METHOD FOR STARTING UP A REACTOR FOR PREPARING PHTHALIC ANHYDRIDE

Abstract

The present invention relates to a process for starting up a reactor for preparation of phthalic anhydride by the catalytic oxidation of ortho-xylene and/or naphthalene, containing a bed of shaped catalyst bodies and within a temperature-controlled salt bath. The industrial production of phthalic anhydride from ortho-xylene and/or naphthalene is affected by selective gas phase oxidation in a shell and tube reactor cooled with a salt bath, which may contain several thousand reactor tubes. There are 4 to 5 different catalyst layers in each reactor, which are introduced into each reactor successively in axial direction.

Claims

1. A process for starting up a reactor for preparation of phthalic anhydride by the catalytic oxidation of ortho-xylene and/or naphthalene, containing a bed of shaped catalyst bodies and within a temperature-controlled salt bath, comprising the steps of: a) calcining the shaped catalyst bodies, in the presence of air and/or O2, at a salt bath temperature exceeding 390° C., b) adjusting the temperature of the salt bath to a temperature between 370° C. and 400° C., c) forming a hotspot in the front third of the catalyst bed in flow direction, by feeding in ortho-xylene and/or naphthalene, d) cooling the salt bath temperature to a salt bath temperature below 360° C. at a rate of greater than 0.5° C./h and increasing the feed of ortho-xylene and/or naphthalene to a loading exceeding 70 g/m.sup.3 (STP), at an air flow rate of 2 m.sup.3 (STP)/h to 4.5 m.sup.3 (STP)/h, wherein, during the feeding with ortho-xylene and/or naphthalene, the absolute reactor inlet pressure does not go below 1435 mbar.

2. The process as claimed in claim 1, wherein the maximum temperature of the shaped catalyst bodies during a) is always in the range between 390° C. and 460° C., preferably in the range between 400° C. and 440° C.

3. The process as claimed in claim 1, wherein step a) is effected for at least 6 h, more preferably at least 24 h.

4. The process as claimed in claim 1, wherein the salt bath temperature in step c) is kept stable between 370° C. and 400° C. for a period of between 1 h and 200 h.

5. The process as claimed in claim 1, wherein the ortho-xylene and/or naphthalene loading in step c) is between 10 g/m.sup.3 (STP) and 40 g/m.sup.3 (STP), at an air flow rate of 2 m.sup.3 (STP)/h to 4.5 m.sup.3 (STP)/h.

6. The process as claimed in claim 1, wherein the cooling in step d) is effected at a rate of > 0.70° C./h, preferably > 1° C./h.

7. The process as claimed in claim 1, wherein the cooling in d) is effected at a rate between 0.5 and 10° C./h, preferably 1° C./h to 10° C./h.

8. The process as claimed in claim 1, wherein the ortho-xylene and/or naphthalene loading on commencement of cooling in step d) is between 10 and 40 g/m.sup.3 (STP) and is increased in step d) to more than 70 g/m.sup.3 (STP), at an air flow rate of constantly between 2 m.sup.3 (STP)/h to 4.5 m.sup.3 (STP)/h.

9. The process as claimed in claim 1, wherein the ortho-xylene and/or naphthale0ne loading in each of steps c) or d) is only sufficiently high that the temperature of the shaped catalyst bodies does not exceed 455° C.

10. The process as claimed in claim 1, wherein the absolute reactor inlet pressure during steps b) and d) does not go below 1450 mbar, more preferably 1460 mbar.

11. The process as claimed in claim 1, wherein the reactor contains four or more catalyst layers consisting of different shaped catalyst bodies.

12. The process as claimed in claim 1, wherein one of the catalyst layers has a lower voidage than the other catalyst layers.

13. The process as claimed in claim 12, wherein the layer having the lower voidage has a voidage below 65%.

Description

[0046] FIG. 1: Evolution with time of salt bath temperature and ortho-xylene feed during process steps b) and c) (absolute reactor inlet pressure in each case < 1435 mbar): (a) comparative test 1, (b) comparative test 2.

[0047] FIG. 2: Evolution with time of salt bath temperature and ortho-xylene feed during process steps b) and c) (absolute reactor inlet pressure in each case > 1435 mbar): (a) inventive test 1, (b) inventive test 2.

[0048] FIG. 3: Evolution with time of air flow rate and absolute reactor inlet pressure during process steps b) and c) (absolute reactor inlet pressure in each case <1435 mbar): (a) comparative test 1, (b) comparative test 2.

[0049] FIG. 4: Evolution with time of air flow rate and absolute reactor inlet pressure during process steps b) and c) (absolute reactor inlet pressure in each case >1435 mbar): (a) inventive test 1, (b) inventive test 2.

[0050] FIG. 5: Evolution with time of phthalic anhydride yield during process steps b) and c) (absolute reactor inlet pressure in each case <1435 mbar): (a) comparative test 1, (b) comparative test 2.

[0051] FIG. 6: Evolution with time of phthalic anhydride yield during process steps b) and c) (absolute reactor inlet pressure in each case >1435 mbar): (a) inventive test 1, (b) inventive test 2.

[0052] FIG. 7: Evolution with time of phthalic anhydride productivity during process steps b) and c) (absolute reactor inlet pressure in each case <1435 mbar): (a) comparative test 1, (b) comparative test 2.

[0053] FIG. 8: Evolution with time of phthalic anhydride productivity during process steps b) and c) (absolute reactor inlet pressure in each case >1435 mbar): (a) inventive test 1, (b) inventive test 2.

EXAMPLES

[0054] Catalytic measurements were conducted on 4 identical catalyst layer arrangements of shaped catalyst bodies, with variation of the absolute reactor inlet pressure between < 1435 mbar (noninventive comparative examples) and >1435 mbar (inventive examples).

[0055] For synthesis of the shaped catalyst bodies used, two different types of steatite rings were used as shaped bodies, designated 8×6×5 ring and 6×5×4 ring. The nomenclature of the geometric dimensions of the rings corresponds to external diameter (De) [mm] x height (H) [mm] x internal diameter (Di) [mm]. The uncoated shaped bodies were introduced into a coating apparatus and coated homogeneously with the active composition, as described in DE 19709589 A1. During the coating operation, an aqueous suspension of the active components (TiO.sub.2, V.sub.2O.sub.5, promoters) and an organic binder (vinyl acetate/ethylene copolymer) was sprayed onto the fluidized inert support heated to 70° C. via multiple nozzles until the desired active composition layer had formed. Table 1 shows an overview of the shaped catalyst bodies produced and the respective active composition.

[0056] For formation of the arrangements of 4 catalyst layers for the respective catalytic measurement, the shaped catalyst bodies were introduced into a salt bath-cooled tube having internal diameter 25 mm and length 4 m. Tables 2 to 5 show an overview of the respective virtually identical fillings as used in the different catalytic tests. For in situ calcination and preforming in process step a), 0.02 to 0.03 m.sup.3 (STP)/h of air was passed in each case through the shaped catalyst bodies in the tube at salt bath temperature 410° C. for more than 48 h. In a centered arrangement within the tube was a 3 mm thermowell with an installed tensile element for temperature measurement.

[0057] In process step b), before the reactor was started up, the salt bath temperature was cooled from the calcining temperature to 390° C.

[0058] At the start of the respective catalytic measurement in step c) (operating time 0 h), at a salt bath temperature of 390° C., air was passed through the tube from the top downward at an air flow rate of 3.3 m.sup.3 (STP)/h and a feed rate of 25 g ortho-xylene/m.sup.3 (STP) (ortho-xylene purity > 98%). The absolute reactor inlet pressure here was adjusted (in accordance with the invention) by means of a valve downstream of the reactor to a value of < 1435 mbar (noninventive) or > 1435 mbar (inventive). After an operating time of at least 41 h, the air flow rate was in each case increased from 3.3 m.sup.3 (STP)/h to 4.0 m.sup.3 (STP)/h, at first keeping the salt bath temperature constant at 390° C. After an operating time of 164 h in comparative example 1, after an operating time of 140 h in comparative example 2, after an operating time of 162 h in inventive example 1, and after an operating time of 94 h in inventive example 2, in process step d), the salt bath temperature in each case was lowered stepwise from 390° C. to 350 to 355° C. within 46 h in comparative example 1, within 21 h in comparative example 2, within 30 h in inventive example 1, and wherein 56 h in inventive example 2. At the same time, the feed of ortho-xylene in the air was increased to the maximum possible extent, but such that a temperature of the shaped catalyst bodies of 455° C. was not exceeded.

[0059] In the catalytic measurements of the invention (absolute reactor inlet pressure > 1435 mbar), it was possible to achieve a maximum feed rate of 80 to 85 g ortho-xylene/m.sup.3 (STP) of air directly with attainment of the minimum salt bath temperature of 350° C. to 355° C. In the noninventive comparative examples (absolute reactor inlet pressure < 1435 mbar), by contrast, the feed rate was limited to 65 g ortho-xylene/m.sup.3 (STP) of air with attainment of the minimum salt bath temperature of 350° C. to 355° C.; an increase in the feed rate above that would have increased the temperature of the shaped catalyst bodies to more than 455° C. Only after operation for a further 250 h was it possible to achieve a feed rate of 80 to 85 g ortho-xylene/m.sup.3 (STP) of air without further lowering the salt bath temperature.

[0060] A detailed overview of the evolution with time of all relevant process parameters in the comparative tests (absolute reactor inlet pressure <1435 mbar) and in the inventive tests (absolute reactor inlet pressure >1435 mbar) during process steps b) and c) is shown in FIG. 1 to FIG. 4.

[0061] In order to ascertain the phthalic anhydride yield and productivity during the catalytic tests, the product stream each case was analyzed with regard to its composition at regular intervals by means of a gas chromatograph (GC 7890B, Agilent) and a non-dispersive IR analyzer (NGA 2000, Rosemount).

[0062] Phthalic anhydride yield was calculated by equation Eq. 2.

[00002]${\text{Y}}_{PA}=\left[139.52-\left(800*\frac{A+B}{B}\right)+\left(1.00-F\right)-\left(1.25*H\right)-\left(1.1*G\right)\right]$

[0063] A = CO.sub.2 in product stream [% by vol.] [0064] B = CO in product stream [% by vol.] [0065] H = maleic anhydride content in product stream [% by wt.] [0066] E = ortho-xylene loading in reactant stream [g/m.sup.3 (STP)] [0067] F = purity of the ortho-xylene used [% by wt.] [0068] G = ortho-xylene slip in product stream [% by wt.] [0069] Y.sub.PA = yield of phthalic anhydride (PA) based on the total weight of the ortho-xylene used [% by wt.]

[0070] As apparent from equation Eq. 2, the phthalic anhydride is directly dependent on the formation of the three most important by-products: CO, CO.sub.2 and maleic anhydride.

[0071] Phthalic anhydride productivity was calculated by equation Eq. 3.

[00003]$P_{PA}=E*Q*\frac{Y_{PA}}{100\%bywt,}$

[0072] E = ortho-xylene loading in the reactant stream [g/m.sup.3 (STP)] [0073] Q = air flow rate [m.sup.3 (STP)/h/tube] [0074] Y.sub.PA = phthalic anhydride (PA) yield based on the total weight of the ortho-xylene used [% by wt.] [0075] P.sub.PA = phthalic anhydride (PA) productivity [g/h/tube]

[0076] The progression of the phthalic anhydride yield and phthalic anhydride productivity over time for the inventive examples (absolute reactor inlet pressure >1435 mbar) and the noninventive comparative examples (absolute reactor inlet pressure < 1435 mbar) is shown in FIG. 5 to FIG. 8.

TABLE-US-00001 Shaped catalyst bodies used Catal yst Ring shape Binder content Proportion of active composition BE T TiO.sub.2 V.sub.2O.sub.5 Promot ers.sup.5 (De x H x Di).sup.1 [mm] [% by wt.].sup.2 [% by wt.].sup.3 [m.sup.2/ g] [% by wt.].sup.4 [% by wt.].sup.4 [% by wt.].sup.5 A0 8×6×5 2.4 9.5 18 87.8 7.5 4.7 A1 6×5×4 2.4 5.6 26 82.7 11.0 6.3 A2 8×6×5 2.4 8.7 18 87.9 7.5 4.6 A3 8×6×5 2.3 8.4 25 89.4 9.4 1.2 .sup.1 De = external diameter, H = height, Di = internal diameter .sup.2 based on the total weight of the shaped catalyst body .sup.3 based on the total weight of the shaped catalyst body without binder .sup.4 based on the active catalyst composition .sup.5 including Sb.sub.2O.sub.3 and small proportions of Nb.sub.2O.sub.5, P and Cs

TABLE-US-00002 Filling parameters for comparative test 1 Catalyst layer Catalyst Fill heights Bulk density Ring shape [cm] [g/cm.sup.3] (De x H x Di) 1 A0 43 0.80 8×6×5 2 A1 151 0.87 6×5×4 3 A2 70 0.86 8×6×5 4 A3 71 0.82 8×6×5

TABLE-US-00003 Filling parameters for comparative test 2 Catalyst layer Catalyst Fill heights Bulk density Ring shape [cm] [g/cm.sup.3] (De x H x Di) [mm] 1 A0 40 0.82 8×6×5 2 A1 151 0.85 6×5×4 3 A2 70 0.83 8×6×5 4 A3 70 0.81 8×6×5

TABLE-US-00004 Filling parameters for inventive test 1 Catalyst layer Catalyst Fill heights Bulk density Ring shape [cm] [g/cm.sup.3] (De x H x Di) [mm] 1 A0 40 0.82 8×6×5 2 A1 150 0.87 6×5×4 3 A2 71 0.81 8×6×5 4 A3 71 0.81 8×6×5

TABLE-US-00005 Filling parameters for inventive test 2 Catalyst layer Catalyst Fill heights Bulk density Ring shape [cm] [g/cm.sup.3] (De x H x Di) [mm] 1 A0 40 0.81 8×6×5 2 A1 151 0.89 6×5×4 3 A2 70 0.87 8×6×5 4 A3 70 0.81 8×6×5