WO2008122790A2

WO2008122790A2 - Heteropolyacid catalysts and biodiesel manufacturing methods using such catalysts

Info

Publication number: WO2008122790A2
Application number: PCT/GB2008/001216
Authority: WO
Inventors: Ian Graham; Tony Larson; Adam Lee; Katabathini Narasimharao; Karen Wilson
Original assignee: University Of York
Priority date: 2007-04-05
Filing date: 2008-04-04
Publication date: 2008-10-16
Also published as: WO2008122790A3

Abstract

There is described a method of manufacturing biodiesel which comprises the esterif ication, transesterif ication or the substantially simultaneous esterif ication and transesterif ication of a feedstock comprising one or more vegetable oils, animal fats and/or mixtures thereof. The manufacturing method may include the use of a heteropolyacid as a catalyst. The preferred heteropolyacids comprise tungsten or molybdenum and a monovalent cation.

Description

CATALYSTS

This invention relates to a novel method of manufacturing and analysing biofuels, novel materials related thereto, their use as catalysts and methods of their preparation.

The depletion of world petroleum reserves and increased environmental concerns has stimulated the search for alternative renewable fuels that are capable of fulfilling an increasing energy demand. Biodiesel fuel (fatty acid methyl esters), synthesized from vegetable oils, has similar physical properties to petrochemical diesel and is considered as the best alternative fuel candidate for use in diesel engines. Biodiesel production involves the catalytic transesterification of long and branched chain triglycerides with alcohols to produce mono-esters and glycerol. Current syntheses employ homogeneous alkaline agents such as potassium or sodium alkoxides or hydroxides. However removal of the soluble base after reaction is a major problem since aqueous quenching results in the formation of stable emulsions and saponification, rendering separation and purification of the methyl ester difficult. As a result biodiesel production by these routes is still not cost-competitive with petrochemical diesel fuel.

Use of a solid base catalyst offers several process advantages including the elimination of a quenching step (and associated contaminated water waste) to isolate the products, and the opportunity to operate a continuous process. Solid bases including zeolites, alkali earth oxides and hydrotalcites have been investigated in transesterification reactions, and we recently reported tuneable solid bases (Li doped CaO and Mg-hydrotalcites) for tributyrin transesterification.

Vegetable oils also contain significant amounts of free fatty acids (even more if used cooking oil is employed) whose presence poses significant processing problems in standard biodiesel manufacture, since the free fatty acid is saponified by homogeneous alkali catalysts leading to a loss of catalyst and increased purification costs. The simplest approach to improving processing of such vegetable oils is to first esterify these free fatty acids to their alkyl esters, e.g. using an acid catalyst.

The development of solid acid catalysts for esterification pre-treatments is therefore crucial for improving the efficiency of biodiesel production. Though considerable literature exists on the esterification of simple aliphatic and aromatic acids by solid acid catalysts such as zeolites and resins, there are few reports on fatty acid esterification.

It has been reported by Rao, et al, in "Zirconium phosphate supported tungsten oxide solid catalysts for the esterification of palmitic acid" , Green Chem., 2006, 8, 790-797, zirconium phosphate supported WO_x solid acid catalysts with W loadings from 1 - 25 wt % show high activity in palmitic acid esterification with methanol.

The ability to perform simultaneous esterification and transesterification with a single catalyst would further simplify the biodiesel manufacturing process. While homogeneous Lewis acids based on carboxylic salts of Cd, Mn, Pb and Zn have shown promise, the only previous reports of solid catalysts capable of performing simultaneous esterification and transesterification require temperatures of 250 ⁰C.

Heteropolyacids (HPAs) are well known solid acids catalysts which are also active towards liquid phase esterification. However they are generally considered to be unsuitable for, e.g. biodiesel production in their native form due to high solubility in polar media. The number of accessible surface acidic sites in HPAs is low in apolar systems where they are insoluble, due to their inherently low surface areas (< 5 m g^"1). The most commonly used heteropolyacid, phosphotungstic acid H₃PW₁₂O₄₀ (HPW), exhibits strong acidity, however the use of pure HPW is not favourable due to their low surface area and high solubility in polar solvents. To overcome these limitations HPWs are usually supported on a carrier such as activated carbon, acidic ion-exchange resin, or SiO₂ or ZrO₂ powders. Supporting HPW on high area solids generally improves its catalytic performance. Literature also reveals that ionic interactions between the support surface and HPW clusters help to generate the active species during the impregnation step. However, because of the weak nature of this interaction it is possible for HPW clusters to leach from the support during reactions in polar solvents.

Okuhara, et al, in J. MoI. Catal. 74 (1992) 247, describe alkali-exchanged heteropoly acids, such as, Cs_xH(i_-X)P W₁₂O₄₀, which exhibit dramatic increases in surface area and profound changes in solubility over the parent HPA. For example, salts with large monovalent ions such as Cs⁺, NH₄ ⁺ and Ag⁺ are insoluble in water. Partial substitution of protons by Cs⁺ also changes the number of available surface acid sites. Izumi, et al, in Catalysis Today, 35 (1997) 183-188, describe ester hydrolysis catalysed by silica included heteropolyacid wherein the heteropolyacid is Cs_{2 5}H₀₅PW₁₂O₄₀ or H₃PW₁₂O₄₀

We have now surprisingly found a monovalent cation, e.g. caesium, doped heteropolyacid materials of the above general formula sin which the monovalent cation content is less than 2.5 are novel and have improved utility as, inter alia, catalysts in liquid phase chemical reactions and especially in the manufacture of biofuels, such as biodiesel.

In particular, we have synthesised a series of insoluble, high surface area Cs_xH_3- _XPW₁₂O_4O solid acids for esterification and transesterification with a view to their application in biofuel, e.g. biodiesel, synthesis.

The various properties of a biodiesel fuel are determined typically by its fatty acid ester content. In addition to the development of effective solid acid catalysts it is desirable to have methods that can efficiently analyse a composition after catalysis. Methods for determining fatty acid species (e.g. triacylglycerol) present in oil have been developed to monitor oil quality and purity. These include normal-phase HPLC. For example, ES2177375 describes the use of HPLC in the detection of triacylglycerol in vegetable oils, in particular olive oil. A further example is described in WO92/ 12421 that describes the analysis of mixtures of triacylglycerols by mass spectroscopy. More recently non-aqueous reverse-phase HPLC (NARP) methods using Cl 8 columns have been used in the analysis of triacylglycerol (Jandera et al. 2004, J. Chromatogr. A 1030:33-41; Holcapek et al. 2003, J. Chromatogr. A 1010:195-215). In our co-pending application WO2006/018621 is described a method for the analysis of plant derived as triacylglycerols that utilises a lipophilic polymer that allows the rapid resolution of triacylglycerols in an oil sample.

This disclosure relates to a process for the combined manufacture and analysis of fatty acid esters formed after the esterification and transesterification of an oil feed stock.

Thus, according to first aspect of the invention we provide a method of manufacturing biodiesel which comprises the esterification, transesterification or the substantially simultaneous esterification and transesterification of a feedstock comprising one or more vegetable oils, animal fats and/or mixtures thereof.

According to a further aspect of the invention there is provided a method for the manufacture and analysis of a biodiesel comprising the steps of: i) performing an esterification, transesterification reaction, or a substantially simultaneous esterification and transesterification reaction, of an oil feedstock comprising one or more vegetable oils, animal fats and/or mixtures thereof; ii) applying the esterified, transesterified feedstock in (i) to a column wherein said column comprises a separation material of a lipophilic polymer; and iii) separating the ester content of the feedstock.

Preferably the method of the invention includes the use of a solid acid as a catalyst in a liquid phase reaction. The solid acid may be a heteropolyacid and especially a heteropolyacid doped with a monovalent cation. The monovalent cation doped heteropolyacid is preferably a material general formula I:

in which M is an atom selected from P or Si; Y is NH₄ ⁺, Na⁺, K⁺ or Cs⁺; X is a metal atom selected from W or Mo; n is an integer 3 or 4; and 0 < x < 3; provided that when M is P, x is < 2.5.

Such materials are also novel per se and therefore according to an alternative aspect of the invention we also provide a material which is a monovalent cation doped heteropolyacid of general formula I :

Y_xH_(n-x)MX₁₂O₄₀ I

in which M, Y, X, n and x are each as defined above.

In a preferred aspect of the invention the atom M is P, i.e. phosphorous. When M is P then n is preferably 3. An especially preferred material is one in M is P, n is 3 and Y is Cs. Thus, the monovalent cation doped heteropolyacid or the heteropolyacid used in the method of the invention may be a material of formula II:

Cs_xH_(3-x)PW₁₂O₄₀ II

in which x is as defined above.

In a separately preferred embodiment of the invention the monovalent cation doped heteropolyacid may be one in which M is Si. When M is Si, then n is 4. An especially preferred material is one in M is Si, n is 4 and Y is Cs.

Thus, in this aspect of the invention the monovalent cation doped heteropolyacid may be a material of formula III:

Cs_xH_(4-x)SiW₁₂O₄₀ III

in which x is as defined above.

The Y, e.g. Cs, content in the material of the invention may vary, thus the Y content, when represented as a function of x, may be from 0.9 to < 2.5, preferably from 0.9 to 2.35 and more preferably from 1.95 to 2.35. Optimal performance as a catalyst is found in the materials of the present invention when x is from 2.0 to 2.3.

According to a further aspect of the present invention we provide a process for the preparation of a material of formula I as hereinbefore described, the process comprising reacting, e.g. a metallotungstic acid with a monovalent cation salt, e.g. a caesium salt.

Any conventionally known monovalent cations may be used, and a more than one species of monovalent cation may be present. Thus, in a preferred aspect of the invention the monovalent cation is selected from one or more of NH₄ ⁺, Na⁺, K⁺ or

Cs⁺. Caesium salts are especially preferred may be utilised in the process of this aspect of the invention. However, preferably the monovalent cation salt, e.g. caesium salt, is a carbonate or a halide, such as fluoride, chloride, bromide or iodide. It is especially preferred that the salt is a halide, such as caesium chloride.

The materials of the invention find utility as catalysts. Therefore, we further provide the materials as hereinbefore described on a support material. It will be understood by one skilled in the art that the word "on" shall include the materials support on the surface of the support or in the pores of a porous support. Indeed the use of a porous support is a preferred aspect of the present invention. When the support is a porous support it may have an average pore diameter of from 40A° to 170 A⁰, preferably from 100A° to 170 A⁰, e.g. 143A⁰.

Alternatively or in addition, the support may have a BET surface area of from 250 m²/g to 1,000 m²/g, preferably from 250 m²/g to 600 m²/g and especially from 350 m²/g to 400 m²/g. In an especially preferred aspect of the invention the support has an average pore diameter of from IOOA⁰ to 170A° and a BET surface area of from 350 m²/g to 400 m²/g, preferably an average pore diameter of 143 A⁰ and a BET surface area of 378 m²/g. Also, the support has a BJH pore volume of from 0.8 cm³/g to 1.4 cnrVg.

A variety of different supports may be used, however, it is preferred that the support is a support with an hydroxyl content of from 0.1 to 1.5 mmol/g. The support may optionally be hydrophilic or hydrophobic however, it is preferred that the support is substantially hydrophobic. The hydrophobicity is preferably due to the presence of alkyl and/or aromatic moieties. Thus, the support may comprise one or more inorganic oxides, such as silica. In an especially preferred embodiment the support is substantially a silica support.

In an especially preferred aspect of the invention the support, e.g. a silica support, is coated with one or more zirconium salts. The whole support may be coated or the support may only be partially coated. The one or more zirconium salts may be selected from zirconium oxide, zirconium phosphate and zirconium sulphate.

According to a yet further aspect of the invention we provide a method of catalysing a chemical reaction which comprises the use of a material of formula I. The materials are especially useful as catalysts for the esterification of organic acids, such as fatty acids, and/or transesterification of organic esters, such as fatty esters. Such fatty acids may comprise linear, branched, saturated and/or unsaturated fatty acids. It is a particular advantage of the materials of the invention that they are suitable for the substantially simultaneous esterification and transesterification of acids/esters. Therefore, as catalysts the materials are useful in the manufacture/refining of biofuels and especially biodiesel. Thus, the materials of the invention are suitable for the esterifϊcation and/or transesterification, and optionally the simultaneous esterifϊcation and transesterification of fatty acids/esters. The said fatty acids and/or esters used as a feedstock for the production of biofuels, such as biodiesel, may originate from a variety of sources, such as, vegetable oil, animal fat and mixtures thereof. When the feedstock is of animal fat origin, it may be, for example, beef tallow. Furthermore, when the feedstock is of vegetable origin it may be selected from one or more of corn oil, palm oil, peanut oil, rapeseed oil, soybean oil, sunflower oil, jatropha oil, hungmai oil, mauha oil, nohr oil, sal oil and mixtures thereof.

Alternatively, or in addition the feedstock may comprise cooking oil, e.g. recycled cooking oil.

In the esterification and/or the transesterification reaction of the present invention a variety of alcohols may be used, for example, a primary or secondary monohydric aliphatic alcohol having one to eight carbon atoms. Such alcohols may be selected from the group consisting of methanol, ethanol, propanol, butanol and amyl alcohol. Preferred alcohols are methanol or ethanol and especially methanol.

In an additional aspect of the invention the caesium doped heteropolyacid may be placed, e.g. grafted, onto a support material or may be in conjunction with a support material. A variety of known support materials may be used in this aspect of the invention, although a preferred support material will be one that increases the exposed catalyst surface area. Thus, the support may preferably be a porous support. Such support materials may be known to the person skilled in the art but a preferred support material would be a silica material.

In a further preferred method of the invention the oils are analysed by column chromatography where said column is a high performance liquid chromatography column (HPLC).

HPLC is a very well known method for the separation and analysis of solute molecules. The separation process is effected through liquid chromatography and relies on the fact that a number of component solute molecules in a sample stream of fluid (mobile phase) flowing through a packed bed of particles (stationary phase) can be efficiently separated from one another with a high degree of resolution. This is based on the fact that individual components in the mobile phase have a different affinity for the stationary phase and therefore a different rate of migration and exit through the column. The efficiency of separation is determined by the amount of spreading through the column which is determined by the column composition. We have found that a column that has a very high lipophilic content is surprisingly effective with respect to the separation of fatty acid esters from a complex mixture such as a biofuel.

In a preferred method of the invention said column comprises a highly lipophilic polymer.

"Highly lipophilic" refers to a separation column that is effective at retaining non- polar compounds such as long-chain hydrocarbons and related structures. "Polymeric" refers to the structure of the base material used to manufacture the packing of the column, for example a polymer based on styrene-benzene, although other column packing materials can be used in the method of the invention, for example silica-based packings.

In a further preferred method of the invention said polymeric material is non- endcapped.

"Non-endcapped" refers to the treatment of the polymeric packing material during manufacture. The polymeric packing material in an "end-capped" column are treated with a blocking compound in the final stages of manufacturer to bind any unwanted active sites that would otherwise affect the separation characteristics of the column.

Polymeric columns do not typically have these active sites (unlike, for example, silica-based columns, where the packings have residual silanol groups), so they are not end-capped.

In a further preferred embodiment of the invention said lipophilic polymer comprises hydrocarbon chains that confer lipophilicity.

In a preferred method of the invention said hydrocarbon chains comprise at least 19 carbon atoms.

In a further preferred method of the invention said hydrocarbon chains comprise at least 20, 21, 22, 23, 24, 25, 26, 27, 28, or 29 carbon atoms. In a preferred embodiment of the invention said hydrocarbon chain comprises at least 30 carbon atoms. Preferably said hydrocarbon chain consists of 30 carbon atoms.

In a preferred method of the invention said method further comprises the steps of: i) detecting and collating the fatty acid ester content of said feedstock; ii) collating the data into a data analysable form; and optionally iii) providing an output for the analysed data.

According to a further aspect of the invention there is provided a kit comprising a catalyst capable of the esterifϊcation and transesterification of an oil feed stock and a column comprising a lipophilic polymer.

Throughout the description and claims of this specification, the words "comprise" and "contain" and variations of the words, for example "comprising" and "comprises", means "including but not limited to", and is not intended to (and does not) exclude other moieties, additives, components, integers or steps.

Throughout the description and claims of this specification, the singular encompasses the plural unless the context otherwise requires. In particular, where the indefinite article is used, the specification is to be understood as contemplating plurality as well as singularity, unless the context requires otherwise.

Features, integers, characteristics, compounds, chemical moieties or groups described in conjunction with a particular aspect, embodiment or example of the invention are to be understood to be applicable to any other aspect, embodiment or example described herein unless incompatible therewith.

Thus, according to a further aspect of the invention we provide a method of manufacturing biodiesel which comprises the esterification, transesterification or the simultaneous esterification and transesterification of a feedstock comprising a vegetable oil, an animal fat and/or mixtures thereof.

In addition we provide the use of a material as hereinbefore described as a catalyst , e.g. as a catalyst for use in the manufacture of biofuels, such as biodiesel.

The invention will now be described by way of example only and with reference to the accompanying schemes, tables and figures' in which Scheme 1 illustrates the proposed structural model for Cs₃PWi₂O_40; Scheme 2 illustrates i. esterification of palmitic acid and ii. transesterification of tributyrin;

Table 1 illustrates the elemental composition of Cs exchanged HPW samples;

Table 2 illustrates the surface area, crystallite size and average pore diameters of Cs exchanged HPW samples; Table 2b illustrates the surface area of Cs exchanged HSiW samples

Table 3 indicates NH₃ adsorption calorimetry data from Cs exchanged HPW samples;

Table 4a indicates the percentage conversion of palmitic acid and tributyrin observed during Cs exchanged HPW catalysed esterification and transesterification; Table 4b indicates the percentage conversion of palmitic acid, tributyrin and glyceryl trioctanoate observed during Cs exchanged HSiW catalysed esterification and transesterification;

Table 5 shows a comparison of TOF and conversion for representative solid acids; and bases in transesterification and esterification reactions;

Table 6 indicates the percentage conversion of tributyrin and palmitic acid after 6 h reaction during one pot esterification and transesterification reactions using a fresh and recycled Cs₂₃HPW catalyst;

Table 7 indicates the percentage conversion of tributyrin after 6 and 24 h using supported Cs_{2 0} HPW/SiO₂ prepared by incipient wetness of a pre-prepared 37wt% HPW/SiO₂ catalyst with CsCl; Table 8 shows the effect of catalyst on lipid composition in vegetable oil.

Figure Ia is a graphical representation of the surface composition of Cs exchanged HPW samples as a function of bulk Cs loading;

Figure Ib is a graphical representation of the surface and bulk Cs: W atomic ratio of

Cs exchanged HPW samples as a function of bulk Cs loading;

Figure 2 illustrates the variation of Cs 3d_5/2 XPS component intensity as a function of bulk Cs loading per Keggin unit in Cs exchanged HPW catalysts; Figure 3 is a Powder XRD of Cs exchanged HPW;

Figure 4 is a DRIFT of Cs exchanged HPW;

Figure 5 is a ³¹P MAS NMR of Cs exchanged HPW;

Figure 6 is a NH₃ calorimetry on Cs exchanged HPW samples;

Figure 8a illustrates the conversion of palmitic acid during esterification with methanol at 60 °C using Cs exchanged HPW catalysts; Figure 8b illustrates the conversion of tributyrin during transesterification with methanol at 60 °C using Cs exchanged HPW catalysts; Figure 9 illustrates the observed product distribution during Cs₂₃H_{0 7}PWi₂O₄₀ catalysed transesterification of tributyrin with methanol at 60 °C; Figure 10 illustrates the calculated TOF using Cs exchanged HPW catalysts in the esterifϊcation of palmitic acid and transesterification of tributyrin with methanol as a function of Cs content;

Figure 11 illustrates the TOF of supported and bulk Cs₂HPW₁₂O₄₀ catalysts in the esterification of palmitic acid with methanol; Figure 12 is a schematic representation of the plant oil processing pipeline.

Example 1

Ia) Catalyst preparation

12-phosphotungstic acid (ex Aldrich) was dried at 100 ⁰C to remove the physically adsorbed water before use. CsCl (ex Aldrich 99.9%) was used as received. Cs_xH_3- ^PW₁₂O₄₀ (x = 1, 2, 2.1, 2.2, 2.3, 2.4, 2.5, and 3) samples were prepared by dropwise addition of predetermined amounts of a 0.02 mol dm^"3 CsCl aqueous-ethanol (50:50 volume ratio) solution to a 0.08 mol dm^"3 ethanol solution of HPW at room temperature. A white precipitate formed, and the solutions were left to dry overnight at room temperature in a vacuum oven to evaporate the ethanol and water. Fine white powders were obtained by oven drying the materials in air at 100 °C. No further pre- treatments were applied to the materials, which were stored in air prior to analysis and reaction testing. Catalysts are abbreviated by reference to their Cs content (e.g. Cs_{2 3} = Cs₂₃H₀₇PWi₂0₄o). A series Of Cs_xH_4-XSiW₁₂O₄₀ samples was also prepared by the same method. The extent of proton exchange by Cs and final composition of the salts was determined by elemental analysis (Tables Ia and b).

Ib) Preparation of Supported Catalysts: The preparation of silica supported Cs salts of tungstophosphoric acid (HPW) was carried out by five different methods (1) Impregnation (2) Sol-Gel (3) Precipitation- Deposition (4) Micro-Emulsion (5) Incipient wetness of preformed 37 wt% HPW/SiO₂. In each of cases 1-4 a material having 50 wt% HPW and composition Cs₂HPW was prepared, while in case 5 a composition of Cs₂HPW was prepared.

(1) Impregnation: For the synthesis, a commercial porous silica-gel (Fisher lOOA) was employed as the support. First, a calculated amount Of Cs₂CO₃ is dissolved in a 20 ml of solvent (50/50 volume ratio of H₂O and Ethanol) and then silica gel is dispersed in the total solution under constant stirring at room temperature. The stirring was continued for 5h and excess of solvent was evaporated by drying in an oven at 120 ⁰C to obtain the Cs modified silica. Then the solids were impregnated with ethanolic HPW solution of the appropriate concentration. Finally, the resulting solids were dried and calcined at 300 ⁰C.

(2) Sol-Gel: A mixture of calculated amounts of TEOS, ethanol, concentrated hydrochloric acid and water was refluxed for 3 h under stirring. After the solution was cooled down to room temperature. An ethanolic solution of calculated amount of HPW was slowly added and stirred for 3 h till the total dissolution of HPW, to the total contents a known amount of an aqueous Cs₂CO₃ is also added and stirring continued for over night. The excess of solvent was evaporated in oven at 120 ⁰C and calcined at 300 ⁰C.

(3) Precipitation-Deposition: For the synthesis, a commercial porous silica-gel (Fisher 100 A) was employed as the support. Calculated amount of silica was dispersed in a solution contained sodium tungstate, dihydrogen sodium phosphate. The pH of total contents was maintained approximately at 2 by adding dilute HCl under constant stirring. The stirring was continued for 5h; to the total contents known amount of aqueous Cs₂CO₃ was added to deposit the Cs salt of HPW on silica support. The precipitated milky white material was filtered, dried at 120 ⁰C and then calcined at 300 ⁰C.

(4) Micro-Emulsion: For the micro-emulsion method, sodium bis(2-ethylhexyl) sulfosuccinate (AOT, Sigma) as surfactant, cyclohexane as an oil phase, tetraethoxysilane (Aldrich, TEOS, 99.99%) as silicon precursor and HPW

(Aldrich) were used. AOT was dissolved in cyclohexane at 0.2 M, and a specified amount of HPW dissolved in water was added to form a well-defined microemulsion phase. And then, TEOS was dropped into the microemulsion phase. After the mixture was stirred for 5 h at room temperature, calculated amount of aqueous Cs₂CO₃ was added to the emulsion. SiO₂ particles with Cs salt of HPW were obtained after 12h of further stirring. The weight ratio of [CsHPW]/[TEOS] was fixed at 1. The obtained nanoparticles in cyclohexane were centrifuged and the particles were then repeatedly rinsed with acetone and finally the occluded surfactant was removed by calcination of the white powder at 300 ⁰C in flow of air. (5) Incipient Wetness: For the synthesis, a commercial porous silica-gel (Fisher

100 A) was employed as the support. First, a calculated amount of HPW is dissolved in a 20 ml of methanol and then silica gel is dispersed in the total solution under constant stirring at room temperature. The stirring was continued for 5h and excess of solvent was evaporated by drying in an oven at

120 ⁰C to obtain the HPW modified silica. The pore volume of the solids was determined then impregnated by incipient wetness with an accurate amount of ethanolic CsCl solution of the appropriate concentration to give the desired Cs:HPW loading. The resulting paste was stirred with a glass rod then the solids were dried at 100 ⁰C.

Ic) Catalyst characterization

Nitrogen porosimetry was performed on a Micromeritics ASAP 2010 instrument. Surface areas were calculated using the BET equation over the range P/P_o = 0.02 - 0.2, where a linear relationship was maintained, while pore size distributions were calculated using the BJH model up to P/P_o = 0.6. DRIFT spectra were obtained using a Bruker Equinox FTIR spectrometer. Powder X-ray diffraction patterns were collected on a Bruker D 8 diffractometer using Cu K_n radiation. The bulk chemical composition of samples was determined with a Hitachi atomic absorption spectrometer. XPS measurements were performed using a Kratos AXIS HSi instrument equipped with a charge neutralizer and Al K₀ X-ray source. Spectra were recorded at normal emission using an analyser pass energy of 20 eV and X-ray power of 225 W. A wide scan across the entire energy range (1100 eV to 0 eV) was also collected at a pass energy of 160 eV to check for impurities and also confirm complete loss of Cl following ion exchange of the HPW samples with CsCl. Energy referencing was employed using the valence band and adventitious carbon. Spectra were Shirley background-subtracted across the energy region and fitted using CasaXPS Version 2.1.9. ³¹P MAS NMR spectra were obtained in single pulse ('ZG') mode (3.5 ms pulse, and 8 s delay between pulses) on a Bruker Avance 400 spectrometer, operating at a frequency of 161.98 MHz. NMR measurements were performed in 2.5 mm outer-diameter rotors, with a sample spin rate of 20 kHz. Spectra were referenced externally to sodium dihydrogen phosphate (0.0 ppm). Typically line broadening of 10 Hz was applied when processing the spectra.

Ammonia adsorption flow calorimetry was performed using a system described previously based on a Setaram 111 differential scanning calorimeter (DSC) and an automated gas flow and switching system, combined with a mass spectrometer detector sampling the down-stream gas flow (Hiden HPR20). In a typical experiment, the catalyst (5-30 mg) was activated at 150°C under a dried helium flow at 5 ml min^"1. Following activation, and maintaining the sample temperature at 150°C, small (typically 1.0 ml but from 0.2 to 5.0 ml) pulses of the probe gas (1 vol % ammonia in helium) at atmospheric pressure were injected at regular intervals into the carrier gas stream from a gas sampling valve. The ammonia concentration downstream of the sample was monitored continuously by mass spectrometry. The pulse interval was chosen to ensure that the ammonia concentration in the carrier gas (including that adsorbed and then desorbed after the pulse had passed) returned to zero to allow the DSC baseline to stabilise. The net amount of ammonia irreversibly adsorbed from each pulse was determined by comparing the MS signal with that recorded during a control experiment through a blank sample tube. The net heat released for each pulse, corresponding to irreversible adsorption of ammonia, was calculated from the DSC thermal curve. From this the molar enthalpy of adsorption of ammonia (ΔH°_adS) was obtained for each successive pulse. The ΔH°_adS values were then plotted against the amount of (irreversibly) adsorbed ammonia per gram of the catalyst, to give a ΔH°_adS/coverage profile for each catalyst. The final uptake of NH₃ and average ΔH°_adS values were determined up to the point the molar enthalpy of adsorption drops numerically below 80 kJ mol^"1, since this is frequently cited as the break point between adsorption on sites of significant acid strength and those with no acid strength.

The bulk and surface composition of the Cs-exchanged HPW samples were first verified by a combination of elemental and XPS analysis. Good agreement between nominal and observed bulk Cs content was observed across the range Cs₁-Cs₃ (Table Ia) (For comparison data for Cs-exchanged HSiW is shown in table Ib which shows a similar variation). Figure Ia shows how the surface composition of Cs-HPW varies with increasing bulk Cs content from 3.9 - 12.1 wt%. There is an almost linear increase in surface Cs and concomitant decrease in surface W, confirming incorporation of Cs into the HPW clusters rather than simple encapsulation of heteropoly acid particles with a Cs overlayer. The surface Cs content is slightly lower than the bulk, which suggests the surface is Cs depleted. This trend is also seen in Figure Ib which shows how the bulk and surface Cs/W atomic ratio of the Cs_xH_3- _xPWj₂θ₄o salts compares to the bulk and nominal values (solid line). The bulk values are in excellent agreement with the theoretical ones; however there is a significant deviation between surface and bulk Cs/W ratio at low Cs content. This deviation could be accounted for by structural models proposed for intermediate Cs_xH_3- _xPW₁₂O₄₀ compositions, wherein Cs₃PW₁₂O_4O core particles are believed to be coated with a surface layer of H₃PW₁₂O₄₀ clusters]. Such layers would be expected to attenuate the Cs signal; however at higher Cs loadings where more Cs₃PW₁₂O_4O is present better agreement would be expected. Indeed by taking the inelastic escape depth for the Cs 3d photoelectron to be 0.7 ran, and the thickness of a H₃PW₁₂O₄₀ cluster of about 1 nm, the bulk Cs signal would be expected to be attenuated by 75 % by an external layer of Keggin clusters, which is consistent with our observed deviation.

Closer inspection of the Cs Sd₅^ state Figure 2 (inset) reveals broadening of the Cs 3d_5/2 peak as the Cs loading increases. Peak deconvolution confirms the presence of two distinct Cs species. The lowest loading Cs₀₉ and Cs_{1 9} samples exhibit a single peak at 724 eV characteristic of Cs⁺. For higher loadings, a second high binding energy state evolves at 724.9 eV which grows rapidly for compositions above Cs_{2 3} as illustrated in Figure 2, where the integrated intensity of these 2 components is shown as a function of Cs per Keggin. This observation could be accounted for by the formation of a Cs₃PW₁₂O₄₀ core in all samples, with Cs⁺ adopting a common octahedral interstitial site in the close packed Keggin lattice as depicted in Scheme 1 (for simplicity, only a few surface HPW units have been drawn). Inspection of the surface terminating layer reveals two Cs⁺ coordination environments, labelled as bridge or terminal sites. Based on the relative number of these sites, we tentatively attribute the low and high binding energy components to the terminal site and bridge sites respectively. The powder XRD patterns of all the samples are shown in Figure 3. H₃PWi₂O₄₀ dried at 100 °C exhibits all the reflections corresponding to a cubic Pn3m crystalline structure. Following initial doping to form CS_O^H_{2 1}PW₁₂O₄₀, a new set of peaks evolve as shoulders on the main HPW reflections. Diffraction peaks corresponding to the free acid disappear as the Cs content increases beyond x = 2. The shift in H₃PW₁₂O₄₀ peaks towards higher angles in the Cs_xH₃-^PW samples is consistent with the body centred cubic structure of Cs₃PW₁₂O₄₀ salts reported in the literature and indicate the presence of a unique crystalline Cs₃PW₁₂O₄₀ phase in all Cs-exchanged materials.

From X-ray peak broadening, it is possible to calculate the average size of the microcrystalline phases using the Scherrer equation D = 0.9λ/(β - β₀) cosθ. The crystallite size (D in diameter) of these Cs salts, summarized in Table 2a, was calculated from this where λ is the X-ray wavelength (Cu K_«) in angstroms (1.54 A), θ is the diffraction angle, β is the line width, and β₀ is the instrumental line width (0.15 °, with all angles in radians). Analysis of the line-widths of the XRD peaks of these Cs salts (JC = 0-2.7) show that the size of the primary crystallites decreases from 48 to 8 nm as the Cs loading increases to Cs_{2 3}. Thereafter the particle size increases to 19.4 nm for the Cs₃ sample indicating the formation of aggregates of the Cs₃PW core at higher Cs loadings.

The surface area of samples increase in line with these morphological changes, rising from 2 m g^"1 for Cso_.9 to 156 m²g^"' Cs₃ observed. This is consistent with the work of

Moffat and co-workers who showed that NH₄ ⁺ and Cs⁺ salts of HPW possess dense porous networks and corresponding higher surface areas than the parent HPW. Our porosimetry measurements support this, revealing is a significant increase in the average pore diameter from 3 to 14 nm for loadings > Cs_{2 0} (Table 2). Misono et al showed that large voids exist between the primary particles (microcrystallites) in these materials. If the size of the primary Cs_{2 7} clusters is about 12 nm, then the closest packed aggregates could form voids of around 3-4 nm. Interparticle voids between larger crystallites would increase the overall apparent average pore diameter consistent with the present study. For comparison surface areas for Cs-HSiW are shown in Table 2b which reveals a similar trend with Cs loading.

Infrared spectra are also an informative fingerprint of the Keggin heteropoly cage structure. The Keggin anion structure consists of a PO₄ tetrahedron surrounded by twelve MO₆ (M = W, Mo) octahedra, which share edges in M₃O₁₃ triads and corners with other triads through bridging oxygens. FTIR spectra of all samples, recorded after drying the samples at 100 °C, are presented in Figure 4. These spectra exhibit similar bands for V₃₅(P-O) = 1080 cm^"1, v(W-O_c-W) = 890 cm^"1 and v(W-O_e-W) = 798 cm^"1 modes. The diffuse reflectance intensity of infra red bands is known to be sensitive to particle size, thus the observed variation in the intensity of these modes may reflect a combination of the presence of smaller Keggin clusters at intermediate Cs substitutions and reduced WO_x content at the highest Cs loadings.

The ³¹P MAS NMR spectra of HPW, Cs₁ and Cs₃ dried at 100 ⁰C are shown in Figure 5. Each exhibits a single peak that changes from δ = -16.7 to -15.6 ppm upon substitution of all 3 protons by Cs. The chemical shift dependence on Cs content per Keggin unit is presented in the inset. A small peak at -14.6 ppm was observed in the parent and Csj samples and is attributed to HPW with lower H₂O content. These observations are in good accord with the work of Dias et al who also found a single peak for dried Cs_xH_3-1PW₁₂O₄₀ materials, but multiple peaks after thermal treatment at 300 °C. The ³¹P NMR chemical shifts correlate with the water content, since the degree of hydration decreases with Cs content in the secondary structure of the polyoxometalate.

The acidity of the Cs_xH_3-xPW₁₂0₄o series was studied by NH₃ flow calorimetry. Figure 6 shows the NH₃ adsorption profiles, and the corresponding uptakes and heats of adsorption are summarised in Table 3. Cs incorporation lowers the number of titratable acid sites in quantitative agreement with the theoretical degree of H⁺ exchange. The average heat of NH₃ adsorption only falls slightly with Cs doping up to loadings of Cs₂₃, suggesting that the residual acid strength of the Cs-doped HPW materials is not strongly perturbed during proton exchange. However the acidity of heavily substituted (>Cs₂₄) samples is significantly lower in accordance with previous measurements.

Example 2

2a) Reactivity

Esterification and transesterification were performed in a stirred batch reactor with samples withdrawn periodically for analysis on a Shimadzu GC17A Gas

Chromatograph fitted with a DBl capillary column (film thickness 0.25 mm, i.d. 0.32 mm, length 30 m), and AOC 2Oi auto sampler. Esterification was performed at 60 °C using 50 mg of catalyst, 0.01 mol of palmitic acid (Aldrich 98 %) and 0.3036 mol

(12.5 cm³) methanol (Fisher 98 %) with 2.5 mmol (0.587 cm³) of hexyl ether (Aldrich 97 %) as internal standard. Transesterification was performed at 60 °C using 50 mg of catalyst, 0.01 mol (3 cm³) of glyceryl tributyrate (Aldrich 98 %) (or 0.01 mol glyceryl trioctanoate (Aldrich 98%)) and 0.3036 mol (12.5 cm³) methanol with 2.5 mmol (0.587 cm³) hexyl ether as internal standard. Catalyst samples were separated from the reaction mixture for recycling by centrifugation.

Reactions were run for 6 h with initial rates determined at conversions < 30 %, with reactions continued for 24 h. Catalyst selectivity and overall mass balances (closure was >98 %) were determined using reactant and product response factors derived from multipoint calibration curves.

Catalyst stability was verified by performing leaching tests in hot methanol, where Cs_xH_3-xPW catalysts were refluxed for 6 hours in methanol after which the solid was removed. The presence of soluble HPW species in the recovered methanol was subsequently investigated by assessing the activity of the residual solvent in both esterification and transesterification reactions.

A selection of commercial solid acid catalysts (SO₄/ZrO₂ (MEL), Nafion (Aldrich) and ZSM-5 (Zeolyst)) were also screened in both transesterification and esterification reactions for comparison with the best Cs_xH_3-xPW catalyst. SO₄/ZrO₂ and ZSM-5 were activated respectively at 550 °C and 500 °C for 3 hours, then stored in air prior to use.

2b) Catalyst activity

The insoluble nature of Cs-doped HPW (with x >1) in polar media make them attractive for use in esterification and transesterification reactions pertinent to biodiesel synthesis. The activity of Cs salts and pure HPW were therefore evaluated in the esterifi cation of palmitic acid (a major saturated fatty acid found in palm oil) and transesterification of tributyrin (a natural constituent of butter) with methanol (scheme 2). Figures 8a and b show the resultant palmitic acid and tributyrin conversions. The limiting conversions after 6 or 24 hrs reaction are given in Table 4a. In both reactions catalyst performance increases with Cs loading up to Cs_{2. J}-Cs₂₃, before decreasing with higher degrees of Cs exchange_. Table 4b shows comparable data for the series of Cs-doped HSiW samples and reveals a similar variation in catalyst activity with loading. The overall catalyst activity of the Cs-HSiW series is however less than that of Cs-HPW.

Figure 9 shows the time-dependent evolution of products and selectivity towards methyl butyrate for the optimal Cs_{2 3}-H_{0 7}PW catalyst. Transesterification proceeds via progressive reaction of the triglyceride ester groups resulting in the formation of di- and monoglyceride intermediates, which are themselves ultimately converted into the methyl ester and glycerol. While the reaction is much slower than our previous observations with hydrotalcite catalysts (24 c.f. 8 hrs), the methyl ester yield from 10 mmol of tributyrin is improved at 24 mmol versus 15 mmol.

Figure 10 compares TOF for both reactions across the catalysts series. Note that while the parent HPW exhibits excellent activity for both transformations, it is of course completely soluble in these reaction media and therefore unable to compete with the many process advantages offered by its heterogeneous counterparts. Low levels of Cs exchange confer comparatively poor activity in both palmitic acid esterification and tributyrin transesterification. However further Cs incorporation promotes both reactions (especially transesterification) with maximal TOF achieved for Cs₂₃ loadings. The best esterification rates of 81 mmol h^"1 g^"1 are far superior to that obtained for ZrPO₄-tethered phosophotungstic acid of 22.5 mmol hr^'1 g^"1, while the best transesterification rate of 6.4 mmol hr^"1 g^"1 compares reasonably well with that for MgO at 15 mmol hr^"1 g^"1. However it should be noted that tailored solid bases such as Li-doped CaO and MgAl hydrotalcites can outperform the present solid acid system in tributyrin transesterification. Further Cs addition induces a dramatic fall in general catalytic activity, which we associate with the loss of Bronsted acid sites for stoichiometrics approaching Cs₃ (apparent from both calorimetry data). The most active Cs₂₃PW catalyst facilitated complete palmitic acid conversion in only 6 h (Table 4a). Further comparison of Cs₂₃ catalyst is made with other solid acid catalysts in Table 5, from which it can be seen that in esterification of palmitic acid the Cs₂₃ sample outperforms SO₄/ZrO₂, Nation and H ZSM-5. Likewise all three commercial catalysts exhibit poor activity in the transesterification of tributyrin with methanol, when compared to Cs₂₃.

Leaching studies were performed as detailed in the experimental to determine whether soluble HPW components were released into the reaction. As predicted from our model wherein low Cs loading samples comprise Cs₃ particles capped by a pure HPW layer, the Cs₁ evidenced some dissolution of an active acid component following reflux in hot methanol. Indeed powder XRD revealed the loss of HPW reflections from the surviving solid isolated after this methanol treatment of the Csj salt. In contrast materials with higher Cs contents were stable to methanol reflux, confirming their heterogeneous mode of action. Having demonstrated the Cs-doped HPWs were separately active in both esterification and transesterification, a single pot reaction was conducted with the Cs₂₃ catalyst to determine whether both palmitic acid esterification and tributyrin esterification could be simultaneously undertaken in a 'one pot' reaction, as desirable for a commercial biodiesel process. Table 6 shows the conversions of both acid and triglyceride following addition of 0.5 mmol palmitic acid to a transesterification reaction. 100 % palmitic acid conversion was observed alongside 50.2 % tributyrin conversion after 6 h reaction, with 90 % selectivity to the methyl ester achieved. In the absence of palmitic acid, tributyrin conversion was comparable confirming simultaneous esterification and transesterification is possible without loss of activity or selectivity in the (secondary) transesterification process. The Cs₂₃ catalyst could also be recycled at least 3 times with negligible loss of activity. While the overall transesterification reaction is slower than possible via a solid base catalysed reaction (Table 5), there are obvious process advantages to operating a single bed reactor with one catalyst formulation such as offered by the present Cs-exchanged HPW solid acids including, inter alia, being able to operate a continuous process.

2b) Activity of supported catalysts in palmitic acid esterification:

The influence of supporting Cs₂HPW on catalyst activity in palmitic acid esterification was also investigated. Materials were prepared by impregnation or precipitation of Cs₂HPW onto a mesoporous amorphous SiO₂ support and by sol-gel or microemulsion methods whereby the Cs₂HPW salt was incorporated into a silica matrix. A comparison of catalyst activity is made in Figure 11 which shows the TOF, normalised to the HPW content for the different prepared catalysts. This reveals that the sol-gel, micro-emulsion and impregnation methods give a substantial increase in catalyst activity over the bulk salts, with the order of activity observed to be

Micro-emulsion > Sol-gel > Impregnation > Bulk > Precipitation deposition.

This increased activity is most likely attributed to the presence of highly dispersed Cs₂HPW clusters in the silica matrix.

2c) Activity of supported catalysts in tributyrin trans-esterification: The influence of supporting Cs₂HPW on catalyst activity in tributyrin transesterification was also investigated. Materials were prepared by incipient wetness whereby a 37wt% HPW/SiO₂ materials was impregnated by pore-filling with an ethanolic solution of CsCl. Sufficient CsCl solution, of the required concentration, was added to just wet the powder forming a paste so that HPW was retained in the pores of the silica support. Catalyst activity is reported in Table 7 which shows the TOF, normalised to the HPW content. The use of incipient wetness can thus be used to generate a catalyst which is active for transesterification of triglycerides. Subsequent hot filtration tests were used to confirm the heterogeneous of the catalyst. Following initial reaction for 1 hour the catalyst was removed by filtration after which the reaction was continuously monitored for any further conversion. Table 7 shows that the conversion of trybutyrin is negligible indicating that there is no leaching of HPW from the support. LC conditions (method under evaluation)

Instrumentation: Thermo Separations Products SCMlOOO degasser; P4000 gradient pump; AS3000 autosampler maintained at 4°C, fitted with a 100 μL stainless steel loop and operated in pull-loop mode . Mobile Phases: A = 60% ethanol 40% acetonitrile 0.2 % formic acid (v/v); B = tetrahydrofuran 0.2% formic acid. Gradient

Profile: 0-5 min 100% A; 5-25 min to 70% A 30% B; 25-30 min to 100% B; 30-35 min 100% B; 35-35.1 min to 100% A; 35.1-40 min 100% A; 1 mL-min-1. 1 min equilibration time between runs. Column: YMC-Pack YMC C30, 250 x 4.6 mm, part number CT99S053546; maintained at 30 °C. Injection conditions: 10 μL injection volume; needle rinse solvent = methanol.

LC conditions (existing method - Holcapek et al. (2003) J. Chromatography A 1010:195-215)

Instrumentation: As above. Mobile Phases: A = Water; B = acetonitrile; C = 2- propanol. Gradient Profile: 0 min 30% A 70% B; 0-20 min to 100% B; 20-36 min 100% B; 36-132 min to 40% B 60% C; 132-135 min to 30% a 70% B; 135-40 min 30% A 70% B. 1 min equilibration time between runs. Column: Two Waters Nova- Pak 100 x 3.9 mm Cl 8 columns connected in series, part number WAT086344; maintained at 40°C. Injection conditions: As above. Mass Spectroscopy conditions

Instrumentation: Thermo LCQ ion trap mass spectrometer. Source = atmospheric pressure chemical ionization; positive ionization mode; vaporizer temperature 500°C; N2 sheath flow 60 units; N2 aux flow 60 units; corona discharge current 5 μA; capillary temperature 150°C; capillary voltage 15 V.

Data collection: LC flow diverted to MS 5-35 min; full scan MS data 500-1500 m/z with automatic gain control on; data dependent fragmentation at normalised collision energy of 35% to identify TAG fatty acid components. Peak integration: total ion current or extracted ions corresponding to TAG ammonium adducts used for peak integration using the ICIS algorithm in the software package Xcalibur 1.2 (Thermo).

Data analysis

Triplicate samples were injected from separate vials. Sample amounts were calculated as extracted ion peak area relative to internal standard peak area for all peaks, or absolute peak area for non-co eluting peaks detected in total ion current mode. Peak tables were exported from Xcalibur for principal components analysis using SPSS 11.0 software, using a correlation matrix and varimax with kaiser normalization rotation. Example 3

TAG profiling of heterogeneously catalysed Rapeseed oil transesterification.

Transesterification of rapeseed oil was performed at 80°C using 21 ml Methanol, 8.9g Rapeseed oil and 200 mg of supported Cs₂H₂Si W₁₂O₄₀/SiO₂ (containing 37 wt% heteropoly acid (prep according to method lb-5)). Reactions were run for 24 h prior to TAG profiling.

Table 8. Effect of catalyst on lipid composition in vegetable oil Samples of supermarket-bought vegetable oil were profiled in triplicate by LCMS before and after solid-phase catalytic reactions. The major triacylglycerol (TAG) and diacylglycerol (DAG) components were identified, and the changes in their distribution determined to assess whether there was any preferential transesterification of particular functionalities. TAG and DAG distributions are expressed relative to 18:3-18:2-18:3 and 18:2-18:2-OH components respectively. The TAG distribution within the fresh and reacted oil remains very similar, showing the HPA catalysts converts almost all TAGs with equal efficiency (i.e. there is no specificity for conversion of a particular chain length TAG in the oil).

In contrast there does appear to be a difference in the distribution of the DAG intermediates, with the accumulation of 18:1-18:1 -OH and 18:1 -OH- 18:1 favoured. Table Ia: Elemental composition of Cs exchanged HPW samples

AAS analysis by dissolving the salts in standard NaOH solution. 1 XPS analysis.

Table Ib: Elemental composition of Cs exchanged HPW samples

' AAS analysis by dissolving the salts in standard NaOH solution. ' XPS analysis.

Table 2a: Surface area, crystallite size and average pore diameters of Cs exchanged HPW samples

a BET equation b Estimated from XRD line width of peak -26° Table 2b: Surface areas of Cs exchanged HSiW samples

' BET equation

Table 3: NH₃ adsorption calorimetry data from Cs exchanged HPW samples

a Calculated assuming a formula H₃PWi ₂O₄₀.6H₂O and a 1 : 1 interaction between each H⁺ and NH₃, with each Cs atom replacing 2 H₂O.

Table 4a: Conversion of palmitic acid and tributyrin observed during Cs exchanged HPW catalysed esterification and transesterification.

^a After 6 hours reaction. Selectivity to methyl palmitate > 98% in all cases ^b After 6 hours reaction, (brackets) after 24 hour reaction c Glycerol formation omitted from selectivity calculation.

Table 4b: Conversion of palmitic acid, tributyrin and glyceryl trioctanoate observed during Cs exchanged HSiW catalysed esterifϊcation and transesterifϊcation.

a After 6. hours reaction. Selectivity to methyl palmitate > 98% in all cases b After 6 hours reaction, (brackets) after 24 hour reaction c Glycerol formation omitted from selectivity calculation. dHot filtration tests reveal some homogeneous contribution from Csi all others are insoluble

Table 5: Comparison of TOF and conversion for representative solid acids and bases in transesterification and esterification reactions

after 3h reaction (brackets after 24hrs)

Table 6: Conversion of tributyrin and palmitic acid after 6 h reaction during one pot esterification and transesterification reactions using a fresh and recycled Cs_{2 3} HPW catalyst.

Table 7 indicates the percentage conversion of tributyrin after 6 and 24 h using supported Cs_{2 0} HPW/SiO₂ prepared by incipient wetness of a pre-prepared 37wt% HPW/SiO₂ catalyst with CsCl.

a Hot filtration test performed by filtering the catalyst from the reaction after lhr then restarting the reaction and checking for further conversion resulting from dissolved homogeneous species. Table 8. Effect of catalyst on lipid composition in vegetable oil

TAG and DAG Composition³

Oil species Ratio Ratio post HPA/Silica^b

Pre-reaction catalysed transesterification

Triacylglycerols 18:1-18:2-18:1 24.1 19.5

18:1-18:1-18:1 17.1 16.9

18:1-18:3-18:1 14.2 12.3

18:1-18:3-18:2 9.6 8.1

18:2-18:1-18:2 8.1 6.7

18:1-16:0-18:1 7.3 6.7

18:2-18:1-16:0 6.0 5.7

18:3-18:1-18:3 3.8 3.4

18:0-18:2-18:1 3.4 2.9

16:0-18:3-18:1 3.4 2.9

18:1-18:0-18:1 0.8 2.2

18:1-20:1-18:1 2.3 1.9

18:2-18:0-18:1 0.7 1.1

18:0-18:1-18:2 1.2 0.5

18:2-18:3-18:2 1.5 1.5

18:2-16:0-18:2 1.5 1.3

16:0-18:2-18:3 1.4 1.5

18:3-18:2-18:3 1.0 1.0

Diacylglycerols 18:2-18:1 -OH 0.0 0.1

18:1-18:1-OH 2.2 4.2

18:1-OH-18:1 1.2 3.0

18:2-18:2-OH 1.0 1.0

18:2-18:3-OH 0.8 1.8

18:3-18:1-01-1 0.3 0.4 aNote the given order of the acyl chains in a lipid species name is arbitrary and does not imply a regiospecific definition. bHP A/Silica defined as Cs₂H₂SiWi₂O₄(ZSiO₂ (containing 37 wt% heteropoly acid (prep according to method lb-5)).

04δ4P.WO.Spec(6)

Claims

1. A method of manufacturing biodiesel which comprises the esterification, transesterification or the substantially simultaneous esterification and transesterification of a feedstock comprising one or more vegetable oils, animal fats and/or mixtures thereof.

2. A method for the manufacture and analysis of a biodiesel comprising the steps of: i) performing an esterification, transesterification reaction, or a substantially simultaneous esterification and transesterification reaction, of an oil feedstock comprising one or more vegetable oils, animal fats and/or mixtures thereof; ii) applying the esterified, transesterified feedstock in (i) to a column wherein said column comprises a separation material of a lipophilic polymer; and iii) separating the ester content of the feedstock.

3. A method according to claims 1 or 2 wherein the method includes the use of a solid acid as a catalyst in a liquid phase reaction.

4. A method according to claim 3 wherein the solid acid is a heteropolyacid.

5. A method according to claim 4 wherein the heteropolyacid is a monovalent cation doped heteropolyacid.

6. A method according to claim 5 wherein the monovalent cation doped heteropolyacid is of general formula I:

Y_xH_(n-x)MX₁₂O₄₀ I

7. A method according to claim 6 wherein M is P.

8. A method according to claim 6 wherein n is 3.

9. A method according to claim 6 wherein X is W.

10. A method according to claim 6 wherein the material of formula I is:

Cs_xHo-X)PW₁₂O₄₀ II

in which x is as defined in claim 6.

11. A method according to claim 6 wherein M is Si.

12. A method according to claim 6 wherein n is 4.

13. A method according to claim 6 wherein the material of formula I is:

Cs_xH_(4-X)SiW₁₂0₄o III

in which x is as defined in claim 6.

14. A method according to claim 6 wherein x is from 0.9 to < 3.0; provided that when M is P, x is < 2.5.

15. A method according to claim 14 wherein x is from 0.9 to 2.35.

16. A method according to claim 15 wherein x is from 1.95 to 2.35.

17. A method according to claim 16 wherein x is from 2 to 2.3.

18. A method according to claim 6 characterised in that the heteropolyacid is in conjunction with a support material.

19. A method according to claim 18 characterised in that the support material is a solid support.

20. A method according to claim 18 characterised in that the support material is a solgel.

21. A method according to claim 20 wherein the support material provides increased exposed catalyst surface area.

22. A method according to claim 20 wherein the support is a porous support.

23. A method according to claim 22 wherein the support has an average pore diameter of from 40 A⁰ to 170 A⁰.

24. A method according to claim 23 wherein the support has an average pore diameter of from 100 A⁰ to 170A°.

25. A method according to claim 20 wherein the support has a BET surface area of from 250 m²/g to 1,000 m²/g.

26. A method according to claim 25 wherein the support has a BET surface area of from 250 m²/g to 600 m²/g.

27. A method according to claim 26 wherein the support has a BET surface area of from 350 m²/g to 400 m²/g.

28. A method according to claim 24 wherein the support has an average pore diameter of from IOOA⁰ to 170A° and a BET surface area of from 350 m²/g to 400 m²/g.

29. A method according to claim 28 wherein the support has an average pore diameter of 143 A⁰ and a BET surface area of 378 m²/g.

30. A method according to claim 22 wherein the support has a BJH pore volume of from 0.8 cm³/g to 1.6 cm³/g.

31. A method according to claim 30 wherein the support has a BJH pore volume of from 1.0 cm³/g to 1.4 cm³/g.

32. A method according to claim 18 wherein the hydroxyl content of the support is from 0.1 to 1.5 mmol/g.

33. A method according to claim 18 wherein the support comprises one or more inorganic oxides.

34. A method according to claim 18 wherein the support comprises silica.

35. A method according to claim 34 wherein the support is substantially a silica support.

36. A method according to claim 18 wherein the support is substantially hydrophobic.

37. A method according to claim 36 wherein the hydrophobicity is due to the presence of alkyl or aromatic moieties.

38. A method according to claim 34 wherein at least a portion of the silica is coated with one or more zirconium salts.

39. A method according to claim 38 wherein the one or more zirconium salts are selected from zirconium oxide, zirconium phosphate and zirconium sulphate.

40. A method according to claim 38 wherein the monovalent cation is selected from one or more OfNH₄ ⁺, Na⁺, K⁺ and Cs⁺.

41. A monovalent cation doped heteropolyacid of general formula I:

Y_xH_(n-x)MX₁₂O₄₀ I

in which M is an atom selected from P or Si;

Y is NH₄ ⁺, Na⁺, K⁺ or Cs⁺;

X is a metal atom selected from W or Mo; n is an integer 3 or 4; and

0 < x < 3.0; provided that when M is P, x is < 2.5.

42. A monovalent cation doped heteropolyacid according to claim 41 wherein M is P.

43. A monovalent cation doped heteropolyacid according to claim 41 wherein n is

3.

44. A monovalent cation doped heteropolyacid according to claim 41 wherein X is W.

45. A monovalent cation doped heteropolyacid according to claim 41 wherein the material of formula I is:

Cs_xHo_-X)PW₁₂O₄₀ II

in which x is as defined in claim 41.

46. A monovalent cation doped heteropolyacid according to claim 41 wherein x is from 0.9 to < 3.0; provided that when M is P, x is < 2.5.

47. A monovalent cation doped heteropolyacid according to claim 41 wherein x is from 2.0 to < 2.5.

48. A monovalent cation doped heteropolyacid according to claim 41 in conjunction with a support material.

49. The use of a monovalent cation doped heteropolyacid according to claim 41 as a catalyst in a liquid phase chemical reaction.

50. The use according to claim 49 wherein the heteropolyacid is in conjunction with a support material.

51. The use according to claim 49 wherein the chemical reaction comprises the esterification of an organic acid.

52. The use according to claim 51 wherein the organic acid is a fatty acid.

53. The use according to claim 52 wherein x is from 0.9 to < 3.0; provided that when M is P, x is < 2.5.

54. The use according to claim 49 wherein the chemical reaction comprises the transesterification of an organic ester.

55. The use according to claim 54 wherein the organic ester is a fatty ester.

56. The use according to claim 54 wherein x is from 2.0 to < 2.5.

57. The use according to claims 51 or 54 wherein the chemical reaction comprises substantially simultaneous esterification and transesterification.

58. The use according to claim 57 wherein x is from 2.0 to < 2.5.

59. The use according to claim 58 wherein x is from 2.0 to 2.3.

60. The use according to claims 51, 54 or 57 wherein feedstock for the reaction is selected from one or more vegetable oils, animal fats and mixtures thereof.

61. The use according to claim 60 wherein the feedstock is selected from the group consisting of beef tallow, corn oil, palm oil, peanut oil, rapeseed oil, soybean oil, sunflower oil, jatropha oil, hungmai oil, mauha oil, nohr oil, sal oil and mixtures thereof.

62. The use according to claim 60 wherein the feedstock comprises used cooking oil.

63. The use according to claims 51, 54 or 57 wherein the alcohol used in the esterification or the transesterification reaction is a primary or secondary monohydric aliphatic alcohol having one to eight carbon atoms.

64. The use according to claim 63 wherein the alcohol is selected from the group consisting of methanol, ethanol, propanol, butanol and amyl alcohol.

65. The use according to claim 64 wherein the alcohol is methanol or ethanol.

66. The use according to claim 65 wherein the alcohol is methanol.

67. The use according to claim 57 which comprises the manufacture of a biofuel.

68. The use according to claim 67 wherein the biofuel is biodiesel.

69. The use according to claim 68 wherein the alcohol used in the esterification, transesterifϊcation and/or the substantially simultaneous esterification and transesterification is methanol.

70. A process for the preparation of a material according to claim 41 which comprises the ionic exchange of a heteropolyacid with a salt of a monovalent cation.

71. A process for the preparation of a material according to claim 48 which comprises mixing a material according to claim 40 with a support material and/or a support material precursor..

72. A process according to claim 70 or 71 wherein the monovalent cation salt is a caesium salt.

73. A process according to claim 72 wherein the caesium salt is a caesium halide.

74. A process according to claim 73 wherein the caesium halide is caesium chloride.

75. A method according to claims 1 or 2 wherein the method is a continuous

54

3THΪOTE SHEET (EULE 2& process.

76. The use of a monovalent cation doped heteropolyacid according to claim 41 in the manufacture of a supported liquid phase catalyst.

77. A method according to claim 2 wherein said column is a high performance liquid chromatography column (HPLC).

78. A method according to claim 77 wherein said column comprises a highly lipophilic polymer.

79. A method according to claim 2 wherein said lipophilic polymer is non- endcapped.

80. A method according to claim 2 wherein said lipophilic polymer comprises a hydrocarbon chain that confers lipophilicity.

81. A method according to claim 80 wherein said hydrocarbon chain comprises at least 19 carbon atoms.

82. A method according to claim 81 wherein said hydrocarbon chain comprises at least 20, 21, 22, 23, 24, 25, 26, 27, 28, or 29 carbon atoms.

83. A method according to claim 80 wherein said hydrocarbon chain comprises at least 30 carbon atoms.

84. A method according to claim 80 wherein said hydrocarbon chain consists of 30 carbon atoms.

85. A method according to claim 2 wherein said method further comprises the steps of: i) detecting and collating the fatty acid ester content of said feedstock; ii) collating the data into a data analysable form; and optionally iii) providing an output for the analysed data.

86. The method, monovalent cation doped heteropolyacid or use substantially as hereinbefore described with reference to the accompanying examples and figures.

0464P.WO.Spec(6)