Overview on C2 and C3 Selective Hydrogenation in Ethylene Plants

Edgar L. Mohundro Consultant

Prepared for Presentation at the American Institute of Chemical Engineers 15 th Ethylene Produces Conference 2003 Spring National Meeting New Orleans, LA March-April 3, 2003

Session: Ethylene Plant Technology C2 and C3 Hydrogenation Technology Review 64a

Unpublished AIChE shall not be responsible for statements or opinions contained in papers or primed in its publications

531

Introduction
This paper will provide an overview on C2 and C3 selective hydrogenation operations in ethylene plants. The primary focus will be on selective hydrogenation of acetylene. The reason for this is that most C3 selective hydrogenation units installed since the mid-late 1970's are liquid phase units. For the most part, the details about the design and operation of liquid phase processes are proprietary, licensed technology. This is an overview paper, providing background information for the succeeding papers in this session. The following papers will cover in more detail some experiences and know-how with acetylene and/or C3 selective hydrogenation within ethylene plants. Acetylene and C3 hydrogenation units are purification steps for ethylene and propylene. Trace levels of acetylene, methyl acetylene (MA), and propadiene (PD) are formed in the steam cracking furnaces. The amount of acetylene and MAPD formed is a function of the furnace feed, along with furnace design and operating conditions. Acetylene will fractionate with ethylene; MAPD will fractionate with propylene. These components are only tolerable at < 1 ppm in polymer grade ethylene and propylene. The overwhelmingly dominant method for removal of acetylene and MAPD is through selective hydrogenation. With few exceptions, acetylene hydrogenation is carried out, in vapor phase, adiabatic fixed bed reactors. The number of beds ranges from 1 to 4, with intercooling. Acetylene hydrogenation is also done with isothermal tubular reactors. Liquid phase C3 hydrogenation became the "process-of-choice" around the mid-70' s. The dominant process is an adiabatic, fixed-bed, liquid phase system. Some C3 liquid phase hydrogenation units have isothermal, tubular reactors. Many vapor phase C3 hydrogenation units are still in service. During past Ethylene Producers Conferences (EPC), there have been several papers that deal with acetylene and C3 hydrogenation and the impact of contaminants on these operations. These are good reference documents. They are given in the "List Of References" for this paper. This paper will cover the following items"

Acetylene Hydrogenation
" ' " " " " " " " Options for locating hydrogenation units within ethylene plant sequences Pros/Cons for Each Location Operating Strategy General Design Comments Promoted Catalysts Theory of Acetylene Hydrogenation Green Oil Comments on Contaminants Regeneration Considerations

532

General Comments on C3 Hydrogenation Safety Considerations

Acetylene Hydrogenation
The basic requirements for acetylene hydrogenation units include the following" Safe operation; e.g., no exothermic runaway reactions; no "hot spots" during operation Maintain acetylene in ethylene product below specification. The typical maximum is 5 ppm; however, ethylene customers may expect the acetylene level to be maintained below 1 ppm; perhaps, below 0.5 ppm. Do not be the cause of interruptions to continuous operation of the ethylene plant; e.g., a target for continuous operation may be 5+ years.

Some secondary requirements or "very strong wishes" are: Operating with a net gain of ethylene; the higher, the better. "Low investment cost"---within the context of the overall cost considerations for ethylene plant investment, project execution, feedstocks, energy, etc. Good catalyst life and reasonable catalyst cost

L o c a t i o n w i t h i n the E t h y l e n e P l a n t P r o c e s s S e q u e n c e An acetylene hydrogenation unit can be located at one of several locations within the ethylene plant sequence. There are two major designations for these locations: These are: Front-end Tail-end or Back-end

In the Front-End category, there are several possibilities. These are: DeC2 overhead in a DeC2-1st process sequence (Figure 1) DeC3 overhead in a DeC3-1st process sequence (Figure 2) Raw Gas: within the process gas compressor circuit, downstream of the caustic tower (Figure 3)

533

A subset for types of front-end hydrogenation units deals with the type of catalyst. The dominant catalyst is Pd-based, with the "modem" catalyst being bimetallic, palladium promoted with a second metal; e.g., with silver. Another type of catalyst is sulfided NiCo. This catalyst was introduced 40+ years ago for front-end hydrogenation units. This catalyst operates with inlet temperatures that range from about 350 F (~175 C) to 450 F (~230 C). Sud Chemic still offers this catalyst, C-36, as a replacement catalyst for the "high temperature" units still in operation. This paper will not cover this type of selective hydrogenation unit. Major debits for this process are" The NiCo catalyst must be presulfided and maintained in a sulfided state via continuous injection of sulfur, in form of H2S, ethyl or n-butyl mercaptan, or DMDS, Typically, this hydrogenation unit operates with an ethylene loss.

The Tail-end or Back-end hydrogenation unit is associated with a demethanizer-first (DeC 1-1 st) ethylene recovery sequence. This unit is located upstream of the ethylene/ethane splitter, on the net D e C 2 0 / H . (Figure 4) At the tail-end location, hydrogen is injected into the reactor feed; whereas, with front-end hydrogenation units, all the hydrogen produced in the cracking furnaces is in the reactor feed.

Comments about Acetylene Hydrogenation Ethylene Plant Process Sequences

at V a r i o u s L o c a t i o n s w i t h i n

Front-End: Overhead of a Front-end DeC2 The feed to this hydrogenation unit contains the C2 and lighter material. The feed is very rich in hydrogen and ethylene, particularly if the furnace feed is ethane. The only unsaturated reactants in the feed to this reactor are acetylene and ethylene. Some pros and cons about this location include the following:

Pros: Clean stream; i.e., no traces of BD to cause catalyst fouling Do not need a spare reactor High concentration of H2 suppresses green oil formation Reactors are smaller, relative to the front-end D e C 3 0 / H system

534

Cons" High concentration of H2 may result in the operation being sensitive to initiation of ethylene hydrogenation; this may result in exothermic runaways Stability of operation is sensitive to CO fluctuations. This is because there is no MAPD in the feed to provide additional moderation.

Front-End: Overhead of a Front-end DeC3 The feed to this acetylene hydrogenation unit contains the C3 and lighter material. The feed is somewhat rich in hydrogen and ethylene. The primary difference between an acetylene hydrogenation unit in this location and one located in the front-end D e C 2 0 / H is the presence of MAPD in the feed. Some pros and cons about this location include the following"

Pros" Clean stream; however, there is the potential for the feed to contain traces of BD Moderately high concentration of H2; this suppresses green oil formation Presence of MAPD provides additional moderation that inhibits hydrogenation of ethylene, and hence, mitigates against exothermic runaways About 50 % of the MAPD is hydrogenated; essentially all of this is converted to propylene. This also means that any downstream C3 hydrogenation facilities will be smaller.

Cons" Acetylene plus MAPD impose a higher "load" on the reactor. This translates to higher overall heat rise, increasing the rate of catalyst fouling. Traces of BD may be in the converter feed; this will bring about additional catalyst fouling. May be difficult to operate without a spare reactor for continuous operation of 5+ years. Reactors are larger than for front-end D e C 2 0 / H units

535

Front-End" Raw Gas Location: This reactor is placed within the process gas compressor circuit, downstream of the caustic tower, but upstream of the main driers. Hence, the feed to this acetylene hydrogenation unit is wet C5 and lighter, containing some C6's. Historically~ this unit was designed for operation with sulfided NiCo catalysts. This type acetylene hydrogenation unit was installed in many ethane crackers designed/built in the mid 60' s/early 70's. For many years, Pd or Pd/promoter catalysts were not considered for this service. Some reasons for this include concerns that Pd-based catalysts were not "sufficiently rugged" for this severe service. In addition, it is known that water inhibits the activity of Pd catalysts, and it is known that water tends to increase the formation of green oil across Pd catalysts. However, our industry know-how is now such that ethylene plants are being designed and/or retrofitted with Pd/promoter catalysts in raw gas acetylene hydrogenation units. Hence, there have been some recent grass-roots designs with Pd/promoter catalysts in raw gas units, and there have been some conversions from sulfided NiCo to Pd/promoter.

Pros: This operation will result in hydrogenation of C4 acetylenes and BD, hence minimizing the potential for fouling at bottom of downstream DeC2 and/or DEC3. Converts the majority of C3 and C4 acetylenes and diolefins into olefins, making the final C3 and C4 streams more attractive for fuel gas, and/or for recycle cracking. Contributes to other cost savings within the ethylene plant; for illustration, this allows for the elimination of a set of guard driers.

Cons: Catalyst deactivation can be relatively rapid, especially for the first bed This is primarily due to the presence of C4 and C5+ acetylene and diolefins. Hence, even with Pd/promoter catalysts, a spare reactor system is needed for continuous operation of 5+ years. May be difficult to maintain operation with an ethylene gain, especially if catalyst is sulfided NiCo Some acid species may be formed across the reactor. This may result in having to deal with downstream corrosion concerns. For sulfided NiCo catalyst, it is necessary to have a spare reactor system, along with facilities for insitu regeneration, reduction, and presulfiding. There are environmental concerns with handling sulfiding agents such as n-butyl mercaptan, ethyl mercaptan, DMDS, etc.

536

Tail-End Location The primary characteristics that differentiate this system from front-end systems are: The feed is C2's, the O/H of a backend DEC2. Hydrogen is controllably injected to each bed; typically at molar H2/C2H2 ratios between 1.2 to 2.2. Green Oil, oligomers of acetylene, causes fairly rapid catalyst deactivation. There will be spare reactors with facilities for insitu catalyst regeneration and reduction. (Figure 5) Each reactor may be regenerated about 1-4 times a year.

Tail-end acetylene hydrogenation is one of the final steps in the recovery and purification of ethylene to produce polymer grade ethylene. This C2-stream may contain 0.5 to 2.5% acetylene. Some concerns/considerations with tail-end acetylene hydrogenation include: Acetylene will be upgraded to ethylene and downgraded to ethane and green oil (oligomers of acetylene, mostly C4's with small amounts of C6 to C20+ material) Tail-end units have downstream green oil (C6+) removal facilities Unreacted H2 in the reactor effluent must be removed via a pasteurization section at the top of the ethylene/ethane splitter or via a downstream secondary demethanizer The character of the catalyst changes throughout the cycle. This change is primarily due to heavy green oil lay down on the catalytic sites. This change is temporary. The catalyst is "rejuvenated" via insitu steam/air regeneration followed by reduction with hydrogen. The character of the catalyst undergoes slow "permanent" changes through online operation and on being subjected to repeated regeneration. Catalyst life is 5+ years; perhaps, even 20+ years. Economic considerations drive operators toward optimizing tail-end hydrogenation units to maximize ethylene gain; however, this can make the operations vulnerable to acetylene breakthrough. An acetylene off-spec incident may result in an substantial production loss of ethylene; e.g., perhaps 1000+ Tons This is because it takes several hours to eliminate acetylene from the ethylene/ethane splitter inventory.

A presentation to the 1999 EPC (Ref 34) is a good source for more information on the advantages and/or concerns with locations for the acetylene hydrogenation unit within ethylene plant sequences.

537

Operating Strategy
The operating strategy for acetylene hydrogenation units is primarily based on staying within the operating window for selective stable performance. (Ref. 5, 6, 34, 35, 36) This means that acetylene will be hydrogenated, completely, and the hydrogenation of ethylene will be minimized. If the catalyst activity is too low, then acetylene leakage will occur. If the catalyst activity is too high, excessive ethylene hydrogenation will occur. The "location and size" of the operating window are affected by several parameters. These include" Inlet temperature Amount of hydrogen ,/ For front-end units, this parameter is the hydrogen partial pressure. " For tail-end units, this parameter is the H2/C2H2 ratio. Inlet acetylene, ---and inlet MAPD for front-end units located in D e C 3 0 / H Level of CO Space velocity .... more of a problem for front-end units Trace catalyst poisons Catalyst type, e.g., promoted catalysts have larger operating window Condition of catalyst; e.g., new, old, history of runaways, amount of time online, etc.

Front-end Acetylene Hydrogenation: For front-end units, the only control parameter is the inlet temperature for each bed. Bed outlet acetylene level is used for feedback control of the inlet temperature. CO fluctuations in the feed present a control problem for these units. In addition, the operation becomes unstable at low space velocities; hence, the flow through the reactors must be maintained above a specified minimum.

Tail-end Acetylene Hydrogenation: There are two primary control parameters for the reactors at this location: Inlet temperature and the H2/C2H2 ratio. Occasional changes may be made to the amount of CO being injected into the reactor feed. Movement "above" the operating window can be effected not only by increasing inlet temperature, but also by injecting too much hydrogen. High temperatures and/or high hydrogen injection rates will lead to nonselective performance, and consequently, acetylene leakage.

More information on modeling and control of acetylene hydrogenation units can be found in References 21 to 26.

538

General Design Comments"

For front-end adiabatic units, the following provides illustrative design parameters" Space velocity for each bed: 8000-12000 V/V/Hr--Gas Volumetric Hourly Rate @ Std Cond./Cat. Vol. No.of beds in series, with intercooling: " Feed acetylene at/below 5000-6000 ppm: two beds " Feed acetylene between about 6000-7500 ppm: three beds " Feed acetylene between 7500-10,000 ppm: four beds For tail-end adiabatic units, the following design parameters may apply" Space velocity for each bed: about 4000 V/V/Hr No. of Beds in series, with intercooling: " Feed acetylene at/below 5000 ppm: one bed " Feed acetylene between 0.5 and 1.7%: two beds " Feed acetylene between 1.7-2.5 %: three beds

More information on design/operating parameters for acetylene hydrogenation parameters can be obtained from the catalyst vendors and/or from References 34 through 36.

Promoted or Bimetallic Acetylene Hydrogenation Catalysts

The "modem" or current generation of selective hydrogenation catalysts are often referred to as being promoted or bimetallic. (Ref. 8) Dow first reported the use of palladium/silver bimetallic catalysts for acetylene hydrogenation in 1957. (Ref. 1) However, Chevron Phillips and Sud Chemie gave the commercial impetus to bimetallic catalysts for acetylene hydrogenation by introducing a Pd/Ag catalyst into commercial service on the O/H from a front-end DEC2. (Refs. 2, 5, 6) This was done in the early '80's. This work was aimed at achieving more stable, selective catalyst performance in service on a front-end DeC20/H stream. Chevron Phillips continued research on acetylene hydrogenation catalysts such that they now offer proprietary catalysts, designated "E Series" catalysts. (Ref. 5, 10, 34) These catalysts have been commercialized in several services; e.g., frontend DeC2 O/H, front-end DeC3 O/H, and raw gas. Sud Chemie offer a "family" of Pd/Ag catalysts for all front-end services and for tail-end acetylene hydrogenation. (Ref. 6, 35, 36) Sud Chemie's front-end catalyst is designated G-83C. Their tail-end catalysts are" G-58C, G-58D, G-58E, G-58H, and G-58I.

539

Kataleuna offer a from-end Pd/promoter catalyst designated 7741R. Their catalyst for tail-end service is 7741 A/F. Kataleuna do not provide information about the type of promoter used in formulating these catalysts. (Ref.34) Axens offer LT-279 for front-end and tail-end applications. Axens do not provide details on the promoter used in formulating this catalyst. (Ref. 3, 8) BASF offer HO-21 for tail-end acetylene hydrogenation. This catalyst is Pd, promoted with Ag. The support is silica oxide. Details on the physical and chemical properties for these catalysts can be obtained from the vendors.

Here are some points on catalyst design taken from a paper by Chevron Phillips at the 1993 Ethylene Producers Conference: (Ref 5) The most active and selective catalysts occur when almost all of the Pd is within 300 microns of the surface. The addition of silver to the "Pd skin" catalyst offers a trade of a small loss in catalyst activity for a large increase in selectivity activity. There is an optimum silver loading" too much silver, while increasing selectivity reduces the activity to unacceptable low levels. Silver reduces the impact of variability in Pd distribution within the catalyst pellet.

Comments on the Theory of Acetylene Hydrogenation

From the standpoint of an industrial operation, the theory of acetylene hydrogenation is, at best, imprecise. It is not possible to draw upon a well-formulated, well-understood body of science and apply such information to the industrial environment. There are a number of reasons for this: The primary one being the conversion of acetylene to green oil. This causes a change in the performance of the catalyst throughout a cycle. The change is much more rapid for a tail-end hydrogenation unit.

540

Another reason is that most reported scientific information is for academic conditions; that is, with "new" catalysts, operating at low pressures, for short periods of time on synthetic feeds. The concentration of palladium deposited on the support is so small, e.g., 150-500 ppm; hence, it is very difficult to characterize these catalysts. The catalysts regenerations. undergo changes time onstream and through successive

Reactions The reactions that can occur in tail-end acetylene hydrogenation units and acetylene hydrogenation units located in the O/H of a front-end DeC2 are listed below: C2H2 + H2 ~ C2H4 C2H4 + H2 ~ C2H6 2C2H2 + H2 ~ C4H6 C4H6 +H2 ~ C4H8

(~)

nC2H2 +nil2 --+ Green Oil (Oligomers of C2H2 ) Reaction O is desirable whereas reactions thru , are undesirable.

The amount of C2H2 converted to C6+ in a front-end D e C 2 0 / H acetylene hydrogenation unit is much, much less than is the case for hydrogenation of C2H2 across a tail-end unit. In addition to reactions 1 through 5, the following reactions occur in an acetylene hydrogenation unit located on the O/H of a front-end DEC3. C3H4 + H2 ~ C3H6 C3H6 + H2 ~ C3H8

nC3H4 +nil2 --+ Green Oil (Oligomers of C3H4)

541

Reaction is desirable whereas reactions thru are undesirable.

Reactions and may not occur to a level of commercial significance in acetylene hydrogenation units located on the O/H of a front-end DEC3. Reactions , , and also occur in C3 hydrogenation reactors.

Relative Adsorption According to literature the rate of hydrogenation of ethylene in an ethylene-hydrogen mixture is 10-100 times faster than the rate of acetylene hydrogenation under identical conditions of temperature, pressure and catalyst. Thus the selectivity of acetylene hydrogenation can only be explained by the preferential adsorption of acetylene molecules on the surface of the palladium.

The rate of adsorption of hydrocarbons can be classified in the following decreasing order:

CO > Acetylenes >> Diolefins >> Olefins

Hence, kinetics is controlled by relative adsorption rates. The preferential adsorption of acetylene over ethylene means that provided there are sufficient acetylene molecules to cover all the active sites then ethylene hydrogenation to ethane is minimized. However, in an industrial operation, near the bottom of the final bed, where the acetylene concentration is reduced to about 1-5 ppm, a point is reached in the reactor bed where there is an insufficient number of acetylene molecules to maintain total active site coverage. Fortunately, carbon monoxide provides a means for blocking or inhibiting the adsorption of ethylene, since it has a higher rate of adsorption than that of ethylene. The CO molecules tend to occupy the active sites in the acetylene deficient regions. For front-end acetylene hydrogenation units, CO will be in the reactor feed at a level of about 200-2000 ppm. For tail-end units, trace amounts of CO (0.1-5 ppm) can be injected into the reactor feed. Usually, this CO comes from bypassing a small stream of untreated hydrogen around the PSA unit or methanator.

542

Oligomers (Green Oil) Formation The following discussion pertains, primarily, to tail-end acetylene hydrogenation. There is unanimous agreement in the literature that oligomers (green oil), C4's and higher, are derived from acetylene and not ethylene. The subject of oligomers formation is one of high commercial importance but which is only discussed in patent literature and few published articles. The amount of green oil formation is primarily a function of the concentration of acetylene being converted. Hence, the rate of formation of green oil is higher on the lead bed where higher acetylene concentrations exist. Discussion of fundamental causes of green oil formation engenders a lot of speculation. It doesn't form without the presence of hydrogen, and the amount formed decreases as the hydrogen partial pressure increases. Most researches agree that the precursor to all types of green oil produced in the selective hydrogenation of acetylene is 1,3-butadiene. This molecule is produced from hydrogenation of an acetylene dimer. Catalyst deactivation by green oil becomes a major problem at very low hydrogen to acetylene ratios. Furthermore, the amount of green oil formation has been reported to be related to: surface acidity of the support, palladium crystallite size, mean pore size and pore size distribution, feed water content, and higher operating temperatures. The net result of heavy green oil formation is a shortening of the catalyst cycle length and decrease in the catalyst selectivity. About 10-20 % of the acetylene is converted to C4 and heavier green oil. (Ref 3) The overwhelming majority is C4's. The next major constituent is C6's. A major improvement from the promoted catalysts is that the amount of heavy green oil polymers is substantially reduced. Hence, the rate of deactivation of the catalyst reduced. Two industrial studies were found that dealt with green oil formation with Pd-only catalysts in industrial units. These studies were done using slipstream test units in ethylene plants in China and Russia (Ref. 11, 12). The Russian study reported that the green oils were composed of 23 wt % olefins, 2.5 wt % diolefins and 74.5 wt % aromatic compounds (Ref. 12). The study carried out in China described the green oils as being long chained olefinic compounds with a general formula corresponding to CnH(1.8-1.9)n where n=14-17. These authors found that the quantity of aromatic compounds was negligible and that one quarter of the olefins was alpha olefins (Ref. 11). Most of the published science on selective hydrogenation is based on studies with Pdonly catalysts. However, the latest generations of industrial acetylene converter catalysts are promoted with another metal; in many cases this is Ag. The following information comes from limited published information on the effect of Ag moderation on the Pd catalytic sites.

543

Axens(formerly Procatalyse) reported at the 1995 Spring AIChE Meeting that Ag moderation reduces C6+ formation but has no effect on formation of C4's. (Ref. 3) Axens compared the formation of green oil across a Pd-only catalyst with a Pd/Ag catalyst. Both catalysts were on an alumina support. The feed acetylene was 2.6 %. Here are some highlights from that paper. Oligomerization occurs via a first step of dimerization of acetylene on palladium sites and goes further through polymerization mechanisms to high molecular weight compounds. The C4s remain gaseous; the C6+ compounds are liquids and remain on the catalyst surface, reacting further to make even heavier hydrocarbons. The conversion of acetylene over a Pd-on-alumina catalyst showed that aboutl 5% of the converted acetylene went to C4's and about 30 % of the converted acetylene went to C6+. The conversion of acetylene over the Pd/Ag catalyst showed that about 15% of the converted acetylene went to C4's and the amount of acetylene converted to C6+ was substantially reduced. The amount and type of C4' s formed was independent of the addition of silver to the catalyst. The amount of acetylene converted to ethylene was increased on addition of the silver. An excerpt from a Sud Chemie brochure on the effect of Ag on green oil formation is as follows: (Ref 35) "This decrease in the production of green oil is accomplished by the use of Ag in conjunction with the palladium hydrogenation metal. In essence, Ag acts like an atomic spacer to increase the distance between the active palladium sites. This lessens the chance for two molecules of acetylene to combine to form the 1,3-butadiene precursor, leading directly to less potential polymer formation." Sud Chemie has reported laboratory data to illustrate the effect of Ag on reducing the amount of heavy green oil formed during hydrogenation of acetylene. This illustration is provided in Figure 6. The amount of polymer formation with the promoted catalyst is only about 40% of the amount produced with the Pd-only catalyst.

Catalyst Poisons and Inhibitors of Pd Catalysts"

Typical ethylene plant contaminants that can cause problems for acetylene and C3 selective hydrogenation catalysts are CO, H20, H2S, and mercaptans. Methanol also causes problems. Other Pd catalyst poisons frequently encountered include COS, arsine,

544

and Hg. A thorough discussion on catalyst poisoning is beyond the scope of this paper. For more information on this subject, see Reference articles 10 and 13 through 20. Some comments on these poisons are as follows: CO is a temporary inhibitor. It is important in providing selectivity/stability to the front-end operations. However, it can cause problems for front-end operations if the concentration fluctuates widely. It is injected, on-purpose, to control selectivity of tail-end acetylene hydrogenation catalysts. The effects from H20 and methanol are similar. H20 is considered to be a green oil promoter. H2S and COS are severe poisons at <1 ppm. COS is not removed via caustic. It causes problems for acetylene converters in the raw gas and front-end D e C 3 0 / H positions. It is also a poison for C3 hydrogenation catalysts Arsenic, in the form of arsine, can poison catalysts in any of the acetylene and C3 hydrogenation units. Arsine, mostly, tracks along with propylene; however, traces of arsine will track ethane. If arsine and COS are present in the ethylene plant process gas, then adsorption bed(s) will have to be installed to prevent catalyst poisoning. (Refs. 17 and 18)

Comments on Catalyst Regeneration

Typically, front-end catalysts are regenerated exsitu. Special diligence must be exercised in getting Pd/promoter catalysts regenerated. This is especially the case if expectations are for the regenerated catalyst to operate for 3-5+ years. The regeneration company should have experience with regenerating Pd/promoter catalysts. In addition, the performance of the regenerated catalysts should be confirmed via laboratory or bench scale activity/selectivity testing.

Tail-end Pd/promoter catalysts are "rejuvenated" by an insitu steam/air bum, followed by H2 reduction. This is a very important operation. A well executed regeneration and reduction returns the catalyst to essentially same SOR performance as was achieved during the previous cycle. Conversely, an incorrectly executed regeneration/reduction can result in placing back in service a catalyst that performs worse than it did when it was taken offiine. Some comments on insitu regeneration is as follows"

545

Many times the regeneration step is "short-changed". This could mean any or all of the following: " " " " Inadequate facilities Inadequate procedures Inadequate training of the operators and process engineers Insufficient operator attention

Typically, it is recommended that the catalyst be subjected to temperatures between 850-900 F (about 450-480 C). The bum is "controlled" by limiting the amount of air (02) in a steam/air mixture. Typical vendor guidelines call for maintaining air between 1-5% (02 between 0.2-1.0%) during the steam/air bum. Regeneration and the subsequent reduction is usually carried out counter current to the process flow. The operators must be provided with explicit guidelines so that they can achieve a good bum at about 850 F (about 450 C), yet not getting into excessively high temperatures, by "building upon" the exotherm as it progresses through the catalyst bed. Problems with the steam/air step can occur if the console operator's attention gets diverted in the midst of a bum. For example: " High exotherms if air is left in unattended ,/ Unable to restart a bum if air is cut out and bed allowed to cool Having more green oil fuel on the bed helps in attaining a good bum. Longer cycle lengths help with this. Promoted catalysts mitigate against this.

C3 Hydrogenation
Typically, this unit is located on the O/H stream from a backend DEC3. (Figure 4) Most units installed prior to the mid-70's were vapor phase reactors. A vapor phase unit is very similar to a tail-end acetylene hydrogenation unit. The catalyst is similar, but the Pd catalyst may not be promoted. Usually, the facilities include a spare reactor with the capability for insitu regeneration. If the inlet MAPD is below about 1%, the removal of MAPD may be carried out in a single reactor. Above 1 % MAPD, the reactor will be two beds with intercooling. A two-bed reactor system may be able to completely hydrogenate inlet MAPD up to about 3.5%, without any recycle The operating/control strategy and safety facilities/procedures are analogous to those for a two-bed tail-end acetylene hydrogenation unit.

546

A vapor phase C3 hydrogenation unit is easier to operate than a tail-end acetylene hydrogenation unit. The reason for this is that the "operating window" is much larger for C3's. That is, it's much easier to selectively hydrogenate MAPD in the presence of propylene than to selectively hydrogenate acetylene in the presence of ethylene. Liquid phase units can handle higher levels of MAPD in the C3 fraction by recycling reactor product to dilute the MAPD in the DeC30/H. This paper cannot go into the details of liquid phase C3 hydrogenation because this technology is licensed technology. Some overview comments that can be made include the following" Liquid phase reactors and associated equipment is smaller, relative to the reactors and associated equipment required for vapor phase units. Liquid phase units require less energy. The dominant liquid phase C3 hydrogenation process is licensed by IFP.( Ref. 33) Typically, this is a two-stage process. (Figure 7) Adiabatic, fixed bed reactors are used. Effluent from the first stage is recycled to reduce the concentration of MAPD in the reactor feed. ABB Lummus offer a proprietary C3 liquid phase selective hydrogenation process and catalyst. This process is designated LB Fining. This process was developed jointly by ABB Lummus and the Beijing Research Institute for the Chemical Industry (BRICI). This is a single stage hydrogenation process. A paper on this process was presented to the 1992 Ethylene Producers Conference (Ref. 31). Several ethylene plants still operate Lurgi-Bayer's isothermal liquid phase C3 hydrogenation process. This process employed only one isothermal tubular reactor. (Ref. 37) CD Tech offers a CDHydro process for selective hydrogenation of C3. This is a catalytic distillation process whereby the catalyst bed is located within a backend DEC3. A paper on this concept was presented at the 1995 Ethylene Producers Conference. (Ref. 32) The catalysts used in liquid phase hydrogenation contain about 10 times more Pd than is the case for vapor phase C3 hydrogenation. Two catalysts offered by Axens are designated LD-265 (Pd-only) and LD-273 (Pd plus a promoter). Other companies that offer C3 liquid phase hydrogenation catalysts include" Sud Chemie, Englehard, Kataleuna, and BASF.

547

Safety Consideratons
The following is not intended to provide precise guidance on the safety facilities and procedures required for acetylene and C3 hydrogenation units. The following provides an overview on safety considerations. This may serve as background information for some of succeeding papers in this session. For more information, see reference articles 27 through 30. The primary safety concern with acetylene and C3 Hydrogenation is to prevent situations that get the reactor(s) into a runaway exothermic reaction, leading to excessive temperatures and pressures such that there is a loss of containment. It's possible to reach conditions where uncontrollable reactions can begin to occur. The reactions include, hydrocracking, polymerization, and ethylene decomposition.

Front-end Acetylene Hydrogenation Units: A major consideration is that the feed to these reactors is and ethylene (30-50%); hence, the safety facilities preventing and/or mitigating against the consequences hydrogenation in the presence of high concentrations of safety considerations is as follows" very rich in hydrogen (20-50%) and procedures are aimed at from the initiation of ethylene hydrogen. An overview on the

Maintain the feed above a minimum rate: Long contact times will initiate ethylene hydrogenation. This is usually handled by having the reactors in a recycle loop. The catalysts suppliers and ethylene contractors provide guidelines for the minimumallowable feed rate. Install multiple thermocouples in the catalyst beds. Have these T/C's, along with bed outlet T/C's, alarm the operators when temperatures have exceeded the upper limit for normal operating conditions. This alerts the operator to take corrective action, such as reducing inlet temperature to the reactor. As the temperatures increase, the heating medium going to the reactor preheater should be tripped. Safety facilities and procedures should be able to handle the need to isolate and depressure, and purge the reactor with an inert gas. The ethylene producer has responsibility for safety facilities and procedures. Input should come from catalyst suppliers and ethylene contractors.

548

Tail-end Acetylene Hydrogenation Units: For tail-end hydrogenation units, the primary consideration is to avoid high ratios of hydrogen to reactants. The normal control scheme addresses this by maintaining a specified H2/feed ratio for each bed. However, there are a few abnormal situations that could lead to excess hydrogen getting into one or more of the catalyst beds. These abnormal situations would include: instrument failure, operator error, and flawed procedures for dealing unsteady state operations. Safety facilities and procedures may include the following" Multiple thermocouples within each catalyst bed. These T/C's, along with bed outlet T/C's would provide high temperature alarms. The lowest level alarm will alert the operator of temperatures that are above the normal operating range. This gives the operator an opportunity to take corrective action. If the temperatures continue to climb, then the hydrogen valves will be tripped. The main consideration is to prevent any additional hydrogen from entering the reactor when there are high temperatures at any point(s) within the catalyst beds. The reactor feed inlet temperature should be reduced when high catalyst bed and outlet temperatures are detected. If temperatures continue to increase, the reactor will have to be isolated, depressured, and purged/cooled with an inert gas.

C3 Hydrogenation The safety facilities and procedures for a vapor phase C3 hydrogenation unit are analogous to those for a backend acetylene converter. For a liquid phase C3 hydrogenation unit, the safety facilities and procedures have to take into account auto-refrigeration concerns from depressuring equipment containing liquid propylene. The liquid phase system is inherently safer than the vapor phase system due to the heat sink of the liquid, which can absorb considerable amounts of heat by vaporization. However, the liquid phase operation, introduces another concern over localized overheating or "hot spots" caused by flow maldistribution. This concem has been mentioned in two EPC papers: (Ref. 29 and 30)

549

List of References

Promoted Acetylene Hydrogenation Catalysts 1. Dow Chemical Company, US 2.802.889, (1957) 2. M.M. Johnson, D. W. Walker, G. P. Nowack, US patent 4,404,124, (1983)
o

B. Didillion, P. Sarrazin, J. P. Boitiaux; "Advanced Catalyst Design for High Ethylene Production and Maximized Cycle Lengths", AIChE Spring Meeting, Session 43, March 1995, Houston, TX K. Sasaki, J. Kimble, M. Scharre, R. Callejas, "Effects of Selectivity on Acetylene Removal Unit", AIChE(EPC), Session 33, March 1997, Houston TX E. Sughrue and M. Johnson, "Highly Selective Catalysts for Acetylene Hydrogenation", AIChE(EPC), Session 18, April 1993, Houston, TX R.W. Voight, J.S. Meriam, S.A. Blankenship, "Evolution of Catalyst for Selective Hydrogenation of Acetylene", AIChE(EPC), Session 19, February 1996, New Orleans LA M. P. Kaminsky, V.J. Kwasniewski, and P. Lysenko, "Acetylene Hydrogenation Catalysis using Pd/A1203, Finding the Optimum Pd Dispersion/Loading and Space Velocity", AIChE(EPC), Session 68, March 1999, Houston TX J-P Boitiaux, J. Cosyns, M. Derrien, and G. Leger, Hydrocarbon Processing, March 1985 F.E. Hancock, and J.K. Smith, "Design and Performance of Acetylene Hydrogenation Catalysts" AIChE(EPC), Session 86, April, 1992, New Orleans, LA

10. D. B. Tiedtke, T. T. P. Cheung, J. Leger, S. A. Zisman, J. J. Bergmeister, Go Ao Delzer, "Chemicals Influencing the Activity of Palladium-Based Catalyst for the Hydrogenation of Acetylene to Ethylene in Acetylene Converters" AIChE (EPC), Session 80, March 2001, Houston, TX.

550

Green Oil Formation 11. L. Yajun, Z. Jing and M. Xueru, "Study on the Formation of Polymers During the Hydrogenation of Acetylene in Ethylene-Ethane Fractions", Chem. Ind. Eng. Soc. China Am. Inst. Chem. Eng. 82, (2) 688, (1982). 12. Z. E. Gandman, M. E. Aerov, V. A. Men'shchikov and V. S. Getmantsev, "Formation of Polymers in Hydrotreating an Ethane-Ethylene Fraction of Pyrolysis Gas to Remove Acetylene", International Chem. Engineering, (Vol. 15, No. 1), Page 183, January 1975.

Catalyst Poisons and Ethylene Plant Contaminants 13. M.P. Kaminsky, "Deactivation Mechanisms for Pd/A1203 Acetylene Hydrogenation Catalyts", AIChE(EPC), Session 26, March 1994, Atlanta GA 14. S. Bailey, F.E. Hancock, J.S.,Booth "The Application of EPR to the Study of Acetylene Removal and Deactivation Reactions", AIChE(EPC), Session 65, February 1996, New Orleans LA 15. F.E. Hancock and J.S. Booth, "A Mechanistic Study of the Deactivation by Hydrocarbon Fouling of Front End Acetylene Hydrogenation Catalysts and the Effect of Carbon Monoxide Incorporation", AIChE(EPC), Session 26, April 1994, Atlanta GA 16. K.J. Enqvist, "Experience on Arsine Contaminant in Ethylene Unit Feed Stock and Disposal of Arsine Contaminated Catalyst/Adsorbent", AIChE(EPC), Session 23, March 1995, Houston, TX 17. D.R. McPhaul and J. A. Reid, "Ethylene Plant Feedstock Contaminants Treatment", AIChE(EPC), Session 20, March 1995, Houston TX 18. D.R. McPhaul and J. A. Reid, "Contaminant Rejection Technology Update", AIChE(EPC), Session 21, February 1996, New Orleans LA 19. Baumgartner and S. Coleman, "Principal Feedstock Contaminants in Ethylene Plants", AIChE(EPC), Session 19, February 1996, New Orleans LA 20. J. Reid, "An Overview of Hydrogenation Catalyst Poisons in Olefin Units" AIChE(EPC), Session 80, March 2001, Houston, TX

551

Control and Modeling of Acetylene Hydrogenation Units 21. E.L. Sughrue, R.L. Hair. R.J. Callejas, "The use of Models in the Operation of Acetylene Hydrogenation" AIChE(EPC), Session 21, April 1995, Houston, TX 22. A.N.R. Bos, and K.R. Westertrep, "The Behaviour of an Adiabatic Packed Bed Reactor: The Selective Hydrogenation of Acetylene/Ethylene", Proceedings of the International Conference of Unsteady State Processes in Catalysis, June 990, Novosibirsk 23. M. W. Brown, A. Penlidis, and G.R. Sullivan, "Control Policies for an Industrial Acetylene Hydrogenation Reactor", Canadian Journal of Chemical Rngineering, Vol. 69, February 1991 24. W.K. Lam, S.C.K. Cua, K.F. McNulty, " Benefit of Acetylene Reactor Modeling",AIChE(EPC) Session 20, March 1990. 25. M. Szukiewicz, K. Kaczmarski, R. Petrus, "Modelling of Fixed Bed Reactor: Two Models of Industrial Reactor for Selective Hydrogenation of Acetylene", 12th International Congress of Chemical and Process Engineering, August, 1996, Prague. 26. M. Nasi, M. Alikoski, and C White, " Advanced Control of Acetylene Hydrogenation Reactors", Hydrocarbon Processing, June 1985.

Safety Considerations in C2 and C3 Hydrogenation Operations 27. M. Edison, "Runaway Reactor Survey", AIChE(EPC), Session 35, March 1997, Houston TX 28. K.J. Enqvist, "The Operating Errors and the Deficiencies in the Design or in the Safety Systems that Lead to the Acetylene Converter Temperature Runaway", AIChE(EPC), Session 65, March 1999, Houston TX 29. J.L. Bancroft, "Temperature Monitoring Systems for Detecting Hot Spots in Exothermic Reactors", AIChE(EPC), Session 65, March 1999, Houston TX 30. J. A Reid, "An Overview of Olefin Unit Accidents and Safety Lessons Learned", AIChE(EPC), Session 65, March 1999, Houston TX

552

C3 Hydrogenation 31. A.E. Krumins, H.L.leung, AND L.X.Xiang, "LB Fining for Liquid Phase MAPD Hydrogenation", AIChE(EPC), Session 86, April 1992, New Orleans, LA 32. G. Gildert, W. Groten, K. Rock, D. Weidert, "Combining Hydrogenation with Distillation in Ethylene Plants", AIChE(EPC), Session 20, March 1995, Houston TX 33. M.L. Derrien, J.W. Andrews, P. Bonnifay, and J. Leonard, "The IFP Selective Hydrogenation Process", Chemical Engineering Progress, Vol. 70, No. 1, January 1974

Overview on Acetylene Hydrogenation 34. V. Kaiser, JP. Laugier, R. DiCinto, M. Picciotti, "Acetylene Removal from Cracked Gas: A Technology Overview", AIChE(EPC), Session 68, March 1999, Houston TX 35. Sud Chemie Technical Bulletin on C2-Tailend Hydrogenation 36. Sud Chemie Operating Instructions Bulletin for G-83C 37. M.L. Derrien, "Selective Hydrogen Applied to the refining of Petrochemical Raw Materials Produced by Steam Cracking", Studies in Surface Science, Vol. 27, Catalytic Hydrogenation, Elsevier.

553

Figure 1

H2/CH4

Demethanizer

Front-End Acetylene Hydrogenation

O1 O1

Ethylene

Ethane/Ethylene Splitter

4:=

""F

Deethanizer

Ethane

HC Feed Cracking Furnaces

"--~ Depropanizer

C3+

Cool/Scrub "-" Heavies

Drawing Courtesy of Sud Chemie

Fi_qure 2 Acetylene Hydroqenation on OIH of Front End DeC3

H2/CH4
.__=

Demethanizer
C2

O1 O1 O1

Front-End Acetylene Hydrogenation

Deethanizer

Depropanizer

C3

--~ C4+ CoollS_.~crub

Heavies

Cracking Furnaces

Drawing Courtesy of Sud Chemie

Figure 3
H21CH4

Demethanizer
C2=

C2 Purification
O1 O1 03

Raw Gas

Acetylene Hydrogenation

Iii i ' i i i
I

Deethanizer

"

C3+

iii!iiiiiiiiiiiiiii!~.~ .....

Cool/Scrub

HC Feed

,---,r

Cracking Furnaces

Heavies

Drawing Courtesy of Sud Chemie

Figure 4 Tail-End Acetylene Hydrogenation Unit

CH4, CO, H2

Demethanizer

C2
T

C2 Selective Hydrogenation
O1 O1 ~J

= $ c2
Deethanizer

C3
C 3 Selective

Hydrogenation Depropanizer HC Feed

"F

c3

Cracking
Furnace

I CoollScru~b
"'-F

Heavies

C4+

Drawing Courtesy of Sud Chemie

Figure 5 2 - B e d Tail-end A c e t y l e n e H y d r o g e n a t i o n R e a c t o r S y s t e m
ii~iiiiiiiiiiiiilili-iiiiiiiiiii~i i:~ i~]iii

ii~iiiiiiiiiiii~iiiii~iiiiiiiiiiiiiiiiiiii

Lead Reactor
01 01

Bed l':i Cataly=!iiiiii ......i~T,l~,~,

i!i!ii Bed !ii!!iiiii!!ii .......

:::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::
. . . . . . .

iiiiiiiiiiiiiii

i,,:~,,,~::~,~!~:==~:~,~,:~,,:~=
ii:i;ii;i;ii~i:iii;i;i~i;?iii;ii;i:i:i~i;i~i;i;i5

. . . . . .

i~!~,~.~.~
:.!:i'.i:.!:!:::::.:.!:.:.!'):?):.:. 'i!)).)::.:. :)).]:.:i?i!:.!:. .................................................................... '..:::.:.

~ ' ~
I

! ~IG~

.............................. ..........................................

.. >........~.~. ....

~ ~ . ~ I ~ ~

........................................

ii ~atalys~ iii

!? ii~ata lY~ iiiiiiiii ~J [iiiiii} I

Guard Reactor

=e. i!i"iliiii!!,i ......

ii

i!iliilii ......! ~i

ii

iii!iiiii.:....... :i i~i
:i:i:.!i!

iiiii !iii.i.i.iii
~
|

i~ii:ii(ii!i:iiiii!!iii:!!i!iiii:ii.i!ii:.]ii::i.~

. . . . . .

......i ~::i:=:!!! i
Steam

Drawing Courtesy of Sud Chemie

Normal Operation

< 1 ppmv C2H2

Regeneration

~ ..... ii!iii!i~ii~Q~ii!i!iii!i!!i!

Figure 6 Green Oil Formation: Sud Chemie Comparison Between Standard and Promoted Tail-end Pd Catalysts (Ref. 35)

Graph No 1

GREEN OIL and POLYMER FORMATION 200 hours Laboratory Tests

Grams
/

24.

0~

~ Standard Catalysts

Promoted Catalysts

Test conditions: 1% C2H2; 1.5 % H2 SV = 5000 Nm3/m3/h T = 50 "C

559

Figure 7 !FP C3 Hydrogenation Unit

to

Hydrogen

steam c r a c k e r

r
......... i-i-7 2\~L;.L .... ill

MAIN
REACTOR

FINISHING REACTOR

01 03 O

,L

I d~u~-~

-.

C3 F e e d s t o c k