Process for converting furfuryl alcohol to levulinic acid esters

Abstract

The present invention relates to a method for converting furfuryl alcohol to levulinate, the method comprising contacting furfuryl alcohol, an alcohol or a mixture thereof, and a homogeneous catalyst to form a reaction mixture at a first reaction temperature in the range of 125°C to 180°C to form the levulinate, and forming the levulinate in the reaction mixture, the method being characterized in that the first homogeneous catalyst is a sulfonic acid catalyst.

Description

Invention Field

[0001] This invention provides a method for converting furfuryl alcohol into levulinate, wherein the method generates a small amount of ether byproducts, such as dialkyl ethers. Background Technology

[0002] It is well known that levulinic acid or levulinate esters are produced from furfuryl alcohol and alcohol in the presence of an acidic catalyst.

[0003] For example, U.S. Patent No. 2,738,367 discloses a method in which furfuryl alcohol is converted to levulinic acid in water via a strongly acidic ion exchange resin at a temperature between 30°C and 100°C, with both water and ion exchange resin in large excess compared to the amount of furfuryl alcohol. The catalyst disclosed in US 2,738,367 is a heterogeneous cation exchange resin sold under the trade names "Amberlite IR-120" and "Amberlite IR-105". The reaction is carried out by slowly adding furfuryl alcohol to the reaction mixture, and after the furfuryl alcohol is fed, the reaction continues in a batch mode under the same reaction conditions. A disadvantage of this method is that the heterogeneous catalyst becomes passivated.

[0004] PCT Patent Publication WO 2010 / 102203 (International Patent Application No. PCT / US2010 / 026358) (awarded to Segetis) discloses a method in which a mixture 2 containing furfuryl alcohol and alkanol is added to a mixture 1 containing alkanol, alkyl levulinate (from levulinic acid and the same alkanol), and a protic acid catalyst. The protic acid catalyst is disclosed to be a strong protic acid catalyst selected from "hydrochloric acid, hydrobromic acid, hydroiodic acid, sulfuric acid, pyrosulfonic acid, perchloric acid, benzenesulfonic acid, methanesulfonic acid, trifluoromethanesulfonic acid, p-toluenesulfonic acid, 1-naphthalenesulfonic acid, 2-naphthalenesulfonic acid, chlorosulfonic acid, fluorosulfonyl and its analogues"; alkyl-aromatic sulfonic acids, aromatic sulfonic acids, or aliphatic sulfonic acids; or a weak protic acid selected from phosphoric acid, pyrophosphate, polyphosphoric acid, aminosulfonic acid, alkali metal salts of sulfuric acid (e.g., sodium bisulfate or potassium bisulfate), chloric acid, bromic acid, perbromic acid, iodic acid, and periodic acid. However, only examples using sulfuric acid as a homogeneous catalyst were taught.

[0005] PCT patent publication WO 2007 / 023173 discloses a method in which a porous, strongly acidic ion exchange resin is used as a heterogeneous catalyst to convert furfuryl alcohol with water or alkanol to levulinic acid or alkyl levulinate, respectively. The catalyst lifetime is increased compared to gel-type acidic ion exchange resins. However, once again, the heterogeneous catalyst will become passivated. To suppress the undesirable formation of diethyl ether, Lange et al. (ChemSusChem, Wiley-VCH Verlag, Weinheim; www.chemsuschem.org), 2009, 2, pp. 437-441, “Conversion of Furfuryl Alcohol Into Ethyl Levulinate Using Solid Acid Catalysts”) proposed using a porous, heterogeneous catalyst comprising a strongly acidic ion exchange resin, wherein the catalyst has pores with an average pore size in the range of 1 nm to 1000 nm, such as a macromesh ion exchange resin. Lange et al. and WO2007 / 023173 (hereinafter Lange et al. 2009) investigated various heterogeneous acids, including sulfuric acid, as catalysts for the conversion of furfuryl alcohol to ethyl levulinate. They found that all heterogeneous catalysts were less reactive than sulfuric acid at each acidic center (the strongest acid) (see Figure 5), with the exception of Amberlyst 46, which followed the properties of sulfuric acid itself. Since Amberlyst 46 is a resin containing benzenesulfonic acid groups and is weaker than sulfuric acid, acid concentration did not appear to be critical for the yield of ethyl levulinate under the test reaction conditions. However, Amberlyst 46 is primarily sulfonated on its surface, while the other tested resins also have sulfonic acid groups internally. Therefore, accessibility appears to be important for achieving high yields of ethyl levulinate under the test reaction conditions, and under the same conditions, the source of the acid is not critical to the yield, as long as the acidic centers are readily accessible. Based on those results, skilled technicians would expect that homogeneous sulfonic acids would yield the same amount of acidic protons as sulfuric acid itself when added via a catalyst. Furthermore, Lange et al. (2009) showed that at high ethyl levulinate yields (beyond 84%), the yield of diethyl ethers increased rapidly, exceeding 3 mol% for sulfuric acid (and therefore exceeding 8 mol% for furfuryl alcohol), and even higher for resins containing sulfonic acids (the combination of Figures 5 and 6). Based on these results, a skilled technician would not expect that homogeneous sulfonic acids would improve the yield of alkyl levulinate while simultaneously producing small amounts of dialkyl ethers.

[0006] The number of available articles on homogeneous catalysts for converting furfuryl alcohol to alkyl levulinate using alkanols is limited, not to mention the use of homogeneous sulfonic acids.

[0007] An article using sulfuric acid as a homogeneous catalyst is by Peng et al. (Facile and Efficient Conversion of Furfuryl Alcohol into n-Butyl Levulinate Catalyzed by Extremely Low Acid Concentration, BioResources 2014 9(3), 3825-3834). They used a batch process with sulfuric acid concentrations between 0.01 wt% and 1 wt%, reflecting the low acid concentrations in the title of the article. Yields were slightly above 80%, with better and the same yields obtained at reaction temperatures of 110 °C and 120 °C. Yields up to 97% were obtained, but that was at a furfuryl alcohol concentration of 1 wt% (butanol in batch reactions exceeds 100 times molar excess, which is not economically viable). They did not compare with other homogeneous catalysts, such as homogeneous sulfonic acids, but merely demonstrated in their work that sulfuric acid is not the preferred catalyst for the production of n-butyl levulinate.

[0008] In another article by Zhou et al. (Ethane-Bridged Organosilica Nanotubes Functionalized with Arenesulfonic Acid and Phenyl Groups for the Efficient Conversion of Levulinic Acid or Furfuryl Alcohol to Ethyl LevulinateChem.Cat.Chem.2016(8)2037-2048), the ethanol decomposition of furfuryl alcohol on sulfonic acid-functionalized silica was investigated, and the performance was compared with Amberlyst 15 and sulfuric acid. In this case, although the yield obtained on Amberlyst-15 was higher than that on sulfuric acid, the conversion frequency per acidic center was only slightly higher than that on sulfuric acid, and the performance of sulfonic acid-functionalized silica was the best. Homogeneous sulfonic acid was not used.

[0009] p-Toluenesulfonic acid (p-TSA) was used as a reference material to compare the performance of sulfonated graphene oxide with homogeneous catalysts such as sulfuric acid and p-TSA (Zhu, Chen, Xue, Wu, Wang, and Fan, “Graphene Oxide: An Efficient Acid Catalyst for Alcoholysis and Esterification Reactions”, ChemCatChemCommunications (Wiley-VCH Verlag, Weinheim), 2014, 6, 3080-3083, see Table 1 entries 17 and 18). The reaction was carried out in batches at 80 °C. An aliphatic alcohol was used in excess by more than 20 times the weight of furfuryl alcohol. Under pTSA, 94.6% ethyl levulinate (EL) and 0.6 wt% ethyl methyl furanyl ether (EMF, an intermediate) and 0.5 wt% triethoxypentanone (TEP, another intermediate) were obtained. After all intermediates are converted, approximately 95.6 wt% ethyl levulinate will be obtained (TEP has a higher molecular weight than EL, so 1.1 wt% TEP yields less than 1 wt% EL). In the case of sulfuric acid, 92.5 wt% ethyl levulinate is obtained with 1.7 wt% EMF and 0.1 wt% TEP. The use of less than 10 times the amount of catalyst in the sulfuric acid setting is likely the reason for the slightly lower yield. Because less catalyst is used, the reaction proceeds more slowly, and more intermediates are not yet converted to levulinate. If all intermediates are converted, then approximately 94.2 wt%–94.3 wt% ethyl levulinate will be obtained. Essentially, the two final values ​​for ethyl levulinate are very similar due to the limited accuracy of the catalytic and analytical methods. In practice, extremely low amounts of substrate are used (0.4 g furfuryl alcohol in 10 g of the total mixture, mainly ethanol, reflecting an uneconomical ethanol / furfuryl alcohol ratio), and no internal standard is added to the reaction or used in the GC method. Therefore, it must be assumed that all products are seen in the GC, which is clearly not the case for the obtained heavy components. Thus, the final yield must be an overestimation of the true yield, and the yields of pTSA and sulfuric acid in Table 1 are essentially similar. The conditions (batch feed, higher amounts of catalyst compared to furfuryl alcohol (FFOH / p-TSA = 4 g / g and 0.67 g / g / h, FFOH / H2SO4 = 40 g / g and 6.67 g / g / h), higher amounts of alcohol compared to furfuryl alcohol (EtOH / FFOH = 24 g / g or 51 mol / mol)) are significantly different from our conditions (batch or continuous feed, FFOH / p-TSA = optionally <200 g / g / h; optionally >2 g / g / h, further optionally >5 g / g / h, even further optionally >10 g / g / h), and EtOH / FFOH molar ratio between approximately 1:1 and 5:1).Zhu et al. did not provide any evidence that p-TSA significantly outperformed sulfuric acid, possibly due to the lower reaction temperature used. With an equivalent amount of graphene oxide (0.1 g), 95.5% ethyl levulinate (EL) and 0.3 wt% ethyl methyl furanyl ether (EMF, an intermediate) and 0.1 wt% triethoxypentanone (TEP, another intermediate) were obtained. Zhu et al. concluded that graphene oxide is a “highly efficient, selective, and reusable acid catalyst for the production of alkyl levulinate.”

[0010] Under substantially similar conditions, Rode et al. (Hengne, Kamble, and Rode, “Single pot conversion of furfuryl alcohol to levulinic esters and γ-valerolactone in the presence of sulfonic acid functionalized ILs and metal catalysts”, GreenChemistry (The Royal Society of Chemistry), 2013, 15, 2540-2547) found that using sulfuric acid (0.3 g), 5 g furfuryl alcohol (FFOH / catalyst = 16.667 g / g or 8.333 g / g / h), 95 ml methanol (MeOH / FFOH molar ratio = 46:1, 130 °C, 2 h reaction time), the conversion was 68%, with selectivity of 74% methyl levulinate, 16% methyl furfuryl ether, and 10% other products (see entry 2 in Table 1). Rode et al. focused on sulfonic acid-functionalized ionic liquids (denoted as SO3H-ILs) and carbon-supported Ru, Re, Ir, and Ag catalysts. Using the ionic catalyst liquid [BMIm-SH][HSO4] (1-butylsulfonic acid, 3-methylimidazolium hydrogen sulfate ionic liquid) derived from BMIm-SH and pTSA and containing MIM cations and p-TSA anions, 95% conversion was achieved, yielding 76% methyl levulinate, 11% methyl furfuryl ether, and 13% other products (see entry 12 in Table 1). Clearly, the pTSA ionic liquid is not the same as pTSA; it is a pTSA salt prepared by reacting pTSA (an acid) with a base. Using the sulfuric acid ionic liquid yielded 99% conversion, 95% methyl levulinate, 2% methyl furfuryl ether, and 3% other products (see entry 13 in Table 1). Under an ionic background, sulfuric acid performs slightly better than pTSA (see entries 13 and 12 in Table 1, performed at 130 °C). Therefore, no indication was found that p-TSA might outperform sulfuric acid, but ionic liquids are certainly other molecules. No homogeneous sulfonic acids as defined herein were used for the conversion of furfuryl alcohol to levulinate. Table 1 describes the different catalysts used in Rode. As can be seen from Table 1, sulfuric acid was used as the homogeneous acid and four heterogeneous acids (Amberlyst-15, SO4-ZrO2, 20% DTP-SiO2, 20% DTP-MMT), and the other catalysts were ionic liquids utilizing methylimidazolium, NMP, and BMIm-SH as bases and sulfuric acid, pTSA, trihydrofluoric acid, and chlorosulfonic acid as salts.

[0011] Ionic liquids are salts whose properties are completely different from the acids and bases that make them. They are a new class of pure ionic salt-like materials that are liquid at exceptionally low temperatures. Currently, the "official" definition uses the boiling point of water as a reference point: "These are ionic compounds that are liquid below 100°C." More generally, they have melting points below room temperature; some even below 0°C. These new materials are liquids over a wide temperature range (300-400°C) from the melting point to the decomposition temperature of these compounds. The reason for this difference becomes apparent when comparing a typical ionic liquid such as 1-ethyl-3-methylimidazolium sulfate (mp < -20°C) with a typical inorganic salt such as table salt (NaCl, mp 801°C). Ionic liquids have significantly lower symmetry! Furthermore, the cation and anion charges are distributed resonantly across a larger molecular volume. Therefore, the solidification of ionic liquids occurs at lower temperatures. In some cases, especially if long aliphatic side chains are involved, a glass transition is observed instead of a melting point.

[0012] In Rode's article, an acidic ionic liquid was used, but as mentioned earlier, the properties of the ionic liquid utilizing pTSA differ from those of pTSA. Therefore, it is impossible to extrapolate the results of BMIm-SH / pTSA to pTSA or vice versa.

[0013] The behavior of the ionic liquid as a catalyst in this paper differs from that of pTSA. Rode found that, "However, the selectivity of alkyl levulinate decreased from 99% to 85% as the alcohol chain length increased from ethanol to n-butanol." See page 2542. Figure 2 Below. In comparison, the yields of ethanol and butanol were found to be comparable.

[0014] p-TSA is heterogeneous via a hydrothermal reaction with glucose and acrylic acid (Yang, Zhang, Ao, and Zhang, "Hydrothermal Carbon Enriched with Sulfonic and Carboxyl Groups as an Efficient Solid Acid Catalyst for Butanolysis of Furfuryl Alcohol", Catalysis Communications (Elsevier) 123 (2019) 109-113). The heterogeneous catalyst obtained was compared with p-TSA for the reaction between furfuryl alcohol and butanol. At 120 °C, with 0.196 g furfuryl alcohol, 5.92 g butanol (ButOH / FFOH molar ratio = 40:1), and 0.1 g catalyst (FFOH / p-TSA = 2 g / g or 0.5 g / g / h; 4 h reaction time), 100% conversion was achieved, with a yield of 75% for butyl levulinate and 3.6% for butyl furfuryl ether (Table 2, item 5). Under these conditions, 16% rotten black matter (heavy component) was obtained, see Table S2 entry 5. At a lower temperature (80°C), the conversion was still 100%, but now 94.2 wt% butyl levulinate and 1.5% butyl furfuryl ether were obtained (Table 2 entry 5b). Under these conditions, 2.5% rotten black matter (heavy component) was obtained, see Table S2 entry 5b. Based on those results, skilled technicians will investigate heterogeneous p-TSA at temperatures below 100°C.

[0015] Using p-toluenesulfonic acid as a catalyst, the yield of furfuryl alcohol to levulinic acid via water conversion was lower than that obtained using most of the heterogeneous catalysts tested in the article (An, Song, Sun, Zhang, Zhang, and Guo, "Conversion of Furfuryl Alcohol to Levulinic Acid in Aqueous Solution Catalyzed by Shell Thickness-Controlled Arenesulfonic Acid Functionalized Ethyl-Bridged Organosilica Hollow Nanospheres" ACS Sustainable Chem. Eng. 2018, 6, 3, 3113-3123), and the TOF (transformation frequency) of p-TSA was the lowest to date (Figures 8a and b). The authors state (page 3120): "As shown in Figure 8a, the LA yield increases rapidly at the start of the p-toluenesulfonic acid-catalyzed FAL hydrolysis reaction, reaching 31.5% after 30 minutes of reaction. However, with further increases in reaction time, the increase in LA yield gradually slows down, and the yield is 46.0% within 120 minutes." The authors also state (page 3120): "Both p-toluenesulfonic acid and Amberlyst-15 are super-strong Brønsted acids (…)" (acid), with very high acid center densities (6846 and 4800 μeq g). -1This ensures that they have a significantly high reaction rate in the initial stage of FAL hydrolysis to LA. However, their strong Brønsted acid properties can also promote FAL polymerization in the current reaction system, causing unwanted brown-black oligomers or tarnish (Figures S2 and S3 and Scheme 1). Therefore, the selectivity of p-toluenesulfonic acid or Amberlyst-15 for LA decreases as the reaction proceeds, accompanied by a slow increase or even decrease in LA yield. This is confirmed in the literature, as p-TSA has been found to be a highly efficient catalyst for furfuryl alcohol polymerization (Article 1: M. Principe, P. Ortiz, R. Martínez, “An NMR study of poly(furfuryl alcohol) prepared with p-toluenesulfonic acid”, Polym. Int. 48 (1999) 637-641 and Article 2: Martha Principe, Ricardo Martínez, Pedro Ortiz and Jacques Rieumont, “The Polymerization of Furfuryl Alcohol with p-toluenesulfonic Acid: Photocross-Linkeable Feature of the Polymer Polímeros”). e Tecnologia, Vol. 10, No. 1, pp. 8-14, 2000. Based on these articles, skilled technicians have no incentive to choose homogeneous catalysts such as p-TSA as catalysts because oligomerization is expected to be an important side reaction with relatively low yields.

[0016] Previous methods have all produced ethers, such as dialkyl ethers, as byproducts, although this is not usually confirmed. It is assumed that ethers not only reduce the recovery rate of alcohols (due to excess alcohol usage, which therefore requires recycling for economic reasons), but also pose safety concerns in the plant, as alcohols are often recovered via distillation, and ethers will automatically recycle along with the alcohols, accumulating in the process. Furthermore, ethers, such as dialkyl ethers, have low autoignition temperatures (160°C and 175°C for diethyl ether and dibutyl ether, respectively), thus posing potential safety issues that need to be kept in mind during plant operation and construction.

[0017] Yang et al. (Catalysis Communications, 123 (2019), pp. 109-112) reported on the use of hydrothermal carbon enriched with sulfonic and carboxyl groups as a solid acid catalyst for the butanol decomposition of furfuryl alcohol. Entry 5 of Table 2 compares Yang's catalyst with pTSA in a Parr batch reactor under reaction conditions containing 0.196 FAL, 5.92 g n-butanol, 0.1 g catalyst, 120 °C, and 4 h. Table 2 reports a yield of 75% for butyl levulinate and 3.6% for butoxymethylfuran (BMF) at 120 °C. The remaining product of 100% conversion is uncertain (refer to Table S2, except for up to 16.0% blackening). Entry 5 of Table 2 provides further yields at 80 °C, of ​​94.2% and 1.5%, respectively, indicating that the yield of butyl levulinate decreases as the temperature rises to 120 °C. Therefore, in Yang's Table 2, pTSA was used at two different temperatures—120°C and 80°C. Complete conversion was achieved at both temperatures under Yang's conditions, but at 80°C, the yield and therefore selectivity (94.2%) were significantly higher than at 120°C (75%). In Yang, the furfuryl alcohol / pTSA ratio was 0.196 g / 0.1 g = 1.92, while in this study, the catalyst was between 0.1 wt% and 5 wt% of the total reaction mixture, and the alcohol / furfuryl alcohol molar ratio was between 1 / 1 and 5 / 1. Therefore, even with 5 / 1 and 5 wt% catalyst, furfuryl alcohol was still at least 3 times more abundant than the catalyst. The better conditions in this study would be 1 wt% catalyst and an alcohol / furfuryl alcohol ratio of 2; in such a case, the furfuryl alcohol / catalyst ratio would be closer to 33. Ultimately, from Yang, a skilled technician would expect to find the optimal temperature at a lower temperature, then 120°C, not at the higher reaction temperature required in this study.

[0018] Garcia-Suarez et al. (Applied Catalysis A, General, 569 (2019), pp. 170-174) discussed the Pd catalysis of certain ester products. Item 1 in Table 2 used furfuryl alcohol as the substrate. The catalysts were 0.025 mmol Pd(OAc)₂, 1 mmol DTBPMB (1,2-bis(di-tert-butylphosphinomethyl)benzene) ligand (0.125 mmol), and MSA (methanesulfonic acid). The remaining reaction conditions were substrate (1 mmol), 1-hexene (1.8 mmol), and MeOH (5 mL), at 120 °C for 20 hours. Although the conversion of furfuryl alcohol was >99%, the yield was only 19.3% (methyl heptanoate (MH), 21.4% methyl levulinate (ML), and 19.5% γ-valerol (GVL). The remaining substances with >99% conversion were not determined. The amount of catalyst used (1 mmol) was equal to that of furfuryl alcohol (1 mmol), and the amount of methanol (5 ml) was in large excess, as it was approximately 200 mmol (MW = 32 and density 0.792 g / ml). Furthermore, the selectivity for the acid conversion of furfuryl alcohol to the desired product was rather low, at 21.4 + 19.5 = 40.9%. % Methyl heptanoate originates from the reaction of 1-hexene with carbon monoxide from furfuryl alcohol. Therefore, at least 19.3% of furfuryl alcohol decomposes to produce carbon monoxide, as item 6 shows that methyl levulinate does not produce carbon monoxide. Therefore, based on this experiment and combined with the Yang mentioned above, a skilled technician would not convert furfuryl alcohol with pTSA at 120°C, but instead would use a lower temperature, such as 80°C. Of course, the low yield of the desirable acid decomposition products in item 1 of Table 2 may be a result of the co-present Pd in ​​the reaction, but a skilled technician is discouraged from carrying out the reaction at 120°C in the absence of Pd.

[0019] The objective of this invention is to optimize the yield of levulinic ester while minimizing the generation of ether byproducts. Summary of the Invention

[0020] It has now been found that the yield achieved by converting furfuryl alcohol to homogeneous sulfonic acid in levulinate using alcohol at temperatures above 125°C is unexpectedly better than that achieved with sulfuric acid.

[0021] According to the present invention, a method for converting furfuryl alcohol to levulinate is provided, the method comprising contacting furfuryl alcohol, an alcohol or a mixture thereof, and a homogeneous catalyst to form a reaction mixture at a first reaction temperature in the range of about 125°C to about 180°C to form levulinate, and forming levulinate in the reaction mixture, the method being characterized in that the first homogeneous catalyst is a sulfonic acid catalyst.

[0022] Optionally, the first homogeneous catalyst consists of a sulfonic acid catalyst. In other words, no other homogeneous or heterogeneous catalysts exist.

[0023] Optionally, the reaction is carried out at a temperature between about 125°C or 130°C and 170°C, about 125°C or 130°C and 160°C, about 125°C or 130°C and about 150°C, and further optionally about 140°C.

[0024] Optionally, the sulfonic acid catalyst is selected from the group consisting of alkyl-aromatic sulfonic acids, aromatic sulfonic acids, halosulfonic acids, and aliphatic sulfonic acids; optionally, the sulfonic acid catalyst is p-toluenesulfonic acid or its monohydrate, 7,7-dimethyl-2-oxobicyclo[2.2.1]heptane-1-yl)methanesulfonic acid (camphorsulfonic acid) or methanesulfonic acid; or hydrates of each thereof. Salts of sulfonic acid catalysts are excluded.

[0025] Optionally, the sulfonic acid catalyst is an alkyl-aromatic sulfonic acid or aromatic sulfonic acid selected from the group consisting of: benzenesulfonic acid, p-toluenesulfonic acid, dodecylbenzenesulfonic acid, 1-naphthalenesulfonic acid, and 2-naphthalenesulfonic acid; or the sulfonic acid catalyst is an aliphatic sulfonic acid, optionally a straight-chain, branched, or cyclic alkyl sulfonic acid, such as methanesulfonic acid, trifluoromethanesulfonic acid, octanesulfonic acid, perfluorooctanesulfonic acid, and 7,7-dimethyl-2-oxobicyclo[2.2.1]heptane-1-yl)methanesulfonic acid (camphorsulfonic acid); or a halosulfonic acid selected from chlorosulfonic acid, 3-chloromethanesulfonic acid, 3-fluoromethanesulfonic acid, and fluorosulfonic acid; or hydrates of each thereof. Salts of such sulfonic acid catalysts are excluded.

[0026] Further optionally, the sulfonic acid catalyst is p-toluenesulfonic acid or p-toluenesulfonic acid monohydrate; or camphorsulfonic acid or camphorsulfonic acid monohydrate; or methanesulfonic acid or its hydrate.

[0027] The sulfonic acid catalyst used in this invention is not a sulfonic acid-functionalized ionic liquid.

[0028] Optionally, the sulfonic acid catalyst comprises about 0.1% to about 5% (w / w), about 0.1% to about 3% (w / w), or preferably about 0.5% to about 3% (w / w) of the reaction mixture; further preferably about 0.5% to about 1.5% (w / w), and even more preferably about 1% (w / w); and / or the molar ratio of alcohol to furfuryl alcohol in the reaction mixture is between about 1:1 and 5:1, or preferably between about 1.1:1 and 3:1; further preferably between about 1.2:1 and 2.7:1; even more preferably about 2.2:1 or about 1.5:1; and / or wherein furfuryl alcohol is fed into the reaction mixture at a feed rate of up to 200 g of furfuryl alcohol per gram of catalyst per hour, or preferably up to 100 g / g / h, and even more preferably up to 50 g / g / h.

[0029] Optionally, the alcohol is a primary or secondary alcohol selected from C1-24 straight-chain or branched alcohols; further optionally selected from methanol, ethanol, n-propanol, isopropanol, n-butanol, isobutanol, 2-butanol, n-pentanol, isopentanol, n-hexanol, n-octanol, n-decanol or 2-ethylhex-1-ol; or alkoxy-alkanol, optionally alkoxyethanol, such as β-methoxyethanol or β-ethoxyethanol; or a combination of two or more of these.

[0030] Alternatively, the alcohol is selected from alicyclic alcohols, optionally cyclohexanol, cyclopentanol, tetrahydrofurfuryl alcohol, and 5-methyl-2-(propane-2-yl)cyclohex-1-ol; or unsaturated alicyclic alcohols, optionally propan-2-en-1-ol, 3,7-dimethyloct-2,6-dien-1-ol, and propan-2-yn-1-ol; or diols as primary alcohols, optionally ethylene glycol, 1,3-propanediol, and 1,4-butanediol; or combinations of two or more of these.

[0031] Alternatively, the alcohol is a mixture of one or more primary and secondary alcohols; and / or a mixture of one or more alicyclic alcohols; and / or a mixture of one or more unsaturated alicyclic alcohols; optionally selected from fusel alcohols and Guerbet alcohols; and a mixture of one or more diols.

[0032] Optionally, the method is carried out in a batch feed mode; or in a continuous stirred tank reactor (CSTR), optionally followed by a plug flow reactor (PFR).

[0033] Optionally, the contact is carried out in a batch-feed reactor.

[0034] When carried out in a batch feed mode, the reaction mixture may be formed as follows: (i) a first mixture comprising an alcohol or a mixture thereof, and optionally a homogeneous sulfonic acid catalyst, is provided in a batch feed reactor; (ii) a second mixture is continuously or intermittently fed to the first mixture in the batch feed reactor, the second mixture comprising furfuryl alcohol and additional amounts of alcohol or mixture thereof; and a third mixture is continuously or intermittently fed to the batch feed reactor, separate from the second mixture, wherein if the first mixture contains a homogeneous sulfonic acid catalyst, the third mixture contains (iii) additional amounts of a homogeneous sulfonic acid catalyst; or if the first mixture does not contain a homogeneous sulfonic acid catalyst, the third mixture contains a homogeneous sulfonic acid catalyst.

[0035] Optionally, a third mixture, fed continuously or intermittently to a batch feed reactor, separate from the second mixture, begins at the same time as the continuous or intermittent feeding of the second mixture. The third mixture may be fed to the batch feed reactor after the second mixture has been continuously or intermittently fed. However, it is undesirable to wait too long, because if the catalyst concentration in the batch feed reactor drops too low, the furfuryl alcohol concentration in the batch feed reactor will increase, negatively impacting the yield.

[0036] When the method is carried out in a batch feed mode, optionally, the first mixture is heated to a first reaction temperature before the second and third mixtures are mixed with the first mixture. Further optionally, the alcohol is heated to the first reaction temperature before the catalyst is added to form the first mixture. Alternatively or additionally, the alcohol may be heated above its boiling point.

[0037] When the method is carried out in a batch feed mode, after the second mixture and the first mixture have been completely mixed, the first temperature may be maintained for up to about 2 hours, optionally about 5 minutes to 60 minutes; or alternatively, after the second mixture and the first mixture have been completely mixed, the first temperature may be raised to a second reaction temperature. Optionally, the reaction mixture may be maintained at the second reaction temperature for 5 minutes to 120 minutes.

[0038] Alternatively, the method is continuous and carried out in a continuous stirred tank reactor. Optionally, multiple CSTRs are connected in series, wherein, more optionally, the first CSTR causes at least 90% conversion of furfuryl alcohol.

[0039] When the method is carried out in continuous mode, the continuous stirred tank reactor may be filled with the reaction mixture as follows: (i) a first mixture comprising an alcohol or a mixture thereof, and optionally a homogeneous sulfonic acid catalyst; (ii) a second mixture continuously or intermittently fed into the reactor of the first mixture comprising furfuryl alcohol, and additional amounts of an alcohol or a mixture thereof; and a third mixture, separately from the second mixture, continuously or intermittently fed into the reactor, wherein if the first mixture contains a homogeneous sulfonic acid catalyst, the third mixture contains (iii) additional amounts of a homogeneous sulfonic acid catalyst; or if the first mixture does not contain a homogeneous sulfonic acid catalyst, the third mixture contains a homogeneous sulfonic acid catalyst.

[0040] Optionally, the third mixture, fed continuously or intermittently to a continuous stirred tank reactor, is separate from the second mixture and begins at the same time as the continuous or intermittent feeding of the second mixture. The third mixture may be fed to the continuous stirred tank reactor after the second mixture has been continuously or intermittently fed. However, it is not desirable to wait too long, as if the catalyst concentration in the reactor drops too low, the furfuryl alcohol concentration in the reactor will increase, negatively impacting the yield.

[0041] When the method is carried out in continuous mode, the first mixture is heated to a first reaction temperature before the second and third mixtures are mixed with the first mixture.

[0042] When the process is carried out in continuous mode, a sulfonic acid catalyst is selected such that the sulfonic acid catalyst accounts for about 0.1% to about 5% (w / w) of the reaction mixture, optionally about 0.5% to about 3% (w / w); further optionally about 0.5% to about 1.5% (w / w), and even more optionally about 1% (w / w).

[0043] Optionally, and regardless of whether the method is carried out in a batch feed mode or a continuous mode, the second mixture is added to the reaction mixture at a feed rate of up to 200 g of furfuryl alcohol per gram of catalyst per hour, optionally up to 100 g / g / h, and more optionally up to 50 g / g / h.

[0044] Optionally, alternatively, or otherwise, and regardless of whether the method is carried out in a batch feed mode or a continuous mode, the second mixture is mixed with the reaction mixture at a feed rate of at least 2 g furfuryl alcohol per gram of catalyst per hour, optionally at least 5 g / g / h, further optionally at least 10 g / g / h, and even more optionally at least 20 g / g / h.

[0045] Optionally, and regardless of whether the method is carried out in a batch feed mode or a continuous mode, the total molar ratio of alkanol to furfuryl alcohol contained in the first and second mixtures is between about 1:1 and 5:1, optionally between 1.1:1 and 3:1; further optionally between about 1.2:1 and 2.7:1; and even more optionally about 2.2:1 or about 1.5:1.

[0046] Optionally, and regardless of whether the method is carried out in a batch feed mode or a continuous mode, the molar ratio of alcohol or a mixture thereof to furfuryl alcohol in the reaction mixture is at least 20:1, optionally at least 50:1, further optionally at least 100:1, and even more optionally at least 200:1.

[0047] When the method is carried out in continuous mode, once the continuous stirred tank reactor has been filled to the required volume, a certain volume of the reaction mixture is withdrawn from the continuous stirred tank reactor; and the required volume is replenished by continuously or intermittently feeding a second mixture into the reactor; and a third mixture is fed separately, continuously or intermittently into the reactor, wherein the withdrawn volume is optionally fed into a second reactor connected in series with the continuous stirred tank reactor, optionally a plug flow reactor, a second CSTR, or a batch feed reactor. Thus, the continuous feeding of the second and third mixtures continues, and simultaneously, a volume equal to the total volume of the two mixtures together is continuously discharged from the CSTR into the second reactor, optionally into a plug flow reactor, a second CSTR, or a batch feed reactor. In fact, continuous feeding and discharging are the definition of a CSTR, so when the continuous stirred tank reactor has been filled to the required volume, the entire system switches from batch feed to CSTR operation mode.

[0048] Optionally, if the method is carried out in batches or in CSTRs followed by a second reactor, such as one or more CSTRs, or followed by PFRs, then higher temperatures can be applied in one or more second reactors after the first reactor.

[0049] Optionally, the extracted volume is fed into a second reactor, such as a plug flow reactor, connected in series with the continuous stirred tank reactor, and the method is carried out in the second reactor at a second reaction temperature that is the same as or 5°C to 15°C higher than the first reaction temperature in the continuous stirred tank reactor; and / or the residence time in the second reactor is in the range of 5 minutes to 120 minutes, optionally in the range of about 10 minutes to 60 minutes.

[0050] Optionally, the reaction mixture is maintained at atmospheric pressure, or, if the reactor is closed, at autogenous pressure (e.g., a pressure increase of up to 10 bar for methanol). Further optionally, the reaction mixture is maintained at the autogenous pressure of the fluid (i.e., not externally applied, but solely by the airtight nature of the vessel and...). The relationship between the pressure generated by P and T (its state equation).

[0051] Optionally, the reaction mixture is maintained at the first reaction temperature without further addition of furfuryl alcohol until the furfuryl alcohol concentration, by weight of the total reaction mixture, is less than 1% (w / w), optionally less than 0.5% (w / w), further optionally less than 0.25% (w / w), and even more optionally less than 0.01% (w / w). Optionally or additionally, the addition rate is such that the furfuryl alcohol concentration in the reactor alcohol, by weight of the total reaction mixture, is less than 1% (w / w), optionally less than 0.5% (w / w), further optionally less than 0.25% (w / w), and even more optionally less than 0.01% (w / w).

[0052] Optionally, the method further includes collecting the formed levulinate, wherein the collection comprises evaporating the alcohol, distilling the levulinate; recovering or neutralizing and removing the catalyst; decanting the soluble reaction product from the insoluble reaction product, filtering the insoluble reaction product from the soluble reaction product, liquid-liquid extraction, or a combination thereof; wherein, further optionally, the distillation comprises vacuum distillation, scraped-film evaporation, or falling-film evaporation. Still further optionally, the collection is carried out in two steps, wherein the first step comprises evaporating the alcohol and the second step comprises evaporating the levulinate; or wherein the first step comprises evaporating the alcohol and a first portion of the levulinate, and the second step comprises removing a second portion of the levulinate.

[0053] Optionally, the byproduct containing levulinic acid is esterified to form levulinic acid ester.

[0054] Optionally, at the end of the reaction according to the method of the invention and before purification of its reaction product, the crude reaction mixture contains levulinate in a yield of at least about 70 mol% based on furfuryl alcohol. In some embodiments, the crude reaction mixture contains levulinate in a yield of at least about 80 mol% based on furfuryl alcohol. In some embodiments, the crude reaction mixture contains levulinate in a yield of up to about 85 mol% based on furfuryl alcohol. In some embodiments, the crude reaction mixture contains levulinate in a yield of up to about 90 mol% based on furfuryl alcohol.

[0055] Optionally, at the end of the reaction according to the method of the invention and before purification of its reaction product, the crude reaction mixture contains total levulinic acid (i.e., levulinate plus levulinic acid) in a yield of up to about 70 mol% based on furfuryl alcohol. In some embodiments, the crude reaction mixture contains total levulinic acid (i.e., levulinate plus levulinic acid) in a yield of up to about 80 mol% based on furfuryl alcohol. In some embodiments, the crude reaction mixture contains total levulinic acid (i.e., levulinate plus levulinic acid) in a yield of up to about 85 mol% based on furfuryl alcohol. In some embodiments, the crude reaction mixture contains total levulinic acid (i.e., levulinate plus levulinic acid) in a yield of up to about 90 mol% based on furfuryl alcohol. In some embodiments, the crude reaction mixture contains total levulinic acid (i.e., levulinate plus levulinic acid) in a yield of up to about 95 mol% based on furfuryl alcohol. In some embodiments, the soluble material in the crude reaction mixture is determined by analytical methods such as proton NMR, HPLC or GC / MS to be free of furfuryl alcohol, containing only alkanols, levulinic acid, dialkyl ethers and levulinic acid esters, and trace amounts of heavy components and other byproducts (e.g., furfuryl alcohol dimers, butyl-m-ol condensates of levulinic acid esters).

[0056] Optionally, at the end of the reaction according to the method of the invention and before / after purification of its reaction products, observable reaction byproducts are tarry or oily residues, which in some embodiments account for about 20% by weight or less of the added furfuryl alcohol. In some embodiments, the tarry or oily residues account for 10% by weight or less of the added furfuryl alcohol. In some embodiments, the tarry or oily residues account for 5% by weight or less of the added furfuryl alcohol. The content of the tarry / oily residues is measured by evaporating the alkanol and levulinate from the reaction vessel and weighing the remaining material or an aliquot thereof.

[0057] Optionally, at the end of the reaction according to the method of the invention and before purifying its reaction product, the crude reaction mixture contains up to 8 mol% ether based on furfuryl alcohol, which is freshly prepared once and added to the recycled amount of dialkyl ether already present in the feed. In some embodiments, the crude reaction mixture is prepared once to yield up to about 5 mol% ether based on furfuryl alcohol. In some embodiments, the crude reaction mixture is prepared once to yield ether in a yield of up to about 2 mol% based on furfuryl alcohol. In some embodiments, the crude reaction mixture is prepared once to yield ether in a yield of up to about 1 mol% based on furfuryl alcohol. The dialkyl ether is recycled along with excess alcohol, so the amount prepared once in the reactor is important. The amount of dialkyl ether in the recycled alcohol is controlled by a purge stream to ensure safe operation of the method.

[0058] Under the reaction conditions of the method according to the invention, a higher yield of levulinic ester is obtained, and the amount of ether produced is less than that produced using industrial homogeneous inorganic acids such as sulfuric acid.

[0059] Furthermore, homogeneous sulfonic acid catalysts are cheaper than heterogeneous acid catalysts, and again, they are less prone to passivation, as heterogeneous acid catalysts are passivated due to the deposition of heavy components and are often not economically recyclable. The dosage of homogeneous acid catalysts is often 1 to 20 times lower than that of heterogeneous catalysts. Therefore, the higher productivity, lower dosage, and lower economic cost of homogeneous catalysts make post-reaction recycling unprofitable, and they are often deactivated by neutralization with alkali.

[0060] For example, each gram of heterogeneous catalyst yields approximately 5-10g of product, after which the catalyst cannot be recycled. In this paper, for example, more than 50g of product can be obtained per gram of catalyst, and as already mentioned, homogeneous catalysts are also less expensive than heterogeneous catalysts.

[0061] Detailed embodiments of the present invention

[0062] In the method according to the invention, in order to obtain a high yield of levulinic ester and a low amount of ether byproducts, it is important to maintain a consistently low concentration of furfuryl alcohol and a consistently high molar ratio of alcohol to furfuryl alcohol. This can be achieved by the following:

[0063] 1. Batch feeding reactor. In this case, at least a portion of the alcohol and optionally at least a portion of the sulfonic acid homogeneous catalyst are loaded into the reactor, while a mixture of furfuryl alcohol and optionally alcohol is simultaneously fed into the reactor. Optionally, a portion of the sulfonic acid homogeneous catalyst is also fed into the reactor in a separate feed line.

[0064] 2. A CSTR that converts furfuryl alcohol to at least 90%, more preferably 95%, and even more preferably 98%, wherein furfuryl alcohol and a homogeneous catalyst are separately fed to the CSTR and the molar ratio of alcohol to furfuryl alcohol is between 1:1 and 5:1, more preferably between 1.2:1 and 3:1, and even more preferably between 1.5:1 and 2.7:1. The CSTR may be followed by a second CSTR, or multiple CSTRs in series, or a PFR (plug flow reactor or tube).

[0065] Preferably, the conversion reaction is carried out in the liquid phase. The advantage of a liquid-phase reaction is that any liquid byproducts that may form, such as oligomerization products of furfuryl alcohol and / or levulinate, will remain in solution. Therefore, preferably, the liquid reaction mixture containing furfuryl alcohol and the alcohol is contacted with a first homogeneous catalyst.

[0066] To limit the amount of byproducts (especially oligomeric condensates of furfuryl alcohol) formed in the contact step of claim 1 herein, the concentration of furfuryl alcohol is kept below 1 wt%, preferably below 0.5 wt%, based on the total weight of the reaction mixture. Typically, the concentration of furfuryl alcohol in the first CSTR or in the reactor of a batch-feed mechanism is below 0.1 wt%, preferably below 0.01 wt%. The concentration must be low enough to avoid furfuryl alcohol polymerization (must be below 1 wt%), and high enough to obtain a good reaction rate.

[0067] A low concentration of furfuryl alcohol in the reaction mixture can be achieved in several ways. The reaction mixture can be diluted, for example, with an excess of another reactant, i.e., an alcohol, or with a reaction product, i.e., levulinate, or with a diluent that is neither a reactant nor a reaction product, such as sulfolane, γ-valerate, or a carboxylic acid ester. Alternatively or additionally, furfuryl alcohol can be supplied to the reaction mixture in stages, or it can be used in a continuous stirred tank reactor (CSTR) operating at high conversion rates to maintain a sufficiently low concentration of furfuryl alcohol in the reaction mixture.

[0068] The reaction mixture may be maintained in contact with the first homogeneous catalyst at a first reaction temperature in the range of 125°C to 180°C without further addition of furfuryl alcohol to the reaction mixture until the concentration of furfuryl alcohol is less than 0.01 wt%, more preferably less than 0.005 wt%, and even more preferably less than 0.001 wt%, each based on the total weight of the reaction mixture.

[0069] To achieve high conversion of furfuryl alcohol with high selectivity, the alcohol is present in stoichiometric excess in the reaction mixture. Therefore, the molar ratio of alcohol to furfuryl alcohol in the reaction mixture is at least 20:1, more preferably at least 50:1, even more preferably at least 100:1, and even more preferably at least 200:1.

[0070] This can be achieved by feeding an alcohol and furfuryl alcohol to the reaction mixture in a molar ratio (the combined total of the first and second mixtures in both batch-feed reactor and continuous stirred tank reactor embodiments) between about 1:1 and 5:1, optionally between about 1.1:1 and 3:1, further optionally between about 1.2:1 and 2.7:1, and even more optionally between about 2.2:1 or about 1.5:1.

[0071] The method according to the invention can be carried out in a batch feeding or continuous mode. When furfuryl alcohol is continuously fed to the first homogeneous sulfonic acid catalyst, it is preferably supplied at a feeding rate of up to 200 grams of furfuryl alcohol per gram of catalyst per hour, more preferably up to 100 g / g / h, and even more preferably up to 50 g / g / h (grams of furfuryl alcohol per gram of catalyst in the reaction mixture per hour). In a preferred embodiment, the method is carried out in a continuous stirred tank reactor (CSTR). One advantage of such a CSTR reactor is that the furfuryl alcohol fed to the reactor is rapidly mixed into the reaction mixture, thus avoiding high concentrations of furfuryl alcohol approaching the homogeneous sulfonic acid catalyst.

[0072] The method according to the invention is carried out at a first reaction temperature in the range of 125°C to 180°C, preferably in the range of 130°C to 150°C, and more preferably in the range of 140°C. The upper limit depends on the rate of formation of the major byproduct (formed from furfuryl alcohol) and the minor byproduct (formed from levulinate, for example, the Diels-Alder reaction) and avoids excessive ether formation (unless, of course, ether formation is desired).

[0073] When the method is carried out in continuous mode, the volume extracted from the first CSTR can be supplied to a second reactor, which is a second CSTR or a subsequent CSTR, or a plug flow reactor, hereinafter referred to as the "second reactor". The reaction temperature in the second reactor, i.e., the second reaction temperature, can be equal to or higher than the first reaction temperature.

[0074] The pressure at which the reactants come into contact with the catalyst is not critical. Preferably, to avoid evaporation of the reactants, the reactor is sealed and the pressure is at least the autogenous pressure of the reaction mixture at the temperature at which the conversion reaction takes place.

[0075] Preferably, the method according to the invention is a method for converting furfuryl alcohol into alkyl levulinate (levulinate ester) by contacting furfuryl alcohol with an alkanol.

[0076] Alkyl alcohols are preferably alkyl alcohols having up to 24 carbon atoms, and even more preferably selected from methanol, ethanol, isopropanol, n-propanol, n-butanol, 2-butanol, n-pentanol, 2-ethylhex-1-ol; or alicyclic alcohols, preferably selected from cyclohexanol; or combinations of two or more of them. Ethanol and n-butanol are particularly preferred. Attached Figure Description

[0077] Figure 1 A schematic diagram of the experimental setup used in Examples 1, 2 and 4 is shown for the batch feeding experiments.

[0078] Figure 2 and 3A schematic diagram of an experimental setup is shown, with a continuous stirred tank reactor in the background and a plug flow reactor in the background. Figure 2 In this process, furfuryl alcohol and butanol are provided, premixed in the desired ratio at T1. Figure 3 In this process, furfuryl alcohol and butanol are fed separately to SM1 (static mixer 1) at a feed rate that achieves the required ratio in SM1. Figure 3 The device shown is the device used in Example 3.

[0079] Figure 4 A schematic diagram of the experimental equipment for the continuous operation device is shown. Detailed Implementation

[0080] The figure mentions butanol as an alcohol. However, it should be understood that butanol is merely one example of an alcohol that can be used to carry out the present invention.

[0081] Figure 1 A schematic diagram of the experimental setup for the batch feed experiment is shown. The main components are: (T1) a feed furfuryl alcohol / butanol tank, (P) a pump, and (R1) a stirred reactor.

[0082] Figure 2 A schematic diagram of a continuous stirred tank reactor is shown. The main components are: (T1) feed furfuryl alcohol / butanol tank, (T3) catalyst tank, (R1) stirred reactor, and (PFR) plug flow reactor.

[0083] Figure 3 A schematic diagram of the replacement of the continuous stirred tank reactor is shown. The main components are: (T1) furfuryl alcohol feed tank, (T2) butanol feed tank, (SM1) static mixer for mixing furfuryl alcohol and butanol feeds, (T3) catalyst tank, (R1) stirred reactor, and (PFR) plug flow reactor.

[0084] This article references Figures 1 to 3 The description states that several variations are possible.

[0085] Change A: Therefore, using Figure 1The experimental setup was used in Examples 1, 2, and 4 in a batch feed mode. Reactor R1 (1000 mL) was loaded with 219.14 g butanol (2.94 mol, 99.4% purity) and 0.032 mol of acid catalyst (6.05 g p-toluenesulfonic acid monohydrate (p-TSA, 99% purity)). The butanol and catalyst were then mixed (600 rpm), and the mixture was heated to 130°C. Once the reaction temperature was reached, 205.79 g butanol (2.76 mol) was fed via pipe 1 and pump (P). A mixture of butanol (99.4% purity) and 205.54 g of furfuryl alcohol (FFOH) (2.05 mol, 98% purity) (molar ratio 1.35:1) was prepared. The mixture was continuously fed over a period of 410 minutes to ensure that the furfuryl alcohol concentration in the reactor remained below 0.2 wt%. The reaction temperature in reactor R1 was maintained at 130 °C. The final total molar ratio of butanol / furfuryl alcohol was 2.78; the molar ratio of furfuryl alcohol / pTSA was 65.2 and the weight ratio was 33.6.

[0086] After the feed is complete, the reactor (R1) is maintained at the reaction temperature for 2 hours to allow the final trace amounts of furfuryl alcohol and all intermediates to be converted into butyl levulinate.

[0087] The examples in this article have used variation A, which uses a premix of butanol and furfuryl alcohol, except for example 3, which uses variation C.

[0088] Change B (see Figure 2 Similar to variation A. After loading reactor R1 with 219.14 g butanol (2.94 mol, 99.4% purity) and 0.032 mol of acid catalyst (6.05 g p-toluenesulfonic acid monohydrate (p-TSA, 99% purity), via pump and pipe 3), the butanol and catalyst are mixed (600 rpm), and the mixture is heated to 130°C. Once the reaction temperature is reached, a mixture of 205.79 g butanol (2.76 mol, 99.4% purity) and 205.54 g furfuryl alcohol (FFOH) (2.05 mol, 98% purity) (molar ratio 1.35:1) is fed via pipe 1, pump, and pipe 4. R1 can be a batch-feed reactor. Alternatively, when loaded, R1 can be converted to a continuous stirred tank reactor, optionally fed forward into a plug flow reactor (PFR).

[0089] Change C (see Figure 3The reaction is similar to variation A. After loading 219.14 g butanol (2.94 mol, 99.4% purity, from T2, via pipes 2 and 4 (and a pump and static mixer (SM1))) and 0.032 mol of acid catalyst (6.05 g p-toluenesulfonic acid monohydrate (p-TSA, 99% purity), from T3, via pipe 3) into reactor R1, the butanol and catalyst are mixed (600 rpm), and the mixture is heated to 130°C. Once the reaction temperature is reached, a mixture of 205.79 g butanol (2.76 mol, 99.4% purity) and 205.54 g furfuryl alcohol (FFOH) (2.05 mol, 98% purity) (molar ratio 1.35:1) is fed via pipes 2 and 1, static mixer 1 (SM1), and pipe 4. R1 can be a batch-fed reactor. Alternatively, when loading, R1 can be converted into a continuous stirred tank reactor, optionally fed forward into a plug flow reactor.

[0090] Variation D: This variation is the same as variations A through C, but after the feed is complete, the reactor (R1) is maintained at a temperature 15°C higher than the reaction temperature for 0.5 hours to allow the final trace amounts of furfuryl alcohol and all intermediates to convert to butyl levulinate. This reduces the cycle time in batch feed experiments.

[0091] Change E: Same as change A, but with reference to... Figure 3 The differences are as follows.

[0092] Reactor R1 (1000 mL) was loaded with 219.14 g of butanol (2.94 mol, 99.4% purity). The butanol and catalyst were then mixed (600 rpm) and heated to 130 °C. Once the reaction temperature was reached, a mixture of 205.79 g of butanol (2.76 mol, 99.4% purity, from T2 via tube 2) and 205.54 g of furfuryl alcohol (from T1 via tube 1) (2.05 mol, 98% purity) (molar ratio 1.35:1) was started. Both were mixed in a static mixer SM1 and fed into reactor R1. Simultaneously with the start of the mixture feeding, 0.032 mol of acid (6.05 g of p-toluenesulfonic acid monohydrate (p-TSA, 99% purity)) was started from T3 via a separate tube 3. The mixture and catalyst were continuously fed over a period of 410 minutes, ensuring that the furfuryl alcohol concentration in the reactor remained below 0.2 wt%. All other conditions are the same as those for variation A or B.

[0093] The final total molar ratio of butanol / furfuryl alcohol is 2.78; the molar ratio of furfuryl alcohol / pTSA is 65.2 and the weight ratio is 33.6.

[0094] After the feed is complete, reactor R1 can be maintained at the reaction temperature for 2 hours to allow the final trace amounts of furfuryl alcohol and all intermediates to convert to butyl levulinate. The expected advantage of change E is that it will produce a significantly reduced amount of dibutyl ether.

[0095] Figure 4 A schematic diagram of the experimental apparatus for a continuous operation device is shown. The main components are: (T1) a furfuryl alcohol feed tank, (T2) a butanol feed tank, (SM1) a static mixer for mixing furfuryl alcohol and butanol feeds, (T3) a catalyst tank, and (R1) a continuous stirred tank reactor (CSTR). R1 ​​is connected to a plug flow reactor (PFR), which is essentially a long tube. The PFR is connected to an alcohol distillation column (C1), the top of which can be recirculated to the butanol tank (T2) via pipe 7, and the bottom fed to the levulinic acid column (C2) via pipe 8. Levulinic acid esters are collected from the top of the column in T5, and heavy components are collected from the bottom in T4.

[0096] Initiation and about continuous mode reactions Figure 1 The batch feeds described are identical, and the operating modes can be variations of each of A to E. Once reactor R1 is loaded, reactor (R1) begins to function as a CSTR reactor. That is, a mixture of furfuryl alcohol and butanol in, for example, a molar ratio of 1.3 is continuously fed to R1, and simultaneously, optionally dissolved in an alcohol (e.g., butanol), catalyst pTSA is continuously fed to R1 via a separate tube, such that the concentration of pTSA in reactor R1 is maintained at 1 wt%, while, via tube 5, the same volume as the two streams is removed from the stirred reactor R1 to a plug flow reactor (PFR). In CSTR R1, the residence time is long enough for furfuryl alcohol conversion to exceed 99%. At 140°C, the time is 2.4 hours (144 minutes). Finally, less than 1% of furfuryl alcohol and intermediates are completely converted to butyl levulinate in the plug flow reactor (PFR) at 140°C within a 30-minute residence time. The stream then enters the distillation column C1 as described above.

[0097] definition:

[0098] A "heterogeneous catalyst" is a solid catalyst, and therefore exists in a different phase from the reaction mixture, which is a liquid phase. In contrast, a homogeneous catalyst is soluble in the reaction mixture; therefore, homogeneous catalysis refers to a reaction in which the catalyst exists in the same phase as the reactants, primarily in a dissolved state. In this paper, the reaction mixture is liquid, and the catalyst is soluble in the liquid reaction phase.

[0099] As used in this article, the “selectivity” of the levulinate product is calculated using the following equation:

[0100] Selectivity (mol%) = (moles of alkyl levulinate at the end of the reaction / (moles of furfuryl alcohol fed to the reactor - moles of furfuryl alcohol at the end of the reaction), in mol%.

[0101] Therefore, when 100% conversion is achieved, there is no unreacted furfuryl alcohol remaining in the reaction mixture at the end of the reaction.

[0102] "Sulfonic acid catalyst" is a homogeneous catalyst. However, as used herein, "sulfonic acid catalyst" can be a hydrate, but not a salt. Specifically, "sulfonic acid catalyst" excludes salts, such as sulfonic acid-functionalized ionic liquids.

[0103] A “strong protic acid” is defined as a protonated acid with a dissociation constant or Ka value of at least about 55 at 25°C and 1 atm. A “weak protic acid” is defined as a protonated acid with a Ka value less than 55.

[0104] The characteristic of sulfonic acid RSO₂OH is its tetrahedral sulfur center, meaning that sulfur is located at the center of four atoms: three oxygen atoms and one carbon atom. The pKa values ​​for p-toluenesulfonic acid and methanesulfonic acid are -2.8 and -1.9, respectively. However, due to their strong acidity, their pKa values ​​cannot be directly measured, and the values ​​usually cited should be considered indirect estimates with significant uncertainty. For example, multiple sources have reported pKa values ​​for methanesulfonic acid as high as -0.6 or as low as -6.5.

[0105] Strong acids have a pKa < -1.74, therefore p-TSA is a strong acid. It should be understood that Ka 10 equals pKa⁻¹, and Ka 55 equals pKa⁻¹.74, because it is a logarithmic scale.

[0106] Catalyst performance is measured by activity, conversion, selectivity, and productivity, and, in the case of expensive and / or heterogeneous catalysts, also by catalyst lifetime. Activity is the amount of substrate that can be converted per unit amount of catalyst and per unit time, and is directly related to productivity, which is the amount of product produced per unit volume and per unit time. Low activity can be compensated for by increasing the amount of catalyst, but this comes with higher catalyst costs. High selectivity is important because low selectivity means the need for waste feedstock and costly, energy-intensive product separation. High conversion is required to minimize or avoid separation costs so that the starting material can be recycled. If the catalyst is expensive, often heterogeneous catalysts or homogeneous catalysts containing precious metals, then catalyst lifetime and recycling are important for economic competitiveness. Instead of lifetime, a better measure is the amount of product produced per unit amount of catalyst under industrial conditions.

[0107] Example

[0108] Material

[0109] All chemicals were purchased from Sigma Aldrich.

[0110] Analytical methods

[0111] GC Analysis

[0112] Product concentrations were analyzed by gas chromatography-FID equipped with a flame ionization detector. GC-FID analysis was performed using a capillary column: column - Restek Rxi - 5 ms, 30 m, 0.25 mm ID, 0.25 μm film thickness, and the following conditions were used:

[0113] The flow rate of He was 1.5 mL / min;

[0114] Start at 40°C and maintain that temperature for 2 minutes;

[0115] • The temperature was increased at a rate of 20°C / min until it reached 300°C, and then maintained at that temperature for 5 minutes.

[0116] • FID at 300℃; and 1,2,4-trimethylbenzene was used as an internal standard for GC.

[0117] Sample preparation procedure

[0118] i. Weigh 150-300 mg of sample, add 3 g of MTBE solvent, record the weight, filter, and run on a GC.

[0119] ii. Quantification via external calibration curves. Several concentrations (typically 5) of each component of interest are prepared in MTBE, ranging from 5 mg / g to 50 mg / g. These concentrations are run on a GC, and the area counts of each component are integrated, and calibration curves are generated using GC software.

[0120] iii. Next, by inputting the weight of the GC formulation, the components in the unknown sample are quantified using a calibration curve, and then the weight of each component is generated for each component found in the chromatogram.

[0121] iv. Check the instrument calibration accuracy and, if necessary, adjust the results by running a control sample of butyl levulinate of known purity, with one control sample for every five samples in the queue.

[0122] The yield and selectivity of the product were calculated by comparing its amount with that of the reference standard and corrected for the correction factors experimentally determined for dialkyl ethers, alkanols, furfuryl alcohols and alkyl levulinates.

[0123] The correction factors for intermediates and heavy components are estimated based on the O / C ratio (oxygen to carbon ratio, as it has a significant impact on the sensitivity of FID (flame ionization detector)) or the expected O / C ratio for the unknown compound.

[0124] Water Analysis

[0125] Water content was analyzed on a Mettler Toledo DL31 instrument using the Karl Fisher method (ASTM E203-16: Standard Test Method for Water by Karl Fisher Titration of Volumetric Method).

[0126] The method of the present invention will be further illustrated by means of the following non-limiting embodiments.

[0127] Example 1A and Comparative Example 1B

[0128] Batch feeding experiment

[0129] Figure 1 A schematic diagram of the experimental setup for the batch feed experiment is shown. The main components are: (T1) butanol / furfuryl alcohol feed tank, (R1) reactor, and (P) pump unit, such as a pump from Scientific Systems, HPLC series 1 piston pump, 0-10 ml / min.

[0130] Example 1A uses Figure 1 The experimental equipment was delivered in batches.

[0131] 219.14 g butanol (2.94 mol, 99.4% purity) and 0.032 mol acidic centers (or protons) (6.05 g p-toluenesulfonic acid monohydrate (p-TSA, 99% purity)) were manually added to reactor R1 (1000 mL). The reactor was closed and purged with nitrogen for 5 minutes to remove most of the air (reducing oxidation side reactions) and minimize the effect of atmospheric humidity on the conditions inside the reactor. The mixture was then heated to 130 °C and the pressure was autogenous, approximately 2 bar. Over a period of 410 minutes, a mixture of 205.79 g butanol (2.76 mol, 99.4% purity) and 205.54 g furfuryl alcohol (2.05 mol, 98% purity) (molar ratio 1.35:1) was slowly and continuously fed into the reactor (1000 mL) (4.97 g furfuryl alcohol per gram of catalyst or 940 g furfuryl alcohol per mole of catalyst per hour). The total molar ratio of butanol to furfuryl alcohol in the reaction mixture is 2.78. The molar ratio of furfuryl alcohol to pTSA is 65.2, and the weight ratio is 33.6 (both referring to the reaction mixture). The total weight of the reaction mixture is 636.52 g, and the catalyst in the reaction mixture is 0.95 wt%. The sulfonic acid catalyst comprises approximately 0.95% (w / w) of the reaction mixture.

[0132] Comparative Example 1B also uses Figure 1 The experimental equipment was delivered in batches.

[0133] 218.23 g butanol (2.93 mol, 99.4% purity) and 0.032 mol acidic centers (or protons) (3.156 g sulfuric acid, 98% purity) were manually added to reactor R1 (1000 mL). The reactor was closed and purged with nitrogen for 5 minutes to remove most of the air (oxygen and humidity). The mixture was then heated to 130°C and the pressure was autogenous, approximately 2 bar. Over a period of 410 minutes (this 410-minute feed time is referred to as approximately 7 hours in Example 3), a mixture of 202.44 g butanol (2.71 mol, 99.4% purity) and 205.44 g furfuryl alcohol (2.05 mol, 98% purity) (molar ratio 1.32:1) was slowly and continuously fed into reactor R1 (1000 mL) (9.52 g furfuryl alcohol per gram of catalyst or 940 g furfuryl alcohol per mole of catalyst per hour in the reaction mixture). The total molar ratio of butanol to furfuryl alcohol in the reaction mixture is 2.75. The molar ratio of furfuryl alcohol to sulfuric acid is 65.1, and the weight ratio is also 65.1, because they have the same molecular weight (both refer to the reaction mixture).

[0134] For each of Example 1A and Comparative Example 1B, the alcohol and acid catalysts were initially fed into reactor R1 at room temperature. Reactor R1 was closed and purged with nitrogen instead of air. The mixture was then heated until the reaction temperature was reached (130°C for both Example 1A and Comparative Example 1B). Once the reaction temperature of 130°C was reached, a furfuryl alcohol / butanol mixture was fed into reactor R1 and stirred at 600 rpm for 410 minutes; the reaction temperature was maintained at 130°C. Once the feeding was completed, the reaction continued at the given reaction temperature of 130°C for 1 hour. Reactor R1 was then cooled, and the samples were analyzed using GC-FID with an internal standard. It is assumed that under these conditions, sulfuric acid dissociates by only one proton, therefore, the same amount of protons are present in both reactions. The two reactions have the same 940 g of furfuryl alcohol per mole of catalyst per hour, meaning that when sulfuric acid dissociates by only one proton (which is very likely), they have the same number of acidic (or proton) centers.

[0135] Results - Example 1A and Comparative Example 1B

[0136] Table 1

[0137]

[0138] *If 100 mol of furfuryl alcohol is fed and 10 mol of butanol is ultimately converted to dibutyl ether, then the conversion rate is 10 mol%.

[0139] Example 1A showed significantly lower formation of dibutyl ether compared to Comparative Example 1B, while the molar conversion of furfuryl alcohol to butyl levulinate was higher. Trace amounts of levulinic acid were found, but due to their small quantity and peak tailing, they could not be integrated. In both cases, the amount was less than 0.2 wt%. All intermediates were converted after the reaction was stopped.

[0140] The selectivity of the butyl levulinate product is calculated by the ratio of the number of moles of butyl levulinate to (the initial number of moles of furfuryl alcohol – the final number of moles of furfuryl alcohol) × 100. If all furfuryl alcohol is eventually converted (as in all experiments), then it will equal the yield.

[0141] Example 1A shows a higher yield of butyl levulinate compared to Comparative Example 1B, at 90.84% ​​versus 84.44% (because the conversion of furfuryl alcohol was 100% in all experiments, so yield equals selectivity).

[0142] These results are surprising because, based on previous literature (Lange et al. 2009), it was expected that pTSA and sulfuric acid would yield the same butyl levulinate yield and selectivity at the initial reaction temperature of 130 °C, since the same amount of protons are present and the protons are readily accessible.

[0143] Comparative Example 2A, Comparative Example 2B, and Comparative Example 2C

[0144] Except as described below, the experiments were conducted in batches in the same manner as in Example 1A and Comparative Example 1B:

[0145] The current reaction temperature is 116℃.

[0146] The addition time is 5 hours, and

[0147] • By weight, approximately twice the amount of pTSA is used compared to sulfuric acid, so that, on a molar basis, the same amount of acidic centers exist when sulfuric acid dissociates with only one proton, thus requiring the same amount of catalyst.

[0148] Comparative Experiment 2A was conducted in a batch feed mode. Initially, 435.11 g of butanol (5.84 mol, 99.4% purity) and 0.0708 mol of acid (13.74 g of p-toluenesulfonic acid monohydrate (p-TSA, 99% purity)) were manually fed into reactor R1 (2000 mL). The reactor was closed and purged with nitrogen for 5 minutes to remove most of the air and humidity. The mixture was then heated to approximately the boiling point of butanol (116 °C) and the pressure was autogenous, approximately 1 bar (at 116 °C). A mixture of 404.03 g butanol (5.42 mol, 99.4% purity) and 411.34 g furfuryl alcohol (4.11 mol, 98% purity) (molar ratio 1.32:1) was slowly and continuously fed into the reactor over a period of 300 minutes (feed rate of 5.99 g furfuryl alcohol per gram of catalyst or 1162 g furfuryl alcohol per mole of catalyst per hour). The overall molar ratio of butanol / furfuryl alcohol in the reaction mixture was 2.74. The molar ratio of furfuryl alcohol / pTSA was 58.04, and the weight ratio was 29.93 (both referring to the reaction mixture).

[0149] Comparative Experiment 2B was also conducted in a batch feed mode. 435.9 g of butanol (5.55 mol, 99.4% purity) and 0.063 mol of acid (6.28 g of sulfuric acid, 98% purity) were fed into a reactor (2000 mL). The reactor was closed and purged with nitrogen for 5 minutes to remove most of the air and humidity. The mixture was then heated to 116 °C and the pressure was autogenous, approximately 1 bar (at 116 °C). Over 5 hours, a mixture of 404.01 g of butanol (5.14 mol, 99.4% purity) and 412.17 g of furfuryl alcohol (4.12 mol, 98% purity) (molar ratio 1.25:1) was slowly fed into the reactor (2000 mL) (feed rate of 13.13 g furfuryl alcohol per gram of catalyst or 1308 g furfuryl alcohol per mol of catalyst per hour). The total molar ratio of butanol / furfuryl alcohol in the reaction mixture was 2.60. The molar ratio of furfuryl alcohol to sulfuric acid is 65.66, and the weight ratio is 65.66 (both referencing the reaction mixture).

[0150] Comparative experiment 2C was also conducted in a batch feed mode. 435.65 g of butanol (5.54 mol, 99.4% purity) and 0.063 mol of acid (6.26 g of sulfuric acid, 98% purity) were fed into a reactor (2000 mL). The reactor was closed and purged with nitrogen for 5 minutes to remove most of the air and humidity. The mixture was then heated to 116 °C and the pressure was autogenous, approximately 1 bar (at 116 °C). Over 5 hours, a mixture of 404.74 g of butanol (5.15 mol, 99.4% purity) and 412.5 g of furfuryl alcohol (4.12 mol, 98% purity) (molar ratio 1.25:1) was slowly fed into the reactor (2000 mL) (feed rate of 13.18 g furfuryl alcohol per gram of catalyst or 1310 g furfuryl alcohol per mol of catalyst per hour). The total molar ratio of butanol / furfuryl alcohol in the reaction mixture was 2.60. The molar ratio of furfuryl alcohol to sulfuric acid is 65.88, and the weight ratio is 65.88 (both referencing the reaction mixture).

[0151] Initially, the alcohol and acid catalysts were manually fed into the reactor at room temperature. The reactor was then closed and purged with nitrogen instead of air. The mixture was then heated until the reaction temperature was reached. Once the reaction temperature was reached, a furfuryl alcohol / butanol mixture was fed into the reactor and stirred at 600 rpm for 300 minutes (5 hours) while maintaining the reaction temperature. Once feeding was completed, the reaction was continued at the given reaction temperature for 1 hour. The reactor was then cooled, and the sample was analyzed using GC-FID with an internal standard. It is assumed that under those conditions, sulfuric acid dissociates by only one proton, therefore pTSA contains slightly more protons, but sulfuric acid is a stronger acid.

[0152] Comparative experiments 2B and 2C have already been included, demonstrating that under almost identical reaction conditions, the results show similar effects.

[0153] Results - Comparative Examples 2A, 2B, and 2C

[0154] Table 2

[0155] Example Yield The number of moles of butanol converted into dibutyl ether Butyl levulinate Expressed as the number of moles of butanol and based on the initial number of moles of furfuryl alcohol.* (mol%) (mol%) 2A 76.67 0.55 2B 86.07 2.29 2C 86.32 2.15

[0156] *If 100 mol of furfuryl alcohol is fed and 10 mol of butanol is ultimately converted to dibutyl ether, then the conversion rate is 10 mol%.

[0157] Comparative Example 2A showed a significantly lower formation of dibutyl ether compared to Comparative Examples 2B and 2C. However, the yield of butyl levulinate in Comparative Example 2A was also lower than that in Comparative Examples 2B and 2C. All intermediates were converted after the reaction was stopped.

[0158] Therefore, at lower temperatures, sulfuric acid is more selective than pTSA as a catalyst, even though the molar amount of protons in the reaction is slightly higher under pTSA.

[0159] Comparative Example 2A shows that 116°C is too low for the reaction with pTSA (compared to Example 1A).

[0160] In contrast, the selectivity of sulfuric acid is greater at 116°C than at 130°C (see Comparative Example 1B).

[0161] The effect of temperature on sulfuric acid differs from that on pTSA, which was unexpected.

[0162] Examples 3A to 3G

[0163] In Examples 3A to 3G, the reaction conditions (pTSA concentration, butanol / furfuryl alcohol ratio in the feed and total reaction mixture, and temperature) were varied according to the procedure of Example 1A (1L reactor). The desired molar ratio of butanol / furfuryl alcohol was introduced into T1, using... Figure 1 The experimental equipment. Alternatively, the experimental equipment is Figure 3 The experimental setup, for Example 3F, allows for independent variations in the feeds of furfuryl alcohol and butanol.

[0164] Results - Examples 3A to 3G

[0165] Table 3

[0166]

[0167]

[0168] *If 100 mol of furfuryl alcohol is fed and 10 mol of butanol is ultimately converted to dibutyl ether, then the conversion rate is 10 mol%.

[0169] Example 3A' is a reproduction of Example 3A. However, due to calibration errors in the original Example 3A, the data in Example 3A' are more representative. The data in Example 3A' show that as the temperature increases from 140°C (Example 3B), the yield of butyl levulinate increases slightly, while the yield of dibutyl ether doubles.

[0170] Example 3H was carried out at 160°C for 7 hours.

[0171] Under these laboratory-scale conditions, Example 3B is currently the optimal reaction condition and is highlighted in bold. This is because the yield of butyl levulinate (BL) increases with temperature, but the yield of dibutyl ether byproducts increases even more with temperature. Therefore, at 160°C (Example 3H), the yield of butyl levulinate is excellent, but a large amount of dibutyl ether is produced.

[0172] The only difference between Examples 3B and 1A is their reaction temperature. The yield and selectivity are both better at the initial reaction temperature of 140°C.

[0173] The only difference between Examples 3C and 3D is their feed time. When the mixture of butanol and furfuryl alcohol is fed into the reactor over 10 hours, both the yield and selectivity are better than those over 4 hours. However, the lower feed rate increases the amount of dibutyl ether produced, and the 10-hour feed time does not improve the yield of butyl levulinate. Therefore, under these conditions, a reaction time of 7 hours is considered optimal.

[0174] The difference between Examples 3E and 3F lies in the molar ratio of butanol to furfuryl alcohol in the feed to the reactor. For each of Examples 3E and 3F, the mixture of butanol and furfuryl alcohol is fed to the reactor over 7 hours, and the reactor is maintained at the first reaction temperature for 2 hours after the addition of both butanol and furfuryl alcohol has been completed. The yield and selectivity are slightly better when the molar ratio in the feed is 1.3:1 (butanol:furfuryl alcohol).

[0175] Examples show that excellent yields are obtained at temperatures between 125°C and 170°C. However, increasing the temperature also increases the formation of dialkyl ethers, and above 170°C, the amount of dialkyl ethers formed becomes unacceptably high. Example 2 previously showed that the yield at 116°C (Comparative Example 2A) was significantly lower. Higher temperatures increase the yield of dibutyl ethers. Slower feeding of furfuryl alcohol to the reaction mixture (Example 3C) does not increase the yield of butyl levulinate, but increases the yield of dibutyl ethers. Faster feeding of furfuryl alcohol decreases the yield (Example 3D). Dibutyl ether yields are lower when furfuryl alcohol is fed faster, but higher when fed slower (Example 3C). If the amount of butanol in the feed (Example 3F) or in the reactor (Example 3E) is reduced compared to furfuryl alcohol, the yield of butyl levulinate remains high and the change in dibutyl ether yield is not significant. Reducing the amount of catalyst by 50% does indeed reduce the amount of butyl levulinate, but the reduction is very small, and the reduction of dibutyl ether is also very small (Example 3G).

[0176] Examples 3I and 3J were carried out at 125°C and 170°C, respectively. At 170°C, some butene began to form and the yield of butyl levulinate decreased slightly (compared to Example 3H, which was carried out at 160°C). However, the yield at 170°C remained acceptable. At 125°C, the yield of butyl levulinate was comparable to that observed in Example 1A (carried out at 130°C).

[0177] Example 3B is encouraged to represent the optimal butyl levulinate selectivity (91.72%) at a reaction temperature of 140°C using p-toluenesulfonic acid as a sulfonic acid catalyst, with acceptable dibutyl ether formation (6.36%).

[0178] Example 4

[0179] The experimental equipment is Figure 2 The experimental equipment. Alternatively, the experimental equipment is Figure 3 The experimental equipment allows for independent variation of the feeds for furfuryl alcohol and butanol.

[0180] As in Examples 3B to 3F, the mol / mol ratio (butanol / FFOH) was varied to 1.5, 1.8, and 2.2, the temperature was fixed at 140°C, the catalyst concentration was 1 wt% pTSA monohydrate, and the residence times for CSTR and PFR were used respectively:

[0181] · 6.5 hours and 20.8 minutes

[0182] ·5 hours and 16.15 minutes, and

[0183] · 3.5 hours and 11.3 minutes

[0184] The CSTR / PFR experiments were based on batch feed experiments, with additional variations in temperature and acid concentration. In the batch feed, a 7-hour reaction feed time achieved a 91.72% yield (Example 3B), after which the batch was reacted for another 1 or 2 hours to convert all intermediates to butyl levulinate. A mol / mol ratio of 2.7 was found to be optimal for the reaction yield. In charge manufacturing, the reaction was scaled up and achieved a reactor yield approaching 94%.

[0185] Some CSTR-PFR experiments were also conducted, in which the residence time of CSTR was 6.5 hours and the residence time of PFR was 1 hour.

[0186] The yield of butyl levulinate was found to increase with the following:

[0187] ·stay time

[0188] • Temperature (up to 140℃)

[0189] Butanol / furfuryl alcohol molar ratio

[0190] pTSA concentration

[0191] For dibutyl ethers, the yield is strongly dependent on:

[0192] ·temperature

[0193] • Butanol / furfuryl alcohol molar ratio (140℃, 1wt% pTSA.H2O, 5-hour CSTR residence time)

[0194] A 2.2 molar ratio of butanol / furfuryl alcohol yields 1.24 kg / 100 kg of furfuryl alcohol.

[0195] A 1.8 molar ratio of butanol / furfuryl alcohol yields 0.44 kg / 100 kg of furfuryl alcohol.

[0196] A 1.5 molar ratio of butanol / furfuryl alcohol yields 0.34 kg / 100 kg of furfuryl alcohol.

[0197] pTSA concentration

[0198] • Slightly dependent on residence time (140℃, 1wt% pTSA·H2O, 2.2mol / mol butanol / FFOH)

[0199] A residence time of 6.5 hours yielded 1.8 kg / 100 kg of furfuryl alcohol.

[0200] A 5-hour residence time yielded 1.24 kg / 100 kg of furfuryl alcohol.

[0201] A residence time of 3.5 hours yielded 0.94 kg / 100 kg of furfuryl alcohol.

[0202] During feeding, the droplets from the two streams should not mix with each other. Furthermore, the droplets should mix rapidly at the molecular level within the reactor to avoid localized high concentrations of furfuryl alcohol and pTSA.

[0203] Example 5

[0204] This example illustrates the reaction of furfuryl alcohol with ethanol to produce ethyl levulinate.

[0205] Example 5 was performed identically to Example 1A (using...) Figure 1 The experimental equipment was similar in molar quantity (ethanol / furfuryl alcohol molar ratio was 1.32:1 (feed) and 2.83:1 (total)), but the weight was different because the molecular weight of ethanol is different from that of butanol, and obviously the autogenous pressure is also different, now it is about 6 bar.

[0206] Anhydrous ethanol (less than 0.2 wt% water) and p-toluenesulfonic acid (1 wt% in the final reaction mixture) were fed into the reactor at room temperature. The reactor (autoclave) was closed and purged with nitrogen instead of air. The mixture was then heated until the reaction temperature was reached. Once the reaction temperature (135 °C) was reached, the furfuryl alcohol / ethanol mixture was fed into the autoclave and stirred at 600 rpm while maintaining the reaction temperature for a period of 420 minutes (7 hours; feed rate of approximately 5 g furfuryl alcohol per gram of catalyst per hour). Once feeding was complete, the reaction was continued at the given reaction temperature for 1 hour or 2 hours (480 minutes or 540 minutes in total). The reactor was then cooled, and the sample was analyzed using internal standard by GC-FID.

[0207] Examples 5B to 5C were carried out similarly, with the following: an alkanol / furfuryl alcohol molar ratio of 1.3:1 (feed) and a final reaction mixture ratio of 2.7:1 (total), a 7-hour feed time, and 2 hours after the reaction (total time 540 minutes) and a final reaction mixture of 1 wt% pTSA.

[0208] Results of Example 5 - Table 4

[0209]

[0210] *If 100 mol of furfuryl alcohol is fed and 10 mol of ethanol is ultimately converted to diethyl ether, then the 10 mol% conversion rate is...

[0211] The experiment originated from a laboratory, but the 60 minutes were not included after the gradual addition.

[0212] 2-Octanol is an example of a secondary alcohol used in this paper.

[0213] Ethylhexanol is an example of a branched alcohol used in this article.

[0214] Example 5 shows that excellent yields were obtained with other alcohols, such as another short-chain primary alcohol (ethanol), a branched long-chain primary alcohol (ethylhexanol), and a secondary alcohol (2-octanol).

[0215] Example 6

[0216] Example 6 uses Figure 1 The experimental setup was conducted in a batch-feed mode. This embodiment compares various sulfonic acid catalysts, including methanesulfonic acid as an example of an aliphatic sulfonic acid and camphorsulfonic acid as an example of a cyclic aliphatic sulfonic acid.

[0217] This embodiment also compares sulfonic acid with ionic liquid catalysts - [BMIm-SH][HSO4] (1-butylsulfonic acid 3-methylimidazolium sulfuric acid) and tFA (trifluoroacetic acid) and Bi-TFA (bismuth trifluoromethanesulfonate).

[0218] Examples 6D, 6E and 6F are comparative examples.

[0219] In all embodiments, 216 g of butanol (2.91 mol, 99.4% purity) and 1 wt% (approximately 6.25 g) of the final total mixture of acid catalyst were manually added to reactor R1 (1000 mL). The reactor was closed and purged with nitrogen for 5 minutes to remove most of the air (reducing oxidation side reactions) and to minimize the effect of atmospheric humidity on the conditions inside the reactor. The mixture was then heated to 140 °C and the pressure was autogenous. After 420 minutes, a mixture of 200 g of butanol (2.71 mol, 99.4% purity) and 206 g of furfuryl alcohol (2.05 mol, 98% purity) (molar ratio 1.31:1) was slowly and continuously fed into the reactor (1000 mL). The total molar ratio of butanol / furfuryl alcohol in the reaction mixture was 2.73. After stopping the feeding, the reaction was carried out at the reaction temperature and with continuous stirring for 2 hours, and then the final sample was taken.

[0220] The catalysts used are pTSA (p-toluenesulfonic acid, CSA (camphorsulfonic acid), MSA (methanesulfonic acid), [BMIm-SH][HSO4] (1-butylsulfonic acid, 3-methylimidazolium hydrogen sulfate ionic liquid), Bi-TFA (bismuth trifluoromethanesulfonate, Lewis acid) and tFA (trifluoroacetic acid).

[0221]

[0222] *Based on the following calculation: when 100 mol furfuryl alcohol is fed and 10 mol butanol is finally converted to dibutyl ether, divide 10 by 100, or 10 mol% conversion.

[0223] **No retention time - 0 hours settling time**

[0224] Examples 3B, 6B, and 6C show that all sulfonic acids yielded excellent yields of butyl levulinate. Compared to pTSA, methanesulfonic acid yielded higher amounts of dibutyl ether, while camphorsulfonic acid yielded lower amounts of dibutyl ether (although camphorsulfonic acid yielded lower amounts under different conditions).

[0225] The ionic liquid [BMIm-SH][HSO4] (catalog number: 57457-100G-F; lot number: BCCD9481) purchased from Sigma Aldrich exhibited unexpectedly low yields (approximately 60%), but with extremely low dibutyl ether formation (0.47%) (Comparative Example 6D). These results are unexpected, as Rode et al. (Single pot conversion of furfuryl alcohol to levulinic esters and γ-valerolactone in the presence of sulfonic acid functionalized ILs and metal catalysts, Green Chem., 2013, 15, 2540-2547) reported high yields. The reason for this discrepancy is speculative. The experimental results do not disclose whether standards (internal or external) were used, therefore, if standards were not used and if heavy components not visible in the GC process were obtained, the calculated yield would be overestimated. However, as mentioned earlier, this is speculative. Clearly, the low yield makes it economically uncompetitive with homogeneous sulfonic acids. Furthermore, ionic liquids are more expensive to manufacture and have high viscosity (Mauro CCRibeiro, High Viscosity of Imidazolium Ionic Liquids with the Hydrogen Sulfate Anion: A Raman Spectroscopy Study J. Phys. Chem. B2012, 116, 24, 7281-7290), making them more difficult to handle.

[0226] Professor Jerome's team used metal trifluoromethanesulfonates; among them, bismuth trifluoromethanesulfonate (a homogeneous catalyst, but not a homogeneous sulfonic acid catalyst) performed best (US Patent No. 10,590,060; Alban Chappaz, Francois Jerome, Karine De Oliveira Vigier, Eric Muller, Jonathan Lai, Matthieu Corbet, Didier Morvan, Process for the preparation of levulinate esters, assigned to University of Poitiers). Using Bi-TFA, excellent yields were observed (Comparative Example 6E, 81.6%). However, the yield using Bi-TFA was significantly lower than the yield observed with homogeneous sulfonic acid, and the yield of dibutyl ether was much higher under the same conditions than the yield observed with homogeneous sulfonic acid.

[0227] Trifluoroacetic acid (Comparative Example 6F) exhibits low butyl levulinate yield and low dibutyl ether formation. However, the butyl levulinate yield is too low to be economically viable.

[0228] Example 7

[0229] This embodiment is used for... Figure 1 The experimental equipment was used to produce ethylhexyl levulinate on a large scale.

[0230] In a stainless steel feed tank, 1120 kg of ethylhexanol and 720 kg of furfuryl alcohol were mixed (molar ratio EH / FFA = 1.3). 1360 kg of ethylhexanol, along with 30 kg of p-toluenesulfonic acid, was then charged into a 5000 L reactor. The reactor was heated to 140 °C, and the ethylhexanol / furfuryl alcohol mixture was fed at 135 °C. The mixture was fed for 7 hours, after which the reactor was maintained at 140 °C (+ or -5 °C) for 2 hours, and then cooled to 60 °C. The catalyst was neutralized with soda ash, and after batch distillation, 1465 kg of ethylhexyl levulinate was recovered, corresponding to an overall recovery yield of 87.44% (the reaction yield itself is likely higher). Additionally, 33.2 kg of diethylhexyl ether was identified, corresponding to 3.7% ethylhexanol molarly relative to the amount of furfuryl alcohol present.

[0231] Example 8

[0232] according to Figure 2Example 8 was conducted in the CSTR followed by PFR format. For this purpose, both the furfuryl alcohol and butanol mixture (T3) and the butanol and pTSA (4.85% solution) mixture (T1) were prepared in large batches. Therefore, during testing under one condition, it was not necessary to refill the batch tanks (T1 and T3). Clearly, with additional pumps and batch tanks, as well as a static mixer, it is possible to achieve the following: Figure 3 The same experiment was conducted using the equipment described in the document.

[0233] The effective volume of the CSTR is 750 ml, and the volume of the plug flow reactor is 40 ml, for which 1 / 2-inch tubing is used. The reaction is started in batch feed mode (according to Example 1), and once the effective reactor volume is reached, the system is switched to CSTR-PFR mode. Samples are taken when steady-state conditions are reached. Initially, the reactor volume needs to be replaced 8 times. Then, new conditions close to the previous conditions are selected so that the new steady-state conditions are reached when the reaction volume is replaced 3 times. Once steady-state conditions are reached, sample points are taken at intervals of one residence time in the reactor, and after taking 3 samples, the average value is calculated and recorded here. If there are outliers, additional samples are taken and the outliers are omitted.

[0234] The mol / mol ratio (butanol / FFOH) (butanol including butanol used to dilute pTSA) was varied to 1.5, 1.8 and 2.2, the temperature was fixed at 140 °C, and the catalyst concentration was 1 wt% pTSA monohydrate.

[0235] Examples 8A to 8G

[0236]

[0237] Examples 8A, 8B and 8D show that reducing residence time reduces the yield of butyl levulinate and the formation of dibutyl ether.

[0238] Examples 8B and 8C, as well as Examples 8A and 8E, show that reducing the molar ratio of butanol / furfuryl alcohol more strongly reduces the yield of butyl levulinate and the formation of dibutyl ether.

Claims

1. A method for converting furfuryl alcohol to levulinate, the method comprising contacting furfuryl alcohol, an alcohol or a mixture thereof, and a homogeneous catalyst to form a reaction mixture at a first reaction temperature in the range of 125°C to 180°C to form the levulinate, and forming the levulinate in the reaction mixture. The alcohol mentioned therein is a primary or secondary alcohol selected from C1-24 straight-chain or branched alcohols; alkoxy-alkanols; alicyclic alcohols; unsaturated alicyclic alcohols; diols that are primary alcohols; or combinations of two or more of them; The method is characterized by: The homogeneous catalyst is a sulfonic acid catalyst or its hydrate, wherein the sulfonic acid catalyst is selected from the group consisting of alkyl-aromatic sulfonic acids, aromatic sulfonic acids, and aliphatic straight-chain or cyclic sulfonic acids, provided that salts of the sulfonic acid catalyst are excluded; and The molar ratio of the alcohol to furfuryl alcohol added to the reaction mixture is between 1:1 and 5:

1.

2. The method of claim 1, wherein the first reaction temperature is between 125°C and 170°C.

3. The method of claim 1, wherein the first reaction temperature is 140°C.

4. The method of claim 1, wherein the sulfonic acid catalyst or its hydrate is selected from p-toluenesulfonic acid or its monohydrate, methanesulfonic acid or its hydrate, or 7,7-dimethyl-2-oxobicyclo[2.2.1]heptane-1-yl)methanesulfonic acid or its monohydrate.

5. The method of claim 1, wherein the sulfonic acid catalyst comprises 0.1% w / w to 5% w / w of the reaction mixture; and / or wherein the molar ratio of the alcohol to the furfuryl alcohol added to the reaction mixture is between 1.1:1 and 3:1; and / or wherein the furfuryl alcohol is fed into the reaction mixture at a rate of up to 200 grams of furfuryl alcohol per gram of catalyst per hour.

6. The method of claim 5, wherein the sulfonic acid catalyst comprises 1% w / w of the reaction mixture; and / or wherein the molar ratio of the alcohol to the furfuryl alcohol added to the reaction mixture is 2.2:1 or 1.5:1; and / or wherein the furfuryl alcohol is fed to the reaction mixture at a feed rate of up to 50 g / g / h.

7. The method of claim 1, wherein the alcohol is selected from methanol, ethanol, n-propanol, isopropanol, n-butanol, isobutanol, 2-butanol, n-pentanol, isopentanol, n-hexanol, n-octanol, n-decanol or 2-ethylhex-1-ol; β-methoxyethanol or β-ethoxyethanol; or a combination of two or more thereof.

8. The method of claim 1, wherein the alcohol is selected from cyclohexanol, cyclopentanol, tetrahydrofurfuryl alcohol, and 5-methyl-2-(propane-2-yl)cyclohex-1-ol; or prop-2-en-1-ol, 3,7-dimethyloct-2,6-dien-1-ol, and prop-2-yn-1-ol; or ethylene glycol, 1,3-propanediol, and 1,4-butanediol; or a combination of two or more of these.

9. The method of claim 1, wherein the alcohol is a mixture of one or more primary or secondary alcohols; and / or a mixture of one or more alicyclic alcohols; and / or a mixture of one or more unsaturated alicyclic alcohols; and / or a mixture of one or more diols.

10. The method of claim 1, wherein the alcohol is replaced with an alcohol selected from fusel alcohols and Guerbert alcohols.

11. The method of claim 1, wherein the contact is carried out in a batch-feed reactor.

12. The method of claim 11, wherein the reaction mixture is formed by the following steps: In the batch-feed reactor, (i) a first mixture comprising the alcohol or a mixture thereof, and optionally the homogeneous catalyst; (ii) The second mixture is continuously or intermittently fed into the batch feed reactor of (i) the first mixture, wherein the second mixture comprises the furfuryl alcohol and additional amounts of the alcohol or mixture thereof; as well as Separate from the second mixture, a third mixture is continuously or intermittently fed to the batch feed reactor, wherein if the first mixture contains the homogeneous catalyst, the third mixture contains (iii) an additional amount of the homogeneous catalyst; or if the first mixture does not contain the homogeneous catalyst, the third mixture contains the homogeneous catalyst.

13. The method of claim 12, wherein the first reaction temperature is maintained for up to 2 hours after the second mixture and the first mixture have been fully mixed; or alternatively, the first reaction temperature is raised to a second reaction temperature after the second mixture and the first mixture have been fully mixed.

14. The method of claim 12, wherein after the second mixture and the first mixture have been fully mixed, the first reaction temperature is maintained for 5 to 60 minutes; or alternatively, after the second mixture and the first mixture have been fully mixed, the first reaction temperature is raised to a second reaction temperature and maintained at the second reaction temperature for 5 to 120 minutes.

15. The method of claim 1, wherein the method is continuous and carried out in a continuous stirred tank reactor.

16. The method of claim 15, wherein the continuous stirred tank reactor is filled with the reaction mixture by the following steps: The reactor is provided with (i) a first mixture comprising the alcohol or a mixture thereof, and optionally the homogeneous catalyst; (ii) The second mixture is continuously or intermittently fed into the reactor (i) the first mixture, the second mixture comprising the furfuryl alcohol, and additional amounts of the alcohol or mixtures thereof; and Separate from the second mixture, a third mixture is fed continuously or intermittently to the reactor, wherein if the first mixture contains the homogeneous catalyst, the third mixture contains (iii) an additional amount of the homogeneous catalyst; or if the first mixture does not contain the homogeneous catalyst, the third mixture contains the homogeneous catalyst.

17. The method of claim 16, wherein the first mixture is heated to the first reaction temperature before the second mixture and the third mixture are mixed with the first mixture.

18. The method of any one of claims 12 to 17, wherein the sulfonic acid catalyst is selected such that the sulfonic acid catalyst accounts for 0.1% to 5% w / w of the reaction mixture.

19. The method of any one of claims 12 to 17, wherein the sulfonic acid catalyst is selected such that the sulfonic acid catalyst accounts for 1% w / w of the reaction mixture.

20. The method of any one of claims 12 to 17, wherein the second mixture is added to the reaction mixture at a feed rate of up to 200 grams of furfuryl alcohol per gram of catalyst per hour in the reaction mixture.

21. The method of claim 20, wherein the feed rate is at most 50 g / g / h.

22. The method of any one of claims 12 to 17, wherein the total molar ratio of the alcohol to the furfuryl alcohol contained in the first mixture and the second mixture is between 1:1 and 5:

1.

23. The method of any one of claims 12 to 17, wherein the total molar ratio of the alcohol to the furfuryl alcohol contained in the first mixture and the second mixture is 2.2:1 or 1.5:

1.

24. The method of any one of claims 16 to 17, wherein when the continuous stirred tank reactor has been filled to the required volume, a certain volume of the reaction mixture is withdrawn from the continuous stirred tank reactor; and the required volume is replenished by continuously or intermittently feeding the second mixture into the reactor; and the third mixture is continuously or intermittently fed into the reactor separately, wherein the withdrawn volume is optionally fed into a second reactor connected in series with the continuous stirred tank reactor.

25. The method of claim 24, wherein the second reactor is a plug flow reactor.

26. The method of claim 24, wherein the method is carried out in the second reactor at a second reaction temperature, the second reaction temperature being the same as or 5°C to 15°C higher than the first reaction temperature in the continuous stirred tank reactor; and / or the residence time in the second reactor is in the range of 5 minutes to 120 minutes.

27. The method of claim 26, wherein the residence time in the second reactor is in the range of 10 minutes to 60 minutes.

Images