Gas separation: processes – Solid sorption – Including reduction of pressure
Reexamination Certificate
2000-11-09
2002-10-29
Spitzer, Robert H. (Department: 1724)
Gas separation: processes
Solid sorption
Including reduction of pressure
C095S119000, C095S130000, C096S130000, C096S132000, C096S144000
Reexamination Certificate
active
06471748
ABSTRACT:
FIELD OF THE INVENTION
The invention relates to pressure swing adsorption (PSA) processes and apparatus, more particularly to the use of high performance adsorbents in PSA processes and systems through the novel deployment of such adsorbents in layers.
BACKGROUND
Cryogenic methods have dominated air separation processes for many years where high purity O
2
, N
2
and or Ar are desired. More recently, both membrane and adsorption processes have become important commercially. In particular, PSA, including superatmospheric adsorption/desorption processes, subatmospheric vacuum swing adsorption (VSA) and transatmospheric vacuum pressure swing adsorption (VPSA) processes are well known in the art. Such methods are typically used to produce oxygen having a purity between about 90 to 95%. There is an increasing need for this purity O
2
in such diverse industries as steel making, glass making and pulp and paper production. Single plant oxygen capacity for such adsorption processes now exceeds 100 tons-per-day contained O
2
(TPDO), and applications continue to arise demanding even greater capacities. At these production and purity levels, O
2
product cost is lower by adsorption than by cryogenic methods, while for larger capacities, economies of scale currently favor the cryogenic methods. Nevertheless, there continues to be considerable economic incentive to extend the production range of adsorption processes for air separation. This must be accomplished by improving performance while reducing the cost of power and capital.
A typical adsorption system for the production of O
2
includes one or more adsorber vessels containing a layer of pretreatment adsorbent for removing atmospheric contaminants followed by a main adsorbent. The pretreatment adsorbent can be any material primarily effective in removing H
2
O and CO
2
, e.g. zeolites, activated alumina, silica gel, activated carbon and other such adsorbents. The main adsorbent material, which usually represents at least 90% of the total volume of adsorbent in the vessel is N
2
—selective, typically from the type A or type X family of zeolites. While many different adsorption cycles have been developed for O
2
production, all pressure swing cycles contain the four basic steps of pressurization, adsorption, depressurization and desorption. When multiple beds are used, the beds are sequenced out of phase for the different cycle steps in order to maintain a constant flow of product. One of many examples of such processes illustrating these basic features is given by Batta in U.S. Pat. No. 3,636,679.
There has been significant development of the various PSA, VSA and VPSA methods for air separation over the past thirty years, with major advances occurring during the last decade. Commercialization of these processes and continued extension of the production range can be attributed primarily to improvements in the adsorbents and process cycles, with advances in adsorber design contributing to a lesser degree. Highly exchanged lithium molecular sieve adsorbents, as illustrated by Chao in U.S. Pat. No. 4,859,217, are representative of advanced adsorbents for O
2
production. A historical review of both adsorbent and process cycle development may be found in Kumar (Sep. Sci. and Technology, 1996).
The increase in N
2
/O
2
selectivity and N
2
working capacity associated with N
2
-selective advanced adsorbents is largely responsible for the improvements in O
2
recovery and reduction in power and bed size factor (BSF). Such adsorbents, however, often have higher heats of adsorption, are more difficult to manufacture and may have poorer mass transfer characteristics, all resulting in a higher adsorbent cost. While many new adsorbents have been developed claiming improved properties for air separation, only a few have been implemented successfully in commercial processes. Advanced adsorbents often fail or fall short of expectations since process performance is projected on the basis of adsorbent equilibrium properties and isothermal-process conditions.
Collins in U.S. Pat. No. 4,026,680 teaches that adiabatic operation intensifies the thermal effects in the adsorbent bed inlet zone. In particular he teaches that there is a “sharply depressed temperature zone,” (hereinafter referred to as a “cold zone”), in the adsorption bed inlet end. This zone is as much as 100° F. below the feed gas temperature. Such a zone results in a thermal gradient over the length of the adsorbent bed of approximately the same magnitude (e.g. about 100° F.). Collins suggests that the cold zone arises from the coupling of an “inadvertent heat-regenerative step” at the inlet end of the bed with the thermal cycling resulting from the adsorption/desorption steps of the process. The regenerative effect may be partly the result of the adsorption of water vapor and carbon dioxide in a pretreatment zone located ahead of the main adsorbent.
The thermal cycling that occurs in an adiabatic process results in an adverse thermal swing, i.e. the adsorption step occurs at a higher temperature than the desorption step. This thermal swing tends to increase with increasing adsorbate/adsorbent heats of adsorption and increasing ratio of adsorption to desorption pressure. These gradients and swings in bed temperature result in various parts of the adsorbent bed functioning at different temperatures. The N
2
/O
2
selectivity and N
2
working capacity of any particular adsorbent may not be effectively utilized over such wide ranges in bed temperature. Dynamic adsorbent properties that vary strongly with temperature are also likely to result in process instability when operating conditions, such as ambient temperature, change.
Considerable attention has been given to eliminating or minimizing the cold zone in adiabatic adsorbers since Collins. Earlier suggestions included raising the feed temperature using external heating or through partial bypass of the feed compressor aftercooler.
Collins proposed the use of heat conducting rods or plates extending the length of the bed for the same purpose. Others have extended this concept by replacing the rods or plates with hollow tubes filled with liquid to provide heat transfer by convection between the warmer product end and the colder feed end of the adsorber. For example, the cold zone temperature is increased from −70° C. to near 0° C. in Fraysse et al. (U.S. Pat. No. 5,520,721) by supplying a heat flux to a passage between the pretreatment and main adsorbents. The primary intent in all of these methods is to elevate the minimum temperature near the feed inlet of the adsorber using direct and/or indirect heat exchange. The entire bed temperature is elevated along with the cold zone temperature when the feed is heated, however, and the overall size of the thermal gradient in the bed remains relatively unaffected.
Another approach attempts to match an adsorbent with a temperature that is most efficient for the desired separation. Typical of such teachings, an adsorbent bed is divided into layers that are maintained at different temperatures using embedded heat exchangers to affect distinct separations.
Armond (EP 0512781 A1) claims to inhibit the effect of the cold zone by selecting two unspecified adsorbents with high removal efficiency for N
2
, at −35° C. to −45° C. and at ambient temperature, respectively. The low temperature material is located near the feed inlet (but downstream of the pretreatment adsorbent) and is followed by the second material.
A main adsorbent, containing at least two layers, has been disclosed by Watson et al. (U.S. Pat. No. 5,529,610) for O
2
production. Watson teaches that no commercially available adsorbent functions optimally over the large temperature gradient (as much as 100° F.) that exists in the main adsorbent region of the bed. NaX zeolite, comprising from 20% to 70% of the total adsorbent volume, is chosen for the lowest temperature region of the bed due to its low capacity and high selectivity at such temperatures. The second layer is preferably CaX zeolite, although other high capacity
Follett Robert J.
Praxair Technology Inc.
Spitzer Robert H.
LandOfFree
Multilayer adsorbent beds for PSA gas separation does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Multilayer adsorbent beds for PSA gas separation, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Multilayer adsorbent beds for PSA gas separation will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2956687