Bimetallic titania-based electrocatalysts deposited on ionic conductors for hydrodesulfurization reactions

Abstract

This invention relates to a method for preparing a bimetallic titania-based catalyst for use in hydrodesulfurization reactions.

Claims

1. A method of preparing a CoMo/TiO.sub.2 catalyst using co-impregnation under Equilibrium Deposition Filtration (EDF), which comprises the sequential steps of: a) first dissolving a molybdenum salt in water and adjusting the pH of the solution to an acid value; b) followed by dissolving a titania compound in the solution of step a) and adjusting the pH of the solution to an acid value; c) followed by dissolving a cobalt salt in the solution of step b) to adsorb said cobalt onto said titania compound with said molybdenum and adjusting the pH of the solution to an acid value; d) evaporating the solution of step c); e) collecting a CoMo/TiO.sub.2 solid; and f) calcining the CoMo/TiO.sub.2 to form a catalyst.

2. The method according to claim 1, wherein the pH in step a) is adjusted to 4.3.

3. The method according to claim 2, wherein the pH of the solution in step a) is adjusted with HNO.sub.3.

4. The method according to claim 1, wherein the pH in step b) is adjusted to 4.3.

5. The method according to claim 4, wherein the pH of the solution in step b) is adjusted to 4.3.

6. The method according to claim 1, wherein the evaporation in step e) is conducted at a temperature of about 40° C. to about 45° C. and at a pressure of 30 mbar.

7. The method according to claim 1, wherein the calcinations in step e) is conducted at about 500° C. for about 2 hours.

8. The method according to claim 1, wherein the final composition of the catalyst is about 14.2 wt. % MoO.sub.3, 3.2 wt. % CoO and 82.6 wt. % TiO.sub.2.

9. The method of claim 1, wherein the acid value of (a), (b), and (c) is 4.3.

10. The method of claim 1, wherein said Mo salt is ammonium heptamolybdate, said titania compound is TiO.sub.2 and said cobalt salt is cobalt nitrate.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) FIG. 1 depicts the transient effect of a constant applied negative potential (−1.5V) on the rate of H.sub.2S formation, the conversion of thiophene and the current. T=250° C., sample: S1.

(2) FIG. 2 depicts the transient effect of a constant applied positive (1.5V) and negative (−1.2V) potential on the rate of H.sub.2S formation, the conversion of thiophene and the current. T=500° C., sample: S2.

(3) FIG. 3 depicts the transient effect of a constant applied negative current (−1 μA) on the H.sub.2S formation catalytic rate, the conversion of thiophene and the catalyst reference potential difference. T=300° C., sample: S3.

DETAILED DESCRIPTION OF THE INVENTION

(4) The NEMCA effect on hydrodesulfurization catalysts can best be described as an electrochemically induced and controlled promotion effect of catalytic surfaces generated by electrolyte charge carrier spillover to/from the electrolyte onto the catalyst surface.

(5) In one embodiment of the present invention, a catalyst composition is provided for the removal of sulfur from petroleum hydrocarbon oils. The catalyst composition is useful in the removal of sulfur from middle distillates produced at distillation temperatures ranging from 200° C. to about 450° C., for example diesel fuel.

(6) In the hydrodesulfurization (HDS) process of the present invention, the catalyst can consist of at least one metal from the metals of Group VIII of the Periodic Table and at least one metal selected from the metals of Group VI of the Periodic Table, which are used as the active metals to be supported on the support.

(7) Examples of the Group VIII metals include cobalt (Co) and nickel (Ni), while examples of the Group VIA metals include molybdenum (Mo) and tungsten (W). The combination of the Group VIII metal and the Group VI metal is preferably Mo—Co, Ni—Mo, Co—W, Ni—W, Co—Ni—Mo, or Co—Ni—W, and most preferably Mo—Co or Ni—Mo.

(8) The content of the Group VI metal in terms of its oxide is preferably in the range of about 1% to 30% by mass, more preferably 3% to 25%, by mass, and most preferably 5% to 20% by mass, based on the mass of the catalyst. If less than 20% by mass is employed, it would not be sufficiently active to desulfurize sufficiently, and if a mass greater than 30% were employed, it would condense resulting in reduced desulfuzation

(9) The supporting ratio of the Group VIII metal and the Group VI metal is a molar ratio defined by [Group VIII metal oxide]/[Group VI metal oxide] ranging from 0.105 to 0.265, preferably 0.125 to 0.25, and most preferably from 0.15 to 0.23. A molar ratio of less than 0.105 would result in a catalyst having inadequate desulfurization activity. A molar ratio of greater than 0.265 would result in a catalyst lacking sufficient hydrogenation activity and reduced desulfurization activity.

(10) The total content of the Group VIII metal and the Group VI metal is preferably 22% by mass or greater, more preferably 23% by mass or greater, and most preferably 25% by mass or greater in terms of oxide based on the mass of the catalyst. A mass of less than 22% would result in a catalyst which exerts insufficient desulfurization activity.

(11) The preferred catalyst support for use in accordance with the present invention is titanium dioxide (TiO.sub.2). While alumina is the most widely used support material for commercial hydrodesulfurization (HDS) catalysts due to its good mechanical properties, titania based catalysts have been found to be more successful and more suitable when the HDS process is based on electrochemical promotion.

(12) Other supports can also be employed provided they are ion conductors. Exemplary of such supports are alumina, ceria, silica, zirconia, RuO.sub.2, CZl and BCN18.

(13) The solid electrolyte supports which can be utilized in the process of the present invention are O.sup.2− ionic conductors, exemplary of which is YSZ (8% mol Yttria Stabilized Zirconia) and low temperature (<400° C.) proton conductors, exemplary of which is BCN 18 (Ba.sub.3 CA.sub.1.18 Nb.sub.1.82O.sub.9-a).

(14) Other solid electrolytes include β.sup.11-Al.sub.2O.sub.3, β-Al.sub.2O.sub.3, Li.sup.+ and K.sup.+ conducting β-Al.sub.2O.sub.3.

(15) The level of the voltage applied is usually in the range of ±0.5V to ±2V, preferably about ±1.5V.

(16) As adverted to previously, the phenomenon of electrochemical promotion of catalysis (EPOC or NEMCA effect) has been utilized in the process of the present invention for the in situ modification of the HDS activity of bimetallic Mo—Co catalyst-electrodes at low temperatures and atmospheric pressure.

(17) The phenomenon of electrochemical promotion of catalysis has been investigated using a variety of metal catalysts (or conductive metal oxides), solid electrolyte supports and catalytic reactions. In electrochemical promotion, the conductive catalyst-electrode is in contact with an ionic conductor and the catalyst is electrochemically promoted by applying a current or potential between the catalyst film and a counter or reference electrode, respectively. Numerous surface science and electrochemical techniques have shown that EPOC is due to electrochemically controlled migration (reverse spillover or backspillover) of promoting or poisoning ionic species (O.sup.2− in the case of YSZ, TiO.sub.2; and CeO.sub.2, Na.sup.+ or K.sup.+ in the case of β″-Al.sub.2O.sub.3, protons in the case of Nafion, CZI (CaZr.sub.0.9In.sub.0.1O.sub.3-α) and BCN18 (Ba.sub.3Ca.sub.1.18Nb.sub.1.82O.sub.9-α), etc.) between the ionic or mixed ionic-electronic conductor—support and the gas exposed catalyst surface, through the catalyst-gas-electrolyte three phase boundaries (TPBs).

(18) Two parameters are commonly used to quantify the magnitude of the EPOC effect:

(19) 1. the rate enhancement ratio, ρ, defined from:
ρ=r/r.sub.o  (1)

(20) where r is the electropromoted catalytic rate and r.sub.o the open-circuit, i.e. normal catalytic rate.

(21) 2. the apparent Faradaic efficiency, Λ, defined from:
Λ=(r−r.sub.o)/(I/nF)  (2)

(22) where n is the charge of the ionic species and F the Faraday's constant.

(23) A reaction exhibits electrochemical promotion when |Λ|>1, while electrocatalysis is limited to |Λ|≦1.

(24) The selectivity of the reaction to the produced hydrocarbon (HC) species has been calculated by
S.sub.i=r.sub.i/r.sub.th  (3)

(25) where, r.sub.i is the formation rate of each HC product and r.sub.th the consumption rate of thiophene, equal to Σr.sub.i within ±2%.

(26) The preferred method for preparing MoCo/T.sub.1O.sub.2 for use in accordance with the present invention involves wet impregnation at various pH values. The wet impregnation method was used for the co-deposition of Mo and Co with a ratio of 15 wt % MoO.sub.3 to 3 wt % CoO. The addition of 82 wt % TiO.sub.2 (anatase) ensures a fine dispersion of the catalyst with only one monolayer of mostly Mo at the surface of the TiO.sub.2 matrix. As indicated, wet impregnation can be carried out at different pH values. At more acidic concentrations of the solution (pH=4 or 4.3) the formation of polymeric Mo.sub.7O.sub.24 is to be expected, while at neutral conditions (pH=6.7), only monomeric MoO.sub.4 species will be obtained.

(27) The polymeric species (mainly [Mo.sub.7O.sub.24].sub.6− and [HMo.sub.7O.sub.24].sub.5−) are deposited through electrostatic adsorption on TiO.sub.2 surface. The monomeric species (mainly [MoO.sub.4].sub.2−) are adhered through the formation of hydrogen bonds and inner sphere complexes. Because the wet impregnation method is used to prepare the catalysts, only a small amount of the species is deposited through the above deposition modes. Thus, the greater amount of the molybdenum species is deposited through bulk deposition. This means that the species are deposited through precipitation during the water evaporation step.

Preparative Example 1

(28) Wet Impregnation Method at pH 4.3

(29) The preparation of the CoMo/TiO.sub.2 (pH=4.3) catalyst is performed using the “co-impregnation under EDF conditions” method. According to this method, 0.7361 g ammonium heptamolybdate [(NH.sub.4).sub.6Mo7O24*4H.sub.2O] were dissolved in about 100 ml of triple-distilled water, in a 500 ml round flask. The pH was adjusted to 4.3 by adding concentrated HNO.sub.3 solution dropwise. To this solution, 3.495 g of TiO2 were added and due to the Mo adsorption on TiO.sub.2 surface and the PZC(=6.5) of the titania used, the pH rose to between 8 and 8.5. To this suspension, 0.52018 g of cobalt nitrate [Co(NO.sub.3)2*6H.sub.2O] were then added. Before the addition of the cobalt nitrate salt the pH was adjusted again to 4.3 in order to avoid Co(NO.sub.3)OH precipitation. As the adsorption of cobalt species altered the solution pH to a value of 3.0-3.5, the latter had to be adjusted again to 4.3 using a concentrated ammonia solution (NH.sub.4OH).

(30) The round flask containing the suspension was then placed in a rotary evaporator and stirred for about 30 min. After a final adjustment to pH 4.3, the evaporation at T=40-45° C. and pressure 30 mbar was started. When the material was dried, it was transferred in a porcelain crucible and calcined at 500° C. for 2 hours in air. The final composition of the catalyst was 14.2 (10) wt % MoO.sub.3 (Mo), 3.2 (2.6) wt % Coo (Co) and 82.6 (87.4) wt % TiO.sub.2.

Preparative Example 2

(31) Co-Deposited Catalyst Powder

(32) A few drops of triple-distilled water was added to a small amount of the final powder of the wet impregnation method of Preparative Example 1 to form a thick paste, which was then spread at the surface of the proton ion conductor. The catalyst film was dried at 120° C. for 30 minutes and was then calcined at 500° C. for 2 hours.

Preparative Example 3

(33) Procedure for Preparing Sputtered Mo Films

(34) Thin Molybdenum (Mo) coatings are produced by a de magnetron sputter process. By this process homogeneous, well adhered, thin metal or metal oxide coatings are produced. For the production of such coatings, suitable for electrochemical promotion, several favorable deposition parameters need to be defined.

(35) Deposition Parameter for the Mo Catalyst-Electrodes

(36) The sputtering parameters for the Mo coating on the CIZ electrolyte substrate with a mass of

(37) m=0.0001 g are:

(38) p=8.86-10.53 mTorr (˜40 ccm Ar)

(39) P=330-403 Watt

(40) I=0.8-1.09 Amp

(41) V=327 Volt

(42) Deposition Time: 10 minutes

(43) The sputtering parameters for the Mo coating on the CIZ electrolyte substrate with a mass of

(44) m=0.0135 g are:

(45) p=5.57-6.74 mTorr

(46) P-306-390 Watt

(47) I=0.8-1.03 Amp

(48) V=340 Volt

(49) Deposition Time: 40 minutes

Preparative Example 4

(50) Procedure for Preparation of MoCo Deposited on Sputtered Mo Films

(51) A thin layer of Mo was sputtered on the CZI electrolyte and calcined at 500° C. The films were reduced in H.sub.2 and presulfation was carried out at 500° C.

(52) The MoCo deposited catalyst was prepared in the following way:

(53) 1. To a round bottom flask, 100 ml of triple distilled water was added.

(54) 2. 3 g of (NH.sub.4).sub.6Mo.sub.7O.sub.24*4H.sub.2O was added to the 100 ml H.sub.2O.

Preparative Example 5

(55) Presulfation

(56) Presulfation of the catalyst-electrolyte assembly was carried out with all catalyst-electrolyte assemblies in Examples I and II in the following manner. A stream of Ar (20 ml/min) was passed through the reactor, discussed in Preparative Example 8, while the reactor was heated to 500° C. and was maintained for 30 min at 500° C. A stream of 15 Vol % H.sub.2S and 85 Vol % H.sub.2 was supplied to the reactor at 500° C. for 2 hours. After completing the presulfation, a stream of Ar was supplied for 1 hour at 400° C. Cooling/heating to the desired reaction temperature.

(57) Preparative Example 6

(58) Two Different Solid Electrolyte Supports were Used:

(59) a. YSZ (8% mol Yttria Stabilized Zirconia), an O.sup.2− ionic conductor.

(60) b. BCN18 (Ba.sub.3Ca.sub.1.18Nb.sub.1.82O.sub.9-α) a low temperature (<400° C.) proton conductor.

(61) Both of the solid electrolyte pellets had a thickness of 2 mm and diameter of 18 mm.

Preparative Example 7

(62) Preparation of Metal and Metal Oxide Interlayers

(63) Thin Mo and TiO.sub.2 layers were deposited over the ionic conductor supports, before the catalyst deposition, to increase the intimate contact or adhesion between the catalyst-electrode, also known as the working electrode, and the support, and to facilitate the migration of the backspillover species onto the catalyst surface. The enhancing role of TiO.sub.2 interlayers has been reported in previous EPOC studies, both for oxidation and hydrogenation reactions.

(64) The Mo/electrolyte and TiO.sub.2/electrolyte type thin electrodes were prepared by metal sputtering as described previously. A magnetron sputtering system was used. High purity Ar and O.sub.2 were used as sputtering and reactive gas, respectively. The discharge characteristics were controlled using a variable DC power supply (1 kV and 2 A). Pure Mo (99.95%) and Ti (99.95%) were used as sputtering targets.

(65) TiO.sub.2 thin layers were deposited only over the YSZ substrates with 600 W target power, which enabled a 0.5 nm/min deposition rate and led to a c.a. 90 nm thickness film after 3 hours of deposition. The substrate temperature was stable during the deposition at 250° C. Also, a post-deposition annealing of the deposited TiO.sub.2 layer was performed in air at 600° C. for 60 minutes, resulting in a 60% rutile and 40% anatase structure. The TiO.sub.2 layers structure characterization has been described previously.

(66) Molybdenum interlayers were deposited in each case holding the target power stable at 280 W, which enabled a 20 nm/min deposition rate, achieving a 100 nm thickness Mo layer after 5 minutes of deposition. The substrate temperature was kept stable at 50° C.

(67) On the other side of the pellets, Au counter and reference electrodes were prepared in each case by application of a metalorganic paste (Metalor, Gold resinate, A1118), followed by drying at 400° C. for 90 minutes and calcination at 650° C. for 30 minutes. Blank experiments using Au also for the working electrode showed that Au is practically catalytically inactive for the HDS reaction, establishing that the observed rate values correspond only to the rate on the catalyst-electrode and not to the counter or reference electrodes. Similar blank experiments with TiO.sub.2 powder showed that TiO.sub.2 also is inactive under the mentioned conditions.

Preparative Example 8

(68) Reactor Operation

(69) An atmospheric pressure electrochemically promoted single chamber reactor was used equipped with a solid electrolyte pellet, on which three electrodes were deposited, namely, the working-catalyst electrode, the counter electrode and the reference electrode. Gold wires were used for the electrical connections between the electro catalytic element and the external power supply unit. A three-bore, ultra high vacuum feed-through unit, special for high temperature H.sub.2 and H.sub.2S environment, was used for the gastight introduction of the gold wires in the reactor. The temperature was measured and controlled by a type-K thermocouple placed in a stainless steel close-end tube in the proximity of the catalyst electrode.

(70) The feed gas composition and total gas flow rate, F.sub.T, was controlled by four mass flowmeters (Brooks smart mass flow and controller, B5878). Reactants were Messer-Griesheim certified standards of pure (99.99%) H.sub.2 and H.sub.2S, while thiophene was introduced using a saturator (P.sub.th.sup.25° C.=3 kPa) with Ar or H.sub.2 carrier gas. Pure (99.99%) argon was fed through the fourth flowmeter in order to further adjust total gas flow rate and inlet gas composition at desired levels. H.sub.2 partial pressure was held constant at 97 kPa, while thiophene partial pressure could be varied from 0.5 to 5 kPa. The total gas flow rate was constant at 30 and 60 cm.sup.3/min. Reactants and products were analyzed by on-line gas chromatography (Shimadzu 10A, equipped with a Porapaq-QS column at 50° C. for the separation of thiophene and the produced C.sub.xH.sub.y) in conjunction with a continuous analysis H.sub.2S colorimeter (Applied Analytics Inc.). Constant currents or potentials were applied using an AMEL 2053 galvanostat-potentiostat.

(71) TABLE-US-00001 TABLE 1 Electrochemical Cells Used in Examples Counter/ Working Reference Sample Cell Electrode Interlayer Electrolyte Electrode S1 MoCo/TiO.sub.2/YSZ/Au MoCo TiO.sub.2 YSZ Au (impregnation) (sputtering) S2 MoCo—TiO.sub.2/ MoCo/TiO.sub.2 Mo YSZ Au Mo/YSZ/Au (dispersed- (sputtering) impregnation) S3 MoCo—TiO.sub.2/ MoCo/TiO.sub.2 Mo BCN18 Au Mo/BCN18/Au (dispersed- (sputtering) impregnation)

Example I. The Use of O2− Ionic Conductor Support (YSZ)

(72) As can be seen by an examination of FIG. 1, it shows the transient effect of a constant applied negative potential (−1.5V) on the rate of H.sub.2S formation (reaction (4)), the conversion of thiophene and the current at 250° C., using sample S1, i.e. Mo—Co/TiO.sub.2(sp)/YSZ/Au, identified more fully in Table 1 and the preparative examples.
C.sub.4H.sub.4S+H.sub.2.fwdarw.C.sub.xH.sub.y+H.sub.2S  (4)

(73) As shown, under open circuit, i.e. normal catalytic conditions, the conversion of thiophene is ˜0.5%. Negative potential application (−1.5V) causes a 3.9-fold increase of the catalytic rate (ρ=3.9), where thiophene conversion reaches ˜2%, while the apparent Faradaic efficiency is 0.2. However, in the present instance where oxygen is not present in the gas mixture, any current or potential-induced catalytic rate change suggests electrochemical promotion, even when |Λ|<1. On the other hand, positive potential application has no effect on the catalytic rate. The electrophilic behavior observed here, i.e. rate increase by electrode potential or work function decrease, appears when the electron acceptor (i.e. thiophene) species is weakly adsorbed on the catalytic surface and the electron donor (H.sub.2) strongly adsorbed. This agrees with the positive order dependence of the reaction rate on P.sub.th, proposed in the literature. Upon the application of negative potential, i.e. O.sup.2− pumping from the catalytic electrode, the thiophene-catalyst bond strength increases and causes higher thiophene coverage of the catalytic surface.

(74) After current interruption the catalytic rate reversibly returns to its initial open-circuit steady-state value. This indicates that the surface sulfur species, formed during the sulfation pretreatment step, were not consumed upon negative polarization, which would cause catalyst partial deactivation.

(75) As can be seen by an examination of FIG. 2, it shows the transient effect of a constant applied positive (1.5V) and negative (−1.2V) potential on the rate of H.sub.2S formation (reaction (4) above), the conversion of thiophene and the current at 500° C., using sample S2 identified more fully in Table 1 and the preparative examples, where the CoMo—TiO.sub.2 dispersed catalyst coating is supported over a thin sputter deposited Mo film interfaced with YSZ.

(76) As shown, under open-circuit conditions the conversion of thiophene is ˜0.7%, similar to that obtained using sample S1 (FIG. 1) at 250° C. Positive potential application (1.5V) causes an 1.5-fold increase of the catalytic rate, while the apparent Faradaic efficiency is 0.1. This is in contrast to the behavior of sample S1, where positive polarization had no effect on the rate. On the other hand, application of a negative potential (−1.2V) causes a similar less pronounced effect on the rate where ρ=1.5, indicating the inverted volcano behaviour of the system. This change from electrophilic to inverted volcano behavior, utilizing similar samples (S1 and S2), S2 is mainly due to the high operating temperature for sample S2. At elevated temperatures both electron acceptor and electron donor species are weakly bonded on the catalytic surface, which results in the observed inverted volcano behavior of FIG. 2. Similar change of the EPOC behaviour by temperature increase has been reported in previous studies on the model reaction of C.sub.2H.sub.4 oxidation over Pt/YSZ, where the behaviour changes from electrophobic to inverted volcano.

(77) After positive current interruption, the catalytic rate decreases to a value lower (?) than the initial open-circuit steady-state value, indicating a deactivation of the catalyst.

Example II. The Use of Proton Conductor Support (BCN18)

(78) As can be seen by an examination of FIG. 3, it shows the transient effect of a constant applied negative current (−1 μA) on the H.sub.2S formation catalytic rate, the conversion of thiophene and the catalyst electrode—reference potential difference at 300° C. using sample S3, identified more fully in Table 1 and the preparative examples.

(79) As shown, negative current application causes a 10% increase of the catalytic rate (ρ=1.1), while the apparent Faradaic efficiency equals 585, i.e. each proton present on the catalyst surface, which establishes an effective double layer, causes the reaction of up to 585 adsorbed from the gas phase hydrogen species. After the negative current interruption, the catalytic rate decreases and stabilizes at a value lower than the initial value. This “poisoning” effect of the negative polarization can be attributed to possible hydrogenation of the catalytically active surface sulfur groups supplied by the backspillover proton species upon negative polarization.

(80) While certain representative embodiments and details have been shown for purposes of illustrating the invention, it will be apparent to those skilled in the art that various changes in the methods and the electrochemical cell disclosed herein may be made without departing from the scope of the invention, which is defined in the appended claims.