Background of the Invention Naphtha streams, especially those that are products of a cracking process such as fluidized catalytic cracking, contain sulfur contaminants which are undesirable. For example, gasolines which are blended naphtha streams are restricted in the permissible level of sulfur contaminants because of the effect such contaminants have on the functioning of catalytic converters. While sulfur contaminated naphthas can be desulfurized by a great many hydrodesulfurization (HDS) catalysts and processes, often hydrotreating also results in severe octane loss due to extensive reduction of the olefins in the naphtha stream. Numerous attempts, of course, have been made to devise catalysts and processes which will favor hydrodesulfurization (HDS) over olefin hydrogenation ; and although some success has been achieved in obtaining greater selectivity often the selectivity gain is obtained at the expense of activity loss.

Thus, there remains a need for improved catalysts and processes for hydrodesulfurization of cracked naphtha with minimum hydrogenation of olefins.

Summary of the Invention Briefly stated, a process is provided for reducing the sulfur content of a hydrocarbon feedstock containing an olefinic component which comprises contacting the feedstock with a sulfided catalyst and hydrogen under hydro- desulfurization conditions, the catalyst comprising (i) at least one non-noble metal of Group VIII ; (ii) at least one metal of Group VIB ; and (iii) at least one metal of Group IB, IIB and IVA, on an inorganic oxide support thereby effecting the hydrodesulfurization of the feedstock without substantially hydrogenating the olefmic component.

Detailed Description of the Invention The feedstock treated according to the invention typically is one commonly designated as a cracked naphtha or gasoline blend stock. A fluid catalytic cracked (FCC) naphtha is a specific example of a suitable feedstock capable of being processed in accord with the invention.

The sulfided catalyst suitable for the practice of the invention comprises (i) at least one non-noble metal of Group VIII ; (ii) at least one metal of Group VIB ; and (iii) at least one metal of Group IB, IIB and IVA, on an inorganic oxide support. Typically the Group VIII metal is present in an amount ranging from about 0. 1 to about 15 wt% ; the Group VIB metal from about 0. 1 to about 40 wt% and the Group IB, IIB and IVA metals from about 0. 01 to about 10 wt% based on the total weight of the catalyst. Representative examples of suitable catalysts include Co-Mo-Cu, Co-Mo-Zn, Co-Mo-Sn, Co-Mo-Cu-Zn, Co-Mo-Sn-Zn and the like.

The support of the catalyst includes inorganic oxides such as alumina, silica, titania, magnesia, silica-alumina and mixtures of these. Alumina is a preferred support, and aluminas characterized as large pore aluminas are more preferred providing superior activity and activity maintenance. Typically large pore aluminas have a surface area greater than about 100 m2/g, a pore volume greater than about 0. 60 ml/g and an average pore diameter greater than about 105 Angstroms. Preferred aluminas have a surface area greater than 170 m2/g and an average pore diameter greater than 115 Angstroms.

The catalyst metals are deposited on the support by techniques well known in the art. The order in which the metals are deposited on the support can vary widely. For example, the metals may be deposited simultaneously, sequentially, or two metals may be deposited simultaneously and the third metal separately either prior to or after the deposition of the other two metals. Prefer- ably the metals are introduced to the support by the incipient wetness method.

After depositing the metal well known techniques for drying and calcining may be employed. Thus drying and calcining may be conducted after each metal addition or after complete metal addition. Drying and calcining may be conducted, for example, in air at 100°C to about 600°C. Similarly, known techniques for activation of the catalyst are employed. Thus the sulfiding treatment of the catalyst may be achieved with blends of hydrogen sulfide and hydrogen or hydrogen sulfide precursors in the presence of hydrogen.

In the practice of the invention the feedstock is contacted with the sulfided catalyst under hydrodesulfurization conditions. These conditions will vary depending upon the feed and the catalyst ; however, suitable conditions are set forth in Table 1.

TABLE 1 Condition Broad Range Preferred Range More Preferred Temueratllre, °C 200-400 250-375 275-300 Pressure, psig10-100050-800100-600 Hydrogen, SCFB 100-6000 500-4000 1000-3000 LHSV 0.1-20 1-15 2-10 Hydrogen Pulity, v°/O 60-100 80-99 85-95 The following Examples will serve to further illustrate the present invention.

Example 1. (Comparative) A commercial HDS catalyst containing 4. 0 wt% CoO and 15. 0 wt% Mo03 was activated by treating with 10% hydrogen sulfide in hydrogen. The catalyst was evaluated on a feed comprising about 33 wt% each of n-heptane, octene-1, m-xylene, 2000 wppm sulfur as 2-methylthiophene, and 20 wppm nitrogen as aniline. The results of this evaluation are presented in Table 2.

Example 2. The commercial HDS catalyst of Example 1 was impregnated with copper nitrate to incorporate about 2 wt% Cu. After pretreatrnent and activation the Co-Mo-Cu catalyst was tested on the feed of Example 1. The results are summarized in Table 2.

Example 3. The commercial HDS catalyst of Example 1 was impregnated with tin chloride to incorporate about 3 wt% Sn. After pretreatment and activation the Co-Mo-Zn catalyst was tested on the feed of Example 1. The results are summarized in Table 2.

TABLE 2 Selective HDS With Co-Mo, Co-Mo-Cu, and Co-Mo-Sn Catalysts 275-300°C, 200 psig, 2000 SCF/B, 7-10 LHSV Exam le Catal st HDS, wt% OS, wt% Selectivi Factor 1 Co-Mo 98 90 0. 8 2 Co-Mo-Cu 97 83 2. 0 3 Co-Mo-Sn 98 80 2. 5 Reaction conditions for the catalysts of Table 2 were selected to permit comparison of relative selectivity at equivalently high levels of HDS. OS is a measure of the degree of olefin saturation, and the selectivity factor is calculated from the rates of HDS and OS. Table 1 illustrates that the catalysts of this invention modified by the addition of Cu and Sn are substantially more selective than the base case catalyst.

Examples 4 (Comparative) and 5-8. A Co-Mo HDS catalyst was synthesized by impregnating alumina with cobalt carbonate and ammonium heptamolybdate. After being dried and calcined at 400°C for 3 hrs, the Co-Mo catalyst was impregnated with copper nitrate to prepare a series of Co-Mo-Cu catalysts containing about 3 wt% CoO, 11 wt% MoO3 and 1-6 wt% Cu. The catalysts were tested as described in Example 1, and the results are summarized in Table 3.

TABLE 3 Selective HDS With Co-Mo, and Co-Mo-Cu Catalysts 275°C, 200 psig, 2000 SCF/B, 10 LHSV Example Cu, wt% HDS, wt% OS, wt% Selectivity Factor 4 0. 0 94 85 1. 5 50. 996802. 1 6 1. 6 96 82 2. 0 7 3. 6 96 80 2. 0 8 5. 3 96 83 1. 9 The data of Table 3 illustrate that the catalysts of this invention are more selective than their Cu-free analog.

Example 9 (Comparative). The catalyst of Example 4 and the catalyst of Example 7 were tested as in Example 1 at process conditions providing a common level of HDS at a common temperature. The data shown in Table 4 confirm that the Co-Mo-Cu catalyst of this invention is intrinsically more selective and that the selectivity credit is not an artifact of the reaction conditions.

TABLE 4 285°C, 200 psig, 2000 SCF/B Catalyst Co-Mo Co-Mo-Cu LHSV 10 4 Selectivity Factor 1. 2 1. 7 OS, Wt% 90 84 HDS, Wt% 94 96 Example 10 (Comparative). A vendor HDS catalyst containing about 2 wt% CoO, 7 wt% MoO3, and 0. 6 wt% K was pretreated and tested as in Example 1. The results of the test are presented in Table 5.

Example 11. The catalyst of Example 10 was modified by the addition of about 0. 9 wt% Cu. The catalyst was activated and tested as described in Example 1. The data from the test are included in Table 5.

TABLE 5 Selective HDS With Co-Mo and Co-Mo-Cu Catalysts 275°C, 200 psig, 2000 SCF/B, 10 LSHV Example 10 11 Catalyst Co-Mo Co-Mo-Cu HDS, Wt% 98 96 OS, Wt% 94 86 Selectivity Factor 1. 5 _ 1. 7 The data of Tables 2, 3 and 5 show that the Co-Mo-Cu catalysts of this invention are more selective than the reference catalysts independent of metals loadings.

Example 12. The catalyst of Example 7 was activated and tested as described in Example 1. The catalyst was subjected to high severity process conditions favoring selectivity by operating at high temperatures and low pressures. Representative data at selected periods of this test are presented in Table 6.

TABLE 6 Effect of Reaction Conditions on Selectivity and Stability 2000 SCF/B, 4 LHSV Balance Period 1 2 3 4 5 6 Hr. on Oil 38 43 52 133 232 235 Temperature, °C 285 300 325 310 300 300 Pressure, psig 200 50 50 25 15 1 50 HDS, Wt% 96 96 99 89 69 90 OS, Wt% 84 81 89 57 32 58 Selectivity Factor 1. 7 2. 0 2. 2 2. 6 3. 0 2. 7 Table 6 illustrates that as reaction temperature increases and pressure decreases, olefin saturation is less favorable, and the selectivity of the reaction increases.

Comparison of balances 2 and 6 at common conditions shows that after extended operations at high severity conditions the catalyst of this invention retains high HDS activity, decreased olefin saturation activity, and a substantially higher selectivity factor. The data indicate that the catalyst of this invention resists deactivation at high severity conditions that favor HDS over OS.

Example 13 (Comparative). A vendor catalyst containing about 1 wt% CoO and 4 wt% MoO3 was used to process a 200-450°F cat naphtha at 500°F, 235 psig H2, 2600 SCF/B, 6. 5 LHSV. The results are summarized in Table 7.

Example 14. The catalyst of Example 2 was used to process a 200-450 °F cat naphtha at 500°F, 235 psig H2, 2600 SCF/B, 6. 5 LHSV. The results are summarized in Table 7.

TABLE 7 Selective HDS of 200-450°F Cat Naphtha With Co-Mo and Co-Mo-Cu Catalysts 500°F, 235 psig H2 2600 SCFB, 6. 5 LSHV Example 13 14 Catalyst Co-Mo Co-Mo-Cu HDS, Wot% 68 80 OS, Wt% 16 17 Selectivity Factor 0. 7 0. 9 Activity @ 150 hr 250 250 Actiwty @ 300 hr 180 250 The catalyst of this invention is more selective than the reference catalyst for the processing of this feedstock. Table 7 confirms the superior stability of the catalyst of this invention which experienced no deactivation over a 150 hr period while the reference catalyst activity decreased by about 30%.

Example 15. (Comparative) The catalyst of Example 4 was used to process a 200-450°F cat naphtha. The results are summarized in Table 8.

Example 16. The catalyst of Example 6 was used to process a 200-450°F cat naphtha. The results are summarized in Table 8.

TABLE 8 HDS of 200-450°F Cat Naphtha With Co-Mo and Co-Mo-Cu Catalysts Example 16 15 16 Catalyst Co-Mo Co-Mo-Cu Co-Mo Co-Mo-Cu Conditions 273°C, 300 si 293°C, 125 psig Activity 200 hr 80 39---- Activity (800 hr 80 39 Activity 900 hr----480 170 Activity 1100ho 380 160 The data show that at the initial process conditions neither catalyst experienced deactivation through 800 hr on oil. When high temperature, low pressure conditions favoring selectivity were imposed the catalyst of this invention resisted deactivation while the reference catalyst experienced 20% deactivation within 200 hours.

Example 17. (Comparative) A Co-Mo HDS catalyst was synthesized by impregnating a large pore alumina with cobalt carbonate and ammonium heptamolybdate. After being dried and calcined at 400°C for 3 hr, the catalyst was tested as in Example 1. The results are summarized in Table 9.

Example 18. A Co-Mo HDS catalyst was synthesized by impregnating a large pore alumina with cobalt carbonate and ammonium heptamolybdate. After being dried and calcined at 400°C for 3 hr, the catalyst was impregnated with copper nitrate to prepare a Co-Mo-Cu catalyst containing about 3 wt% CoO, 11 wt% Mo03 and 4 wt% Cu. The catalyst was tested as in Example 1. The results are summarized in Table 9.

TABLE 9 Selective HDS With Co-Mo, and Co-Mo-Cu Catalysts 275°C, 200 psig, 2000 SCF/B, 10 LHSV Example Cu, Wt% HDS, Wt% OS, Wt% Selectivity Factor 4 0. 0 94 85 1. 5 17 0.0 99 97 1.3 7 3. 6 1 96 80 2. 18 3.8 94 70 2.3 The results illustrate that the Co-Mo catalyst prepared on the large pore alumina is more active than its conventional alumina analog. The Co-Mo-Cu catalyst on the large pore alumina is more selective at comparable activity than its conven- tional alumina analog.