Interesterification catalyst and process

Abstract

A process for the production of an ester product from a mixture of at least two different ester compounds includes the steps of mixing together at least two different starting ester compounds to form a first ester mixture; and contacting the first ester mixture with a catalyst including from 30-60% of calcium oxide and at least one second metal oxide at a temperature of at least 180° C., for a duration of at least one hour, with mixing, to form a second ester mixture having a melting point which is lower than the melting point of the first ester mixture.

Claims

1. An interesterification process for the production of an ester product from a mixture of at least two different ester compounds comprises the steps of: a) mixing together at least two different starting ester compounds to form a first ester mixture; and b) contacting said first ester mixture with a catalyst comprising from 30-60 weight % of calcium oxide and at least one second metal oxide such that an interesterification reaction takes place to form a second ester mixture having a melting point which is lower than the melting point of said first ester mixture, said second metal oxide is selected from the group consisting of an oxide of a Group 2A metal other than calcium, an oxide of a transition metal, lanthana, silica, alumina and a metal aluminate.

2. The interesterification process as claimed in claim 1, wherein at least one of said at least two different starting ester compounds is a triglyceride.

3. The interesterification process as claimed in claim 1, wherein at least one of the starting ester compounds comprises an ester of a carboxylic acid containing from 12 to 24 carbon atoms.

4. The interesterification process as claimed in claim 1, wherein the catalyst is pre-dispersed in at least one of said starting ester compounds.

5. The interesterification process as claimed in claim 1, wherein the second metal oxide comprises magnesium oxide.

6. The interesterification process as claimed in claim 1, wherein the catalyst further comprises from 1-5 weight % of an alkali metal.

7. The interesterification process as claimed in claim 1, wherein the catalyst has a surface area less than 20 m.sup.2/g.

8. The interesterification process as claimed in claim 1, wherein the interesterification takes place at a temperature between 0 and 300° C.

9. The interesterification process as claimed in claim 1, wherein the catalyst is separated from the second ester mixture and added to a first ester mixture for use in a subsequent process.

10. The interesterification process according to claim 1 wherein said catalyst comprises from 30-60 weight % of calcium oxide, from 40-70 weight % of magnesium oxide and from 0 to 5 weight % of sodium or potassium.

11. The interesterification process as claimed in claim 2, wherein at least one of the starting ester compounds comprises an ester of a carboxylic acid containing from 12 to 24 carbon atoms.

12. The interesterification process as claimed in claim 2, wherein the catalyst is pre-dispersed in at least one of said starting ester compounds.

Description

EXAMPLE 1

(1) 17.70 g of Ca(NO.sub.3).sub.2.4H.sub.2O and 18.81 g of Mn(NO.sub.3).sub.2.4H.sub.2O were dissolved in 273 mL of demineralised H.sub.2O. In addition, 28.6 g of Na.sub.2CO.sub.3 (anhydrous) was dissolved in 337 mL of demineralised H.sub.2O, and this solution was heated to 60° C. while stirring at 600 rpm. The solution containing the calcium and manganese salts was then added to the Na.sub.2CO.sub.3 solution in a drop-wise manner, which lead to the immediate formation of a light brown-coloured precipitate. After the completion of metal salt addition the reaction was left stirring at 60° C. for 1 hour. The precipitate was then collected by filtration and re-dispersed in hot water. This procedure was performed in order to remove as much of the residual sodium from the sample as possible. The precipitate was re-collected and the washing step was repeated a further two times, before the precursor was dried overnight at 80° C.

(2) Thermogravimetric analysis (TGA) performed using a TA Instruments SDT2960 instrument showed a major mass loss at around 725° C., suggesting oxide formation at this temperature. Samples were subsequently calcined at 800° C. for 2 hours in air.

(3) X-ray diffraction (XRD) spectra, collected using a Bruker D8 Advance instrument with a Cu source, showed the calcined material consisted of phase-pure perovskite.

(4) Inductively coupled plasma atomic emission spectroscopy (ICP-AES), run on a Perkin Elmer Optima 3300 RL instrument, showed 28.4% Ca, 38.8% Mn and 1.8% Na. BET surface area data, collected on a Quantachrome Autosorb-1 instrument, gave a surface area of 2.3 m.sup.2/g.

(5) The analytical methods described in this Example were used to characterise the materials made in the subsequent examples, unless stated otherwise. All catalysts were stored in an argon atmosphere in a glove-box prior to use or characterisation.

EXAMPLE 2

(6) A material was prepared using the same procedures as in Example 1 except that 10.6 g Sr(NO.sub.3).sub.2 and 12.57 g of Mn(NO.sub.3).sub.2.4H.sub.2O were dissolved in 273 mL H.sub.2O and 19.1 g Na.sub.2CO.sub.3 was dissolved in 225 mL H.sub.2O. XRD showed the calcined material was comprised predominantly of the perovskite phase SrMnO.sub.3 with a small amount of SrCO.sub.3 present. ICP-AES data showed the material 43.4% Sr, 29.7% Mn and 1% Na. The BET surface area was 2.6 m.sup.2/g.

EXAMPLE 3

(7) 20.8 g calcium acetate monohydrate was dissolved in 131.5 mL H.sub.2O and this solution was added to 10 g magnesium oxide powder (Sigma Aldrich, surface area 72.2 m.sup.2/g). The resulting slurry was stirred and heated at 80° C. until dry and then further dried at 80° C. overnight. The TGA data showed several mass losses, with the loss in mass appearing to end after 650° C. The sample was calcined at 780° C. for 8 hours and XRD showed the material consisted of crystalline CaO and MgO. ICP-AES gave the expected amounts of Ca and Mg, 24.3 and 30.6% respectively. A BET surface area of 10.7 m.sup.2/g was recorded. The overall loading of CaO was equal to 40 wt % (wt CaO/(wt CaO+wt MgO)).

EXAMPLE 4

(8) 11.8 g of Ca(NO.sub.3).sub.2.4H.sub.2O and 26.93 g of Mg(NO.sub.3).sub.2.6H.sub.2O were dissolved in 1 L of H.sub.2O. A second solution contained 29.53 g of Na.sub.2CO.sub.3 dissolved in 1 L H.sub.2O. The concentrations of the solutions were therefore 0.1550 M (metal) and 0.2786 M (base). Both solutions were heated to 60° C. while being magnetically stirred and then the solutions were pumped into a flash precipitation reactor at 20 mL/min, with stirring inside the reactor set to 2000 rpm. The resulting mixture was collected and kept at room temperature for 1 day. The precipitate was collected by filtration, re-slurried, washed with warm water and filtered again. This washing procedure was performed a further 3 times before the precipitate was dried at 80° C. overnight. The dried powder was calcined in air at 800° C. for 2 hours inside a tube furnace and then cooled to room temperature under a flow of argon to avoid the formation of bulk carbonate and the adsorption of CO.sub.2 to the catalyst surface.

(9) XRD showed a phase-pure CaO—MgO material with no CaCO.sub.3. ICP-AES analysis showed 34.2% Ca, 31.7% Mg and 0.4% Na. The BET surface area was 12.6 m.sup.2/g.

EXAMPLE 5

(10) A material was prepared in the same manner as and with the same quantities as described in Example 4, however the calcium and magnesium nitrate salt solution was added drop-wise into a magnetically stirred sodium carbonate solution. Upon completion of addition the precipitate was aged for 1 hour at 60° C. and then filtered, re-slurried in warm water and re-filtered. The filtration process was repeated a further 3 times, after which the precipitate was then aged at 80° C. overnight. Calcination was performed in a tube furnace under flow of air at 800° C. for 2 hours with the sample then cooled to room temperature under Ar. ICP-AES analysis showed 33.6% Ca, 30.8% Mg and 0.6% Na. The BET surface area was 12.7 m.sup.2/g.

EXAMPLES 6-8

(11) Materials were prepared using the method described in Example 5 using different amounts of calcium nitrate and sodium carbonate while maintaining the same overall metal concentration. Materials containing 30 wt % CaO, 50 wt % CaO and 60 wt % CaO were prepared by this method which had measured surface areas of 54.3 m.sup.2/g, 7.9 m.sup.2/g and 6.7 m.sup.2/g respectively.

EXAMPLE 9

(12) A material containing 40 wt % CaO and MgO was also prepared as described in Example 5, by freeze drying instead of standard drying. A surface area of 49.5 m.sup.2/g was measured using the BET method.

EXAMPLE 10-11

(13) Materials containing 40 wt % and 50 wt % CaO, respectively, were prepared using the general method described in Example 5 except that the solids concentration was increased four times by doubling the required amounts of calcium nitrate, magnesium nitrate and sodium carbonate and halving the amount of water. Surface areas of 18.6 m.sup.2/g (40 wt % CaO) and 9.0 m.sup.2/g (50 wt % CaO) were recorded using the BET method.

EXAMPLE 12

(14) Calcium carbonate was deposited by precipitation onto a pre-formed MgO support as follows: 10 g of MgO (heavy, BDH Chemicals) was dispersed in 192 mL of H.sub.2O, to which 22.66 g of Na.sub.2CO.sub.3 was added. This dispersion was heated to 60° C. A solution was made consisting of 28.07 g of calcium nitrate tetrahydrate in 192 mL of H.sub.2O and this solution was added to the MgO dispersion in a dropwise manner with stirring. Formation of a white precipitate was observed on addition, and this was left stirring at 60° C. for 1 hour. The solids were collected by vacuum filtration and washed with warm water (250 mL) a total of 4 times by re-slurrying and filtering. The washed solids were then dried overnight at 80° C. and calcined in air (800° C. for 2 hours, 10° C./min ramp rate) with cooling under a flow of Ar. A surface area of 4.8 m.sup.2/g was recorded using the BET method, while ICP-AES showed 29.5% Ca, 33.9% Mg and 0.69% Na.

EXAMPLES 13-18

(15) Materials having the same nominal 40 wt % CaO loading were prepared using, instead of the heavy MgO solid material, MgO (light, from Alfa Aesar)), LiAlO.sub.2 (supplied by Alfa Aesar), ZrO.sub.2, SiO.sub.2 (P432), Al.sub.2O.sub.3 (PURALOX™ HP14/150) and La.sub.2O.sub.3 (prepared by precipitating lanthanum carbonate precursor from lanthanum nitrate using sodium carbonate, followed by calcination).

EXAMPLE 19

(16) A material was prepared by mixing 11.9 g of CaCO.sub.3 and 10 g of MgO (light, Alfa Aesar) in 500 mL H.sub.2O (solid concentration of 0.044 g/mL). The mixture of suspended particles was then spray dried using a Buchi B-290 Mini Spray Drier at a rate of 15 mL/min with an inlet temperature of 180° C. and an air flow rate of 670 L/hour. The spray dried powder was calcined in air at 800° C. for 2 hours inside a tube furnace and then cooled to room temperature under a flow of argon.

EXAMPLES 20-23

(17) Materials containing 40% CaO and either 0.5, 1, 2 or 5 wt % sodium, respectively, were prepared by the method described in Example 19 by adding sodium carbonate to the calcium and magnesium mixture. 0.3842 g Na.sub.2CO.sub.3 was used to provide 1% sodium in the final material and other concentrations of sodium were made by modifying the amount of sodium carbonate added.

EXAMPLE 24

(18) A material containing 40% CaO and 2% sodium was prepared using the method of Example 22 but using a calcination temperature of 715° C. The surface area of the final catalyst was 19.5 m.sup.2/g.

EXAMPLES 25-29

(19) Materials containing 40% CaO and either 0, 0.5, 1, 2 or 5 wt % sodium, respectively, were prepared by the method described in Examples 19-23 but using a dispersion of LiAlO.sub.2 (Alfa Aesar) instead of MgO.

EXAMPLE 30

(20) A material containing 40% CaO and 2% sodium was prepared using the method of Example 22 with the addition of a wet-milling step prior to spray drying. The powders were dispersed in 50 mL H.sub.2O and milled using a Fritsch Pulverisette planetary ball mill with 10 mm ZrO.sub.2 beads at a 1:10 powder to beads mass ratio, at 400 rpm for 1 hour, with 10 minute pauses after every 15 minutes. The mixture was then made up to 500 mL with water and spray dried as in Example 19. A surface area of 9.8 m.sup.2/g was recorded for the calcined material.

EXAMPLE 31

(21) A material was prepared by spray drying an aqueous solution of calcium, magnesium and sodium acetates. 30 g magnesium acetate tetrahydrate, 11.82 g calcium acetate monohydrate and 1.11 g sodium acetate trihydrate were dissolved in 300 mL water. The resulting solution was spray dried at a rate of 9 mL/min with an inlet temperature of 130° C. and an air flow rate of 670 L/hour. The powder was then calcined in a tube furnace in air at 800° C. for 2 hours. The furnace was ramped at 10° C./min to 300° C., then at 5° C./min to 400° C. and again at 10° C./min to 800° C., before being allowed to cool under a flow of argon. A BET surface area of 6.5 m.sup.2/g was recorded.

COMPARATIVE EXAMPLE A

(22) An interesterification catalyst comprising K.sub.2CO.sub.3 and MgO was prepared. 5 g of K.sub.2CO.sub.3 was dissolved in 17 mL H.sub.2O and added to 19.84 g of MgO. The catalyst was dried overnight at room temperature and then at 110° C. for 16 hours. Calcination was performed at 500° C. for 2 hours in air. This procedure is intended to produce a catalyst as described in Example 3 of U.S. Pat. No. 6,072,064 for the purpose of comparison.

EXAMPLE 32: INTERESTERIFICATION REACTION

(23) The materials made as described in Examples 1-31 were tested as interesterification catalysts. A slurry-phase interesterification reaction was performed using the powdered catalyst under the following conditions:

(24) Raw materials: soybean oil and palm stearin at a weight ratio 4:1, (12.5 g of soybean oil and 3.125 g palm stearin).

(25) Amount of catalyst: 10 wt %, (1.57 g)

(26) Reaction temperature: set temperature 225° C., giving an oil temperature of 205° C.

(27) Reaction time: 5 hours

(28) Pressure: atmospheric pressure

(29) Stirring rate: set to 600 rpm

(30) Atmosphere: reactions performed under a flow of argon gas, to avoid oxidation of the oils. Samples of the oil were taken at 1 hour intervals and these were analysed by differential scanning calorimetry (DSC) using a Mettler Toledo DSC822e instrument from −60° C. to 60° C. The activity of the catalyst was assessed by determining the partial areas under the DSC peak(s) and, by assuming the oil was 100% solid at −60° C., and 0% solid at 60° C., it was possible to determine the percentage of oil that was solid at a particular temperature T. The melting point was considered to be the temperature at which the oil was 5% solid and 95% liquid. The activity of each catalyst may then be compared by comparing the melting point of the oil mixture following reaction. The starting soybean oil and palm stearin mixture has a melting point of 51.5° C. measured by this method and this drops to 30.9° C. following interesterification using sodium methoxide. The results are shown in Table 1.

COMPARATIVE EXAMPLE B

(31) Sulphated tin oxide (SnO.sub.2—SO.sub.4.sup.2−) was prepared as according to the method described by Jitputti et al (J. Jitputti, B. Kitiyanan, P. Rangsunvigit, K. Bunyakiat, A. Attanatho, P. Jenvanitpanjakul, Chem. Eng. J., 2006, 116, 61). This reference describes the promising activity of this material for the transesterification of triglycerides with methanol. The sample was tested for the interesterification of soybean oil and palm stearin as described in Example 32. No change was found in the melting properties of the oils after the reaction indicating that Comparative Example B showed no activity as an interesterification catalyst.

COMPARATIVE EXAMPLE C

(32) Sulphated zirconium oxide (ZrO.sub.2—SO.sub.4.sup.2−) was prepared as according to the method described by Jitputti et al (J. Jitputti, B. Kitiyanan, P. Rangsunvigit, K. Bunyakiat, A. Attanatho, P. Jenvanitpanjakul, Chem. Eng. J., 2006, 116, 61). The samples was tested for the interesterification of soybean oil and palm stearin as described in Example 32. No change was found in the melting properties of the oils after the reaction indicating that Comparative Example C showed no activity as an interesterification catalyst.

COMPARATIVE EXAMPLE D

(33) According to the publications of S. Yan et al, (S. Yan, S. O. Salley, K. Y. S. Ng, Appl. Catal. A, 2009, 353, 203), ZnO—La.sub.2O.sub.3 is a highly active catalyst for the transesterification of triglycerides and methanol to fatty acid methyl esters and glycerol. Samples of ‘ZnO—La.sub.2O.sub.3’ (shown by XRD in both the referenced publication and our own work to be comprised of a mixture of lanthanum phases, primarily La.sub.2O.sub.2CO.sub.3) at Zn/La ratios of 3:1 and 6:1 were prepared according to the method described by the authors. The samples were tested in the interesterification process according to Example 32 and showed no activity, with the melting point recorded at 51.1° C. before and after the reaction.

COMPARATIVE EXAMPLE E

(34) Georgogianni et al. (K. G. Georgogianni, A. P. Katsoulidis, P. J. Pomonis, M. G. Kontominas, Fuel Process. Technol., 2009, 90, 671) describes several mixed oxide systems for the transesterification of soybean oil with methanol. The Mg—Al hydrotalcite (Mg/Al ratio of 3:1) was found to be the most basic and most active of their materials, achieving a 96% conversion of soybean oil to fatty acid methyl esters under their reaction conditions. We prepared the Mg—Al hydrotalcite catalyst according to the description in the referenced journal article at the Mg/Al ratio of 3:1. The catalyst showed no activity for the interesterification process described by Example 32, with the oil melting point remaining unchanged at 51.5° C. after the reaction.

(35) Comparison Examples B-E show that catalysts which have previously been demonstrated to be active for transesterification may not be useful interesterification catalysts.

EXAMPLE 33

(36) The amount of catalyst metal leaching from the catalyst into the oil during interesterification was determined by analysing the product oil by ICP-AES following removal of the catalyst after the reaction. The results are shown in Table 2.

(37) TABLE-US-00001 TABLE 2 Ca Mn Na Mg (ppm) (ppm) (ppm) (ppm) Oil before interesterification 30 <10 <10 Oil after interesterification 14 133 19 Catalyst 1 Oil after interesterification 30 <10 <10 Catalyst 5

(38) This shows that the CaO—MgO catalyst has superior resistance to metal leaching compared with the CaMnO.sub.3 catalyst. This is a particular advantage because the presence of metal species in edible oils is generally undesirable.

(39) TABLE-US-00002 TABLE 1 Surface Melting Catalyst CaO Other area point Example Catalyst type (wt %) comments (m.sup.2/g) (° C.) 1 CaMnO.sub.3 28.4% Ca 2.3 30.9 2 SrMnO.sub.3 — 43.4% Sr 2.6 36.7 3 CaO—MgO 40 10.7 43.8 4 CaO—MgO 40 12.6 32.9 5 CaO—MgO 40 12.7 33.4 6 CaO—MgO 30 54.3 43.6 7 CaO—MgO 50 7.9 34.5 8 CaO—MgO 60 6.7 37.6 9 CaO—MgO 40 49.5 37.8 10 CaO—MgO 40 18.6 33.1 11 CaO—MgO 50 9.0 35.7 12 CaO—MgO 40 4.8 34.6 13 CaO—MgO 40 Light MgO 4.2 34.9 14 CaO—LiAlO.sub.2 40 15.9 34.7 15 CaO—ZrO.sub.2 40 10.1 35.3 16 CaO—SiO.sub.2 40 27.7 46.5 17 CaO—Al.sub.2O.sub.3 40 77.9 42.4 18 CaO—La.sub.2O.sub.3 40 3.7 38.0 19 CaO—MgO 40 7.9 41.3 20 CaO—MgO 40 0.5% Na 5.0 37.9 21 CaO—MgO 40 1.% Na 37.8 22 CaO—MgO 40 2% Na 2.6 33.2 23 CaO—MgO 40 5% Na 5.1 31.6 24 CaO—MgO 40 2% Na 19.5 32.1 25 CaO—LiAlO.sub.2 40 5.5 38.9 26 CaO—LiAlO.sub.2 40 0.5% Na 37.3 27 CaO—LiAlO.sub.2 40 1.% Na 36.6 28 CaO—LiAlO.sub.2 40 2% Na 2.4 37.7 29 CaO—LiAlO.sub.2 40 5% Na 1.9 36.6 30 CaO—MgO 40 2% Na 9.8 34.0 31 CaO—MgO 40 2% Na 6.5 32.9 Comp A K.sub.2CO.sub.3—MgO *2% catalyst 32.9 loading *Catalyst testing of Comparative Example A was performed as described except that the catalyst loading was 2 wt % and the reaction time was 1 hour.

EXAMPLE 34

(40) Catalyst samples were collected after use in the interesterification reaction and used in four further interesterification reactions, being separated from the oil mixture each time before use in a subsequent reaction. The melting points after 5 hours of reaction time for each use or “cycle” are shown in Table 3. The catalyst Comparison A was used at 2% catalyst loading and the reaction time was 1 hour, as described above.

(41) Examination of catalyst of Comparison Example A withdrawn from the first reaction cycle showed it to be associated with a waxy mixture of catalyst and metal soap material. The loss of metal to soap formation may explain the reduced activity of the catalyst in subsequent cycles. The catalyst of Example 5 also showed some evidence of soap formation after cycle 3. Soap formation was not observed in the catalyst of Example 22 until after cycle 5. Soaps were noticed in the Example 30 reactions during cycle 4 and the test was stopped for that reason.

(42) TABLE-US-00003 TABLE 3 MP MP MP Cycle 1 MP Cycle 2 Cycle 3 MP Cycle 4 Cycle 5 Catalyst (° C.) (° C.) (° C.) (° C.) (° C.) Example 5 33.4 32.0 30.0 Example 22 33.2 33.1 33.3 32.9 32.8 Example 30 34.8 34.3 34.9 32.6 Example 31 32.9 32.8 31.8 32.7 Example 37* 33.0 33.1 35.5 40.3 43.5 Comparison 32.9 35.7 38.5 42.3 46.2 Example A *Note: 90 minute reaction time

EXAMPLE 35 (COMPARISON)

(43) Calcium carbonate was precipitated from a solution of calcium nitrate using a solution of ammonium carbonate as the base solution. The solids were separated by filtration, washed and dried and calcined at 800° C. to convert them to calcium oxide. The calcium oxide solids were used as a catalyst in the interesterification reaction described in Example 32. After 5 hours the melting point of the reaction mixture was 47.1° C.

EXAMPLE 36 (COMPARISON)

(44) Particles of magnesium oxide, of the type used in the preparation of Example 12 were used as a catalyst in the interesterification reaction described in Example 32. After 5 hours the melting point of the reaction mixture was 48.2° C.

EXAMPLE 37

(45) A material was prepared by mixing 11.9 g of CaCO.sub.3 and 10 g of MgO (light, Alfa Aesar) and 1 g of K.sub.2CO.sub.3 in 500 mL H.sub.2O (solid concentration of 0.044 g/mL). The mixture was calculated to give the same no of moles of K in the mixture as the number of moles of Na in Example 22. The mixture of suspended particles was then spray dried using a Buchi B-290 Mini Spray Drier at a rate of 15 mL/min with an inlet temperature of 180° C. and an air flow rate of 670 L/hour. The spray dried powder was calcined in air at 800° C. for 2 hours inside a tube furnace and then cooled to room temperature under a flow of argon.

(46) The catalyst was tested in the interesterification reaction as described in Example 32, except that the reaction was run for only 90 minutes instead of for 5 hours. The catalyst was then collected and re-used in subsequent interesterification reactions (each 90 minutes) as described in Example 34. The melting point of the mixture after each cycle is shown in Table 3.

EXAMPLE 38

(47) The spray dried catalyst described by Example 22 was formed into pellets. The CaO—MgO powder formed as described in Example 22 was mixed in a Turbula® shaker-mixer with 1 wt % graphite as binder. The mixture was compressed using an Alexanderwerk® Roller Compactor WP120 hydraulic press. The compacted material was then ground and sieved to achieve a size distribution of 250-600 microns, and mixed again in the shaker mixer with 1 wt % graphite. The mixture was then compressed into cylindrical tablets of 3 mm diameter and 3.3 mm length using a Dott Bonapace® CPR-6 hydraulic press. The tablets had an average density of 2.35 g/cm.sup.3 and an average crush strength (applied along the length of the pellet) of 6.5 kgf.

(48) The catalyst pellets were tested in a Harshaw Falling Basket Catalyst Reactor using a high temperature bolted closure stirred batch reactor made by Autoclave Engineers. The test was performed as follows: 16.2 g (12 mL) of catalyst pellets were loaded evenly into four sample baskets. The autoclave was filled with the feedstock, comprising 360 g (400 mL) soybean oil and 90 g (100 mL) palm stearin. The system was held at 0.1 bar gauge pressure, which was the lowest measurable pressure for the system, with a flow of N.sub.2 maintained through the reactor for the duration of the test. The autoclave was heated to 100° C. to melt the oil, and the reactor was then stirred at 250 rpm. The autoclave was then heated to 210° C. at which point the catalyst basket was dropped into the oil and the reaction begun. The reaction was stopped after 24 hours and the melting point of the oil was determined by DSC to be 36.3° C. at an equivalent LHSV of 1.74 hr.sup.−1.