Ruthenium on low acidity zirconia support

Abstract

The invention relates to a catalyst comprising 0.1-5 wt% ruthenium and 0.1-2.0 wt% sodium on a zirconia support. Such catalysts have low levels of acid sites on the zirconia support and are useful in a range of applications, particularly those requiring a support with low acidity. Also disclosed is a process for preparing the catalyst by impregnation of a zirconia support with a co-impregnation support comprising ruthenium oxalate and sodium citrate.

Description

[0001] P102547W001 1

[0002] Ruthenium on low acidity zirconia support

[0003] Field of the Invention

[0004] The present invention relates to heterogeneous catalysts comprising ruthenium on a zirconia support and a method for their manufacture.

[0005] Background

[0006] Ruthenium supported on a zirconia support (hereafter “Ru / ZrO ) has been investigated for a variety of different applications in heterogeneous catalysis.

[0007] For example, Ru / ZrC>2 is known as a catalyst for the water-gas shift reaction (CO + H2O — > H2 + CO2). WO99 / 64150A1 describes an autothermal catalyst comprising ruthenium on a zirconia carrier. W002 / 066380A1 describes an alkali-metal modified zirconia-support catalyst and its use as a water-gas shift catalyst. In the example ZrO2 extrudates are ground and sieved, then co-impregnated with an aqueous solution of ruthenium trichloride hydrate and potassium nitrate. The impregnated material was dried then calcined to obtain the catalyst.

[0008] W02006 / 093364A1 describes Ru / ZrO2 as a catalyst for the hydrogenation of sugars to sugar alcohols. In the examples the catalyst is prepared by impregnation of a ZrO2 support with ruthenium chloride to achieve a Ru loading of 3 wt%. The supported catalyst was dried then reduced (350 °C for 6 h in a H2-containing atmosphere). The reduced catalyst was then washed with aqueous ammonia then re-dried.

[0009] W02006 / 114831A1 describes a catalyst for reducing the content of SO3 present in combustion exhaust gas. The catalyst comprises Ru and / or Ir supported on a substrate, and at least 50 wt.% of the Ru and / or Ir is present within a 150 pm thick surface layer of the substrate. ZrC>2 is mentioned as an option for the substrate.

[0010] WO2016 / 135268A1 describes a method for preparing a chemical compound from a starting compound in a contaminant-containing feedstock, in which the contaminant is present in an amount of at least 0.01 wt.% based on the weight of the feedstock. The catalyst used in the method comprises ruthenium on a zirconium oxide support. The examples use catalysts P102547W001 2 which are prepared by adding a precursor solution comprising RuNO(NOs)s in deionized water to monoclinic ZrC>2 support, evaporating the water, followed by steps of drying, calcining and reduction using H2.

[0011] CN119076050A describes the preparation of ruthenium-based solid acid catalysts having ruthenium on a mixed zirconium dioxide and HZSM-5 molecular sieve support and their use in the hydrogenation of carbon dioxide to produce ethanol. In Example 2 the catalyst 1 %Ru-5%S2O8272ZrC>2 is prepared by a method which involves adding an ammonium persulfate-impregnated zirconia support to a solution of ruthenium trichloride, then adding a solution of sodium borohydride. The mixture is filtered then dried to obtain the catalyst.

[0012] DE102009012415A1 describes Ru / ZrC>2 as a catalyst for the gasification of biomass. The catalyst is a shell catalyst with an “open pore” catalyst support. In the example 10 g of ZrC>2 pellets (3 mm diameter) were sequentially impregnated with an aqueous solution of ruthenium trichloride having a volume equal to the pore volume, then with a 3.5M sodium hydroxide solution in order to effect precipitation of ruthenium hydroxide. The resulting catalyst had an average shell thickness of 459 pm. The resulting shell catalyst was then washed with deionised water until chloride-free and then dried.

[0013] EP2103593A1 describes a process for the conversion of an aliphatic alcohol into an aliphatic amine which is carried out in the presence of ammonia, hydrogen and a catalyst. Example 1 uses a Ru / ZrC>2 catalyst.

[0014] EP2446963A1 describes a catalyst for the selective hydrogenation or aromatic or polyunsaturated compounds in the presence of a Ru / ZrC>2 catalyst, which is promoted with Fe2O3 and at least one other metal oxide promoted selected from an oxide of a group IB, I IB and 11 IA metal.

[0015] Some important properties which impact the suitability of a ZrC>2 support for a given reaction are the BET surface area, median pore size and the acidity. The production of a ZrC>2 support typically involves a final stage in which the material is calcined. Calcination temperature correlates inversely with BET surface area and positively with median pore size i.e. the higher the calcination temperature the lower the BET surface area and the larger the median pore size. Calcination temperature also correlates inversely with the P102547W001 3 acidity i.e. the higher the calcination temperature the lower the acidity. To the inventors’ best knowledge it is not possible to produce ZrC>2 supports having simultaneously a low acidity and a high BET surface area just by controlling the calcination temperature.

[0016] It is known to support Ru on SiC which inherently has very low acidity. Ru / SiC catalysts are used in a variety of reactions such as: catalytic wet air oxidation, as described in CN107442148A and EP3131855A2; selective hydrogenation of phthalates, as described in CN108499586A; and the selective hydrogenation of esters, as described in US2007 / 255070A1 . Although Ru / SiC catalysts have received attention there are a number of drawbacks to the use of SiC in some reactions. Firstly, whilst SiC is relatively inert, it can undergo oxidation at high temperatures which can lead to a deterioration in the support, ultimately requiring replacement of the catalyst. Secondly, SiC has very low surface functionality meaning it is difficult to tune the properties of a SiC support to make it more suitable for a given reaction. Thirdly, SiC is relatively expensive compared to oxidic supports. These problems could potentially be solved by replacing the SiC support with a ZrC>2 support having a BET surface area, median pore size and acidity which is identical to, or at least closely matched to, the incumbent SiC support. However, as was explained above, because the ZrC>2 calcination temperature impacts these parameters, it is not possible to manufacture ZrC>2 having simultaneously a lower acidity and a high BET surface area purely by controlling the calcination temperature.

[0017] A possible approach to decouple the acidity from the BET surface area and median pore size would be to calcine the ZrC>2 support to a temperature in order to achieve the desired BET surface area and median pore size, and then modify the acidity by chemical treatment of the calcined support. However, this would add additional steps to the process which is particularly disfavoured at commercial production scales. There is therefore a need for a method which can produce a Ru / ZrC>2 catalyst in which the acidity of the ZrC>2 support can be controlled independently of the BET surface area and median pore size, without adding additional steps to the catalyst manufacturing process.

[0018] Summary of the Invention

[0019] The present inventors have found that this problem can be solved by impregnating a ZrC>2 support with a solution comprising ruthenium oxalate and sodium citrate. P102547W001 4

[0020] In a first aspect the invention relates to a method for manufacturing a catalyst comprising ruthenium on a zirconia support, comprising the steps of:

[0021] (i) preparing a co-impregnation solution comprising ruthenium oxalate and sodium citrate in water;

[0022] (ii) adding the co-impregnation solution to a zirconia support; and

[0023] (iii) drying the product of step (ii).

[0024] An important step in the process is the co-impregnation using a solution prepared from ruthenium oxalate and sodium citrate. Ruthenium chloride is widely used in the preparation of ruthenium catalysts, but a problem with its use is that HCI is generated during a subsequent activation step with H2. This either requires steps to be taken to abate HCI generated during activation of the catalyst (if using H2 as reducing agent), or to carry out reduction in the aqueous phase e.g. with aqueous hydrazine. These problems are avoided when using ruthenium oxalate because this salt is free of chloride residues.

[0025] The role of the sodium ions is to reduce the acidity of the support. Combining ruthenium oxalate and sodium ions in aqueous solution has the potential to form sodium oxalate. Sodium oxalate has a solubility at room temperature which is approximately 35 g / L at room temperature (see the article “A Review of Sodium Oxalate Solubility in Water” Ind. Eng. Chem. Res. 2023, 62, 19394-19401) and therefore has the potential to precipitate out at commercial production scales where high concentrations are preferred. The inventors trialled a variety of different sodium salts and found that they were incompatible with ruthenium oxalate under the conditions tested. However, it was surprisingly found that sodium citrate and ruthenium oxalate could co-exist in solution across a variety of concentrations without forming a precipitate.

[0026] In a second aspect the invention relates to an oxidic catalyst obtained or obtainable by a method according to the first aspect.

[0027] An additional step (iv) of reducing the product of step (iii) in a ^-containing atmosphere at a temperature of at least 450 °C may be carried out in order to produce a reduced catalyst. In a third aspect the invention relates to a reduced catalyst obtained or obtainable by a method according to the first aspect. P102547W001 5

[0028] In a fourth aspect the invention relates to a catalyst comprising ruthenium on a zirconia support, wherein: the catalyst comprises 0.1-5 wt% ruthenium based on the total weight of the catalyst; the catalyst comprises 0.1 -2.0 wt% sodium based on the total weight of the catalyst.

[0029] The catalyst according to the fourth aspect may be an oxidic catalyst or, if a reduction step is carried out, a reduced catalyst. If the catalyst is reduced in a (^-containing atmosphere at a temperature of at least 450 °C then the resulting reduced catalyst has a ruthenium particle size of 20-30 nm. of the Figures

[0030] Figure 1 shows the impact of calcination temperature on support acidity.

[0031] Figure 2 shows the impact of calcination temperature on support BET surface area.

[0032] Figure 3 shows the impact of calcination temperature on support median pore diameter.

[0033] Figure 4 shows the correlation between the BET surface area and acidity of a range of in house and commercial zirconia supports.

[0034] Figure 5 shows the correlation between the BET surface area and acidity of a range of in house and commercial zirconia supports, now also including sodium-doped supports.

[0035] Figure 6 shows the correlation between ruthenium loading and metal area for samples prepared using a ZrC>2 support which had been calcined at 400 °C.

[0036] Figure 7 shows the correlation between the support median pore diameter and the ruthenium loading obtained after reducing the Ru / ZrC>2 catalyst in H2 / N2 at 280 °C.

[0037] Detailed description of the Invention

[0038] Sub-headings are included for convenience only and are not intended to limit the disclosure. P102547W001 6

[0039] Manufacturing method

[0040] Step (i)

[0041] In step (i) a co-impregnation solution is prepared comprising ruthenium oxalate and sodium citrate in water. The co-impregnation solution can be prepared by dissolving ruthenium oxalate and sodium citrate in water in any sequence; the order in which the salts are dissolved has no impact on the stability of the resulting co-impregnation solution. The ruthenium oxalate and sodium oxalate may be added as solids or as a solution. Ruthenium oxalate is normally commercially available in the form of a solution rather than as a solid, and therefore in practice the sodium citrate (either solid or as a solution) will normally be added to a solution of ruthenium oxalate.

[0042] The sodium citrate may be monosodium citrate (NaCeHyO?), disodium citrate (ix^CeHeO?), trisodium citrate (NasCeHsO?) or a mixture of two or more thereof. It is preferred that the sodium citrate is trisodium citrate because, for a given sodium loading, the use of trisodium citrate introduces fewer citrate residues needing to be destroyed by the reduction step.

[0043] It is preferred that no salts other that sodium citrate and ruthenium oxalate are used to prepare the co-impregnation solution. A benefit of this is that the resulting catalysts are chloride-free and can therefore be reduced in a ^-containing atmosphere without concerns over producing HCI.

[0044] The volume of water used should be large enough to ensure that the sodium citrate and ruthenium oxalate are fully dissolved. The volume of water used is preferably at least 90% of the absorption volume of the support in order to achieve good homogeneity, preferably at least 95%. The volume of water used should preferably be no more than 110% of the absorption volume of the support, in order to avoid losing ruthenium. Preferred volumes are therefore 90-110%, preferably 95-110%, in each case relative to the absorption volume of the support. These volumes are particularly preferred when an incipient wetness impregnation method is carried out. The absorption volume can be measured in advance by submerging a sample of the support in water, leaving the support submerged for a duration which is equal to intended contact time between the co-impregnation solution and P102547W001 7 the support in step (ii), decanting the excess solution, and then calculating the mass increase.

[0045] If the co-impregnation solution is dilute then the ruthenium oxalate and sodium citrate may dissolve at room temperature. However, particularly when the co-impregnation is relatively concentrated, it is preferred that the co-impregnation solution is heated to ensure that the ruthenium oxalate and sodium citrate are fully dissolved. Heating to a temperature of 50-90 °C is preferred, more preferably 50-60 °C. Heating is preferably continued until the coimpregnation solution is added to the support in step (ii).

[0046] Step (ii)

[0047] In step (ii) the co-impregnation solution is added to a zirconia support. The zirconia support preferably has a BET surface area of at least 20 m2 / g. Whilst there is no particular maximum in the BET surface area, a practical maximum is 100 m2 / g because this is a typical BET surface area achieved by calcining zirconia at a “low” temperature of 400 °C (see Examples). A preferred range is 20-100 m2 / g, which encompasses most commercially available zirconias.

[0048] The shape and dimensions of the support will depend on many factors including the required geometric surface area of the catalyst, the required crush strength, and the acceptable pressure drop. It is preferred that the support has a spherical, cylindrical, or multi-lobed shape. Supports having a trilobe shape are preferred because they provide a good balance between high geometric surface area and low pressure drop.

[0049] The co-impregnation solution is preferably added to the support by incipient wetness impregnation because of the simplicity, low cost and low amounts of waste associate with this technique.

[0050] Step (Hi)

[0051] In step (iii) the product of step (ii) is dried. The role of this step is to remove water from the pores of the support. Typical and preferred drying conditions are a temperature of 95-115 °C. Drying is normally carried out in air because this is most convenient practically. The P102547W001 8 resulting material formed after step (iii) may be described as an oxidic catalyst because, whilst some ruthenium (III) may be reduced to ruthenium (0) during decomposition of ruthenium oxalate, some ruthenium is normally oxidised back to ruthenium (III) or ruthenium (IV) during the drying step.

[0052] Step (iv) (optional)

[0053] Step (iv) is an optional step in which the product of step (iv) is reduced in a ^-containing atmosphere at a temperature of at least 450 °C. Using a ^-containing atmosphere for the reduction rather than a solution-based reduction is possible because of the fact that ruthenium oxalate is used as the ruthenium source, thereby avoiding the incorporation of chloride residues. Step (iv) is preferably carried out using a mixture of H2 in an inert gas, preferably a mixture of H2 / N2, preferably 0.5-10 vol% H2 with balance N2.

[0054] In order to ensure that both the citrate and oxalate residues are fully removed, it is necessary to ensure that the reduction is carried out at a temperature of at least 450 °C. A temperature of 450-600 °C is preferred. As will be more fully discussed under the “Examples” heading, the inventors have surprisingly found that under these conditions the reduced catalyst has a ruthenium particle size of 20-30 nm, more typically approximately 25 nm, irrespective of the median pore size of the support. The reasons for this are not fully understood, but it may indicate that the citrate is acting as a template during the reduction step. Ruthenium particle sizes are calculated from the ruthenium content measured by inductively coupled plasma (ICP) spectroscopy, and the metal area as measured by the procedure detailed in the examples section, via the equation ICP Ru content (wt. %) x 6)

[0055] Ru particle size size nm) = 10,000 x 1237 x Metal area 0 2 / #))

[0056] The above equation is derived from equation (12.42) in Characterisation of Porous Solids and Powders: Surface Area, Pore Size and Density Chapter 12, Chemisorption, Site Specific Gas Adsorption page 229 (ISBN 978-90-481-6633-6).

[0057] Catalysts P102547W001 9

[0058] The invention also relates to a catalyst as defined in the fourth aspect of the invention. The catalysts may be oxidic or in reduced form.

[0059] The catalyst comprises 0.1-5 wt% ruthenium based on the total weight of the catalyst. The content of ruthenium is measured by ICP spectroscopy. Loadings of ruthenium toward the higher end of this range may require steps (ii) and (iii) to be carried out multiple times to achieve the desired loading. A ruthenium loading of 0.1-2 wt% is preferred because it covers most practical interest and can normally be achieved through a single impregnation step.

[0060] The catalyst comprises 0.1-2 wt% sodium based on the total weight of the catalyst. The content of sodium is measured by ICP spectroscopy. A sodium content towards the lower end of this range may be appropriate where a support with a relatively high acidity is desired whereas a sodium content towards the higher end of this range may be appropriate where a support with a very low acidity is desired (i.e. acidity < 0.25 mLNH3 / g) . In practice a loading of 1.5 wt% sodium is normally enough to ensure that the support has very low acidity, without introducing excessive amounts of citrate residues, and is therefore preferred. A loading of 0.3-1.5 wt% sodium is particularly preferred.

[0061] The presence of additional metals besides ruthenium, sodium and zirconium could adversely impact the performance of the catalyst. It is therefore preferred that the content of any metals other than ruthenium, sodium or zirconium is < 0.1 wt.%, preferably < 0.05 wt.%. The content of each metal can be determined by ICP spectroscopy.

[0062] The zirconia support may have a spherical, cylindrical, or multi-lobed shape. Supports having a trilobe shape are preferred because they provide a good balance between high geometric surface area and low pressure drop.

[0063] If the oxidic catalyst formed in step (iii) is subjected to a reduction step (iv) at a temperature of 450 °C or more, the citrate and oxalate residues are removed and the oxidic ruthenium is reduced to particles of Ru(0). Catalysts prepared in this way have a ruthenium particle size of 20-30 nm, typically approximately 25 nm.

[0064] Examples P102547W001 10

[0065] Procedure for measuring acidity

[0066] The acidity was measured by ammonia temperature-programmed desorption (NH3-TPD). The particulate was ground to a powder and charged to a Micromeritics AutoChem 2950HP thermal analyser. A 40 mL / min flow of helium was passed over the particulate at atmospheric pressure while heating to 400 °C at a rate of 20°C / min. A temperature of 400 °C was maintained for 10 minutes before cooling to 120°C. Once at 120°C a 40 mL / min flow of helium and a 40 mL / min flow of 5%v / v ammonia in helium was passed over the particulate for 30 minutes. The physisorbed ammonia was purged using a 40 mL / min flow of helium for 30 minutes. The particulate temperature was increased to 700°C at a rate of 5°C / min and held for 30 minutes, during this time a TCD detector was used to monitor the desorption of ammonia from the particulate. The total amount of ammonia desorbed was calculated by integrating the desorption profile of the first peak in the desorption profile (approximately 125-400 °C), which had been calibrated using known quantities of ammonia in helium. This value was then normalised by the mass of sample to obtain the acidity in mLNH3 / g.

[0067] Procedure for measuring metal area

[0068] Ruthenium metal areas were measured on a Micromeritics™ HTP 6 Station Chemisorption Analyser. For each analysis approximately 0.5g to 1g of sample was charged to a chemisorption tube and a chemisorption run initiated using the following conditions. The samples are reduced with 100% hydrogen at 400°C for 180 minutes with a hydrogen flow rate of 200 SCCM. After the reduction stage is finished all the hydrogen is removed by raising the temperature to 420°C and evacuating for 720 minutes. The sample is then cooled under vacuum to 35 °C and evacuation continued for 30 minutes after achieving a vacuum of <10 umHg. A leak test is carried out prior to analysis. The leak rate should be less than 5 umHg / minute. Any higher than this and the integrity of the sample may be compromised, and lower surface areas may ensue due to re-oxidation of the ruthenium by air leaking into the system. At the analysis temperature of 35 °C the sample is dosed with 100% hydrogen over a range of pressures between 100 and 760 mmHg. At each pressure the chemisorbing hydrogen is allowed to equilibrate, and the volume of hydrogen uptake is measured and recorded automatically. Pressure I uptake pairs constitute a chemisorption isotherm. At the end of the analysis the sample is discharged, and the reduced weight of sample is recorded. P102547W001 11

[0069] The analysis entails the measurement of two isotherms. The first is a measure of the “total” hydrogen taken up by the sample, which includes both chemisorbed and physisorbed hydrogen. The sample is then evacuated to remove the physisorbed (“weak”) component and the isotherm repeated to quantify the amount of physisorbed hydrogen that then readsorbs to the catalyst. The instrument software subtracts the “weak” isotherm from the “total” isotherm to yield the “strong” isotherm. After identifying a plateau region on both isotherms, the data is extrapolated back to zero pressure and the values used to calculate the Ru metal areas. These are referred to as Ru area (Otot) and Ru area (Ostr) respectively (where tot = total and str = strong). The reduced weight is used to express the Ru areas in m2 / g of reduced catalyst.

[0070] Impact of calcination temperature on ZrC properties

[0071] A series of ZrC>2 supports were prepared by extruding a mixture of ZrC>2 powder (RC-100 from Daiichi Kigenso Kagaku Kogyo Co. Ltd.) and a binder to the required shape (trilobe with a diameter of 1.4 or 2.5 mm), drying the resulting extrudates and then calcining the resulting extrudates at temperatures ranging from 400 °C to 900 °C for a duration of 4h in each case. In the following the terminology “ZrC>2-TEMP” is used; for example a ZrC>2 support calcined at 500 °C is referred to as ZrC>2-500. The acidity, BET surface area and median pore diameter is reported in Table 1 and the impact of calcination temperature on these parameters is illustrated in Figures 1 , 2 and 3 respectively. As the calcination temperature was increased from 400 °C to 900 °C there was an approximately linear decrease in both acidity (see Figure 1) and BET surface area (see Figure 2). The median pore diameter was relatively constant up to 600 °C, then increased above 600 °C (see Figure 3). P102547W001 12

[0072] Table 1.

[0073] Comparison against commercial supports

[0074] Commercial ZrC>2 supports from Zircomet, SGN and Daiichi Kigenso Kagaku Kogyo Co. Ltd. (DKKK) were analysed for their acidity, BET surface area and median pore diameter. The results are included in Table 2 together with values for the zirconia supports described above; the results are ordered by increasing BET surface area. The correlation between BET surface area and acidity is shown in Figure 4. All of the supports showed a clear correlation between BET surface area and acidity, namely: Acidity (mLNH3 / g) > 0.015 x (BET surface area (m2 / g)).

[0075] P102547W001 13

[0076] Table 2.

[0077] Impact of doping with sodium

[0078] The ZrC>2-400 and ZrC>2-600 supports were impregnating using a solution of sodium citrate having a volume equal to the absorption volume of the support, followed by drying at 105 °C, then calcining in air at 350 °C in order to remove any carbon associated with citrate decomposition. A calcination temperature of 350 °C was chosen because it was lower than the calcination temperature of any of the supports and therefore should not change the intrinsic acidity of the support. The acidity is reported in Table 3. Figure 5 shows the correlation between BET surface area and acidity, now including the sodium doped supports. It was assumed that the BET surface area and median pore diameter were not impacted by sodium doping. By modifying the support by sodium doping it was possible to lower the acidity and thereby access supports satisfying the condition: Acidity (ml_NH3 / g) < 0.015 x (BET surface area (m2 / g)). P102547W001 14

[0079] Table 3.

[0080] Impregnation with ruthenium oxalate

[0081] Approximately 50 samples were prepared by impregnating zirconia supports ZrC>2-400 to ZrC>2-900 with a solution of ruthenium oxalate, drying and then reducing the resulting catalyst in a stream of 2vol%H2 / 98vol%N2. The temperature of the sample was controlled by means of a furnace. Parameters which were varied included calcination temperature of the zirconia support, diameter of the support, ruthenium loading and temperature during reduction.

[0082] A first surprising finding was that there was an approximately linear relationship between ruthenium loading and metal area. This trend is illustrated in Figure 6 for the examples in Table 4 which were all prepared using a ZrO2 support which had been calcined to 400 °C. The fact that metal area increased linearly with metal loading implies that the particle size remained constant. The relationship also held for ZrC>2 supports calcined at other calcination temperatures. P102547W001 15

[0083] Table 4.

[0084] A second surprising finding was that, at constant ruthenium loading, there was a close correlation between the median pore diameter of the ZrC>2 support and Ru particle size. Ru particle size was determined from metal area by equation:

[0085] (ICP Ru content (wt. %) x 6)

[0086] Ru particle size size (nm) = 10,000 x 1237 x (Metal area 0 2 / #))

[0087] This correlation is illustrated in Figure 7 for the examples in Table 5.

[0088] Table 5.

[0089] 1Measured by ICP spectroscopy P102547W001 16

[0090] Co-impregnation with ruthenium oxalate and sodium citrate

[0091] A series of catalysts were prepared by co-impregnating various zirconia supports with a solution of ruthenium oxalate and sodium citrate, drying and then reducing the resulting catalyst in a ^-containing atmosphere. The ruthenium loading was targeted to be approximately 0.9 wt% so as to allow direct comparison with the results in Table 5 and Figure 7. The loading of sodium was approximately 1 .5 wt% in each example. The samples were then reduced using 2vol.%H2-95vol.%N2. A higher reduction temperature of 540 °C was used because it was found that carrying out reduction at 280 °C was not high enough to decompose the residual citrate. The metal area and Ru particle size of the resulting catalysts are reported in Table 6. Surprisingly, the correlation between Ru particle size and median pore size no longer held and a Ru particle size of approximately 25 nm was observed in each case. The reasons for this are not yet fully understood, but the fact that in the case of catalyst 0.92%Ru-(2.5mm-Zr02-400)-540 the Ru particle size was larger than the median pore diameter shows that the Ru particles are mobile and are able to migrate to the larger pores. The fact that the Ru particle size was relatively constant might be due to citrate acting as a template during the reduction.

[0092] Table 6.

[0093] 2Measured by ICP spectroscopy

Claims

P102547W001 17Claims1. A method for manufacturing a catalyst comprising ruthenium on a zirconia support, comprising the steps of:(i) preparing a co-impregnation solution comprising ruthenium oxalate and sodium citrate in water;(ii) adding the co-impregnation solution to a zirconia support; and(iii) drying the product of step (ii).

2. A method according to claim 1 , wherein the zirconia support used in step (ii) has a BET surface area of at least 20 m2 / g.

3. A method according to claim 1 , wherein the zirconia support used in step (ii) has a BET surface area of 20-100 m2 / g.

4. A method according to any of claims 1 to 3, wherein the zirconia support has a spherical, cylindrical, or multi-lobed shape.

5. A method according to claim 4, wherein the zirconia support has a trilobe shape.

6. A method according to any of claims 1 to 5, wherein the sodium citrate used in step (i) is trisodium citrate (NasCeHsO?).

7. A method according to any of claims 1 to 6, wherein the co-impregnation solution is heated to ensure that the ruthenium oxalate and sodium citrate are fully dissolved.

8. A method according to claim 7, wherein the co-impregnation solution is heated to a temperature of 50-90 °C.

9. A method according to claim 7, wherein the co-impregnation solution is heated to a temperature of 50-60 °C.

10. A method according to any of claims 1 to 9, wherein in step (ii) the co-impregnation solution is applied to the zirconia support by incipient wetness impregnation.P102547W001 1811. A method according to any of claims 1 to 10, wherein step (iii) is carried out at a temperature of 95-115 °C.

12. A method according to any of claims 1 to 11, comprising an additional step (iv) of reducing the product of step (iii) in a ^-containing atmosphere at a temperature of at least 450 °C.

13. A method according to claim 12, wherein step (iv) is carried out at a temperature of 450-600 °C.

14. An oxidic catalyst obtained or obtainable by a method according to any of claims 1 to 11.

15. A reduced catalyst obtained or obtainable by a method according to claim 12 or claim 13.

16. A catalyst comprising ruthenium on a zirconia support, wherein: the catalyst comprises 0.1-5 wt% ruthenium based on the total weight of the catalyst; and the catalyst comprises 0.1 -2.0 wt% sodium based on the total weight of the catalyst.

17. A catalyst according to claim 16, wherein the catalyst comprises 0.1 -2.0 wt% ruthenium based on the total weight of the catalyst.

18. A catalyst according to claim 16 or claim 17, wherein the catalyst comprises 0.1- 1.5 wt% sodium based on the total weight of the catalyst.

19. A catalyst according to claim 18, wherein the catalyst comprises 0.3-1.5 wt% sodium based on the total weight of the catalyst.

20. A catalyst according to any of claims 16 to 19, wherein the zirconia support has a spherical, cylindrical, or multi-lobed shape.

21. A catalyst according to claim 20, wherein the zirconia support has a trilobe shape.P102547W001 1922. A catalyst according to any of claims 16 to 21 , wherein the content of any metals other than ruthenium, sodium or zirconium is < 0.1 wt.%.

23. A catalyst according to any of claims 16 to 22, wherein the catalyst is a reduced catalyst having a ruthenium particle size of 20-30 nm when calculated by the equation(ICP Ru content (wt. %) x 6) Ru particle size size nm) = 10,000 x - - - - -r-1237 x Metal area (m2 / g))

Images