The present invention relates to an improved process of producing crystalline fructose. The process is based on chromatographic fractionation of fructose crystallization run-offs with cation exchange resins in two different ion forms and introduction of the fructose fractions thus obtained into further crystallization for the production of crystalline fructose.
One of the problems associated with conventional fructose crystallization methods relates to the presence of crystallization inhibitors, especially dimeric sugars of fructose, in the crystallization. The formation of dimeric sugars is especially accelerated in prolonged heating of a concentrated fructose solution in an acidic environment. Dimeric sugars having an inhibiting effect on the crystallization of fructose are formed in the enzymatic or acid inversion of sucrose to a mixture of fructose and glucose, in the enzymatic isomerization of glucose to fructose and in the crystallization process itself especially during evaporation. The presence of dimeric sugars and other impurities in the crystallization of fructose leads to lower process yields. The presence and formation of dimeric sugars in the crystallization of fructose is discussed for example in the Example at columns 17 and 18 of
Another problem associated with conventional fructose crystallization methods relates to a low overall yield of crystalline fructose from the fructose feed solution for the reason that fructose has not as a rule been recovered by crystallization from crystallization run-offs because of a low yield due to dimeric crystallization inhibitors such as difructose dianhydrides, which are concentrated into the run-offs. Instead, the run-offs have conventionally been used for preparing liquid fructose syrups, for example.
It is thus an object of the present invention to provide a process of producing crystalline fructose so as to alleviate the above disadvantages, such as an insufficient yield of crystalline fructose due to the presence and formation of crystallization inhibitors in the fructose crystallization leading to inefficient crystallization of fructose from crystallization run-offs. The object of the invention is achieved by a process which is characterized by what is stated in the independent claim. The preferred embodiments of the invention are disclosed in the dependent claims.
The invention is based on the treatment of fructose crystallization run-offs by a chromatographic fractionation with cation exchange resins in two different ion forms and introduction of the fructose fractions thus obtained into further crystallization for the production of further crystalline fructose. The process of the invention provides an improved overall yield of crystalline fructose calculated on the basis of the fructose source as well as on the basis of fructose in the crystallization batch.
The following drawings are illustrative embodiments of the invention and are not meant to limit the scope of the invention as defined in the claims in any way.
The present invention relates to a process of producing crystalline fructose with a high overall yield of crystalline fructose from the starting material of a fructose process and from the crystallization feed.
The process of the invention comprises, as characteristic elements, the following steps:
The crystallization as such is carried out by conventional methods known in the art for the crystallization of fructose, advantageously by cooling crystallization or by other methods such as boiling crystallization, or precipitation crystallization, or a combination thereof. Seeding of the crystallization mass may be used, if desired. The crystallization may be carried out in water, but an alcohol, such as ethanol, or a mixture of water and alcohol, can also be used. The crystallization is preferably carried out at a pH in the range of 4.5 to 5.5. The crystals are separated from the crystallization massecuite for instance by centrifugation or filtering to provide crystalline fructose and a fructose run-off.
Said one or more crystallization steps provide crystalline fructose and a fructose run-off, which can be crystallized to provide further crystalline fructose and a further fructose run-off.
The crystallization run-offs contain high amounts of fructose as well as reasonable amounts of glucose and disaccharides. In a typical embodiment of the invention, the crystallization run-offs may contain 88 to 96% of fructose, 2 to 5% of disaccharides and 1 to 8% of glucose, based on the dry substance content (DS) of the run-offs.
Disaccharides, especially fructose dimers, act as crystallization inhibitors in the crystallization of fructose, whereby they should be removed as efficiently as possible from the crystallization run-offs before subsequent recovery of fructose by further crystallization.
Consequently, in a further step of the process of the invention, at least part of said fructose crystallization run-offs are subjected to chromatographic fractionation in a separation system, which comprises two or more cation exchange resin beds, whereby at least one of said cation exchange resin beds is in a Ca2+ form and at least one of said cation exchange resin beds is in a monovalent cation form.
The resin in a monovalent cation form may be in a Na+ form or K+ form, for example.
In a preferred embodiment of the invention, the chromatographic fractionation by the separation system described above provides from the fructose run-off a fructose fraction, which contains 94 to 98% of fructose, less than 3%, preferably less than 2% of disaccharides and less than 1.5%, preferably less than 1.0% of glucose on DS.
Said disaccharides existing in the fructose run-offs may be fructose dimers, degradation products of starch, such as maltose or isomaltose or a residue of raw material such as saccharose. In connection with the present invention, said disaccharides are especially selected from disaccharides A and disaccharides B, which are mainly composed of fructose dimers, such as difructose dianhydrides. Fructose dimers have been formed in acidic conditions, in a concentrated fructose solution in enzymatic conversions (inversion, isomerization), in acidic inversion of saccharose and during the crystallization process of fructose. Fructose dimers such as difructose dianhydrides include various difructose anhydrides and various diheterolevulosans, which have been formed from the polymerization of fructose during a prolonged treatment of fructose in a concentrated form in acid or alkaline conditions.
Disaccharides may be analyzed by HPLC with a Na+ -form resin using water as an eluant. The elution order of disaccharides in the HPLC-diagram is saccharose (similar retention time also for maltose and isomaltose), disaccharides B, disaccharides A, glucose and fructose. Disaccharides A and B show two separate peaks in the HPLC diagram. HPLC-peaks for disaccharides A and B mainly consist of different difructose dianhydrides, which are considered to be more harmful crystallization inhibitors than the other type of disaccharides such as saccharose, maltose or isomaltose.
The disaccharide A type of difructose dianhydrides is more difficult to remove from a fructose run-off solution by chromatographic separation than disaccharides B. Inventors have surprisingly discovered that the disaccharide A type of difructose dianhydrides can be efficiently separated from fructose solution using a cation exchange resin in a monovalent form in the chromatographic separation. The cation exchange resin may be a strong or weak cation exchange resin. Especially advantageous removal of disaccharides A, disaccharides B and glucose and low loss of fructose can been achieved when using cation exchange resin beds in Ca2+- form and Na+ -form consecutively. The accumulation of crystallization inhibitors to the fructose run-off will be prevented and the separated fructose fraction depleted in fructose dimers can be recycled back to the crystallization to obtain a total yield of crystalline fructose up to 93 to 97% based on the fructose feed to crystallization or 93 to 98% based on the sucrose used for the inversion and isomerization, if sucrose has been used as a basic source of fructose.
The fructose fraction obtained from the chromatographic fractionation of the invention typically comprises less than 1.5%, preferably less than 1.0% of disaccharides A and less than 1.5%, preferably less than 0.8% of disaccharides B on DS. A typical range for disaccharides A is 0.5 to 1% on DS and for disaccharides B 0.5 to 0.8% on DS.
The separation of disaccharides A and disaccharides B from fructose can be seen from
Said at least one other fraction in the chromatographic fractionation of fructose run-off may be a residue fraction enriched in fructose dimers. Consequently, in addition to the fructose fraction, which is depleted in fructose dimers (disaccharides A and B), a residue fraction can be obtained, which is enriched in fructose dimers (disaccharides A and B). The fructose fraction depleted in fructose dimers (disaccharides A and B) typically comprises less than 1.5%, preferably less than 1.0% disaccharides A and less than 1.5%, preferably less than 0.8% disaccharides B on DS. The residue fraction enriched in fructose dimers typically contains 2 to 8%, preferably 4 to 8% disaccharides A and 3 to 10%, preferably 8 to 10% disaccharides B on DS.
The chromatographic separation system used in the process of the present invention comprises two or more cation exchange resin beds, whereby at least one of the resin beds in a Ca2+ form and at least one of the resin beds is in a monovalent cation form.
In one embodiment of the invention, at least 20% of the total length of the resin beds of the separation system is in a Ca2+ form. In another embodiment of the invention, at least 20% of the total length of the resin beds of the separation system is in a monovalent cation form.
The relation between the length of the Ca2+ resin bed/beds and the other resin bed/beds may be adjusted depending on the content of disaccharides A and B and glucose in the fructose crystallization run-off used as the feed for the chromatographic fractionation. Consequently, in a further embodiment of the invention, when the feed contains less than 2% disaccharides A and B on DS and more than 3% glucose on DS, 60 to 80% of the total length of the resin beds of the separation system may be in a Ca2+ form. In another embodiment of the invention, when the feed contains more than 3% disaccharides on DS and less than 2% glucose on DS, 60 to 80% of the total length of the resin beds may be in a monovalent cation form.
Said two or more cation exchange resin beds of the separation system are preferably composed of strongly acid cation exchange resins. The resins have typically a styrene skeleton, which is preferably cross-linked with divinylbenzene.
Said two or more cation exchange resin beds may also comprise a weakly acid cation exchange resin. The weakly acid cation exchange resin may be in a free acid form, for example.
Said two or more cation exchange resin beds may be arranged in series or in parallel.
The chromatographic fractionation in accordance with the present invention may be carried out with a batch process or a simulated moving bed process (SMB process). The simulated moving bed process may be continuous or sequential.
The temperature of the chromatographic fractionation is typically in the range of 20 to 90°C, preferably 40 to 65°C. The pH of the solution to be fractionated is typically in the range of pH 3 to pH 6, preferably in the range of pH 4 to pH 5 to minimize further formation of fructose-based disaccharides.
The chromatographic fractionation provides the fructose fraction with a fructose yield of more than 80%, preferably more than 90% and with an especially preferred fructose yield of more than 95% based on fructose in the fructose crystallization run-off used as the feed in the chromatographic fractionation. In the fructose fraction, the amount of disaccharides A is reduced to a level of less than 60% and the amount of disaccharides B is reduced to a level less of than 40% based on the disaccharide A or B content in the fructose run-off.
The fructose fraction obtained from the chromatographic fractionation has a typical fructose purity of more than 93%, preferably more than 95% and more preferably more than 97%, based on dissolved dry substance (DS). The removal of disaccharides B is as a rule more efficient than the removal of disaccharides A.
The residue fraction obtained from the chromatographic fractionation of the fructose crystallization run-off with the process of the invention contains fructose in an amount of 45 to 65% on DS, glucose in an amount of 10 to 30% on DS, preferably 20 to 30% on DS, disaccharides A in an amount of 2 to 8% on DS, preferably 4 to 8% on DS, disaccharides B in an amount of 3 to 10% on DS, preferably 8 to 10% on DS. The dry substance yield to the residue fraction represents 5 to 15 weight-%, preferably 5 to 8 weight-% of the run-off.
Typical yield of crystalline fructose in a single pass (no previous crystallizations) in the crystallization of a solution containing over 95% fructose on DS is in the range of 40-55%, normally about 45% of the fructose in the feed. Generally the yield of crystalline fructose in a second pass (crystallization of a run-off from a previous crystallization) is less than 90% of the yield of the first pass. Also a much longer time is required for the crystallization of the second pass due to the crystallization inhibitors.
The efficient removal of the crystallization inhibitors with the method of the invention facilitates the overall crystallization yield of the crystalline fructose to be more than 90%, preferably more than 93% and most preferably over 95%, based on the fructose in the crystallization feed.
The fructose fraction depleted in fructose dimers, obtained from the fractionation of the fructose run-off is then introduced into at least one of said one or more crystallization steps for the production of crystalline fructose. This step may be realized in different ways.
In one embodiment of the present invention, the process may comprise the following steps:
In another embodiment of the present invention, the process may comprise the following steps:
Conventional sources for obtaining fructose are as a rule mixtures of glucose and fructose, such as solutions of inverted sucrose, solutions of isomerized glucose (starch based) and mixtures thereof. The fructose source may also be a fructose solution obtained by hydrolysing inulin.
The solution of inverted sucrose is typically obtained by enzymatic or acidic inversion of sucrose to a mixture of fructose and glucose. One process for enzymatic inversion is described in
The solution of isomerized glucose is obtained from starch-based glucose syrup, which is isomerized with an isomerase enzyme to a mixture of glucose and fructose. The enzymatic isomerization is preferably carried out at the optimum pH range for the enzyme, i.e. at a pH of about 8. One process for the isomerization is disclosed in
Even mixtures of inverted sucrose and isomerized glucose can be used as a source in the production of crystalline fructose.
Consequently, the first step in the fructose process typically comprises separation of these mixtures of glucose and fructose into a glucose fraction and a fructose fraction. To achieve a high process yield of crystalline fructose especially from sucrose, the crystallization inhibitors, such as dimeric sugars formed either in the inversion or isomerization process or during the crystallization, should be removed as efficiently as possible by various separation techniques.
In a further embodiment of the present invention, the process may thus also comprise, as a preceding step, chromatographic fractionation of a solution containing glucose and fructose to obtain a glucose fraction and a fructose fraction for producing crystalline fructose.
In a still further embodiment of the invention, the process may also comprise the following steps:
The chromatographic separation of glucose and fructose is typically carried out with a cation exchange resin, preferably with a strongly acid cation exchange resin. The resin is preferably in a divalent cation form, whereby the divalent cation is typically selected from Ca2+, Mg2+ or Sr2+, especially Ca2+. The resin has typically a styrene skeleton, which is preferably cross-linked with divinylbenzene.
In a typical embodiment of this latter embodiment of the invention, the starting solution containing glucose and fructose is selected from inverted sucrose, converted and isomerized starch and isomerized glucose.
In a still further embodiment of the invention, the process may com-comprise the following steps:
In a still further embodiment of the invention, the process may comprise the following steps:
One further embodiment of the process of the invention comprises, as characteristic elements, the following steps:
In a still further embodiment of the process of invention, where the further steps (e) to (i) comprise the following:
In another embodiment of the process of the invention, where the further steps (e) to (i) comprise the following:
In a still further embodiment of the process of the invention, steps (e) to (i) comprise the following:
Furthermore, the fructose fraction obtained from the chromatographic fractionation of a fructose crystallization run-off in accordance with the present invention is as a rule subjected to decolorization before the next crystallization.
The embodiment is especially efficient, if the source of glucose/fructose solution is sucrose, because the feed to the second crystallization is a mixture of fructose run-off (1) and fructose fraction (2) containing a low amount of crystallization inhibitors and a high amount of fructose to facilitate an overall yield of crystalline fructose of up to 98% of the sucrose. The process schema is also advantageous in respect to the required chromatographic system, because the volume of the fructose run-off (2) to be separated is significantly lower (30-40%) than if the total volume of run-off (1) would be separated.
The following examples represent illustrative embodiments of the invention without limiting the scope of the invention in any way.
In the examples and throughout the specification and claims, the following definitions have been used:
The content of fructose and glucose was measured as follows: Liquid chromatography HPLC, Na+ form cation exchange column at a temperature of 85°C with a flow rate of 0.8 ml/min using water as the eluent.
The content of disaccharides A and disaccharides B was measured by HPLC (liquid chromatography) in the following conditions:
The abbreviation "tn" refers to 1000 kg.
"Mother liquor" refers to a fructose crystallization run-off.
The process equipment included four columns connected in series, feed pump, recycling pumps, eluent water pump, heat exchangers, flow control means for the out-coming liquids as well as inlet and product valves for the various process streams. The height of each column was 4.0 m and each column had a diameter of 3.1 m. The columns were packed with a strong acid gel type cation exchange resin (manufactured by Finex) in Ca2+-form. The divinylbenzene content of the resin was 5.5% and the mean bead size of the resin was 0.36 mm.
As a feed, enzymatically inverted sucrose (obtained by a process described in
The liquor concentration was 67.5 g/100 ml and the pH was 4.1. The fructose syrup was composed as set forth below, whereby the percentages are given on a dry substance weight basis.
The fractionation was performed by way of a 12-step SMB sequence as set forth below. The feed and the eluent were used at a temperature of 65°C and water was used as an eluent.
Step 1: 3.5 m3 of feed solution were pumped into the first column at a flow rate of 25 m3/h and a glucose fraction was collected from the same column. Simultaneously 4.45 m3 of water were pumped into the third column at a flow rate of 35 m3/h and fructose fraction was collected from the same column.
Step 2: 2.85 m3 of water were pumped into the third column at a flow rate of 28 m3/h and glucose fraction was collected from the first column.
Step 3: 15.6 m3 were circulated in the column set loop, formed with all columns, at a flow rate of 30 m3/h.
Step 4: 3.5 m3 of feed solution were pumped into the second column at a flow rate of 25 m3/h and a glucose fraction was collected from the same column. Simultaneously 4.45 m3 of water were pumped into the fourth column at a flow rate of 35 m3/h and fructose fraction was collected from the same column.
Step 5: 2.85 m3 of water were pumped into the fourth column at a flow rate of 28 m3/h and glucose fraction was collected from the second column.
Step 6: 15.6 m3 were circulated in the column set loop, formed with all columns, at a flow rate of 30 m3/h.
Step 7: 3.5 m3 of feed solution were pumped into the third column at a flow rate of 25 m3/h and a glucose fraction was collected from the same column. Simultaneously 4.45 m3 of water were pumped into the first column at a flow rate of 35 m3/h and fructose fraction was collected from the same column.
Step 8: 2.85 m3 of water were pumped into the first column at a flow rate of 28 m3/h and glucose fraction was collected from the third column.
Step 9: 15.6 m3 were circulated in the column set loop, formed with all columns, at a flow rate of 30 m3/h.
Step 10: 3.5 m3 of feed solution were pumped into the fourth column at a flow rate of 25 m3/h and a glucose fraction was collected from the same column. Simultaneously 4.45 m3 of water were pumped into the second column at a flow rate of 35 m3/h and fructose fraction was collected from the same column.
Step 11: 2.85 m3 of water were pumped into the second column at a flow rate of 28 m3/h and glucose fraction was collected from the fourth column.
Step 12: 15.6 m3 were circulated in the column set loop, formed with all columns, at a flow rate of 30 m3/h.
After equilibration of the system, the following fractions were drawn from the system: one glucose-enriched fraction from each column and one fructose-enriched fraction from each column. The results including HPLC analyses for the combined fractions are set forth in the table below.
The overall fructose yield calculated from these fractions was 95.2%.
The process equipment included four columns connected in series, feed pump, recycling pumps, eluent water pump, heat exchangers, flow control means for the out-coming liquids as well as inlet and product valves for the various process streams. The height of each column was 4.0 m and each column had a diameter of 0.2 m. The columns were packed with a strong acid gel type cation exchange resin (manufactured by Finex) in Ca2+-form. The divinylbenzene content of the resin was 5.5% and the mean bead size of the resin was 0.36 mm.
As a feed, an enzymatically isomerized glucose solution (
The liquor concentration was 64.2 g/100 ml and the pH was 4.0. The fructose syrup was composed as set forth below, whereby the percentages are given on a dry substance weight basis.
The fractionation was performed by way of a 12-step SMB sequence as set forth below. The feed and the eluent were used at a temperature of 65°C and water was used as an eluent.
Step 1: 17.9 l of feed solution were pumped into the first column at a flow rate of 110 l/h and a glucose fraction was collected from the same column. Simultaneously 17.4 l of water were pumped into the third column at a flow rate of 110 l/h and fructose fraction was collected from the same column.
Step 2: 17.4 l of water were pumped into the third column at a flow rate of 120 l/h and glucose fraction was collected from the first column.
Step 3: 59.5 l were circulated in the column set loop, formed with all columns, at a flow rate of 125 l/h.
Step 4: 17.9 l of feed solution were pumped into the second column at a flow rate of 110 l/h and a glucose fraction was collected from the same column. Simultaneously 17.4 l of water were pumped into the fourth column at a flow rate of 110 l/h and fructose fraction was collected from the same column.
Step 5: 17.4 l of water were pumped into the fourth column at a flow rate of 120 l/h and glucose fraction was collected from the second column.
Step 6: 59.5 l were circulated in the column set loop, formed with all columns, at a flow rate of 125 l/h.
Step 7: 17.9 l of feed solution were pumped into the third column at a flow rate of 110 l/h and a glucose fraction was collected from the same column. Simultaneously 17.4 l of water were pumped into the first column at a flow rate of 110 l/h and fructose fraction was collected from the same column.
Step 8: 17.4 l of water were pumped into the first column at a flow rate of 120 l/h and glucose fraction was collected from the third column.
Step 9: 59.5 l were circulated in the column set loop, formed with all columns, at a flow rate of 125 l/h.
Step 10: 17.9 l of feed solution were pumped into the fourth column at a flow rate of 110 l/h and a glucose fraction was collected from the same column. Simultaneously 17.4 l of water were pumped into the second column at a flow rate of 110 l/h and fructose fraction was collected from the same column.
Step 11: 17.4 l of water were pumped into the second column at a flow rate of 120 l/h and glucose fraction was collected from the fourth column.
Step 12: 59.5 l were circulated in the column set loop, formed with all columns, at a flow rate of 125 l/h.
After equilibration of the system, the following fractions were drawn from the system: one glucose-enriched fraction from each column and one fructose-enriched fraction from each column. The results including HPLC analyses for the combined fractions are set forth in the table below.
The process equipment included a separation column, feed pump, eluent water pump, heat exchangers, flow control means for the out-coming liquid as well as inlet and product valves for the various process streams. The height of the column was 3.3 m and column had a diameter of 2.76 m. The column was packed with a strong acid gel type cation exchange resin (manufactured by Finex) in Na+-form. The divinylbenzene content of the resin was 5.5% and the mean bead size of the resin was 0.35 mm.
As a feed, a fructose crystallization run-off was used and the aim was to separate fructose contained therein
The liquor concentration was 41.9 g/100 ml and the pH was 4.0. The fructose run-off was composed as set forth below, whereby the percentages are given on a dry substance weight basis.
The feed and the eluent were used at a temperature of 65°C and water was used as the eluent. The feed volume was 2.6 m3 and the flow rate for the feed and elution was 3.2 m3/h. Feed interval for the separation was 9.0 m3.
After equilibration of the system with several feeds, the following fractions were drawn from the separation column product valves: residual fraction, two recycle fractions (both sides of the fructose peak) and fructose product fraction. The results including HPLC analyses for the residual fraction, combined recycle fractions and the fructose fraction are set forth in the table below.
The overall fructose yield calculated from these fractions was 94.7%. In the fructose product fraction, the glucose content was reduced to 43.8%, the content of disaccharides A to 56.3% and the content of disaccharides B content to 50.0% compared to the content in the feed.
The process equipment included a separation column, feed pump, eluent water pump, heat exchangers, flow control means for the out-coming liquid as well as inlet and product valves for the various process streams. The height of the column was 3.7 m and column had a diameter of 2.76 m. The column was packed with a strong acid gel type cation exchange resin (manufactured by Finex) in Ca2+-form. The divinylbenzene content of the resin was 5.5% and the mean bead size of the resin was 0.35 mm.
As a feed, a fructose crystallization run-off was used and the aim was to separate fructose contained therein.
The liquor concentration was 41.9 g/100 ml and the pH was 4.0. The fructose run-off was composed as set forth below, whereby the percentages are given on a dry substance weight basis.
The feed and the eluent were used at a temperature of 65°C and water was used as the eluent. The feed volume was 2.5 m3 and the flow rate for the feed and elution was 3.2 m3/h. Feed interval for the separation was 9.0 m3.
After equilibration of the system with several feeds, the following fractions were drawn from the separation column product valves: residual fraction, two recycle fractions (both sides of the fructose peak) and fructose product fraction. The results including HPLC analyses for the residual fraction, combined recycle fractions and the fructose fraction are set forth in the table below.
The overall fructose yield calculated from these fractions was 94.8%. In the fructose product fraction, the glucose content was reduced to 18.8%, the content of disaccharides A to 75.0% and the content of disaccharides B to 35.0% compared to the content in the feed.
The process equipment included two columns connected in series, feed pump, recycling pumps, eluent water pump, heat exchangers, flow control means for the out-coming liquids as well as inlet and product valves for the various process streams. The height of both columns was 4.3 m and both columns had a diameter of 2.76 m. The columns were packed with a strong acid gel type cation exchange resin (manufactured by Finex) in Ca2+-form. The divinylbenzene content of the resin was 5.5% and the mean bead size of the resin was 0.36 mm.
As a feed, a fructose crystallization run-off was used and the aim was to separate the fructose contained therein.
The liquor concentration was 65.5 g/100 ml and the pH was 4.0. The fructose crystallization run off was composed as set forth below, whereby the percentages are given on a dry substance weight basis.
The fractionation was performed by way of a 16-step SMB sequence as set forth below. The feed and the eluent were used at a temperature of 65°C and water was used as an eluent.
Step 1: 1.8 m3 of feed solution were pumped into the first column at a flow rate of 9 m3/h and a recycle fraction was collected from the second column.
Step 2: 0.2 m3 of feed solution were pumped into the first column at a flow rate of 9 m3/h and a fructose fraction was collected from the second column.
Step 3: 4,8 m3 of feed solution were pumped into the first column at a flow rate of 9,5 m3/h and a residual fraction was collected from the same column. Simultaneously 12,3 m3 of water were pumped into the second column at a flow rate of 20 m3/h and fructose fraction was collected from the same column.
Step 4: 12.5 m3 were circulated in the column set loop, formed with all columns, at a flow rate of 12 m3/h.
Step 5: 5.7 m3 of water were pumped into the first column at a flow rate of 12 m3/h and a residual fraction was collected from the second column.
Step 6: 12.5 m3 were circulated in the column set loop, formed with all columns, at a flow rate of 12 m3/h.
Step 7: 5.7 m3 of water were pumped into the second column at a flow rate of 12 m3/h and a residual fraction was collected from the first column.
Step 8: 9.4 m3 were circulated in the column set loop, formed with all columns, at a flow rate of 12 m3/h.
Step 9: 1.8 m3 of feed solution were pumped into the second column at a flow rate of 9 m3/h and a recycle fraction was collected from the first column.
Step 10: 0.2 m3 of feed solution were pumped into the second column at a flow rate of 9 m3/h and a fructose fraction was collected from the first column.
Step 11: 4.8 m3 of feed solution were pumped into the second column at a flow rate of 9.5 m3/h and a residual fraction was collected from the same column. Simultaneously 12.3 m3 of water were pumped into the first column at a flow rate of 20 m3/h and fructose fraction was collected from the same column.
Step 12: 12.5 m3 were circulated in the column set loop, formed with all columns, at a flow rate of 12 m3/h.
Step 13: 5.7 m3 of water were pumped into the second column at a flow rate of 12 m3/h and a residual fraction was collected from the first column.
Step 14: 12.5 m3 were circulated in the column set loop, formed with all columns, at a flow rate of 12 m3/h.
Step 15: 5.7 m3 of water were pumped into the first column at a flow rate of 12 m3/h and a residual fraction was collected from the second column.
Step 16: 9.4 m3 were circulated in the column set loop, formed with all columns, at a flow rate of 12 m3/h.
After equilibration of the system, the following fractions were drawn from the system: three residual fractions from both columns, one fructose-containing fraction and one recycle fraction from both columns. The results including HPLC analyses for combined fractions are set forth in the table below.
The overall fructose yield calculated from these fractions was 95.5%. In the fructose product fraction, the glucose content was reduced to 17.9%, the content of disaccharides A to 84.6% and the content of disaccharides B to 40.0% compared to the content in the feed.
The process equipment included two columns connected in series, feed pump, recycling pumps, eluent water pump, heat exchangers, flow control means for the out-coming liquids as well as inlet and product valves for the various process streams. The height of both columns was 3.95 m and both columns had a diameter of 0.2 m. The columns were packed with a strong acid gel type cation exchange resin (manufactured by Finex) in Na+-form. The divinylbenzene content of the resin was 5.5% and the mean bead size of the resin was 0.36 mm.
As a feed, a fructose crystallization run-off was used and the aim was to separate the fructose contained therein.
The liquor concentration was 66.1 g/100 ml and the pH was 3.8. The fructose crystallization run-off was composed as set forth below, whereby the percentages are given on a dry substance weight basis.
The fractionation was performed by way of a 14-step SMB sequence as set forth below. The feed and the eluent were used at a temperature of 65°C and water was used as an eluent.
Step 1: 13 l of feed solution were pumped into the first column at a flow rate of 90 l/h and a recycle fraction was collected from the second column.
Step 2: 22 l of feed solution were pumped into the first column at a flow rate of 90 l/h and a residual fraction was collected from the same column. Simultaneously 47 l of water were pumped into the second column at a flow rate of 190 l/h and fructose fraction was collected from the same column.
Step 3: 54 l were circulated in the column set loop, formed with all columns, at a flow rate of 90 l/h.
Step 4: 28 l of water were pumped into the first column at a flow rate of 90 l/h and a residual fraction was collected from the second column.
Step 5: 57 l were circulated in the column set loop, formed with all columns, at a flow rate of 90 l/h.
Step 6: 25 l of water were pumped into the second column at a flow rate of 90 l/h and a residual fraction was collected from the first column.
Step 7: 44 l were circulated in the column set loop, formed with all columns, at a flow rate of 90 l/h.
Step 8: 13 l of feed solution were pumped into the second column at a flow rate of 90 l/h and a recycle fraction was collected from the first column.
Step 9: 22 l of feed solution were pumped into the second column at a flow rate of 90 l/h and a residual fraction was collected from the same column. Simultaneously 47 l of water were pumped into the first column at a flow rate of 190 l/h and fructose fraction was collected from the same column.
Step 10: 54 l were circulated in the column set loop, formed with all columns, at a flow rate of 90 l/h.
Step 11: 28 l of water were pumped into the second column at a flow rate of 90 l/h and a residual fraction was collected from the first column.
Step 12: 57 l were circulated in the column set loop, formed with all columns, at a flow rate of 90 l/h.
Step 13: 25 l of water were pumped into the first column at a flow rate of 90 l/h and a residual fraction was collected from the second column.
Step 14: 44 l were circulated in the column set loop, formed with all columns, at a flow rate of 90 l/h.
After equilibration of the system, the following fractions were drawn from the system: three residual fractions from both columns, one fructose-containing fraction and one recycle fraction from both columns. The results including HPLC analyses for combined fractions are set forth in the table below.
The overall fructose yield calculated from these fractions was 96.1%. In the fructose product fraction, the glucose content was reduced to 48.1%, the content of disaccharides A to 36.1% and the content of disaccharides B to 41.2% compared to the content in the feed.
The process equipment included two columns connected in series, feed pump, recycling pumps, eluent water pump, heat exchangers, flow control means for the out-coming liquids as well as inlet and product valves for the various process streams. The height of both columns was 3.9 m and both columns had a diameter of 0.2 m. The columns were packed with a weakly acid gel type cation exchange resin (manufactured by Finex). The divinylbenzene content of the resin was 8% and the mean bead size of the resin was 0.31 mm. The resin was balanced to pH 4.5 with 5-% Na-acetate solution and HCl by circulating solution through the resin bed until the outflow pH was 4.5.
As a feed, fructose crystallization run-off was used and the aim was to separate the fructose contained therein.
The liquor concentration was 68.1 g/100ml and the pH was 3.9. The fructose crystallization run-off was composed as set forth below, whereby the percentages are given on a dry substance weight basis.
The fractionation was performed by way of a 16-step SMB sequence as set forth below. The feed and the eluent were used at a temperature of 65°C and water was used as an eluent.
Step1: 8 l of feed solution were pumped into the first column at a flow rate of 90 l/h and a recycle fraction was collected from the second column.
Step 2: 22 l of feed solution were pumped into the first column at a flow rate of 90 l/h and a residual fraction was collected from the same column. Simultaneously 37 l of water were pumped into the second column at a flow rate of 143 l/h and first 3 l of recycle fraction and then 34 l of fructose fraction were collected from the same column.
Step 3: 5 l of feed solution were pumped into the first column at a flow rate of 90 l/h and a fructose fraction was collected from the second column.
Step 4: 50 l were circulated in the column set loop, formed with all columns, at a flow rate of 90 l/h.
Step 5: 24 l of water were pumped into the first column at a flow rate of 90 l/h and a residual fraction was collected from the second column.
Step 6: 55 l were circulated in the column set loop, formed with all columns, at a flow rate of 90 l/h.
Step 7: 24 l of water were pumped into the second column at a flow rate of 90 l/h and a residual fraction was collected from the first column.
Step 8: 48 l were circulated in the column set loop, formed with all columns, at a flow rate of 90 l/h.
Step 9: 8 l of feed solution were pumped into the second column at a flow rate of 90 l/h and a recycle fraction was collected from the first column.
Step 10: 22 l of feed solution were pumped into the second column at a flow rate of 90 l/h and a residual fraction was collected from the same column. Simultaneously 37 l of water were pumped into the first column at a flow rate of 143 l/h and first 3 l of recycle fraction and the 34 l of fructose fraction were collected from the same column.
Step 11: 5 l of feed solution were pumped into the second column at a flow rate of 90 l/h and a fructose fraction was collected from the first column.
Step 12: 50 l were circulated in the column set loop, formed with all columns, at a flow rate of 90 l/h.
Step 13: 24 l of water were pumped into the second column at a flow rate of 90 l/h and a residual fraction was collected from the first column.
Step 14: 55 l were circulated in the column set loop, formed with all columns, at a flow rate of 90 l/h.
Step 15: 24 l of water were pumped into the first column at a flow rate of 90 l/h and a residual fraction was collected from the second column.
Step 16: 48 l were circulated in the column set loop, formed with all columns, at a flow rate of 90 l/h.
After equilibration of the system, the following fractions were drawn from the system: three residual fractions from both columns, one fructose-containing fraction and one recycle fraction from both columns. The results including HPLC analyses for combined fractions are set forth in the table below.
The overall fructose yield calculated from these fractions was 94.2%. In the fructose product fraction, the glucose content was reduced to 48.0%, the content of disaccharides A to 31.3% and the content of disaccharides B to 25.0% compared to the content in the feed.
The process equipment included two columns connected in series, feed pump, recycling pumps, eluent water pump, heat exchangers, flow control means for the out-coming liquids as well as inlet and product valves for the various process streams. The height of both columns was 3.95 m and both columns had a diameter of 0.2 m. The columns were packed with a strong acid gel type cation exchange resin (manufactured by Finex) and first column was in Na+-ion form and second column in Ca2+-ion form. The divinylbenzene content of the resin was 5.5% and the mean bead size of the resin was 0.36 mm.
As a feed, a fructose crystallization run-off was used and the aim was to separate the fructose contained therein.
The liquor concentration was 66.1 g/100 ml and the pH was 3.8. The fructose crystallization run-off was composed as set forth below, whereby the percentages are given on a dry substance weight basis.
The fractionation was performed by way of a 14-step SMB sequence as set forth below. The feed and the eluent were used at a temperature of 65°C and water was used as an eluent.
Step 1: 10 l of feed solution were pumped into the first column at a flow rate of 90 l/h and a recycle fraction was collected from the second column.
Step 2: 25 l of feed solution were pumped into the first column at a flow rate of 90 l/h and a residual fraction was collected from the same column. Simultaneously 45 l of water were pumped into the second column at a flow rate of 190 l/h and fructose fraction was collected from the same column.
Step 3: 58 l were circulated in the column set loop, formed with all columns, at a flow rate of 90 l/h.
Step 4: 30.5 l of water were pumped into the first column at a flow rate of 90 l/h and a residual fraction was collected from the second column.
Step 5: 55 l were circulated in the column set loop, formed with all columns, at a flow rate of 90 l/h.
Step 6: 27.5 l of water were pumped into the second column at a flow rate of 90 l/h and a residual fraction was collected from the first column.
Step 7: 48 l were circulated in the column set loop, formed with all columns, at a flow rate of 90 l/h.
Step 8: 13 l of feed solution were pumped into the second column at a flow rate of 90 l/h and a recycle fraction was collected from the first column.
Step 9: 22 l of feed solution were pumped into the second column at a flow rate of 90 l/h and a residual fraction was collected from the same column. Simultaneously 45 l of water were pumped into the first column at a flow rate of 190 l/h and fructose fraction was collected from the same column.
Step 10: 55 l were circulated in the column set loop, formed with all columns, at a flow rate of 90 l/h.
Step 11: 28 l of water were pumped into the second column at a flow rate of 90 l/h and a residual fraction was collected from the first column.
Step 12: 58 l were circulated in the column set loop, formed with all columns, at a flow rate of 90 l/h.
Step 13: 30.5 l of water were pumped into the first column at a flow rate of 90 l/h and a residual fraction was collected from the second column.
Step 14: 49 l were circulated in the column set loop, formed with all columns, at a flow rate of 90 l/h.
After equilibration of the system, the following fractions were drawn from the system: three residual fractions from both columns, one fructose containing fraction and one recycle fraction from both columns. The result including HPLC analyses for combined fractions are set forth in the table below.
The overall fructose yield calculated from these fractions was 95.8%. The dry substance yield to the residue fraction was 6.6%. In the fructose product fraction, the glucose content was reduced to 22.2%, the content of disaccharides A to 55.0% and the content of disaccharides B to 38.9% compared to the content in the feed.
Approximately 18.1 tn of crystalline fructose was recovered in 36.4 hours from a fructose feed liquid containing 97.2% fructose, 0.8% glucose and 1.3% disaccharides on DS. The crystallization was performed in a cylindrical cooling crystallizer (30 m3).
Seed crystal magma was prepared in 10 m3 cooling crystallizer by filling the crystallizer with evaporated 90.5% w/w feed liquid, seeding at 57.1°C temperature by adding 1.5 kg seed crystals and cooling the suspension to 40.4°C in 32 hours. The average size of the seed crystals was about 0.03 mm and 90% of the crystals were between 0.02-0.08 mm as analyzed by a PMT-PAMAS particle analyzing system.
About 9 m3 of the seed crystal magma, which had a mean crystal size about 0.2 mm, was placed in the 30 m3 cooling crystallizer. Next about 92 w/w % fructose syrup was added to the crystallizer and mixed with the seed magma. When the crystallizer was filled, the temperature of the mass in the crystallizer was adjusted to 56°C and a total concentration of 91.2% w/w was measured. A crystal yield of about 5-10% of dry substance was obtained. The mass was then cooled to 24.3°C over 26 hours. After cooling, the crystals were separated and washed by a conventional centrifuge, dried in a drum drier, sieved and packed. The fructose yield after cooling was about 58% of the dry substance. The product yield was about 44.5% of dry substance (45.7% of fructose), with a mean crystal size of 0.53 mm and purity of 99.9%.
Approximately 17.5 tn of crystalline fructose was recovered in 45.8 hours from the fructose feed liquid containing 96.7% fructose, 0.5% glucose and 1.9% disaccharides on DS. The crystallization was performed in a 30 m3 cylindrical cooling crystallizer.
Seed crystal magma was prepared in a 10 m3 cooling crystallizer by filling the crystallizer with evaporated 91.0% w/w feed liquid, seeding at 58.1°C temperature by adding 1.5 kg seed crystals and cooling the suspension to 42.2°C in 39 hours. About 9 m3 of the seed crystal magma, which had a mean crystal size about 0.2 mm, was placed in the 30 m3 cooling crystallizer. Next about 92 w/w % fructose syrup was added to the crystallizer and mixed with the seed magma. When the crystallizer was filled, the temperature of the mass in the crystallizer was adjusted to 56°C and a total concentration of 91.8% w/w was measured. The mass was then cooled to 30.0°C over 36 hours. After cooling, the crystals were separated and washed by a conventional centrifuge, dried in a drum drier, sieved and packed. The fructose yield after cooling was about 57% of the dry substance. The product yield was about 43.2% of dry substance with a mean crystal size of 0.51 mm and a purity of 99.9%.
Approximately 16.8 tn of crystalline fructose was recovered in 54.1 hours from a fructose feed liquid containing 96.0% fructose, 1.0% glucose and 2.1% disaccharides on DS. The crystallization was performed in a 30 m3 cylindrical cooling crystallizer.
Seed crystal magma was prepared in 10 m3 cooling crystallizer by filling the crystallizer with evaporated 90.5% w/w feed liquid, seeding at 56.2°C temperature by adding 1.0 kg seed crystals and cooling the suspension to 38.6°C in 47 hours. About 9 m3 of the seed crystal magma, which had a mean crystal size about 0.2 mm, was placed in the 30 m3 cooling crystallizer. Next about 92 w/w % fructose syrup was added to the crystallizer and mixed with the seed magma. When the crystallizer was filled, the temperature of the mass in the crystallizer was adjusted to 56°C and a total concentration of 91.7% w/w was measured. The mass was then cooled to 26.5°C over 44 hours. After cooling, the crystals were separated and washed by a conventional centrifuge, dried in a drum drier, sieved and packed. The fructose yield after cooling was about 53% of the dry substance. The product yield was about 41.4% of dry substance with a mean crystal size of 0.52 mm and a purity of 99.7%.
Approximately 17.7 tn of crystalline fructose was recovered in 41.4 hours from a fructose feed liquid containing 97.2% fructose, 0.5% glucose and 1.7% disaccharides on DS. The crystallization was performed in a 30 m3 cylindrical cooling crystallizer.
Seed crystal magma was prepared in 10 m3 cooling crystallizer by filling the crystallizer with evaporated 91.0% w/w feed liquid, seeding at 58.6°C temperature by adding 1.5 kg seed crystals and cooling the suspension to 40.4°C in 39 hours. About 9 m3 of the seed crystal magma, which had a mean crystal size about 0.2 mm, was placed in the 30 m3 cooling crystallizer. Next about 92 w/w % fructose syrup was added to the crystallizer and mixed with the seed magma. When the crystallizer was filled, the temperature of the mass in the crystallizer was adjusted to 56°C and a total concentration of 91.6% w/w was measured. The mass was then cooled to 28.2°C over 31 hours. After cooling, the crystals were separated and washed by a conventional centrifuge, dried in a drum drier, sieved and packed. The fructose yield after cooling was about 58% of the dry substance. The product yield was about 43.5% of dry substance with a mean crystal size of 0.45 mm and a purity of 99.8%.
Crystalline sucrose (57.0 tn/d) is first dissolved and inverted enzymatically (
Fructose is crystallized (feed purity 97.2% and the mass amount of 59.9 tn/d) with 45.7% fructose yield as described in Example 9. Total fructose process yield calculated from sucrose is 46.7% and the fructose yield calculated from fructose entering the crystallization (fructose in the crystallization feed) is 45.9%. Production volume for fructose is 26.7 tn/d. In this option total mass amount to be crystallized is 59.9 tn/d and no mother liquor separation is used.
Crystalline sucrose (57.0 tn/d) is first dissolved and inverted enzymatically (
Fructose is crystallized (feed purity 97.2% and the mass amount 59.9 tn/d) with 45.7% fructose yield as described in Example 9. Mother liquor (amount 33.2 tn/d) is subjected to the chromatographic separation using Na+ and Ca2+ resins and fractionated with 94% fructose yield as described in example 8. Fructose fraction from the mother liquor separation (amount 30,4 tn/d) is crystallized (feed purity 97.7% and mass amount 30.4 tn/d) with 46% fructose yield. Total fructose process yield calculated from sucrose is 70.8% and the fructose yield calculated from fructose in the crystallization feed is 69.4%. Production volume is 40.4 tn/d. In this option mother liquor amount to be separated is 33.2 tn/d and the total mass amount to be crystallized is 90.3 tn/d.
Crystalline sucrose (57.0 tn/d) is first dissolved and inverted enzymatically (
Fructose is crystallized (feed purity 97.2% and mass amount 59.9 tn/d) with 45.7% fructose yield as described in Example 9. Mother liquor (amount 33.2 tn/d) is subjected to the chromatographic separation using Na+ and Ca2+ resins and fractionated with 94% fructose yield as described in Example 8. Fructose fraction (61.7 tn/d) from the mother liquor separation is crystallized (feed purity 97.9% and the mass amount 61.7 tn/d) with 46% fructose yield and second crystallization mother liquor (amount 33.9 tn/d) is also subjected back to the mother liquor separation. Total fructose process yield calculated from sucrose is 95.5% and the fructose yield calculated from fructose in the crystallization feed is 93.6%. Production volume is 54.5 tn/d. In this option the mother liquor amount to be separated is 67.1 tn/d and total mass amount to be crystallized is 121.6 tn/d.
Crystalline sucrose (57.0 tn/d) is first dissolved and inverted enzymatically (
Fructose is crystallized (feed purity 97.2% and mass amount 59.9 tn/d) with 45.7% fructose yield as described in Example 9. Mother liquor (amount 33.2 tn/d) is subjected to the second crystallization step and 43% fructose yield (feed purity 96.0% and mass amount 71 tn/d) is obtained like described in Example 11. Mother liquor (purity 93.4% and amount 41.6 tn/d) from the second crystallization step is subjected to the chromatographic separation using Na+ and Ca2+ resin beds like described in Examples 5, 6 and 8. Fructose fractions from the mother liquor separation are combined (purity 96.7% and amount 37.8 tn/d) and subjected back to the second crystallization step. Total fructose process yield calculated from sucrose is 98.3% and the fructose yield calculated from fructose in the crystallization feed is 96.4%. Production volume is 56.1 tn/d. In this option mother liquor (=run-off) amount to be separated is 41.6 tn/d and total mass amount to be crystallized is 130.9 tn/d.
The results of Example 16 show that a very high yield of crystalline fructose (98.3% on the basis of sucrose) is achieved by introducing a reasonably low amount of fructose mother liquor (41.6 tn/d) from the second crystallization step to chromatographic fractionation using a combination of Na+ and Ca2+ resins. Consequently, the chromatographic fractionation improves the crystallization.
Crystalline sucrose (57.0 tn/d) is first dissolved and inverted enzymatically (
Fructose is crystallized (feed purity 97.2% and the mass amount 124.8 tn/d) with 44.7% fructose yield as described in Example 12 and mother liquor (amount 70.4 tn/d) is subjected to the chromatographic separation using Ca2+ resin with 94% fructose yield as described in Example 5. Fructose fraction from the mother liquor separation (purity 97.2% and the amount 64.9 tn/d) is subjected back to crystallization. Total fructose process yield calculated from sucrose is 95.2% and the fructose yield calculated from fructose in the crystallization feed is 93.3%. Production volume is 54.3 tn/d. In this option the mother liquor amount to be separated is 70.4 tn/d and the total mass amount to be crystallized is 124.8 tn/d.
In this example (separation with Ca2+ form resin), a high fructose process yield is achieved, but a large amount of mother liquor (70.4 tn/d) must be subjected the chromatographic separation to ensure a high purity of the crystallization mass and a relatively good crystallization yield.
For comparison purposes, the process of this example was modified by using a Na+ form resin and the Ca2+ form resin in the chromatographic separation in consecutive beds (columns) and by returning 50% of the mother liquor from the crystallization directly to the same crystallization step (with a the feed purity of 96.0% and a mass amount of 135.7 tn/d). In this process modification, the need for the mother liquor separation (the amount of the mother liquor to be introduced into the chromatographic separation) can be reduced significantly (to an amount of 39.7 tn/d with a purity of 93.4%), while a relatively high crystallization yield (fructose yield of 43%) is still maintained and the total process yield was improved (the total fructose process yield from sucrose was 98.5% and the fructose yield from fructose in the crystallization feed was 96.5%.
It will be obvious to a person skilled in the art that, as the technology advances, the inventive concept can be implemented in various ways. The invention and its embodiments are not limited to the examples described above but may vary within the scope of the claims.
This list of references cited by the applicant is for the reader's convenience only. It does not form part of the European patent document. Even though great care has been taken in compiling the references, errors or omissions cannot be excluded and the EPO disclaims all liability in this regard.