US3484496A - Desulphurisation and hydrogenation of aromatic hydrocarbons - Google Patents

Description

Dec. 16, 1969 J CARRUTHERS ET AL DESULPHURISATION AND HYDROGENATION OF AROMATIC HYDROCARBONS Filed 001;. 5, was

2 Sheets-Sheet 1 Fl G1 INVENTORS JOHN wmsoR JO'HN CARRUTHERS BY MORGAN, FINNEGAN, DURHAM a PINE ATTORNEYS Dec. 16, 1969 1Y R ET AL 3,484,496

DESULPHURISATION AND HYDROGENATION 0F AROMA'XIC HYDROGARBONS Y 2 Sheets-sheaf. 3

Filed on, s,- 1966 ma J51 V INVENTORS JOHN wmson ATTORNEYS United States Patent US. Cl. 260-667 14 Claims ABSTRACT OF THE DISCLOSURE Catalytic desulphurisation of an aromatic hydrocarboncontaining fraction containing up to 50 p.p.m. wt. sulphur including thiophene sulphur, is effected in the presence of hydrogen without appreciable hydrogenation of the aromatic hydrocarbons in the fraction or production of a substantial amounto f hydrogen sulphide. The fraction, together with hydrogen, is contacted at an elevated temperature and pressure with a supported nickel catalyst, the flow of hydrogen being controlled to maintain the equilibrium hydrogen partial pressure greater than the minimum necessary to prevent catalyst deactivation as sulphur is adsorbed by the catalyst but less than that at which up to mol percent hydrogenation of the aromatic hydrocarbons occurs. The desulphurised fraction still under pressure and containing up to 2 p.p.m. wt. sulphur, is hydrogenated over a supported nickel catalyst in two stages, the bulk of the hydrogenation, 90 to 99% wt., occurring in the first stage with cooling so as to control the stage temperature within desired limits, and the hydrogenation completed in the second stage, without coolmg.

It is also known that sulphur combines with nickel at moderate temperatures and pressures and hence a process in which the aromatics are desulphurised over nickel is potentially feasible. However it has been found that for such a process to be effective for long periods it has to be operated in the presence of hydrogen. This is believed to be due to the fact that the sulphur is present as organic sulphur compounds and that as the sulphur is adsorbed, unsaturated organic radicals are produced which tend to polymerise on the nickel surface and reduce its catalytic activity. If hydrogen is available these unsaturated radicals are hydrogenated to harmless saturated hydrocarbons. The need to have hydrogen present then introduces the risk of hydrogenation of the aromatic hydrocarbons and the practical value of a desulphurisation process over nickel thus turns on whether it is possible to find conditions in which desulphurisation takes place Without appreciable hydrogenation of the aromatic hydrocarbons.

The present invention is concerned with control of the hydrogen partial pressure during at least the initial stage when the nickel has high hydrogenation activity.

According to the present invention therefore a process for the desulphurisation of an aromatic hydrocarbon-conlit) 3,484,496 Patented Dec. 16, 1969 taining fraction containing from 1 to 50 p.p.m. wt. sulphur without appreciable hydrogenation of the aromatic hydrocarbons comprises passing the fraction in the liquid or vapour phase, and in the persence of hydrogen, over supported nickel at an elevated temperature and pressure such that sulphur combines with the nickel but no substantial amount of hydrogen sulphide is produced, the equilibrium hydrogen partial pressure being greater than the minimum necessary to prevent catalyst deactivation but less than that at which up to 10 mol percent hydrogenation of the aromatic hydrocarbons occurs.

In a further aspect the invention consists in a process in which an aromatic hydrocarbon-containing fraction containing up to 50 p.p.m. wt. sulphur is desulphurised by the above-mentioned process, to produce a fraction containing up to 2 p.p.m. wt. sulphur, and this fraction is hydrogenated in two stages both using: supported nickel catalysts, in which not less than wt. and not more than 99% wt. of the fraction is hydrogenated in the first hydrogenation stage, with the hydrogenation reaction being substantially completed in the second hydrogenation stage, the temperature of the first hydrogenation stage being controlled by cooling and the second hydrogenation stage being uncooled.

The aromatic hydrocarbon-containing fraction need not consist wholly of aromatics, but if it does not the preferred other components are saturated hydrocarbons. Preferably, however, the feedstock contains at least wt. of aromatic hydrocarbons. The preferred aromatic hydrocarbon is benzene and subsequent discussion, both of desulphurisation and hydrogenation will be based on the treatment of this substance. The production of cyclohexane by the hydrogenation of benzene is an important industrial process, since cyclohexane is used in the manufacture of nylon, as a solvent and as a reactant in chemical syntheses. Fibre grade cyclohexane, for nylon production, must have a purity of not less than 99.5% and a total aromatic content of less than 500 p.p.m. wt. The present desulphurisation process is capable of dealing with feedstocks containing from 1 to 50 p.p.m. wt., and preferably not more than 10 p.p.m. wt. sulphur in any form, including thiophenic sulphur. Feedstocks containing higher amounts of sulphur may be subjected to any of the known catalytic hydrodesulphurisation processes, using catalysts of poor hydrogenation activity, for example, cobalt and molybdenum oxides on alumina, to reduce the sulphur content to the desired level, or more than one stage of desulphurisation according to the present process may be used and the present desulphurisation process includes such hydrodesulphurisation steps. If a preliminary hydrodesulphurisation is carried out hydrogen sulphide produced must be removed before the feedstock is contacted with the supported nickel material in the subsequent desulphurisation process. The hydrogenation stages of the invention may use an aromatic feedstock containing up to 2 p.p.m. wt. sulphur, although the desulphurisation process is capable of producing a product containing less than 1 p.p.m. wt sulphur.

References to sulphur contents in this specification are to both combined and uncombined sulphur, but are expressed as the element.

Nickel is susceptible to de-activation. by sulphur-containing materials although it has a number of advantages over other substances used for the hydrogenation of aromatic hydrocarbons. The supported nickel catalysts used in the present processes may incorporate any of the known natural or synthetic support materials, such as the refractory oxides of elements of Groups II to V of the Periodic Table, or kieselguhr, pumice, or sepiolite. Sepiolite is the preferred material and the preferred catalyst for both the desulphurisation and hydrogenation processes of the invention is nickel on sepiolite prepared 3 and activated according to the disclosures of British Patent No. 899,652. It is not essential, however, that the same catalyst should be used in the desulphurisation process as in any subsequent hydrogenation process or in both stages of the hydrogenation process.

Nickel on sepiolite prepared and activated according to the above-mentioned British patent may contain from 1 to 50% wt. nickel (expressed as elemental nickel) and more particularly from 5 to 25% wt. Such a catalyst has a high nickel surface area and has high activity and selectivity. It is capable of maintaining its hydrogenation activity in the hydrogenation stages of the invention up to a sulphurznickel atomic ratio of 01:1 and its total sulphur capacity is much higher than this. We have found that sulphur absorption takes place at least up to 0.75:1 sulphurznickel atomic ratio. Since the sulphur capacit of the supported nickel material is high and is known, it is possible to provide a sufficient amount to give an economic catalyst life. It has been found that a life in excess of one year can be obtained with nickel on sepiolite using a feedstock containing 1.3 p.p.m. wt. thiophenic sulphur.

In addition to the requirement in the desulphurisation process that a small amount of hydrogen must be present to prevent deactivation of the catalyst surface, the extent of hydrogenation must be controlled, at least when the catalyst surface is fresh, to prevent an excessive temperature rise occurring, since the hydrogenation of benzene over nickel is an exothermic reaction. The technique now to be described ensures that hydrogenation in the desulphurisation process occurs only to an acceptable extent, and hence that only an acceptable amount of heat is evolved.

As stated earlier it has been found that the presence of hydrogen is necessary if a desulphurisation process using a supported nickel catalyst is to be operated for a reasonable period. The amount of hydrogen required, however, is small, and as will be shown later, is amply provided for by the choice of a suitable hydrogen partial pressure.

Equilibrium constants can be calculated for the reaction S H +3H =C H from the known Free Energies of Formation of benzene and of cyclohexane at various temperatures. On the assumption of various levels of conversion of benzene to cyclohexane it is possible to calculate the hydrogen partial pressure at equilibrium and to construct a graph of hydrogen partial pressure against temperature. This is shown as FIGURE 2 of the drawings. Each curve is for a particular ratio of cyclohexane t benzene in mol percent.

It will be seen from FIGURE 2 that there is a significant hydrogen partial pressure at equilibrium at temperatures above about 400 F. (204 C.) even at low levels of conversion. At a hydrogen partial pressure of only 0.5 p.s.i. it has been calculated that there is at least a hundred times as much hydrogen as the minimum required to prevent de-activation of the catalyst surface by polymerisation. Although not shown in FIGURE 2, a hydrogen partial pressure sufficient to prevent de-activation of the nickel surface may be achieved at temperatures as low as 122 F. (50 C.).

A conversion of mol percent may be regarded as a satisfactory compromise between the need to avoid excessive hydrogenation and the requirement to provide an adequate hydrogen partial pressure, and reference to FIG- URE 2 will show that such a level of conversion will provide a hydrogen partial pressure of 0.5 p.s.i. at 400 F. A conversion of mol percent will be seen to provide a hydrogen partial pressure slightly greater than this at 400 F.

It has been found that pressure control of the hydrogen introduced into the desulphurisation reactor may lead, in the case of a fresh catalyst, to a temperature runaway, slnce in this case by feeding in hydrogen to a predetermlned pressure an excessive amount may be introduced. In the present invention flow control is. therefore used,

The desulphurisation reaction may be conducted in liquid or vapour phase or in mixed (gas/liquid) phase. Vapour phase operation is preferred. If in the liquid phase, upward flow in the reactor is preferably employed. A fixed or fluidised catalyst bed may be used. In the case of a fluidised catalyst high liquid velocities may cause catalyst to be carried over with the product, in which case a settling tank would be necessary to recover the product.

The desulphurisation reaction should preferably be in the vapour phase to allow the hydrogen partial pressure to be at the desired low level. The maximum plant pressure is restricted to the vapour pressure of the benzene at the desired temperature of operation. At 400 F. the plant pressure is thus restricted to above 200 p.s.i.g., and at 450 F. to about 300 p.s.i.g. Operation at fairly high temperatures means that the sulphur capacity and desulphurisation activity of the catalyst are increased and also the equilibrium hydrogen partial pressure is increased.

In practice an upper limit of operating temperature is set by the onset of side-reactions, such as cracking, isomerisation, and ring opening. Of these cracking is the most important. When the catalyst is partially deactivated the operating temperature can be raised without byproduct formation taking place.

The space velocity should be as high as possible consistent with the required level of desulphurisation. Having regard to the above discussion, the extent of hydro genation of the aromatics is limited to not more than 10% vol. and suitable process conditions to achieve this may be chosen from the following ranges:

Temperature, 50 to 290 C. (preferably to 250 C.)

Pressure, 0 to 2000 p.s.i.g. (preferably to 500 p.s.i.g.)

Space velocity, 0.05 to 10 v./v./hr. (preferably 0.2 to 5.0

v./v./hr.

Inlet hydrogen: hydrocarbon ratio on total feed, 0.01:1

to 0.5 :1 molar (preferably 0.05 to 02:1 molar) Temperature rises occurring will be within the range given. Thus the reactor exit temperature must not exceed the upper limit of the range and the inlet temperature must be above the lower limit. Also the upper limit will apply to the increased temperature usable wtihout significant by-product formation when the nickel is sulphided, compared to that usable when it is fresh.

It may be possible to adjust the reaction conditions for the desulphurisation of benzene so that, for example, an equilibrium hydrogen partial pressure of about 10 p.s.i.a. is obtained. However, a hydrogen partial pressure of 0.01 to 10 p.s.i.a. may be used until the sulphurznickel atomic ratio exceeds 0.01:1.

If the product from the reaction has to be separated immediately into normally-liquid product and hydrogen in a separator operating at at least the reaction pressure (eg a conventional high pressure separator), the technique described in our co-pending U.K. patent application No. 41,939/65 entitled Improvements Relating to the Operation of Reactor Systems is particularly useful. The problem is that the amount of hydrogen which dissolves in the benzene in the separator may not be very different from the total amount of hydrogen present and that under these circumstances the build-up of the required reaction pressure is slow and has a finite limit. In the co-pending application a technique of feeding additional hydrogen or other gas to the separator is disclosed, thereby obviating the difficulty.

However if it is desired to make the desulphurisation process here described part of a comprehensive process, 1.e. a process in which the desulphurised material is further processed under pressure and in the presence of hydrogen, this may be done by direct connection without the necessity for a high pressure separator.

Material containing up to 2 p.p.m. wt. sulphur, derived from the desulphurisation process just described, may be hydrogenated over supported nickel in two stages. The

first of these will be designated the main reactor stage and the second the finishing reactor stage.

The material entering the main reactor is preferably in mixed (gas/liquid) phase, and is preferably in the vapour phase on leaving the reactor. It may be in mixed phase or in vapour phase on leaving the reactor, depend ing on the extent of hydrogenation taking place in the reactor and the extent of cooling. The inlet material may possibly be in the vapour phase, but in this case as the outlet temperature is fixed a large amount of recycle cooling would be necessary. Since the bulk of hydrogenation occurs in the first hydrogenation stage i.e. the main reactor, the major part of the heat produced by the hydrogenation reaction is produced in this stage and cooling is therefore necessary to control the stage temperature to within the desired limits. This cooling may be achieved either by liquid recycle or by the use of a cooled tubular reactor. Liquid may be conveniently recycled from the main reactor outlet or from the finishing reactor outlet. The use of liquid recycle means that the linear velocity through the reactor is increased, with consequent enlargement of the reactor to achieve the same contact time. This may be avoided by using a cooled tubular reactor, with the catalyst in the tubes and a cooling agent being passed over them. In this way the temperature rise in the tubes is limited to the required range. In this type of reactor a higher average catalyst bed temperature can be attained for a given level of hydrogenation than is possible with an adiabatic reactor. Suitable cooling agents for the cooled tubular reactor are steam, water under pressure, gas, or indeed any substance which is thermally stable within the temperature range of the process. In such a system the limiting factor is the rate at which heat can be removed to keep the catalyst at a temperature within the acceptable range. If a cooled tubular reactor is used as the main hydrogenation reactor this stage is preferably in the vapour phase throughout, since otherwise distribution difficulties may occur.

The most convenient recycle cooling medium is cyclohexane, and desirably this is, as far as is possible, in the liquid phase at the main reactor inlet, since the heat of vaporisation will assist the cooling effect, and the minimum of cyclohexane to achieve the necessary cooling may be recycled. This means that the total feed to the hydrogenation process, i.e. cyclohexane and fresh desulphurised benzene, enters the main reactor at as low a temperature as possible, provided that the temperature is high enough for the catalyst to be sufiiciently active to give the required degree of hydrogenation. In order to achieve complete conversion to cyclohexane in one stage it would be necessary to restrict the exit temperature to below 300 (1., otherwise the hydrogenation reaction would not go to completion. In fact the, reaction temperature would have to be less than about 200 C., so that the hydrogen partial pressure would be at a suitable level in relation to the total pressure. Moreover, to obtain complete conversion and at the same time to maintain the temperature rise across the reactor at a suitable value would require recycle at a level which would unduly depress the hydrogen partial pressure. The alternatives of using a higher total pressure or a very large excess of hydrogen would be impracticable in a commercial process.

To obviate these difficulties the hydrogenation stages of the invention use an uncooled finishing reactor. From 90 to 99% wt., and preferably about 95% wt. of cyclohexane is produced in the main hydrogenation reactor with the remainder of the conversion taking place in the finishing reactor. At these lower levels of conversion in the main reactor a higher main reactor exit temperature may be employed than if 100% conversion were .attempted in the main reactor, While maintaining a satisfactory hydrogen partial pressure.

A higher hydrogen partial pressure is more easily maintained in the finishing reactor than in the main reactor,

and completion of hydrogenation achieved without an unacceptable temperature rise, even though cooling is not employed. The finishing reactor may operate in mixed (gas/liquid) phase or vapour phase, and its outlet temperature may be, and preferably is, lower than the main reactor outlet temperature, since this is advantageous for high levels of conversion.

It may be desirable to periodically reverse the flow of reactants in the main reactor. This is because any sulphur not removed in the desulphurisation stage would tend to deactivate the catalyst at the inlet side of the reactor, where the temperature, and thus the sulphur capacity of the catalyst, is lower than at the outlet. By flow reversal catalyst of higher sulphur capacity would be exposed to the material from the desulphurisation stage with consequent extension of the main reactor catalyst life.

The hydrogen used in the process of the invention may be commercially pure or it may be a mixed gas derived from a refinery process, such as a steam reformer tail gas, also containing methane, or catalytic reformer off-gas. Gases containing two or more carbon atoms per molecule may be used, provided that reaction conditions, and in particular the temperature, are carefully controlled to avoid, in the presence of fresh nickel, a temperature runaway. Preferably the gas contains at least 50 mole percent hydrogen, and more suitably 70 to 99 mole percent hydrogen. An advantage of the process is that hydrogen produced by steam reforming of natural gas or naphtha may be used without make-up gas compression.

In accordance with conventional practice gas may be recycled to the main reactor, and if a mixed gas is used, for example one containing methane, gas may be purged from the recycle gas stream, or not, as desired, or methane may be removed in the liquid leaving the high pressure Temperature, 77 to 662 F. (25 to 350 C.) (preferably 122 to 572 F. (50 to 300 C.))

Pressure, 25 to 2000 p.s.i.g. (preferably 50 to 500 p.s.i.g.)

Space velocity (fresh feed), 0.25 to 10.0 v./v./hr. (preferably 0.5 to 5.0 v./v./hr.)

Product recycle ratio, 2:1 to 10:1 (preferably 2.5:1 to

Hydrogen recycle rate on total feed, 50 to 5000 s.c.f./b.

(preferably 200 to 1000 s.c.f./b.)

Finishing reactor Temperature, 77 to 662 F. (25 to 350 C.) (preferably 122 to 572 F. (50 to 300 C.))

Pressure, 25 to 2000 p.s.i.g. (preferably 50 to 500 p.s.i.g.)

Space velocity,.0.25 to 10.0 v./v./hr. (preferably 0.5 to

5.0 v./v./hr.)

Hydrogen recycle rate, to 5000 s.c..f./b. (preferably 500 to 2000 s.c.f./b.

The temperature rises occurring in each stage will be within the ranges given, the reactor exit temperatures not exceeding the upper limits of the ranges: set out, and the inlet temperatures being above the lower limits.

References made in this specification to main hydrogenation reactors, and finishing reactors, or these terms suffixed by the word stage include the use of one or more reactors in any stage, or the use of one or more reactors containing more than one stage. It is only required that the catalysts of each stage should be physically separate and that independent control of the process parameters should be possible in each stage.

One embodiment of a combination process in which benzene is desulphurised by the process previously described and the desulphurised benzene is then hydrogenated as described is illustrated in FIGURE 1 of the accompanying drawings as a simplified flowsheet. It should be emphasized that this flowsheet combines desulphurisation and hydrogenation processes although the former is separately viable.

In this flowsheet benzene enters the system via line 1 and pump 2, and hydrogen or a mixed gas containing hydorgcn enters via line 3 and valve 4. The mixture passes into a desulphurisation reactor 6 and thence into the main reactor 7, make-up hydrogen being supplied via line 3 and another valve 4. From main reactor 7, cyclohexane, residual unconverted benzene and excess hydrogen pass into a high pressure separator 8, from which the cyclohexane-benzene mixture is recycled via line and a valve 4 to inlet line 1, if desired, tocontrol the hydrogen partial pressure in the desulphurisation reactor. Gas is recycled via line 9 and cyclohexane-benzene recycled via pump 11 and line 10. A minor proportion of the cyclohexane-benzene passes via line 12 and a valve 4 to the finishing reactor 13 and thence to a high pressure separator 14, from which cyclohexane product passes via line .15 to storage, gas from separator 14 being returned to the main reactor. If a mixed gas is used the inerts may be vented at a point on the line between separator 8 and line 9, since the gas leaving separator 8 will have a higher inerts content than that from separator 14. As shown make-up hydrogen may be supplied via line 3 directly to the finishing reactor.

In the embodiment just described typical process conditions might be a common pressure in each reactor of 200 p.s.i.g., respective temperatures in the desulphurisation reactor, main reactor and finishing reactor of 400- 500 R, (204-260" C.), 200-480 F. (93250 C.) and 320-430 F. (160-220 C.) and a space velocity in each reactor of 1 v./v./hr., nickel sepiolite being the catalyst in both the desulphurisation rand hydrogenation processes. The gas used could be steam reformer hydrogen, containing 95% hydrogen and supplied at 250 p.s.i.g., and 95 cyclohexane could be recycled at a recycle ratio to fresh feed of 3.5 :1 to 4.0:1 by weight to control the temperature of the main reactor. By achieving 95% conversion in the main hydrogenation stage, and completing conversion in the finishing reactor a cyclohexane product containing only 1 p.p.m. wt. benzene could be obtained without significant de-acti-vation of the nickel sepiolite.

In the following examples illustrating the invention Example I refers to the desulphurisation process of the invention and Examples 2 and 3 refer to the subsequent hydrogenation stages.

EXAMPLE 1 The sulphur content of benzene was reduced from 1.3 to 0.2 p.p.m. weight over nickel on sepiolite catalyst under the following conditions:

Pressure p.s.i.g 200 Temperature F. (maximum) 450 Space velocity v./v./hr 1.0 Inlet gas Hydrogen Inlet H zhydrocarbon ratio molar 0.1:1 Secondary gas to high pressure separator Hydrogen Secondary gas rate as H :hydrocarbon ratio molar 0.05:1

During this operation hydrogenation of benzene to cyclohexane was limited to about 3.5% wt. This hydrogenation reaction effected a slight temperature rise of about 30 F. over a short length of catalyst bed. This will be referred to as the hot spot. As the operation progressed the sulphur taken up poisoned the catalyst at the top of the bed and made it inactive for benzene hydrogenation. Consequently, the position of the hydro genation reaction gradually moved down the bed. This was followed by observing the position of the hot spot in the reactor.

The run was continued under these conditions until 1007 hours on stream, at which time the temperature was lowered to 430 F. At this temperature the sulphur content of the product was 0.3 p.p.m. wt. The run was con tinued under these conditions until 4400 hours on stream, by which time the hot spot had travelled 39 percent down the catalyst bed. If this continued at the same rate the catalyst life would be 11,300 hours or 67 weeks.

EXAMPLE 2 Main hydrogenation recycle reactor Desulphurised benzene containing 0.2 to 0.3 p.p.m. weight sulphur derived from the desulphurisation process of the invention was passed to the main hydrogenation reactor where it was converted to to 97 percent weight cyclohexane.

The following initial opertaing condition were used:

Pressure p.s.i.g 200 Inlet temperature F 180 Outlet temperature F 403 Fresh feed space velocity v./v./h 1.0 Product recycle space velocity v./v./h 3.93 Total feed space velocity v./v./h 4.93 Make-up gas Hydrogen Gas recycle rate on fresh feed s.c.f./b 5000 Gas recycle rate on total feed s.c.f./b 1000 H partial pressure at reactor outlet p. s.i.a 92

The sulphur content of the cyclohexane was reduced to an estimated 0.1 p.p.m. weight in this stage. Consequently, the hydrogenation activity of the catalyst declined very slowly because the rate of sulphiding was very low. The rate at which the operating temperature had to be raised to maintain hydrogenation at the required level is as follows:

Estimated Cyelohexane Temperature, F. Catalyst Ratio Content of Su1phur:Nickel Product,

Inlet Outlet Atomic Percent wt.

If the same rate of deactivation were maintained the estimated catalyst life would be 5500 hours, by which time the sulphurznickel ratio would be 0.021: 1.

Finishing hydrogenation stage The residual 3 to 5 percent weight benzene was hydrogenated over nickel on sepiolite catalyst to yield pure cyclohexane under the following conditions.

Pressure p.s.i.g 200 Temperature F 408 Space velocity v./v./hr 1.0 Make-up gas Hydrogen Gas recycle rate s.f.c./b 1165 Hydrogenzhydrocarbon ratio at reactor outlet molar 1:1 Hydrogen partial pressure at reactor outlet -p.s.i.a 108 Under these conditions the benzene content of the product was 1 to 2 p.p.m. by Weight. The run was in progress for 4880 hours without any indication of a decline in the activity of the catalyst. The method of determining sulphur was unable to detect any difference between the sulphur content of feedstock and product. No estimation of catalyst life can be made. From the data available it may be concluded that the catalyst would be active almost indefinitely.

The purity of the cyclohexane product was 99.78 percent weight. When the desulphurisation process referred to in Example 1 was operated at 430 F and the product hydrogenated under the conditions referred to in the present example the purity of the final cyclohexane product was within the range 99.86 to 99.92% wt. However, desulphurising at 450 F. caused by-product formation at a slightly higher level.

EXAMPLE 3 Desulphurised benzene containing 0.4 to 0.9 p.p.m. wt. sulphur derived from the desulphurisation process of the invention was hydrogenated in two stages according to the hydrogenation process of the invention. 95 to 97% conversion to cyclohexane occurred in the main reactor, which was recycle cooled, the operating conditions being identical with those of the corresponding stage of Example 2.

After this operation the cyclohexane product had a sulphur content within the range 0.2 to 0.5 p.p.m. wt. depending on the sulphur content of the feedstock. The rate at which the operating temperature had to be raised to maintain the required degree of hydrogenation is given below:

Cyclohexane Temperature, F. Catalyst Content of Sulphur;Nickel Product, Inlet Outlet Atomic Ratio Percent wt.

If catalyst deactivation continued at the same rate the estimated catalyst life would be 3200 hours, which would correspond to a catalyst sulphurznickel ratio of 0.021:1.

Finishing hydrogenation reactor The residual 3 to percent weight benzene was hydrogenated over nickel on sepiolite catalyst to yield pure cyclohexane under the same conditions as in Example 2.

Under these conditions the sulphur content of the cyclohexane was reduced by about 0.1 p.p.m. weight. Consequently, there was a slow decline in the hydrogenation activity of the catalyst. The rate at which the benzene content of the product increased is given in the table below:

B enzene Content of Catalyst Product, Ratio HOS p.p.m. wt. Sulphur: Nickel Atomic The estimated catalyst life is in excess of 5000 hours. Gas chromatographic analyses of the feedstock and product gave the following results:

What we claim is:

1. A process for the desulphurisation of a benzenecontaining fraction containing from 1 to 50 p.p.m. wt. sulphur including thiophene sulphur, without appreciable hydrogenation of the aromatic hydrocarbons, which comprises passing the fraction in the fluid state, together with hydrogen, over nickel supported on sepiolite at an elevated temperature of from 50 to 250 C. and pressure of from 100 to 500 p.s.i.g. such that sulphur combines with the nickel but no substantial amount of hydrogen sulphide is produced, the equilibrium hydrogen partial pressure being greater than the minimum necessary to prevent catalyst deactivation but less than that at which up to 10 mol percent hydrogenation of the aromatic hydrocarbons occurs.

2. A process as claimed in claim 1 in which the fraction contains at least wt. of aromatic hydrocarbons.

3. A process as claimed in claim 2 in which the fraction is benzene containing not more than 10 p.p.m. wt. sulphur.

4. A process as claimed in claim 1 in which the nickel supported on sepiolite contains from 1 to 50% wt. nickel, expressed as the element.

5. A process as claimed in claim 1 in which the reaction conditions are selected from the following:

Temperature (L. 50-250 Pressure p.s.i.g -500 Space velocity -v./v./hr 0.05-10 Inlet hydrogen: hydrocarbon ratio on total feed mola:r 0.0l:1-0.5:l

6. A process as claimed in claim 5 in which the reaction conditions are selected from the following:

7. A process as claimed in claim 1 in which the equilibrium hydrogen partial pressure is so adjusted that up to 5 mol percent hydrogenation occurs.

m 8. A process in which a benzene-containing fraction containing up to 50 p.p.m. wt. sulphur is desulphurised by a process as claimed in claim 1 to produce a fraction containing up to 2 p.p.m. wt. suphur, and this fraction is hydrogenated in two stages, both using nickel supported on sepiolite as catalyst, in which not less than 90% wt. and not more than 99% wt. of the fraction is hydrogenated in the first hydrogenation stage, with the hydrogenation reaction being substantially completed in the second hydrogenation stage, the temperature of the first hydrogenation stage being controlled by cooling and the second hydrogenation stage being uncooled.

9. A process as claimed in claim 8, in which the material entering the first hydrogenation stage reactor is in mixed phase and this stage is cooled by liquid recycle.

10. A process as claimed in claim 8 in which the second hydrogenation stage reactor outlet temperature is lower than the first hydrogenation stage reactor outlet temperature and the second stage operates in mixed phase.

11. A process as claimed in claim 8 in which the reaction conditions for the first and second hydrogenation stages are selected from the following:

First stage:

Temperature (50-300 C.) 122-572 Pressure p.s.i.g 50-500 Space velocity (fresh feed) v./v./hr 0.5-5.0 Product recycle ratio 2.5: 16:1

Hydrogen recycle rate on total feed s.c.f./b 200-1000 Second stage:

Temperature (50300 C.) 122-572 Pressure p.s.i.g- 50-500 Space velocity v./v./hr 0.5-5.0 Hydrogen recycle rate s.c.f./b- 500-2000 12. A process as claimed in claim 8 in which a mixed gas derived from a refinery process, and containing at least 50 mol percent hydrogen, is used.

13. A process as claimed in claim 12 in which the mixed gas is steam reformer tail gas.

14. A process as claimed in claim 8, in which the desulphurisation process reactor pressure and the pressure in both stages of the hydrogenation process is 200 p.s.i.g., the respective temperatures in the desulphurisation reactor and the first and second hydrogenation stages are 400-482F., 200480F., and 320430F., the space velocity in the desulphurisation process and both stages of hydrogenation is 1 v./v./hr., a mixed gas containing 95% hydrogen, derived from a steam reformer and supplied at 250 p.s.i.g., is used, and 95% cyclohexane is recycled to the first stage reactor at a ratio to fresh feed of 3.5:1 to 4:1.

References Cited UNITED STATES PATENTS 3,190,830 6/1965 Rowland et a1 208143 3,147,210 9/1964 Hass et al 260667 3,202,723 8/1965 Thonon 260667 2,755,317 7/1956 Kassel 260667 1,974,057 9/1934 Steffen et a1. 20859 2,303,075 11/1942 Fry 260667 2,464,539 3/1949 Voorhies 260667 3,341,613 9/1967 Hann 260667 2,459,465 1/1949 Smith 20857 2,426,929 9/ 1947 Greensfelder 260667 3,254,134 5/1966 Smith et al 260667 3,222,274 12/1965 Carl 260667 2,300,877 11/1942 Drennan 208244 3,215,751 11/1965 Bourne et a1. 2606832 DELBERT E. GANTZ, Primary Examiner V. OKEEFE, Assistant Examiner UNITED STATES PATENT OFFICE 35 W CERTIFICATE OF CORRECTION Pa n 3,484,496 Dated December 16, 1969 Inventofla) John Winsor and John Carruthers It is certified that error appears in the above-identified patent and that said Letters Patent are hereby corrected as shown below:

Col. 1, line 21 for "amounto f" read --amount of;

Col. 2, lines 35 and 36 for "nylon" reed -Ny1on--;

G01. 3, line 41 for "85%" read --C H Col. 4, line 15 for "above" read --about--;

Col. 4, line 51 for "0.01:1" read --0.1:1--;

Col. 7, line 11 for "hydorgen" read --hydrogen--;

Col. 8, line 20 for "opertaing" read --opereting--;

Col. 8, line 69 for "content" read ---contents--;

601. 10, lines 63 and 71 for "122-572" read --122 to 572F--.

SIGNED A SEALED Anew Edward M. Fletcher, Jr. smwymt Attesting Officer WILLIAM E. I

I L Comissioner of Paton

Images