compared to example 7a, about 3.5% more energy was required in the two vaporizers in the case of the nonuniform introduction of vapor.

TABLE-US-00007 Vapor stream (15a) 33.33% by weight, Vapor stream (15b) 33.33% by weight, Vapor stream (15c) 33.33% by weight, in each case based on the vapor feed stream 15.

[0173] Local ratios of liquid phase and gaseous phase in the rectification column 40 were changed by the nonuniform distribution of the gaseous phase fed in. In order to fulfil the same separation task in the rectification column 40 as in example 7a, more energy is required than in the case of a uniform distribution of the gaseous phase.

[0174] A comparison of examples 7a and 7b taking into account examples 1 to 6 shows that the energy consumption for the same separation performance is reduced by a uniform distribution of the gaseous phase over the cross-sectional area of a column 2, in particular a rectification column 40, which is achieved by a configuration according to the invention of the inlet pipes 3, 5.

TABLE-US-00008 List of reference symbols 1 Apparatus for carrying out mass transfer processes 2 Column 2A, 2B, 2C Subcolumns 3 First inlet pipe 4 Radial direction 5 Second inlet pipe 6 Inlet pipe diameter 7 Column diameter 8 Height 9 Separation-active internals 10 Feed stream 11 Entry 12 Total stream, liquid phase 12a, 12b, 12c Bottom offtake streams 13 Main stream 13a, 13b Streams of the main stream 14a, 14b Exiting vapor streams 14c, 14d Exiting liquid streams 15 Vapor feed stream 15a, 15b, 15c Vapor streams of the vapor feed stream 16 Runback 16a, 16b, 16c Liquid streams 17 Further inlet pipe diameter 18 Substream 19 Circumference 20 First vaporizer 21 Abscissa, velocity 23 Ordinate, cross-sectional area 25 Third inlet pipe 27 Fourth inlet pipe 28 Section height 29 Column section 30 Second vaporizer 31 Abscissa, relative velocity 33 Ordinate, cumulated proportion by area 35 First curve of the frequency function 37 Second curve of the frequency function 39 Third curve of the frequency function 40 Rectification column 41 Cumulated proportion by area of 95% 43 Cumulated proportion by area of 5% 44 Fourth curve of the frequency function 45 Fifth curve of the frequency function 46 Sixth curve of the frequency function 47 Seventh curve of the frequency function 51 Abscissa, normalized velocity 53 Ordinate, proportion by area 55 First curve of the frequency distribution 59 Second curve of the frequency distribution 60 Eighth curve of the frequency function 62 Ninth curve of the frequency function 64 Tenth curve of the frequency function 66 Eleventh curve of the frequency function 68 Twelfth curve of the frequency function 70 Thirteenth curve of the frequency function 72 Fourteenth curve of the frequency function 74 Fifteenth curve of the frequency function α Angle β Further angle