WO2014095359A1

WO2014095359A1 - Method for the start-up of a plant for the production of propylene

Info

Publication number: WO2014095359A1
Application number: PCT/EP2013/075440
Authority: WO
Inventors: Martin Rothaemel; Thomas Renner; Stéphane Haag; Martin GORNY; Roberta Olindo
Original assignee: L'Air Liquide Société Anonyme pour l'Etude et l'Exploitation des Procédés Georges Claude
Priority date: 2012-12-21
Filing date: 2013-12-03
Publication date: 2014-06-26
Also published as: DE102012112839A1

Abstract

The present invention relates to a method for the start-up of a plant, wherein in the plant in ongoing operation at least one olefin is produced from at least one oxygenate, and wherein the plant includes at least one reactor filled with a solid catalyst with at least one educt inflow. Via the educt inflow, propylene initially is fed in, until the plant has reached its operating conditions with regard to pressure and flow rates. After reaching the operating conditions, the amount of propylene incrementally or continuously is decreased to zero and correspondingly the amount of at least one oxygenate incrementally or continuously is increased such that the flow rate of the educt inflow remains constant. In that in the start-up operation it is omitted to reach the operating condition of the plant with regard to temperature, a temperature increase damaging the catalyst can be avoided.

Description

Method for the Start-up of a Plant

for the Production of Propylene The present invention relates to a method for the start-up of a plant for the production of olefins, in particular propylene, wherein in the plant at least one olefin is produced from at least one oxygenate in a continuous operation, and wherein the plant includes at least one reactor operated with a solid catalyst with at least one educt inflow.

Propene (C3H6), often also referred to as propylene, is one of the most important starting substances of the chemical industry. The demand for the base material propylene is increasing worldwide, wherein propylene just like ethylene mostly is produced from petroleum in a steam cracker in a ratio dependent on the process and the raw materials.

To obtain additional propylene, a number of processes exist, such as the PDH process which proceeds from propane as educt. What is known, however, above all is the so-called MTP process, in which olefins are produced from methanol (MeOH) or dimethyl ether (DME) by catalytic conversion on a zeolitic catalyst. By varying the catalyst under process conditions, the selectivity of the products obtained can be influenced and the product spectrum thus can be shifted towards short-chain olefins (then often also the process name Methanol- to-Olefin (MTO)), towards longer-chain products (then often also the process name Methanol-to-Gasoline (MTG)) or towards propylene.

The fundamentals of an MTP process are described for example in DE 10 2005 048 931 A1 . From an educt mixture containing steam and oxygenates such as methanol and/or dimethyl ether, C₂ to C₄ olefins are produced above all. By a heterogeneously catalyzed reaction in at least one reactor, the educt mixture is converted to a reaction mixture comprising low-molecular olefins and gasoline hydrocarbons. By a suitable separation concept, higher olefins, above all the C₅₊ fraction, can at least partly be recirculated into the reactor as recycling stream and in said reactor for the most part be converted to propylene, whereby the propylene yield is increased.

The MTP process basically is refined to such an extent that it can be carried out on an industrial scale with a conversion of several hundred thousand tons per year.

One problem with such large-scale chemical plants however is the so-called start-up, by which the transfer of the plant from the rest condition into the continuous operating condition is to be understood. Problems arise with the application of the normally hierarchical control technology during the start-up, the mate- rial and energetic interconnections and couplings of the individual components by recirculations and wirings, and the starting of the wanted reaction.

In this connection, US 2005/0038061 A1 describes a method for the start-up of systems which employ molecular sieves as catalyst, as in part also is the case in the MTP process. Emphasis is put on the temperature profile during the startup operation and the avoidance of a contact of the catalyst with water, without especially discussing the medium used in the start-up operation.

US 6,872,867 A describes a method for starting up an MTP process, in which the start-up is made such that the entire system is divided into two zones and the start-up gas not specified in detail circulates in two recirculation zones.

An MTP plant usually is put into operation in that liquefied gas (LPG) is introduced into the plant as start-up medium via the educt inflow(s) and subsequent- ly the operating conditions pressure, temperature and flow rates are adjusted. As soon as all plant components have reached their operating conditions with regard to pressure, temperature and flow rates, which must exist to actually achieve an MTP reaction in the reactor, the oxygenate is introduced via the educt inflow(s).

What is problematic with the choice of liquefied gas as start-up gas is the quality of the available liquefied gas. In some regions of the world, there is obtained liquefied gas which contains potential catalyst poisons, in particular halogens, alkali metals and trimethylamine. The use of liquefied gas as start-up medium therefore involves the risk that already during the start-up of the plant the catalyst is irreversibly damaged by catalyst poisons contained therein. Liquefied gas in the required high degree of purity however is not available at each site or an acquisition in the required large amounts is not economically expedient. US 2009/0187058 A1 in this connection refers to the fact that olefins also can be used as start-up medium of an MTP process, without discussing the particularities in the use of this start-up medium.

It speaks for the use of propylene that propylene mostly is the main product desired by the plant operator, and therefore in most of all cases other plants for the propylene production, in particular a steam cracker, already are present at the site.

What is, however, problematic with the use of propylene is the fact that propyl- ene itself reacts at the catalyst of the MTP plants. There occur strongly exothermal reactions, whereby the catalyst can be damaged irreversibly. On the other hand, a polymerization of the propylene can be effected due to heating as a result of the exothermicity, whereby the plant becomes clogged and can only be cleaned mechanically. Therefore, it is the object of the present invention to provide a method in which an MTP plant can be put into operation with propylene as start-up medium.

In accordance with the invention, this object is solved with the features of claim 1 . For this purpose, propylene initially is fed into the plant via the at least one educt inflow, until the plant has reached its operating conditions with regard to pressure and flow rates. After reaching the operating conditions in the individual plant components, the amount of propylene is lowered to zero incrementally or continuously and correspondingly the amount of at least one oxygenate, which is used as educt in the ongoing operation, is increased incrementally or continuously such that the flow rate of the educt inflow remains constant.

By deliberately refraining from bringing the plant to the operating condition with regard to the temperature, it can be avoided that side reactions of the propylene occur at the catalyst. This is the absolute prerequisite for the use of propylene, as its side reactions are strongly exothermal and thus can lead to a runaway of the reaction and/or to an irreversible damage of the catalyst. At higher temperatures there is also the risk that the propylene will polymerize and the plant becomes clogged. Due to the deliberate decision to not adjust all operating condi- tions of the plant already during the use of the start-up medium, propylene can be used as start-up medium for the start-up of an MTP plant.

The use of propylene as start-up medium for an MTP plant in particular is advantageous when the plant contains at least one purification device downstream of the reactor, which likewise is put into operation. As side reactions of the propylene are not inhibited completely, the stream exiting from the reactor has a composition which is similar to the one in the ongoing operation. Therefore, it is possible to operate the at least one purification device under conditions which are very similar to those of the continuous operation. In particular, it is also possible to operate existing return conduits e.g. of higher olefins, especially C2 to Ce olefins. As a result, a large part of the propylene can be recovered already during the start-up, so that the costs for the start-up medium are lowered distinctly.

As catalyst, all catalysts usable for an MTP process can be employed. In particular silicon aluminum phosphates (SAPO) or zeolites, quite particularly ZSM- 5, as well as metal-doped zeolites can be used. Advantageously, the educt inflow is fed into the reactor with a temperature between 400 and 440 °C, preferably 430 °C, until the plant has reached its operating conditions with regard to pressure and flow rates. At these temperatures, in particular when using a typical zeolite-based catalyst, the rate of the occurring side reactions is controlled such that although the side reactions de- sired for the downstream purification devices occur, both a high exothermicity and an unwanted polymerization can reliably be avoided.

Typically, an MTP reactor consists of a plurality of trays which each include a catalyst bed, wherein the catalyst neither must be poured homogeneously in the tray nor each tray must include the same catalyst. The inflow of the reactor is identical with the inflow of the first tray.

In a preferred aspect of the invention, the outflow of the respective upstream tray forms the inflow of the respective downstream tray. When propylene is used as start-up medium, it is recommendable to each cool the flux between the trays, so that the inlet temperature into the next tray approximately is identical (preferably ± 20°C) to the temperature of the reactor inflow. Preferably, cooling is effected to a temperature between 400 and 440 °C. This provides an additional possibility for adjusting the propylene temperature inside each tray such that undesired, highly exothermal side reactions and/or polymerizations can reliably be avoided.

For adjusting the optimum operating temperature of the reactor, the educt inflow is fed into the reactor with a temperature between 400 and 540 °C, preferably between 430 and 500 °C, as soon as the propylene content is lowered to zero, and thus the risk of undesired side reactions no longer exists. The exit temperature lies between 500 and 540 °C. The adjustment of the operating condition with regard to the temperature thus only will be effected when the use of the start-up medium already is completed. This represents a deviation from the typical start-up behavior, but is possible, since the temperatures during the startup with the start-up medium propylene and the required reaction temperature lie comparatively close to each other. In a particularly preferred embodiment, the temperature during the incremental or continuous decrease of the propylene content in the educt inflow is increased in proportion to the decrease of the propylene content, with the increase being effected such that after lowering the propylene content to zero, the operating temperature of the continuous operation is present. Larger temperature jumps and related undesired side reactions or thermal loads of the material thereby can be avoided.

Furthermore, the use of propylene as start-up medium was found to be advantageous in particular when a zeolite is used as catalyst. This is due to the fact that on the zeolite the propylene is partly converted to C2 to Ce olefins and thus the purification devices downstream of the reactor perform similar tasks as in the continuous operation already during the start-up operation.

In particular, it was found to be favorable to use a regenerable catalyst, wherein regeneration is understood to be the combustion of carbonaceous deposits. Should such deposits occur during the start-up, it is possible due to the regeneration to restore the catalyst such that it can be reused or be used further and thus no additional costs are incurred by a premature refilling of catalyst. Finally, it was found to be favorable that the flow velocity of the educt inflow lies between 0.3 and 0.5 kg(educt)/kg(catalyst)^*h^"1 , preferably at about 0.4 kg(educt)/kg(catalyst)^*h^"1. Thus, a better control of the reactions taking place can be ensured.

Further developments, advantages and possible applications of the invention can also be taken from the following description of an exemplary embodiment and the Figures. All features described and/or illustrated form the subject-matter of the invention per se or in any combination, independent of their inclusion in the claims or their back-reference.

Example

The use of propylene as start-up medium was carried out in a pilot plant with an MTP reactor with six trays, as follows:

0.4 kg (educt)/kg(catalyst)^*h^"1 was adjusted as flow velocity, whereby a good control of exothermal reactions and an avoidance of temperatures above 500 °C could be ensurd. During the entire start-up, the inflow of water was effected at a flow velocity of 0.56 kg(water)/kg(catalyst)^*h^"1. The temperature of the entire plant was adjusted to 430 °C by using tempered nitrogen as purge gas. After the entire plant was flooded with nitrogen, the inflow of water was switched on and 10 min later the actual start-up with propylene was started, in that propylene was fed in with a temperature of 430 °C and this temperature was kept constant during the subsequent 24 operating hours. Every 6 hours, the temperature of the individual six trays (H1 - H6) was determined. Subsequently, the tempera- ture of the propylene was reduced to 400 °C and a further 24 h test was conducted, wherein again every 6 hours a measurement value was taken for each tray. At the beginning of the reaction, a high increase of the temperature in the first tray was noted. The temperature increase of tray 1 , where the temperature increase is about 60 °C at an inlet temperature of 430 °C and 70 °C at an inlet temperature of 400 °C, decreased to a temperature delta for the sixth tray of about 20 °C for an inlet temperature of 430 °C and 15 °C for an inlet tempera- ture of 400 °C. As can be taken from the following Table 1 , the reaction in the first tray is distinctly more exothermal than in the other trays, which is due to the fact that the amount of propylene in the entire stream on entry into the first tray is distinctly higher with 100 % than on entry into the second tray. Analyzing the outflow of the first tray reveals that inside the first tray about 32% of the propyl- ene have already been converted, whereas over all six trays the conversion is about 80%. Thus, a large part of the conversion already takes place in the first tray, which explains the distinctly more significant increase in temperature.

Table 1 : Temperature increase over all trays as a function of time.

The pressure loss over all trays, on the other hand, remains relatively constant and within an acceptable range, as is shown in Table 2. At an inlet temperature of 430 °C, the pressure loss of about 385 mbar is slightly higher than the pressure loss at an inlet temperature of 400 °C, at which the pressure loss is about 365 mbar. Table 2: Pressure loss over all trays as a function of time.

The conversions achieved are shown in Tables 3 and 4. The propylene conversion over all six trays is about 80% both for an inlet temperature of 430 °C and for an inlet temperature of 400 °C. From the combination of exothermicity of the main reaction and relatively lower activation energies of the reactions taking place it can be derived that at both inlet temperatures approximately the same reactions take place. The carbon balance was relatively good with a value of 103%. The carbon balance is calculated as follows:

CB(%) = 1 00 * Cout [mol(C)/h] / C_in [mol(C)/h]. The analysis of the product obtained shows about 3 to 4 % of unknown or undetermined components, which substantially are heavy hydrocarbons or aromatics. Table 3: Propylene conversion and carbon balance as a function of time

The selectivities (S) are calculated according to the following Formula:

The selectivity of C₅₊ components is about 40.3% for an inlet temperature of 430 °C and about 42.8% for an inlet temperature of 400 °C. A lower inlet tempera- ture leads to a higher content of heavy hydrocarbons.

Table 4: Selectivities as a function of time.

The high content of C2-C6 olefins can be recirculated to the MTP reactor and, as soon as the actual MTP reaction starts, likewise for the most part is converted to propylene.

The amount of propylene is determined by a rotameter, which has been cali- brated before the test. A gas meter (l/h) and a mass flow controller (g/h) based on the measurement of the Coriolis force were used to determine the amounts of gas produced from the reactor.

The quality of the propylene was checked by corresponding analyses before commencement of the experiment.

As catalyst, there was not used a fresh catalyst, but a ZSM-5 regenerated already.

The temperature profile over the trays is shown in the drawing, wherein: Fig. 1 shows the temperature profile over the individual trays (H) as a function of time, and

Fig. 2 shows the temperature profile in the first tray in dependence on the position in the catalyst bed.

As shown in Fig. 1 , a distinct temperature increase occurs inside the first tray after switching to propylene as start-up medium, which must be ascribed to the exothermicity of the individual reactions. As had to be expected, the temperature difference occurring in the individual trays decreases in dependence on the position of the respective trays in the circuit of the trays, so that the second highest temperature difference is found in the second tray and the smallest temperature difference occurs in the last one of the six trays. This is related to the fact that all reactions taking place are greatly dependent on the concentra- tion of propylene and the propylene concentration decreases over the individual trays.

Fig. 2 shows the course of the temperature in dependence on the height position in the tray 1 at various reaction times. It is found that in the lower region of the bed the reactions start virtually immediately and are promoted further by the heat released by them. With increasing length of the path covered already in the catalyst fixed bed, the concentration of propylene within the catalyst fixed bed decreases, so that the reaction rates of the exothermal reactions and as a result the temperature also decrease again.

Claims

1 . A method for the start-up of a plant, wherein in the plant in ongoing operation at least one olefin is produced from at least one oxygenate and wherein the plant includes at least one reactor filled with a solid catalyst with at least one educt inflow, characterized in that via the educt inflow propylene first is fed in, until the plant has reached its operating conditions with regard to pressure and flow rates, and that after reaching the operating conditions the amount of the propylene incrementally or continuously is decreased to zero and correspond- ingly the amount of at least one oxygenate incrementally or continuously is increased such that the flow rate of the educt inflow remains constant.

2. The method according to claim 1 , characterized in that the plant contains at least one purification device downstream of the reactor and the purification device also is put into operation.

3. The method according to claim 1 or 2, characterized in that the educt inflow is fed into the reactor with a temperature between 400 and 440 °C, until the plant has reached its operating conditions with regard to pressure and flow rates.

4. The method according to any of the preceding claims, characterized in that the reactor includes at least two series-connected trays, wherein the inflow of the reactor forms the inflow of the tray connected first.

5. The method according to claim 4, characterized in that the outflow of the tray upstream in the series at least partly forms the inflow of the tray downstream in the series.

6. The method according to claim 5, characterized in that the inflow of the downstream tray is cooled to a temperature between 400 and 440 °C.

7. The method according to any of the preceding claims, characterized in that after lowering the propylene content to zero, the educt inflow is fed into the reactor with a temperature between 400 and 540°C.

8. The method according to any of the preceding claims, characterized in that during the incremental or continuous lowering of the propylene content in proportion to the oxygenate content in the educt inflow, the temperature of the educt inflow is brought to the operating temperature of the ongoing operation.

9. The method according to any of the preceding claims, characterized in that the catalyst is a zeolite.

10. The method according to any of the preceding claims, characterized in that the flow velocity of the educt inflow lies between 0.3 and 0.5 kg(educt)/kg(catalyst)^*h^"1.