WO2014137619A1

WO2014137619A1 - Process for acid dehydration of sugar alcohols

Info

Publication number: WO2014137619A1
Application number: PCT/US2014/017563
Authority: WO
Inventors: Kenneth STENSRUD; Erik Hagberg; Stephen Howard; Erin M. ROCKAFELLOW
Original assignee: Archer Daniels Midland Company
Priority date: 2013-03-05
Filing date: 2014-02-21
Publication date: 2014-09-12
Also published as: KR20150123923A; US9630974B2; EP2964743A1; EP2964743A4; CN105026542A; EP2964743B1; JP2016516014A; US20160016970A1; JP6267240B2

Abstract

A process is described for the acid-catalyzed dehydration of a sugar alcohol, wherein the catalyst comprises a water-tolerant Lewis acid. In particular embodiments, the catalyst comprises a homogeneous water-tolerant Lewis acid, especially a homogeneous Lewis acid selected from the group consisting of bismuth (III) triflate, gallium (III) triflate, scandium (III) triflate, aluminum triflate, tin (II) triflate and indium (III) triflate. Such catalysts are effective for dehydrating both of sorbitol and the 1,4-sorbitan dehydration precursor of isosorbide, and bismuth (III) triflate particularly is beneficial for dehydrating mannitol to isomannide.

Description

PROCESS FOR ACID DEHYDRATION OF SUGAR ALCOHOLS

BACKGROUND OF THE INVENTION

|0O011 The present Invention is concerned with processes for making dehydration products from sugar alcohols, and more particularly but without limitation_: to acid-catalyzed processes for making Isohexides, such as Isosorbide, from hexito!s such as sorbitol or from rnonoanhydrohexitois such as 1.4-sorbitan.

[0002] The dehydration products that can be made by the acid- catalyzed double dehydration of sugar alcohols (that is, by removing two waters in succession from a sugar alcohol), in particular, hexitols such as sorbitol, have been the subject of extensive work. Isosorbide, also known as 1 ,4,3.6~dianbydrosorbitol. Is now commercially produced and marketed as a monomer for imparting renewable content to polyesters and polycarbonates _; and has been used as a pharmaceutical intermediate.

[0003] A variety of acid catalysts have been evaluated for use in carrying out the dehydration of sorbitol through certain monoanhvdrohexitoi intermediates (e.g., 1 ,4-sorbitan) to Isosorbide. inorganic acids such as H2SO4, H3PO4, and NCI are readily obtained, inexpensive materials but are difficult to regenerate. In order to avoid the regeneration and attendant disposal problems, solid resin catalysts have been tried. Unfortunately, In the presence of wafer and at the temperatures required for carrying out the dehydration, very few solid acids can demonstrate the activity and stability needed to begin to contemplate a commercially viable process.

[00041 United States Patents Nos. 6,849,748; 7,420,067; 7,439,352; 7,772,412 and 7,982,059 provide examples of prior art methods for producing isohexides (also referred to as anhydrosugar alcohols, anhydrohex!tois, anhydroalditols etc) such as isosorbide, from sorbitol from dextrose.

[0005J Commonly-assigned US Patent No. 6.849,748 to Moore et ai., for example, describes a solvent-free process wherein a sugar alcohol^■··· such as sorbitol - is heated with stirring until molten, and then dehydrated in the presence of a soluble acid or acidic ion exchange resin with stirring, under vacuum (to remove the water product and drive the reaction toward the products) and at an elevated temperature, then the resulting anhydrosugar alcohol Is purified by distillation, followed by melt crystallization and/or redistillation. The final purified product Is isolated by centrifugation or filtration. Enumerated preferred acid catalysts include sulfuric acid, phosphoric acid, p-to!uenesuifonie acid, and p-methanesulfonic acid.

100061 Commonly-assigned US Patent No. 7,420,067 mentions these same acids, as well as acidic ion exchange resins and acidic zeolite powders as additional options. Successive film evaporators, especially wiped film evaporators under vacuum, are described for use In purifying the product Isosorbide.

[0007] US Pat. No, 7,772,412 io Holladay et al. describes a process for making isosorbide wherein sorbitol Is fed to a reactor containing a dehydration catalyst and a hydrogenation co-catalyst, with hydrogen being supplied countercurrentiy to the reactor for removing water as it is formed and for "reducing or eliminating ...o!igomeric or polymeric material in the dehydration product^", to which undesirable color formation had been attributed. Suitable dehydration catalysts include the mineral acid catalysts, solid acid catalysts such as the heteropolyaeids, mesoporous silicas, acid clays, sulfated zirconia, molecular sieve materials, cation exchange resins and zeolites, and combinations of any of these. The hydrogenation catalyst is described as typically being a supported metal or multi-metal catalyst. Palladium in particular is described as especially preferable for the metal, with platinum, nickel, cobalt, ruthenium, rhenium, rhodium, Iridium and iron also being listed.

[00081 US Pat. No. 7,982,059 describes a process for converting aqueous sorbitol to xylite! and isosorbide in the presence of an acid catalyst and without a hydrogenation co-catalyst mo e particularly Involving reacting an aqueous sorbitol solution with an acid zeolite at about 250 degrees Celsius and a pressure maintained at from about 68 bars to about 80 bars to produce the xylite! and isosorbide. [00091 WO 2013/138163 to Binder et al. describes a process for forming one or more dehydration products from an aqueous sugar alcohols solution including one or more alcohols from pentoses and hexoses, wherein the aqueous sugar alcohols solution is subjected to an acid -catalyzed dehydration using a substituted sulfonic acid catalyst soluhilized in the aqueous sugar alcohols solution. In certain preferred embodiments, the dehydration process is conducted rapidly and with rapid cooling of the dehydration products prior to any separation of the residual sugar alcoho s) from the dehydration products in the overall product mixture, in the manner prescribed for the dehydration of aqueous sugar solutions in WO 2013/106136 to Sanborn et al.

SUMMARY OF THE INVENTION

[0010} The following presents a simplified summary of the invention in order to provide a basic understanding of some of its aspects. This summary is not an extensive overview of the invention and is intended neither to Identify key or critical elements of the invention nor to delineate its scope. The sole purpose of this summary is to present some concepts of the invention in a simplified form as a prelude to the more detailed description that Is presented later.

[0011] With this in mind, the present invention in a first aspect concerns a process for the acid-catalyzed dehydration of a sugar alcohol wherein the catalyst comprises a water-tolerant Lewis acid, in particular embodiments, the catalyst comprises a homogeneous water-tolerant Lewis acid, especially a homogeneous Lewis acid selected from the group consisting of bismuth (ill) irifiate, gallium (ill) inflate, scandium (IM) inflate, aluminum inflate, fin (II) inflate and Indium (111) triflate. Such catalysts are effective for dehydrating both of sorbitol and the 1 ,4-sorbiian dehydration precursor of Isosorblde, and bismuth (III) triflate particularly is beneficial for dehydrating mannitoi to iso annide , so that in a second, more particular aspect the present invention concerns an improved process for making an isohexide from a corresponding hexltol. DETAILED DESCRIPTION OF PREFERRED EMBODIMENTS

[0012] A preferred process according to the present invention for dehydrating sorbitol involves mixing sorbitol with from 0.005 mo! percent and greater of a water-tolerant Lewis acid, heating to at least 140 degrees Celsius, and carrying out the acid-catalyzed dehydration of sorbitol isotherma!ly for an hour or longer under a reduced pressure to continuously remove water from the reaction. The water-tolerant Lewis aoid is preferably one or more of bismuth (III) trifiate, gallium (HI) trifiate, scandium (ill) trifiate, aluminum trifiate, tin (II) trifiate and indium (ill) trifiate, and while yields of isosorbide and the 1 ,4-sorbitan precursor of isosorbide obtained from these catalysts can be seen from the examples below to vary somewhat dependent on the catalyst used, the catalyst loading and reaction conditions of temperature and duration, It Is expected that catalyst loadings of not more than 0 1 mol percent, temperatures of not more than 160 degrees Celsius and reaction times of not more than 3 hours will provide commercially acceptable yields of isosorbide, The crude product mixture may then be purified according to any of the known methods for doing so.

[0013] As will be evident from the examples that follow, use of the preferred Lewis acids under these conditions provides a number of benefits, including enhanced yields of Isosorbide and of the 1 ,4-sorbitan precursor of isosorbide as compared to the most effective Brbnsted acid surveyed, namely, sulfuric acid, much reduced cataiyst loadings for achieving a targeted yield of Isosorbide, avoidance of the neutralization requirements posed by the conventional Brbnsted acids before distillation of the crude product mixture and better color of the crystalline isosorbide distillates that may be realized.

|0O14] While some or all of these benefits are expected to be attainable in the dehydration of sugar alcohols generally (where "sugar alcohols" is understood to include partially dehydrated sugar alcohols such as, for example, monoanhydrobexitols from the partial dehydration of hexitols), the extent to which certain benefits or advantages are observed, the pellicular water-tolerant Lewis acid catalysts that prove most effective and the optimum process conditions for carrying out the Lewis aeid-ealaiyzed dehydrations can he expected to vary somewhat from one sugar alcohol to the next. As an example, we found bismuth inflate to be particularly advantageous for catalyzing the dehydration of mannitol to isomannide. Those skilled in the art will be well able, in any event, to determine the optimum features of a process for dehydrating a particular sugar alcohol using a water-tolerant, Lewis acid catalyst as claimed herein by routine experimentation.

[001SJThe present invention is further illustrated by the following., non- limiting examples:

£00163 Comparative Examples 1 -6

10017] For benchmarking the performance of the water -tolerant Lewis acid catalysts of the present invention, a number of Bronsted acids were evaluated for the acid-catalyzed dehydration of sorbitol. In each instance, a three neck, 250 ml. round bottomed flask equipped with a magnetic stir bar was charged with 100 grams of sorbitol (0.549 mol), then was immersed in an oil bath set at 140 degrees Celsius. Once the sorbitol liquefied and attained an internal temperature of 140 degrees as determined by an internal temperature probe, a quantity (2 mol percent in all cases except for phosphoric acid, which was added at 5 mol percent) of the Brdnsted acid In question was introduced by syringe through a rubber septum-capped neck. Under a reduced pressure of less than 5 ton, the reaction was then continued isothermaiiy for 1 hour. After this time, the vacuum was broken, the crude product mixture was cooled and quenched with 50 percent aqueous sodium hydroxide, then was weighed and quantitatively analyzed by gas

chromatography. The results, shown in Table 1 , show that sulfuric acid was the most effective Bronsted acid of those surveyed for dehydrating sorbitol under the indicated conditions, though unidentified side products accounted for about 23 percent of the crude product mixture. Table 1

Acid p a isosGrbide 2J3;;SgrbiiaQ Accountability c n ersion yieid (wt %)

.imoL% ! CmsL%}

Sulfuric -5 100 00 67.72 0 00 9.30 76.35 p-toiiienesuif nic -2 ,3 100.00 22.83 54.90 9.30 87.00

Msthanesulfonie -1.3 100.00 18.20 59.30 8.70 86.20

Oxalic 1 .25 12 88 0 00 3.62 1 .48 92.22

Beta!ns HC! 1 .84 14.30 0.00 4.40 1 .33 91 .1

Phosphoric 2, 14 72.00 5.18 55.SS 3. 15 92.31

1001 SI Comga we^

[0019] The same experimental setup, procedure and conditions were used as in Comparative Examples 1-8, except that 0.1 moi peroent of various Lewis acids (for Examples 1-8) or 0.1 moi percent of sulfuric acid (for

Comparative Example 7} was used. The results were as shown in Table 2, as follows:

Table 2

f00201 Comparative Example 8 and Exampies 7-12

[0021] The same experimental setup, procedure and conditions were used as in Comparative Example 7 and Examples 1-8 (0.1 moi percent of catalyst), except that the reaction was continued for 2 hours at 140 degrees Celsius after introduction of the catalyst, as opposed to 1 hour, The results are shown in Table 3: Table 3

Acid % sor itpi isosgrfcide Accountability.

conversion yisid mgl yield (moi

m %2 M

Si(OT¾ 99.53 23.73 61.06 7.87 94 96 in{GTf)₃ 98 79 19.20 €6.4 ·; 7.74 100.00

Sc{OTf)₃ 99.58 25.68 56 64 8 77 96.43

Ga{OTf}₃ 99.88 31.59 49 67 7 39 9:2.09

Sn{OTf}₃ 94.06 12.70 73.49 8.03 100.00

A!(OTf)₃ 100.00 29.40 63.87 S.0Q 94 10

Sulfuric 83.63 6.33 6S.34 6.23 98.87

100221 Comparative Example 9 and Examples 13-18

[0023] The same experimental setup, procedure and conditions were used as in Comparative Example 8 and Examples 7-12 (0.1 mol percent of catalyst), except that the reaction was continued for 3 hours at 140 degrees Celsius after introduction of the catalyst, as opposed to 2 hours. The results are shown in Table 4:

Table 4

Acid % sorbitol !sosor ide l ^ sor ijan 2,S-aort¾tan Accc^tabJIty.

conversion yjeidlmo! ii imsi

%i % ¾i

BI{OTf);J 99.86 32.46 5109 7.33 92.67

In(OTf)?. 100.00 44 66 36.27 3.24 90.15

Sc{OTf}₃ 100.00 49.36 32.2S 9.01 88,40

Ga(OTf)s 100.00 67.20 3.96 6.96 79.37

Sn(OTf>₃ 100.00 24.02 66 37 7 90 100.00

A!(OTf}₃ 100.00 47.13 31.85 7 70 88.98

Sulfuric 100.00 25.30 60.22 5 60 91.88 [0024] Comparative Example 10 and Examples _.1_.9-24

[0Q2S]The same experimental setup and procedure were used as in previous examples, except that the reaction temperature was increased to 180 degrees Celsius, and the reaction was continued for 1 hour after introduction of the acid catalyst being evaluated (again at 0.1 moi percent). Results were as shown In Table 5:

Table 5

Isosor ide 2_:5_;;Sor ! 3n Accpunia tv con ersion iejdilBS! Mli imo! ;e!d.irnol (wt %j

%1 %i %i

BKOTfh 100.00 62.02 3.83 7 34 76.21 in-iOTfh 100.00 88.40 8 74 7.89 83.63

Sc{OTf}₃ 100.00 32.10 46.21 8 15 89. 3

Gs(OTf}₃ 100.00 64.62 5.07 6.23 77.71

Sn(OTf);; 97.95 17.95 64.64 9.68 97.35

MOTf}₃ 100 00 48 01 28.48 7.64 84.00

Sulfuric 100.00 49.78 26.35 8 70 88.92

[Q028]Comparaj.^

[0027] The acids were evaluated at a lower catalyst load of 0.05 mo! percent, the lower temperature of 140 degrees Celsius and with a reaction time of two hours, with the results shown In Table 8 as fo!io s:

Table 6

[0028] C^^r iye Example 12 and Examples 31-36

[0029] The acids were evaluated at the lower catalyst load of 0.05 mol percent used in Examples 25-30, but at the higher temperature of 180 degrees Celsius and with a reaction time of one hour rather than two after introduction of the catalyst being evaluated, with the results shown In Table 7 as follows:

Table 7

00301 Comparative Example 13 and Examples 37-42

[0031] The acids were evaluated at a still lower catalyst load of 0.01 moi percent, at a temperature of 160 degrees Celsius and with a reaction time of one hour after introduction of the catalyst being evaluated, with the results shown in Table 8 as follows:

Table S

Acid % sor j ol isgsof ide 2_:5;:Scr it n Accountability inversion .vM ims! f*t %;

%i m m

Bi{OTf)₃ 58.13 2.95 48.02 4 9 109 00 in{OTf}₃ 71.67 5.19 58.41 6.23 38.70

SciQTf 40.40 1.74 34.48 3.43 99 46

Gs(OTf)₃ 67.72 4,42 58 30 6 08 99.66

Sn(OTf};j 71.00 4 88 58.58 6.10 99.21 OTi}₃ 64.41 3.56 54.54 5.65 100.00

Sulfuric 26.90 0.00 26 17 0.61 99.88 0032] Comparative Example 14 and Examples 43-44

|0033] The acids were evaluated at a still lower catalyst load of 0.005 mol percent, at a temperature of 60 degrees Celsius and with a reaction time of one hour after introduction of the catalyst being evaluated, with the results shown in Table 9 as follows:

Table 9

a IsGSor ide 2,5;: Accouniabls!^

sorbin. sorbiian

%! 1

Bi(OT¾ 71 .06 4.58 58.47 6.38 100.00 in{Off):₅ 88.69 9.99 68.93 8.26 100.00

Sulfuric 21.10 0.00 19.22 0.87 100,00

[0035] For these examples, two runs were conducted using differing amounts of sulfuric acid (0.1 moi percent for Comparative Example 15 and 1 mol percent for Comparative Example 18) to catalyze the dehydration of manni oi to isomannlde and anhydromannitois, and the results were

compared to a run using 0.1 mol percent of bismuth (111) friflate under the same conditions of 160 degrees Celsius, one hour run time and a reduced pressure of 20 torr.

[0036] For the two sulfuric acid experiments, a three neck 250 ml round bottomed flask equipped with a magnetic stir bar was charged with 100 grams of mannitoi (0.549 mol), then immersed In an oil bath maintained at 160 degrees Celsius. Once the mannitoi liquefied and attained an internal temperature of 160 degrees as measured by an internal temperature probe, a condenser was outfitted onto one of the flask necks and vacuum was initiated. The sulfuric acid was then introduced via syringe through a rubber septum capped neck. After an hour, the vacuum was broken, and the crude product mixture was cooled, weighed and quantitatively analyzed by gas

chromatography. [0037] For the run with the inventive bismuth triflate catalyst, a three neck 250 ml round bottomed flask was charged with the mannitol and with 360 milligrams of the bismuth triflate catalyst, then Immersed in the 180 degree Celsius oil bath Once the mannitol liquefied and the bismuth triflate dissolved in the mannitol and as the mixture achieved an internal

temperature of 180 degrees Celsius, then a condenser was outfitted onto one of the flask necks and vacuum was initiated down to a pressure of 20 ton. After one hour, the vacuum was broken, and the crude product mixture was cooled, weighed and quantitatively analyzed by gas chromatography,

[0038] The results were that 100% conversion of the mannitol was realized in ail three runs, but the yields of iso annide (expressed In moi percents) were much greater using the bismuth triflate catalyst: sulfuric acid at 0.1 mol percent gave only 2 percent of isomannide, whereas at 1 mo! percent addition the Isomannide yield was 25 percent. However, by comparison, the inventive bismuth triflate gave 61 percent of isomannide.

i f

Claims

What is claimed is:

1 . A process for she acid-catalyzed dehydration of a sugar alcohol, comprising contacting a sugar alcohol with a water-tolerant Lewis acid catalyst at a temperature and for a time sufficient to produce water and at least a partially dehydrated sugar alcohol product.

2. A process according to claim 1 , wherein a homogeneous water-tolerant Lewis acid catalyst is used in the process.

3. A process according to claim 2, wherein a catalyst is used that is selected from the group consisting of bismuth (!!!) trif!ate, gallium (!!!) trlf late _: scandium (HI) inflate, aluminum inflate, indium (HI) trifiate, tin (I!) inflate and

combinations of two or more of these.

4. A process according to claim 3_: wherein sorbitol is dehydrated to produce at least some isosorbide.

5. A process according to claim 4, wherein at least 0.005 mo! percent of the

catalyst is used,

6. A process according to claim 5, comprising dehydrating sorbitol ai a

temperature of least 140 degrees Celsius.

7. A process according to claim 8_: wherein the dehydration Is accomplished over a period of at least 1 hour at a temperature of at least 140 degrees Celsius.

8. A process according to claim 7, wherein the dehydration is accomplished

under reduced pressure with continuous removal of water from the product mixture in the course of the dehydration.

9. A process according to claim 3, wherein mannltoi is dehydrated to produce at least some isomannide in the presence of a bismuth trifiate catalyst.

10. A process for producing isosorbide from sorbitol, comprising contacting the sorbitol with an effective amount of homogeneous Lewis acid catalyst selected from the group consisting of bismuth (ill) inflate, gallium (IN) inflate, scandium (III) inflate, aluminum inflate, indium (III) triflaie, tin (II) inflate and combinations of tvvo or more of these, at a temperature and for a time sufficient to produce a product mixture including isosorbide.

1 1. A process according to claim 10, wherein from 0,005 moi percent to 0.1 mol percent of catalyst is used, at a temperature of from 140 degrees to 180 degrees Celsius and over a period of from 1 hour to 3 hours.

12 A process according to claim 1 wherein the dehydration is accomplished urvder reduced pressure with continuous removal of water from the product mixture In the course of the dehydration.