PRIORITY CLAIM OF EARLIER NATIONAL APPLICATION

[0001] This application claims priority to U.S. Application No. 61/286,437 filed

December 15, 2009.

BACKGROUND OF THE INVENTION

[0002] This invention relates to a process for removal of mercury from hydrocarbon streams. More particularly, the invention relates to the removal of mercury from hydrocarbon streams that contain a large proportion of unsaturated hydrocarbons.

[0003] Elemental mercury and mercury compounds contaminate both crude oil derived hydrocarbon streams and natural gas. Due to the health effects of mercury and the adverse impact of mercury in causing corrosion of aluminum containing heat exchangers, it is important to remove mercury from hydrocarbon streams. Mercury amalgamates with aluminum, resulting in mechanical failure and leakage. In addition, mercury can poison some catalysts including hydrogenation catalysts in acetylene converters and converters of other unsaturated hydrocarbons. Since the level of mercury that can be tolerated is not established, it is desired to remove the mercury to a level where it can not be detected with the available analytical capability. This is a level below 1 part per trillion.

[0004] Many mercury removal technologies are available and their application is straightforward in the case of non-reactive feeds such as in saturated hydrocarbons and fractions. However, the ethylene industry presents a challenge due to the production of highly reactive side products. Steam cracker furnaces crack saturated hydrocarbons and produce a large proportion of unsaturated hydrocarbons, mostly olefins but also some acetylene and its homologues. These unsaturated hydrocarbons can deactivate the solid adsorbent masses used for mercury removal due to hydrocarbon polymerization and coke formation. In addition, acetylenic hydrocarbons are known to form explosive acetylides in combination with copper and silver, which are commonly used as active ingredients in solid masses for mercury removal. Thus, special considerations must be taken to remove mercury from cracked gas streams or any other streams that contain or may contain acetylenic hydrocarbons.

[0005] While copper compounds such as copper oxides and sulfides have been used to remove trace quantities of mercury from certain hydrocarbons, the treatment of unsaturated hydrocarbons that may contain acetylenes has proven more difficult due to the potential formation of explosive acetylides. A solution for this problem has now been found.

SUMMARY OF THE INVENTION

[0006] The invention provides for the removal of mercury from a gas stream that contains any amount of acetylide precursors including acetylene or acetylenic hydrocarbons. In particular, the invention provides for the total containment of mercury from cracked gas by combining the dehydration and mercury removal. The cracked gas is dehydrated with molecular sieves containing a layer of a regenerative mercury scavenger such as UOP's HgSiv regenerative mercury removal adsorbent. The water free and Hg-free cracked gas hydrocarbons are then separated in the separation section of the ethylene plant. A portion of a demethanizer overhead (DMO) is returned as a regenerant for the dehydration unit. Then the DMO regenerant leaving the dehydration unit is cooled and directed it to one or more non- regenerative mercury removal beds which contain a composite CuS-alumina porous mass capable of removing mercury without forming explosive acetylides, or reducing the cupric sulfide to cuprous sulfide (a form inactive as a mercury scavenger and also more likely to produce explosive acetylides) or excessive aging. Since the level of mercury that is removed is small, a non-regenerative bed can be used for extended periods of time without the adsorbent being replaced.

DETAILED DESCRIPTION OF THE INVENTION [0007] The present invention can be used to treat hydrocarbons containing acetylene or acetylenic hydrocarbons. Typically, there may be 0.1% of such precursors in the

hydrocarbons being treated. However, it is important to consider that any amount of acetylides can be a problem due to the possible accumulation over time. The hydrocarbon streams may be from cracked gas or stream cracking to make ethylene.

[0008] It was confirmed that a material produced by co-forming of porous basic Cu carbonate and alumina powders followed by thermal treatment of the formed composite beads and subsequent treatment with S to produce CuS-alumina composite, does not have a propensity to form copper acetylides upon treatment with acetylene. Therefore, this material is suited as a scavenger for mercury removal from cracked gas regenerant streams. This material is characterized as having large crystals produced by conodulizing. Furthermore, the CuS-alumina composite produced in the manner described above has a propensity to resist reduction of the CuS active phase by the hydrogen present in the regenerant which produces Cu2S. The latter compound is not suited as a mercury scavenger. Lastly, the CuS-alumina composite resists deterioration since it does not promote the reaction of the reactive components of the regenerant with the material causing coking, residue deposition and pore blocking. These adverse factors could render a mercury scavenger ineffective in service.

[0009] The composite material used in the present invention comprises from 5 to 60 wt% CuS, preferably from 10 to 55 wt% CuS, more preferably from 20-45 wt% CuS, and most preferably 40 wt% CuS. In addition, preferably the composite material contains from 10 parts million to 2 wt% chloride ions and more preferably from 50 ppm to 1 wt% chloride ions.

[0010] An analysis of several different gas streams is shown in the following table for both the cracked gas and the regeneration gas. This analysis does not show the level for mercury and some other impurities.

TABLE

Cylinder

Gas

Sample # 1 2 3 4 AVG mol-% ^; 5

; CRACKED ;

\ GAS, mol-% ;

: Hydrogen 17 25 30 19 23 25 25

; Carbon

0.18 0.13 0.12 0.19 0.2 0.2 0.1 1 : Monoxide

1 Methane 28 25 16 27 24 balance | 25 i Acetylene 0.7 0.5 0.4 0.5 0.5 1 0.7 Ethylene 34 32 33 33 33 40 32

1 Ethane 6 8 15 6 9 - 10 i Propylene 1 1 5 3 10 7 - 5

; sum 96.88 95.63 97.52 95.69 96 100.0 97

\ REGEN

! GAS, mol-% !

! Hydrogen 6 6 30 7 12 10 15 i Carbon

0.4 0.3 0.3 0.3 0.3 0.3 - ; Monoxide

! Methane 94 93 69 92 87 balance 1 76 i Acetylene 0.01 0.01 0.01 0.01 \ 0.01 0.01 -

; Ethylene 0.3 0.2 0.6 0.5 0.4 0.4 7.5

; sum 100.71 99.51 99.91 99.81 I 99.985 100.0 98.5 ;

[0011] In order to test the copper sulfide adsorbent used in the present invention and determine whether it would react with acetylenes to form copper acetylides, the following experiment was conducted under severe conditions. As can be seen above, typically there was 0.5% acetylene in the cracked gas and 0.1% in the regeneration gas. In our experiments, the adsorbent was tested under 100% acetylene to simulate extreme conditions. A sample of copper sulfide adsorbent was ground to increase the surface area to be exposed to acetylene.

The sample was then exposed to 100% acetylene at 1 atmosphere pressure and room temperature for 24 hours. The acetylene was then evacuated from the sample and purged with nitrogen gas. Infrared spectroscopy was then used to examine the sample for copper acetylides. If no copper acetylides were found, then the material as considered to be safe for use as an adsorbent. There was no sign of any exothermic process after 24 hour exposure of either whole beads or ground sulfided copper adsorbent. In addition, there was no sign of copper acetylide formation after 1 week of exposure of the sulfided copper adsorbent in dry acetylene at 40°C. In addition, in an accelerated test that was designed to simulate long term exposure of the adsorbent to acetylene, after 1 week there were still no signs of acetylides being formed.