Membrane-augmented manufacture of cumene

Details Membrane-augmented manufacture of cumene Membrane-augmented manufacture of cumene

2002-05-13

2003-02-25

Padmanabhan, Sreeni (Department: 1621)

Chemistry of hydrocarbon compounds

Purification, separation, or recovery

By membrane, selective septum, or coalescer

C585S446000, C585S450000, C585S819000

Reexamination Certificate

active

06525236

ABSTRACT:

FIELD OF THE INVENTION
The invention concerns manufacturing of propylene derivatives. More specifically, the invention concerns the selective purging of propane and recovery of propylene in the process by using gas separation membranes to treat the reactor vent stream.
BACKGROUND OF THE INVENTION
The United States produces more than 10 billion pounds annually of chemicals derived from propylene. Important derivatives include acrylonitrile, butyl alcohol, propylene oxide, isopropyl alcohol and cumene.
In a typical propylene derivative manufacturing process, propylene and other reagents are introduced into a high-pressure reactor. The raw effluent from the reactor is transferred continuously to one or more separation steps, from which a stream of raw derivative product is withdrawn for further purification. A stream of overhead gases, containing unreacted propylene, is also withdrawn from the separation steps. If the conversion of propylene to product is high, such as 95% or above, these overhead light gases may simply be sent to the fuel line. In many cases, however, propylene conversion per pass is much lower than this, and the overhead gas is recirculated to the reactor. Thus, the propylene feed to the reactor is a combination of fresh propylene and propylene recirculated in the reactor/product separation process loop. The fresh feed is usually chemical-grade propylene, a high-purity reagent that has a propylene content of about 95% or above, the remaining 5% or less being mostly propane, which passes unchanged through the reactor. Although the proportion of inert gas introduced into the reactor loop with the fresh feed in this way is small, the amount circulating builds up quickly, reducing the catalyst activity and reactor productivity. Propane build-up is usually controlled to a steady-state propane content in the loop in the range about 5-30%, by continuously venting a small fraction of the overhead gas from the recirculation loop.
Such a purge operation is unselective however, and, since the vent stream may contain as much as 90 vol % or more of propylene, multiple volumes of propylene may be lost from the loop for every volume of propane that is purged. Even though the volume of gas vented is only a few percent of the volume of fresh feed, the propylene lost in this way may typically amount to a few million pounds per plant per year, with a value of $1 million or more.
Despite its high value, propylene recovery from the purge stream, by separating it from the propane before the propane is vented, is generally not cost effective. Separation of propylene from propane is difficult, because of the similar physical properties, including close boiling points (propylene, −48° C. and propane −42.2° C.). When high-purity propylene is manufactured, it is separated from propane in a C
3
splitter, a large cryogenic distillation column that typically contains 150 or more trays. It is clearly not practical or economic to install such equipment solely for purge treatment. Pressure swing adsorption (PSA), which can make product streams of high purity, has also been considered, but available adsorbents are not very effective, and PSA systems are also costly and energy intensive.
In summary, the concentration of propane in the reactor is regulated by the rate of purging via the vent stream; to achieve maximum productivity from the reactor and a high conversion of propylene to product, the propylene concentration in the reactor should be as high, and the propane content as low, as possible. Without the ability to recover propylene from the purge gas, however, there is an inevitable trade-off between controlling propane concentration in the reactor and losing propylene feedstock in the purge vent stream, by which operators of polypropylene plants are constrained. By purging a chosen percentage of the effluent light overhead gas, the operator makes what is for him, in the circumstances specific to the plant, the most acceptable compromise between the two undesirable extremes of excessive propylene loss and excessive loss of reactor efficiency.
Separation of propylene from propane by means of membranes is discussed extensively in the literature. It is well known that numerous materials and membranes exist with intrinsic selectivity for propylene over propane. These include facilitated transport membranes, polymeric membranes and inorganic membranes.
Membranes are not immediately attractive, however, for propylene recovery from vent gas, because, unlike PSA and cryogenic distillation, membranes are not able to produce a high-purity propylene permeate stream and at the same time achieve high levels of propylene recovery. The reason for this is that a membrane is not a perfectly selective barrier. If the membrane area and time in which the molecules of the gas stream are in contact with the membrane surface are very small, only a very small cut of the total feed flow will permeate. Since propylene permeates faster than propane, most of this small permeate cut will be propylene. That is, the permeate stream will have high propylene purity, and the residue stream will have a composition that is not much changed from the membrane feed composition. In other words, most of the propylene that was present in the feed gas will remain on the feed side, and will be lost when that gas is vented. Propylene recovery can be increased by increasing membrane area and contact time for the gas molecules. However, in this case propane permeation will also be increased. In other words, increasing propylene recovery also results in increasing propane recovery. Thus, little propylene will remain on the feed side to be lost by venting, but the recovered gas will be of low propylene purity.
Furthermore, the propylene/propane selectivity that can be obtained from most membranes under real operating conditions is small. Although literature references cite propylene/propane selectivities even as high as 50 or more, these data have generally been obtained from experiments with pure gases under low feed pressure conditions and with a vacuum on the permeate side of the membrane. With gas mixtures at high pressures, the best propylene/propane selectivity that can be obtained is typically no higher than between 2 and 3.
Despite these inherent difficulties, it has been proposed to apply membrane separation to the recovery of light olefins from reactor vents. U.S. Pat. No. 4,623,704 describes such a process for recovering ethylene from the reactor vent of a polyethylene plant. In this case of polyethylene manufacturing, the reactor is run at very high ethylene, very low ethane levels, so that the vent stream contains 96.5% ethylene, only 2.7% ethane and smaller amounts of methane and nitrogen. The stream is passed across a cellulose triacetate membrane that is selective for ethylene over ethane. Although the membrane selectivity is poor, the membrane produces an upgraded permeate stream, now containing 97.9% ethylene and 1.5% ethane, which is considered sufficiently free of impurities for return to the reactor, and a residue stream containing 89.9% ethylene, .8.5% ethane, which is purged from the reactor loop and used as fuel gas.
A chapter by R. D. Hughes et al., entitled “Olefin Separation by Facilitated Transport Membranes”, in
Recent Developments in Separation Science,
N. N. Li et al. (Eds), CRC Press, 1986, discusses pilot-scale tests of a facilitated-transport membrane module at a polypropylene plant. The module was used to treat vent gas from the reactor with a view to recovering propylene. The test was a technical success for the membranes, in that the module was able to produce a permeate stream typically containing about 97-99% propylene. However, since the membrane process could not produce polymer-grade propylene, the permeate was not recirculated to the reactor, and the process was not pursued.
Thus, recovery of propylene from the propane vent stream of reactors using propylene as a feedstock has been recognized to be desirable for many years. It has also been recognized that the recovered propylene

Affiliated with

Also associated with

Rating

Membrane-augmented manufacture of cumene does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Membrane-augmented manufacture of cumene, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Membrane-augmented manufacture of cumene will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-3166250