The present invention relates to a reactor system and a method for the treatment of a gas stream, such as coker gas, a gasifier outlet gas or a retort gas. More specifi ^¬ cally the reactor system is either a once-through or a recycle reactor system comprising three fixed-bed reactors, in which a gas with the composition up to 20% H ₂ S

up to 35% total olefins

up to 10% di-olefins

up to 12% CO + C0 ₂ + COS

up to 20% H ₂

Reactor system and method for the treatment of a gas stream up to 2% organic sulfur compounds

and balanced by saturated light hydrocarbons is treated with the purpose to lower the total content of sulfur compounds to a maximum value of 10-20 ppm by weight, preferably below 10 ppm by weight.

In the refining industry, sulfur removal or recovery is a very important issue that often does not get the attention it deserves. Sulfur is one of the dominant contaminants in petroleum fractions, and legislation not only limits the permissible sulfur content of finished products, but also limits refinery emissions to the atmosphere. Therefore, sulfur removal and recovery is a vital process for refiner- ies and gas plant operations. In most locations, the sulfur is hydrotreated and thus converted to hydrogen sulfide, which can be scrubbed from the various liquid or gas streams. The hydrogen sulfide collected from the hy- drotreaters and/or gas plants can subsequently be treated, e.g. by the Claus process.

Various gas treatment processes for sulfur removal are de ^¬ scribed in the prior art. Thus, US 8,080,089 describes a method and an apparatus for efficient gas treatment, where SOx compounds are removed such that the concentration of SOx remaining in the gas is between 0 and 10 ppmv. Further, US 7,374,742 discloses a method for removing sulfur species from a gas stream without the use of a sulfur species re ^¬ moval process, such as an amine scrub. The sulfur species are removed by directly subjecting the gas stream to a sul ^¬ fur recovery process, such as a Claus process at high pres ^¬ sure and moderate temperatures, wherein the sulfur recovery process comprises a catalyst which does not comprise acti ^¬ vated carbon.

The reactor system according to the present invention, also called a gas hydrotreater, comprises a catalyst technology with the capability of substantially lowering the sulfur content of mercaptan-rich refinery gases. By treating a gas, such as a coker gas, in the reactor system according to the invention, it has surprisingly turned out that the content of sulfur-containing compounds can be brought down to a maximum value of 10-20 ppm, preferably below 10 ppm by weight after a final chemisorption step (typically amine-based) to remove residual ¾S.

The reactions taking place in the reactor system according to the invention are all gas phase reactions in an environ ^¬ ment with ¾, CO, CO ₂ and ¾S besides various hydrocarbons. As already mentioned, the reactor system according to the invention is a reactor system comprising three consecutive fixed-bed reactors. The composition of the reactor system is shown in the appended figure.

The first reactor (Rl) is required to avoid gum formation of di-olefins later in the process. In the second reactor (R2), all sulfur-containing compounds and olefins are re- acted, but some traces of mercaptans and COS may still be present. These compound traces are treated in the third re ^¬ actor (R3) .

If the amount of di-olefins in the feed gas is above 1000 ppmw, then the first reactor (Rl) is mandatory. In case the amount is below 1000 ppmw, Rl can be omitted, and the feed gas stream is then led directly to the inlet of R2 via a feed/effluent heat exchanger. The main challenge is to treat mercaptans while simultane ^¬ ously handling di-olefins, H ₂ S and COS contents. Thus, the exotherm in the first reactor (Rl), which normally must handle the conversion of di-olefins to mono-olefins , has to be controlled carefully. Because it is possible to control the processes so that only the di-olefin reactions actually take place, then the system layout can be designed with just a simple low-cost fired heater.

As mentioned, the reactions are all gas phase reactions, and the competing exothermic reactions are mono-olefin to saturation and ¾S + alkanes to mercaptans. There is a de ^¬ sired temperature window of around 130-210°C, within which the reactions can be controlled while having a useful plant layout and catalyst system. Thus, the desired temperature window in the first reactor (Rl) can actually result in an economically very advantageous system layout for the treat ^¬ ment of coker gas. The core concept underlying the present invention is to make a more efficient hydrodesulfurization of gas streams, such as coker gas streams. The first reactor Rl (the "hy- drotreater" or pre-hydrogenator) is situated upstream of an amine wash plant and substantially decreases the content of sulfur in the gas prior to entering the main unit R2. As mentioned earlier, the first reactor (Rl) is required to avoid gum formation of di-olefins later in the process. However, this is only necessary if the content of di- olefins in the feed gas stream is above 1000 ppmw. The main unit R2 is a hydrogenating reactor, and reactor R3 is a COS post-treating hydrolysis reactor.

The main technical novelty of this approach lies in a modi ^¬ fication of the pre-treater catalyst to selectively treat di-olefins rather than mono-olefins in order to provide ap ^¬ propriate temperatures in the main reactor R2 in a cost- effective way.

The coker gas, which can be treated according to the inven- tion, is typically a coker sour gas from a coker sponge ab ^¬ sorber. In the hydrotreater plant, non-H ₂ S sulfur compounds are converted into ¾S. Typically this coker sour gas has a composition as indicated in Table 1 below:

Table 1: Coker sour gas composition Component Average mole percentage

hydrogen 9.0

nitrogen 0.2

methane 47.5

carbon monoxide 0.5

carbon dioxide 0.2

ethylene 1.9

ethane 17.8

hydrogen sulfide 13.8

propylene 2.3

propane 4.8

1-butene 0.3

isobutylene 0.1

trans-2 -butene 0.1

cis-2-butene 0.1

isobutane 0.2

n-butane 0.7

isopentane 0.2

N-pentane 0.2

C ₆ H14 + 0.2

In addition the coker sour gas may contain traces (≤ 0.01 mole percent) of carbonyl sulfide, 1 , 3-butadiene, HCN/RCN, benzene, toluene, xylene and ammonia.

The catalysts present in the three reactors of the system according to the invention are catalysts generally employed in hydrodesulfurization (HDS) processes. Such HDS catalysts after activation generally contain mixed sulfides of Co or Ni and Mo or W supported on high-surface-area carriers such as γ-alumina (AI ₂ O3) . The main reason for their wide appli ^¬ cation lies in their high tolerance to ¾S that is produced during hydrotreating reactions. The industrial application of a Co-Mo sulfide catalyst was already reported 70 years ago, and it is still the most common catalyst for HDS reac ^¬ tions .

In the following the treatment method according to the in ^¬ vention will be described in more detail with reference to the figure. As mentioned above, the hydrotreater plant consists of three reactors: A pre-hydrogenator Rl, a hydrogenating reactor R2 and a COS hydrolysis reactor R3.

A feed gas (f) is mixed with hydrogen (h) obtained from battery limit. Subsequently the resulting process gas is optionally (depending on the amount of di-olefins in the feed gas) pre-heated in a first feed/effluent heat exchang ^¬ er and passed through the pre-hydrogenator Rl . Then it is further pre-heated in a second feed/effluent heat exchanger before entering the hydrogenation reactor R2, if necessary after passing a start-up heater. If the content of di- olefins in the feed gas is sufficiently low, then the feed gas/hydrogen mixture is fed directly to the inlet of the hydrogenation reactor R2 via the second feed/effluent heat exchanger.

From the hydrogenation reactor R2 the process gas is cooled in the second feed/effluent heat exchanger. A bypass on the hot side of the second feed/effluent heat exchanger is used to control the exit temperature from R2.

The cooled process gas from the second feed/effluent heat exchanger is mixed with high pressure steam (hps) and then further cooled in the process gas boiler B before being sent to the COS hydrolysis reactor R3, which is a post- treating reactor used for reacting traces of COS present in the outlet stream from R2. The process gas from R3 is either recycled to the inlet of Rl (indicated as a dotted arrow in the figure) or cooled in the first feed/effluent heat exchanger, heating up the gas to the di-olefin pre-hydrogenator Rl . The gas is further cooled in a water cooler, where the steam is condensing. The process condensate is separated from the hydro-treated gas in a process condensate separator.

From the process condensate separator the hydro-treated gas is sent to an amine treatment plant A for removal of H ₂ S. A stream lean in amine is sent through the amine treatment plant and leaves the plant as a stream rich in amine. The product (p) contains 10-20 ppm by weight, preferably below 10 ppm by weight sulfur. The system also includes means to flush out NH ₄ C1 salt

(formed by reaction between HC1 and NH ₃ ) with water to prevent clogging.

Pre-hydrogenation reactions and catalyst

A coker sour gas feedstock contains above 1000 ppmw of di- olefins. Di-olefins in a coker sour gas feed have a high tendency to gum formation due to polymerization or carbon formation at the normal operating temperature of the hydro- genation reactor R2. In order to prevent these problems, the di-olefins in the coker sour gas feed are converted in the pre-hydrogenation reactor Rl containing a hydrogenation catalyst, for example applicant's nickel-molybdenum hydro ^¬ genation catalyst TK-437. TK-437 catalyses the following reactions:

Ri=R ₂ -R ₃ =R ₄ + H ₂ → Ri=R ₂ -R ₃ H-R ₄ H

Ri=R ₂ -R ₃ = 4 + 2H ₂ → HRi-HR ₂ -R ₃ H-R ₄ H

Ri=R ₂ + H ₂ → HRi-HR ₂ where R is a hydrocarbon radical.

If the operating temperature is sufficiently low, the last reaction is unlikely to occur.

For any given feedstock a certain hydrogen flow is required for the hydrogenation reactions. A sufficient amount of hy ^¬ drogen must always be added in order to minimize the risk of polymerization or carbon formation.

The TK-437 catalyst is pre-sulfided and does not need to be sulfided prior to operation.

Another useful catalyst is applicant's molybdenum-based catalyst TK-719, which is especially suitable for olefin containing feeds, where activity grading is needed to pre ^¬ vent formation of gum.

Hydrogenation reactions and catalyst

The hydrogenation reactor R2 is loaded with a nickel- molybdenum hydrogenation catalyst, preferably applicant's

TK-261 catalyst, placed in a single bed in the reactor. TK-261 catalyses the following reactions:

RSH + H ₂ RH + H ₂ S

R1SSR2 + 3H ₂ RiH + R ₂ H + 2H ₂ S

R1SR2 + 2H ₂ RiH + R ₂ H + H ₂ S

(CH) ₄ S + 4H ₂ C4H10 + H ₂ S

COS + H ₂ CO + H ₂ S

C0 ₂ + H ₂ S COS + H ₂ 0

Ri=R ₂ + H ₂ HRi-R ₂ H

The conversion of olefins to saturated hydrocarbons is a strongly exothermic reaction. The temperature rise will ap ^¬ proximately be between 50 and 90 °C depending on the content of olefins in the feedstock.

An excess of minimum 10% hydrogen must be present at the outlet of the hydrogenation reactor R2 to prevent carbon formation or polymerization of the olefins. If the hydrogen flow is insufficient, this can also result in a poor con ^¬ version of organic sulfur compounds and increase the slip of organic sulfur through the unit.

The gas exiting the hydrogenation reactor may contain up 150 ppm of olefins. This residual content of olefins may recombine with the H ₂ S present in the gas at a concentra ^¬ tion around 14 %. The general reaction schemes are:

C _n H _2n + H ₂ S ~ C _n H _2n+1 SH (n = 1-4)

C _n H _2n + H ₂ S ~ 2C _n H _2n+1 SH (n = 3, 4)

C _n H _2n + H ₂ S ~ (CH ₃ ) ₃ CSH The maximum activity of the hydrogenation catalyst depends on the concentration of hydrogen and the temperature at the inlet to the reactor. The recommended outlet temperature of the reactor is 400°C. At temperatures above 400°C, coke can be formed on the catalyst surface, thereby decreasing the activity of the catalyst. Two locations are of importance here: the piping and heat exchangers that take the gas from the exit of the hydro ^¬ genation reactor (R2) to the inlet of the COS hydrolysis reactor (R3) , and the CKA catalyst in R3, which provides a large contact surface area that might enhance the recombi- nation reaction.

The gas leaving the hydrogenation reactor (R2) will be at or very close to equilibrium at 400°C with respect to ole ^¬ fin hydrogenation, organic sulfur hydrogenation, COS hydrolysis and water-gas shift. Equilibrium of the first two would imply that the recombination reactions are in equi ^¬ librium, too, the last two that the COS hydrogenolysis is in equilibrium.

During start-up, the catalyst is heated with once-through natural gas and hydrogen. The catalyst must not be operated above 300 °C with hydrocarbons without hydrogen, because a carbon laydown may otherwise take place and thereby block the catalyst surface. As a result of this, the hydrogena ^¬ tion will be insufficient. The TK-261 catalyst is available as a pre-sulfided or as an oxidized product. A pre-sulfided catalyst does not need to be sulfided before being taken into operation. An oxidized catalyst must be sulfided in situ to obtain its activity.

Operation on a non-sulfided catalyst will increase the risk of hydrocracking, resulting in severe temperature fluctua ^¬ tions. Olefins have a marked tendency to effect carbon for ^¬ mation on the non-sulfided catalyst. The affinity for car ^¬ bon formation is higher at low hydrogen partial pressures and high temperatures. The affinity for carbon formation also depends on the type of olefins.

After sulfiding the catalyst is pyrophoric and thus it should not be exposed to air at temperatures above 70 °C.

COS hydrolysis reaction and catalyst

The COS hydrolysis reactor R3 is loaded with an activated alumina catalyst, preferably applicant's CKA-3 catalyst, placed in a single bed in the reactor. The CKA catalyst is selectively active for the COS hydrolysis reaction:

COS + H ₂ 0 «■ C0 ₂ + H ₂ S The CKA catalyst does not require any activation in connec ^¬ tion with start-up. It is heated in natural gas to a tem ^¬ perature at least 50°C above the dew point of the process gas . During operation the gas must stay around 50°C or more above the dew point to prevent condensation in the pores of the catalyst. Such condensation may damage the catalyst. The COS slip-out from the COS hydrolysis reactor is deter ^¬ mined by equilibrium, and a low COS leakage is favoured by a high steam content and a low temperature. High pressure steam is added to the process gas stream to the COS hydrolysis reactor in order to reduce the slip of COS out of the reactor. Loss of steam will result in a breakthrough of COS through the reactor. A high CO ₂ content in the feed gas will also result in a higher COS slip due to a shift in the equilibrium reaction.

The invention is further illustrated by the examples which follow. The invention is however not in any way limited to these examples.

Example 1

Treatment of a sour semi-coker gas

A sour semi-coker gas from a retort gas plant has the over- all composition as indicated in the following Table 2:

Table 2: Sour semi-coker gas composition

propene 6.8

propadiene trace

∑ C4H10 1.3

∑ C4Hg 4.1

∑ C4H ₆ 1.1

∑ C5H12 0.5

∑ C5H10 1.4

∑ C ₅ Hg 0.6

∑ C ₆ H14 0.1

∑ CeHi2 trace

∑ ΟεΗιο trace

C6+ trace

The above gas is mixed with hydrogen. Subsequently the re ^¬ sulting process gas is pre-heated to 175°C in a first feed/effluent heat exchanger. The pre-heated gas is passed through the pre-hydrogenator and then it is further preheated in a second feed/effluent heat exchanger before en ^¬ tering the hydrogenation reactor at around 290-400°C. From the hydrogenation reactor the process gas is cooled in the second feed/effluent heat exchanger. A bypass on the hot side of the second feed/effluent heat exchanger is used to control the exit temperature from the hydrogenation re ^¬ actor to 400°C.

The process gas from the hydrogenation reactor is mixed with high pressure steam and then further cooled in a process gas boiler before being sent to the COS hydrolysis re ^¬ actor . The process gas from the COS hydrolysis reactor is cooled in the first feed/effluent heat exchanger, heating up the gas to the di-olefin pre-hydrogenator . The gas is further cooled in a water cooler, where the steam is condensing. The process condensate is separated from the hydro-treated gas in the process condensate separator.

From the process condensate separator the hydro-treated gas is sent to an amine treatment plant for removal of ¾S. A stream lean in amine is sent through the amine treatment plant and leaves the plant as a stream rich in amine. The product of the treatment contains less than 10 ppm by weight sulfur.

Example 2

Treatment of a coker sour gas A coker gas with a composition as indicated in Table 1 is subjected to the same treatment as in Example 1. Also in this case the final product of the treatment contains less than 10 ppm by weight sulfur.

Example 3

Gas analysis following plant start-up This example shows the results of analysis of gas samples taken after start-up of the plant. The samples were taken at the following sites:

Sample 1 (SI) : the gas inlet (f) Sample 2 (S2) : the outlet of Rl

Sample 3 (S3) : the outlet of R2

The last sample, Sample 4 (S4) was taken from the product stream.

The results, given as mole ppm compound, are indicated in Table 3 below. Results below 0.4 mole ppm compound (for sulfur dioxide, dimethyl sulfide, thiophene, isobutyl mer- captan, dimethyl disulfide, 2-ethyl thiophene, 2,5-dimethyl thiophene, ethyl methyl disulfide, tetrahydrothiophene, 2- methylthiophene, 3-methylthiophene, 2-methyltetrahydro- thiophene and diethyl disulfide) are not indicated in the table .

For gases, mole ppm equals ppmv.

Table 3: Gas analysis after plant start