Olefin epoxidation process (Shell Oil Company)

PATENT
Olefin epoxidation process (Shell Oil Company)
Publication number
US8546592 B2
Application number
US 13/246,325
Publication date
Oct 1, 2013
Inventors
Wayne Errol Evans, Paul Michael Mcallister, Randall Clayton Yeates, Jeffrey Michael Kobe
Original Assignee
Shell Oil Company
Abstract
A process for the production of an olefin oxide, which process comprises reacting a feed comprising an olefin and oxygen in a reactor tube in the presence of a silver-containing catalyst, wherein the presence of water in the catalyst bed is controlled such that the ratio of the partial pressure of water (PPH2O) divided by the vapor pressure of water (VPH2O) is less than 0.006, preferably less than 0.004.
BACKGROUND OF THE INVENTION
In olefin epoxidation an olefin is reacted with oxygen to form an olefin epoxide, using a catalyst comprising a silver component, usually with one or more further elements deposited therewith on a support. The olefin oxide may be reacted with water, an alcohol, carbon dioxide or an amine to form a 1,2-diol, a 1,2-diol ether, 1,2-carbonate or an alkanolamine. Thus, 1,2-diols, 1,2-diol ethers, 1,2-carbonates and alkanolamines may be produced in a multi-step process comprising olefin epoxidation and converting the formed olefin oxide with water, an alcohol, carbon dioxide or an amine.
The performance of the epoxidation process may be assessed on the basis of the selectivity, the catalyst's activity and stability of operation. The selectivity is the molar fraction of the converted olefin yielding the desired olefin oxide. The catalyst is subject to an ageing-related performance decline during normal operation. The ageing manifests itself by a reduction in the activity of the catalyst. Usually, when a reduction in activity of the catalyst is shown, the reaction temperature is increased in order to compensate for the reduction in activity, however at the expense of selectivity. In the typical operation of a fresh catalyst, the process is operated at a reaction temperature of up to about 250° C. Upon catalyst ageing the reaction temperature may gradually be increased to values substantially above 250° C. until the reaction temperature becomes undesirably high or the selectivity becomes undesirably low, at which point in time the catalyst is deemed to be at the end of its lifetime and would need to be exchanged. It goes without saying that from an economical point of view it is highly desirable to improve the performance of the catalyst and to extend its lifetime as much as possible. Quite modest improvements in the maintenance of selectivity over long periods yield huge dividends in terms of efficiency in the olefin epoxidation process and, if applicable, also in the overall process for the production of a 1,2-diol, a 1,2-diol ether, 1,2-carbonate or an alkanolamine.
Therefore, for decades much research has been devoted to improving the activity, the selectivity and the lifetime of the catalysts, and to find process conditions which enable full exploitation of the catalyst performance. For example, it is well known that low CO₂ levels are useful in improving the selectivity of high selectivity catalysts. See, e.g., U.S. Pat. No. 7,237,677; U.S. Pat. No. 7,193,094; US 2007/0129557; WO 2004/07873; WO 2004/07874; and EP 2,155,708. These patents also disclose that water concentration in the reactor feed should be maintained at a level of at most 0.35 mole percent, preferably less than 0.2 mole percent. Other patents disclose control of the chloride moderator to maintain good activity. See, e.g., U.S. Pat. No. 7,657,331; EP 1,458,698; and US Pub. Pat. App. 2009/0069583. Still further, there are many other patents dealing with the epoxidation process and means to improve the performance of the catalyst in the process. See, e.g., U.S. Pat. Nos. 7,485,597, 7,102,022, 6,717,001, 7,348,444, and US Pub. Pat. App. 2009/0234144.
Notwithstanding the improvements already achieved, there is a desire to further improve the performance of the silver-containing catalysts in the production of an olefin oxide, a 1,2-diol, a 1,2-diol ether, a 1,2-carbonate or an alkanolamine.
SUMMARY OF THE INVENTION
The present invention provides a process for the production of an olefin oxide, which process comprises reacting a feed comprising an olefin and oxygen in the presence of a supported silver-containing catalyst loaded into a reactor tube (i.e., the catalyst bed), wherein the presence of water at any point in the catalyst bed is controlled such that the ratio of the partial pressure of water (PPH₂O) divided by the vapor pressure of water (VPH₂O) is less than 0.006, preferably less than 0.004. As shown in the examples which follow, even the low levels of water that had been considered acceptable in the past are detrimental to the performance of a silver-containing catalyst.
This invention constitutes a means to reduce the rate of selectivity loss by an epoxidation catalyst while in operation that is different from the well-known effects of time and temperature described above in the prior art. In this invention the water vapor concentration in the catalyst bed is reduced to certain levels in order to reduce significantly the rate of selectivity loss and the overall selectivity loss during the catalyst operational cycle. This is different from the prior art view, because the primary effect of reducing water vapor concentration is not a slower decline rate due to lower operating temperature for the catalyst. In this invention, we have found that water causes another ageing mechanism which may actually lower the rate of temperature increase but at the same time cause accelerated selectivity loss. In this invention we have found that significantly lower concentrations of water in the vapor phase at conditions where condensation of liquid water is not possible has resulted in changes in the catalyst that lead to loss of selectivity. The hygroscopic nature of the catalyst or catalyst support results in adsorption of water on the surface of the catalytic material even when conditions are such that liquid water should not be present on the catalyst or internal reactor surfaces. i.e., well above the dew point of water. Thus, the presence of excess water in the vapor phase will suppress selectivity and lead to increased rates of sintering or losses of key water soluble dopants from the catalyst surface.
In the present invention, we have found that the redistribution of key water soluble dopants on the surface of the catalyst can be greatly reduced and therefore the catalyst selectivity loss rates can be reduced significantly by reduction of the ratio of the partial pressure of water (PPH₂O) divided by the vapor pressure of water (VPH₂O) at the inlet and throughout the catalyst bed. Vapor phase water is introduced in a typical commercial reactor in the feed gas at the inlet of the reactor as well as by generation within the reactor due to the complete combustion of a portion of the ethylene fed to the reactor to CO₂ and water. See, e.g., US Pub. Pat. App. 2009/0234144, which disclosure is incorporated in its entirety herein. There are a number of ways by which the ratio of the partial pressure of water (PPH₂O) divided by the vapor pressure of water (VPH₂O) can be reduced. These include:
Increased cooling of the overhead streams coming from the ethylene oxide (“EO”) removal and/or CO₂ removal sections of the plant that return to the reactor.
Diversion of less of the recycle gas through the CO₂ absorber.
Operation of the EO and CO₂ absorbers at lower temperature.
Increasing the Gas Hourly Space Velocity at fixed EO production to reduce the water concentration gradient increase in the reactor.
Reduction in work rate or EO production per unit volume of catalyst to reduce the amount of H₂O formed in the reactor.
Utilization of catalysts with higher selectivity such that the amount of water produced across the catalyst bed is reduced for a given EO production rate.
Reducing the reactor operating pressure so as to reduce the partial pressure of H₂O.
Operation of the reactor at higher temperature than required to increase the vapor pressure of water.
These are some of the means by which the ratio of the partial pressure of water (PPH₂O) divided by the vapor pressure of water (VPH₂O) in the reactor/catalyst bed can be reduced, but this is not an exhaustive list. The concept can be applied to existing plants by making changes in operating variables and/or changes to plant hardware such as heat exchangers, absorbers, and compressors. The concept can be applied to new plants in the design phase as well.
A quantitative analysis method has been developed to determine the level of water vapor which causes accelerated selectivity loss of epoxidation catalysts while in operation. Extensive evaluation of post mortem results of spent catalysts demonstrated that surface concentrations of water soluble dopants as measured by X-ray Photoelectron Spectroscopy (“XPS”) were reduced significantly when the ratio of the water partial pressure in the gas phase to the vapor pressure (PPH₂O/VPH₂O) of water at the location of the sample in the reactor during operation exceeded 0.004. The reduction of surface concentration of these water soluble dopants is directly linked to selectivity loss of the catalyst. Samples which were not exposed to PPH₂O/VPH₂O>0.004 showed much less reduction in the surface concentration of water soluble dopants and much less selectivity loss. It is most preferable that the ratio of PPH₂O/VPH₂O is less than 0.004 over the entire length of the catalyst bed. But advantages are also shown where the ratio is less than 0.004 over a portion of the catalyst bed—for example where the ratio is less than 0.004 over greater than 50% of the reactor tube length (defined as the length from the catalyst bed inlet to the catalyst bed outlet), preferably over greater than 80% of the reactor bed length.
While ratios of PPH₂O/VPH₂O<0.004 are highly desirable, it may not be possible to achieve this in many commercial plants throughout the entire catalyst bed due to hardware limitations, operating constraints, or EO production requirements. This does not preclude a plant from taking advantage of the concept. Reduction of PPH₂O/VPH₂O is expected to be beneficial no matter what the starting point. Thus, if a plant can reduce PPH₂O/VPH₂O from 0.007 to 0.006 one would still expect a beneficial effect. Likewise, it may be possible to increase the portion of the catalyst bed that operates at PPH₂O/VPH₂O<0.004. This will have a beneficial effect as it will reduce the rate of selectivity loss from this portion of the bed and will have a beneficial effect on the rest of the catalyst bed which may operate above this threshold.
The finding that water vapor concentrations can cause accelerated selectivity loss is an unexpected result as the traditional view has been that higher temperatures were the primary driving force behind selectivity loss. Thus, the invention described here is fundamentally different from prior methods attempted to increase initial selectivity and to reduce the rate of selectivity decline for a given catalyst. Another unexpected aspect of the invention is that the moderator level needs to be shifted, usually towards higher levels, as the water level is decreased, in order to maintain optimum performance. This is not intuitive—water level has little overall impact on catalyst activity, and the “Q” factor optimum is primarily a function of temperature, so most operators wouldn't see any reason to change the “Q” factor with water levels. Those that do would change the “Q” factor proportionally to the water level, in order to compensate for H₂O adsorption, but in most situations, this change is in the wrong direction. The “Q” factor is taught in U.S. Pat. No. 7,193,094, which patent is incorporated by reference herein. However, in the '094 patent, the moderator level is primarily a function of the reactor temperature. This current invention reveals that proper control of the “Q” factor requires re-optimization of the moderator level whenever any significant shift in water levels occurs in the catalyst bed.
The invention also provides a method of using an olefin oxide for making a 1,2-diol, a 1,2-diol ether, 1,2-carbonate or an alkanolamine comprising converting the olefin oxide into the 1,2-diol, the 1,2-diol ether, 1,2-carbonate or the alkanolamine, wherein the olefin oxide has been obtained by the process according to this invention.
Free Full Text Source: http://www.google.com/patents/US8546592