Method for obtaining an inulin-containing composition

Abstract

A method for obtaining an inulin-containing composition includes providing an inulin-containing plant material, such as Cichorium intybus roots, and providing the inulin-containing plant material in particulate form, wherein the particles have a particle size distribution such that at most 45 vol % of the particles has a size0.15 mm, and at least 90 vol % of the particles has a size4.0 mm. The optionally dried particulate inulin-containing plant material is then subjected to an extraction step in which the inulin is extracted from the plant material to obtain an inulin-enriched juice and an inulin-depleted pulp, which are then separated, preferably by either vacuum filtration, pressure filtration and/or centrifugation.

Claims

1. A method for obtaining an inulin-containing composition, comprising the steps of: (1) providing an inulin-containing plant material in particulate form, wherein the particles have a particle size distribution such that at most 45 vol % of the particles has a size0.15 mm, and at least 90 vol % of the particles has a size4.0 mm; (2) subjecting the particulate inulin-containing plant material provided in step (1) to an extraction step comprising contacting the particulate inulin-containing plant material with an aqueous extractant and extracting the inulin from the plant material to the extractant to obtain an inulin-enriched juice and an inulin-depleted pulp wherein the extraction step or at least one of the plurality N of extraction steps have a duration per extraction step of between 10 and 300 sec; and wherein the temperature of the extractant in the extraction step or in at least one of the plurality N of extraction steps is provided between 60 C. and 75 C.; and (3) separating the inulin-enriched juice as filtrate from the inulin-depleted pulp as retentate wherein the inulin-containing plant material is dried to a dry matter content of at least 80 wt. %, as measured by ISO 6496 before providing the inulin-containing plant material in said particulate form, or wherein the inulin-containing plant material is not dried before providing the inulin-containing plant material in said particulate form, and said particulate inulin-containing plant material is subjected to a prior separation step to remove a juicy fraction therefrom before said extraction.

2. The method according to claim 1, wherein at most 30 vol % of the particles has a size0.15 mm, and at least 90 vol % of the particles has a size3.0 mm.

3. The method according to claim 1, wherein at most 30 vol % of the particles has a size0.10 mm, and at least 90 vol % of the particles has a size2.0 mm.

4. The method according to claim 1, wherein at most 25 vol % of the particles has a size0.10 mm, and at least 70 vol % of the particles has a size1.0 mm.

5. The method according to claim 1, wherein at least 5 vol %, preferably at least 10 vol % of the particles has a size0.15 or 0.10 mm.

6. The method according to claim 1, wherein the particulate inulin-containing plant material is subjected to a plurality N of extraction steps, wherein N2.

7. The method according to claim 6, wherein the inulin-containing plant material is dried and the aqueous extractant has a carbohydrate content ranging from 0 Brix to 40 Brix.

8. The method according to claim 6, wherein all of the extraction steps, have a duration per extraction step of between 10 and 300 sec, and/or wherein the plurality of extraction steps have a total duration of at most 20 min.

9. The method according to claim 1, wherein the extraction step or at least one of the plurality of extraction steps is carried out in a countercurrent flow, wherein an input material is provided to flow in a direction opposite to the direction of flow of the aqueous extractant.

10. The method according to claim 9, wherein the particulate inulin-containing plant material is subjected to a plurality N of extraction steps, wherein N2, and wherein a retentate obtained in an extraction step n of a continuous extraction is used as input material of a subsequent extraction step n+1, wherein n+1N.

11. The method according to claim 9, wherein the particulate inulin-containing plant material is subjected to a plurality N of extraction steps, wherein N2, and, wherein the filtrate obtained in the extraction step n+1 is used as aqueous extractant of the extraction step n.

12. The method according to claim 9, wherein the particulate inulin-containing plant material is subjected to a plurality N of extraction steps, wherein N2, and, wherein the aqueous extractant of an extraction step n has a higher carbohydrate content measured in Brix than the aqueous extractant of a subsequent extraction step n+1.

13. The method according to claim 9, wherein the particulate inulin-containing plant material is subjected to a plurality N of extraction steps, wherein N2, and, wherein the amount of extraction steps N is at least 4 and the carbohydrate content of the aqueous extractant decreases from 30 Brix in the first extraction step to 0 Brix in a final extraction step.

14. The method according to claim 1, wherein the extraction step or at least one of the plurality N of extraction steps, and separating the filtrate from the retentate is carried out by a vacuum band filter, a pressure filter and/or rotating vacuum filter and/or centrifuge.

15. The method according to claim 1, wherein the weight ratio between the amount of final juice or filtrate and the amount of input material corrected to a typical dry substance of fresh material of 25% in all the extraction steps is between 0.5 and 1.2.

16. The method according to claim 1, further comprising the steps of: (4) adding a flocculation agent to the extractant, for forming at least one floc with at least one contaminant of the extractant; and (5) discharging the floc after the floc formation.

17. The method according to claim 1, wherein the extraction or at least one of the plurality N of extraction steps, and/or at least the step of the optional floc formation, is carried out at a pH of between 3 and 5.

18. The method according to claim 1, wherein providing the inulin-containing plant material in particulate form is carried out by grinding and/or milling of the inulin-containing plant material, optionally followed by sieving.

19. The method according to claim 2, wherein drying the inulin-containing plant material to the dry matter content is carried out by any one of sun-drying or oven drying at a temperature of between 30 C. and 200 C.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) FIGS. 1-8 plot solute concentration [ Brix] over time of various working examples of the present invention;

(2) FIG. 9 plots extraction loss ( Brix/Brix) over time in accordance with a working example of the present invention

(3) FIGS. 10 and 11 plot particle size distribution in accordance with working examples of the present invention;

(4) FIG. 12 schematically illustrates a multi-step filtration experiment in accordance with the present invention;

(5) FIG. 13 plots volume over particle size in accordance with a working example of the present invention; and

(6) FIG. 14 plots rate of extraction over time in accordance with a working example of the present invention.

Experiment 1: Extraction with Different Starting Materials

(7) A first set of samples was prepared from Cichorium intybus plants of which the roots were sliced and dried to a dry matter amount suitable for conservation, corresponding to a dry matter content according to ISO 6496 (1999) as specified above of 88.0%.sub.w. The dried sliced roots were subsequently used as-is for comparison or ground to a particulate starting material and sieved with sieves with a specific pore size to obtain samples according to the invention, as well as for a comparative example with a powder with a particle size distribution not according to the invention. The samples are summarized in Table 1. In this respect, the sieve range of powder is defined by the sieves used to obtain the starting material. For example, a sieve range of powder of 1.0 mm-2.0 mm implies that the sample consists of starting material which was able to pass a sieve with a pore size of 2.0 mm, but was not able to pass a sieve with a pore size of 1.0 mm.

(8) TABLE-US-00001 TABLE 1 Set-up Sieve range T.sub.extr Sample Shape of powder [ C.] Example 1 Particulate <0.63 mm 65 Example 2 Particulate 1.0-2.0 mm 65 Comparative Example A Particulate 2.0-4.0 mm 65 Comparative Example B Dried sliced roots Not applicable 65 Comparative Example C Dried sliced roots Not applicable 65 Example 3 Particulate <0.63 mm 75 Example 4 Particulate 0.63-1.0 mm 75 Example 5 Particulate 1.0-2.0 mm 75 Comparative Example D Particulate 2.0-4.0 mm 75 Comparative Example E Dried sliced roots Not applicable 75 Comparative Example F Dried sliced roots Not applicable 75

(9) 15 grams of each of the samples were brought in contact with 100 grams of water. During extraction, the water was kept at a temperature T.sub.extr as listed in Table 1 and a refractometer was used in order to determine the concentration of dissolved substances in the juice after 2.5 minutes, 5 minutes and continued 5-minute intervals until reaching 1 h.

(10) FIGS. 1 and 2 demonstrate a faster rate of extraction and higher final solute concentration rates after 1 h when using dried powders from Cichorium intybus plants roots which are sieved with sieves with a pore size of 2.0 mm compared to dried powders with a larger pore size, as well as dried and sliced roots.

Experiment 2: Extraction with Different Starting Materials and Different Extraction Periods

(11) Various Cichorium intybus plant root samples were prepared from the same plant material as mentioned in Experiment 1 in a way as summarized in Table 2. Extraction was carried out in an otherwise similar way as in Experiment 1, however this experiment additionally comprises experiments which were completed after 20 minutes (Examples 6A and 8A).

(12) TABLE-US-00002 TABLE 2 Set-up Sieve range T.sub.extr Sample Shape of powder [ C.] Example 6 Particulate <1.0 mm 65 Example 6A Particulate <1.0 mm 65 Example 7 Particulate 1.0-2.0 mm 65 Comparative Example G Particulate 2.0-4.0 mm 65 Example 8 Particulate <1.0 mm 75 Example 8A Particulate <1.0 mm 75 Example 9 Particulate 1.0-2.0 mm 75 Comparative Example H Particulate 2.0-4.0 mm 75

(13) The product of the experiments was separated using vacuum filtration with a vacuum of 800 mbar after extraction into a filtrate, which was analyzed by HPLC, and a retentate. The retentate was extracted for 60 minutes with water at a temperature of 75 C. in order to simulate complete extraction, and again separated into a retentate, which was discarded, and a filtrate, which was analyzed by HPLC.

(14) Table 3 provides an overview of the solute concentration of the extract (from the extraction) and the retentate (deduced from the extract of the extraction simulating complete extraction), as well as the amount of fructooligosaccharides with a degree of polymerization of 5 or higher.

(15) TABLE-US-00003 TABLE 3 Solute concentration and degree of polymerization of obtained extracts [solute].sub.extr [solute].sub.retent DP5+ [ Brix] [ Brix] [g/100 g Brix] Example 6 10.01 1.52 81.3 Example 6A 10.30 1.60 80.6 Example 7 9.38 1.78 79.4 Comparative Example G 8.43 1.70 77.7 Example 8 10.30 1.52 81.9 Example 8A 10.66 1.48 81.5 Example 9 10.00 1.60 81.8 Comparative Example H 10.05 1.86 77.6

(16) The results in this table demonstrate that it is possible to extract more inulin with a relatively high degree of polymerization within a shorter period of time from the samples according to the invention, in which extraction with smaller particles leads to a higher yield in general, as well as in terms of chains with a DP of 5 or more. Furthermore, it is clear that the procedure results in a reduction of the amount of soluble carbohydrates present in the retentate after extraction.

(17) FIGS. 3 and 4 again demonstrate a faster rate of extraction and higher final solute concentration rates after 1 h when using dried powders from Cichorium intybus plants roots which have a particle size lower than 4.0 mm.

Experiment 3: Extraction with Extractant Having High Carbohydrate Content

(18) Various Cichorium intybus plant root samples were prepared from the same plant material as mentioned in Experiment 1 in a way summarized in Table 4. Extraction and complete extraction simulation was carried out in the same way as described in Experiment 2, however the extractant used was water mixed with spray-dried inulin from Cichorium intybus to obtain an aqueous extractant with a starting carbohydrate content of 20 Brix, as confirmed with the refractometer.

(19) TABLE-US-00004 TABLE 4 Set-up Sieve range T.sub.extr Sample Shape of powder [ C.] Example 10 Particulate <1.0 mm 65 Example 11 Particulate 1.0-2.0 mm 65 Comparative Example I Particulate 2.0-4.0 mm 65 Example 12 Particulate <1.0 mm 75 Example 13 Particulate 1.0-2.0 mm 75 Comparative Example J Particulate 2.0-4.0 mm 75

(20) The carbohydrate contents, measured with the refractometer, are plotted in FIGS. 5 and 6 for extraction at 65 C. and 75 C. respectively. These figures demonstrate the possibility of obtaining a juice with a high solute concentration expressed in Brix.

Experiment 4: Countercurrent Simulation with Smaller Fraction for Different

(21) Multiple extractants were provided, starting from water as an aqueous extractant of 0 Brix, and by mixing an amount of spray dried inulin with water in order to obtain further aqueous extractants with carbohydrate contents of 20 Brix, 15 Brix, 10 Brix, and 5 Brix, respectively. Each of the extractants comprises an amount of sulfuric acid 10% w in order to bring the pH of the extractant to 5.5.

(22) Various Cichorium intybus plant root samples were prepared from plants of which the roots were sliced subsequently dried. The dry matter content of the starting material was confirmed to be 88.0% w (ISO 6496 (1999) as specified above). The sliced and dried roots were subsequently ground to a particulate starting material and sieved with sieves with a specific pore size as summarized in Table 7.

(23) TABLE-US-00005 TABLE 7 Set-up Sieve range t.sub.extr Sample Shape of powder [sec] Example 14 Particulate <1.0 mm 60 Example 15 Particulate <1.0 mm 150

(24) Countercurrent extraction was simulated by carrying out the following steps: subjecting 15 grams of one the particulate (starting) material from one sample at a time with 100 grams of the prepared 20 Brix extractant for text seconds, and filtered by vacuum filtration in order to obtain a first filtrate and first retentate. As a second step, the first retentate was contacted with an amount of the prepared 15 Brix extractant for t.sub.extr seconds, totaling with the first retentate to 115 grams, and filtered by vacuum filtration in order to obtain a second filtrate and second retentate. As a third step, the second retentate was contacted with an amount of the prepared 10 Brix extractant for t.sub.extr seconds, totaling with the second retentate to 115 grams, and filtered by vacuum filtration in order to obtain a third filtrate and third retentate. As a fourth step, the third retentate was contacted with an amount of the prepared 5 Brix extractant for t.sub.extr seconds, totaling with the third retentate to 115 grams, and filtered by vacuum filtration in order to obtain a fourth filtrate and fourth retentate. As a fifth step, the fourth retentate was contacted with an amount of water (0 Brix) for t.sub.extr seconds, totaling with the fourth retentate to 115 grams, and filtered by vacuum filtration in order to obtain a final filtrate and final retentate. During each of these simulations, the extraction baths had a bath temperature of 65 C.

(25) After each of the simulations, the concentration of dissolved substances of the filtrate was measured with a refractometer according to the aforementioned procedure. Furthermore, 60 grams of the final retentate of each of the experiments was extracted with 300 grams of water for 60 minutes at a temperature of 75 C. in order to simulate complete extraction, and filtered by vacuum filtration. The amount of dissolved substances in this filtrate (representative for the amount still present after completion of the simulation) was determined with the refractometer. The results are listed in Table 8.

(26) TABLE-US-00006 TABLE 8 Solute concentration Example 14 Example 15 [solute] [solute] Step # [ Brix] [ Brix] 1 (20 Brix) 27.3 26.7 2 (15 Brix) 19.5 20.1 3 (10 Brix) 13.1 13.8 4 (5 Brix) 7.1 7.3 5 (water) 1.8 1.7 Retentate 0.30 0.40

(27) The results in this table demonstrate that the differences in extraction quality between an extraction time of 60 and 150 seconds are within the limit of significance at an extraction bath temperature of 65 C. Countercurrent extraction with 5 steps of 60 seconds is sufficient to obtain a high yield of good quality.

Experiment 5: Countercurrent Simulation with Smaller Fractions

(28) Multiple extractants were provided, starting from water as an aqueous extractant of 0 Brix, and by mixing an amount of spray dried inulin with water in order to obtain further aqueous extractants with carbohydrate contents of 20 Brix, 15 Brix, 10 Brix, and 5 Brix, respectively. Each of the extractants comprises an amount of sulfuric acid 10% w in order to bring the pH of the extractant to 5.5.

(29) Various Cichorium intybus plant root samples were prepared from plants of which the roots were sliced and dried. The dry matter content of the starting material was confirmed to be 91.9% w (ISO 6496 (1999) as specified above). The dried material was subsequently ground to a particulate starting material and sieved with sieves with a specific pore size. The samples are summarized in Table 9.

(30) TABLE-US-00007 TABLE 9 Set-up Sieve range Sample Shape of powder Example 16 Particulate <0.20 mm Example 17 Particulate 0.20-0.63 mm Example 18 Particulate 0.63-1.0 mm

(31) Countercurrent extraction was simulated by carrying out the following steps subjecting 15 grams of the particulate (starting) material from one sample at a time with 100 grams of the prepared 20 Brix extractant for 60 seconds, and filtered by vacuum filtration in order to obtain a first filtrate and first retentate. As a second step, the first retentate was contacted with an amount of the prepared 15 Brix extractant for 60 seconds, totaling with the first retentate to 115 grams, and filtered by vacuum filtration in order to obtain a second filtrate and second retentate. As a third step, the second retentate was contacted with an amount of the prepared 10 Brix extractant for 60 seconds, totaling with the second retentate to 115 grams, and filtered by vacuum filtration in order to obtain a third filtrate and third retentate. As a fourth step, the third retentate was contacted with an amount of the prepared 5 Brix extractant for 60 seconds, totaling with the third retentate to 115 grams, and filtered by vacuum filtration in order to obtain a fourth filtrate and fourth retentate. As a fifth step, the fourth retentate was contacted with an amount of water (0 Brix) for 60 seconds, totaling with the fourth retentate to 115 grams, and filtered by vacuum filtration in order to obtain a final filtrate and final retentate. During each of these simulations, the extraction baths had a bath temperature of 65 C.

(32) After each of the simulations, the concentration of dissolved substances of the filtrate was measured with a refractometer according to the aforementioned procedure, and the time necessary for the filtration to complete was recorded. Furthermore, 60 grams of the final retentate of each of the experiments was extracted with 300 grams of water for 60 minutes at a temperature of 75 C. in order to simulate complete extraction, and filtered by vacuum filtration. The amount of dissolved substances in this filtrate (representative for the amount still present after completion of the simulation) was determined with the refractometer. The results are listed in Table 10.

(33) TABLE-US-00008 TABLE 10 Solute concentration and filtration times Example 16 Example 17 Example 18 [solute] t.sub.filtr [solute] t.sub.filtr [solute] t.sub.filtr Step # [Brix] [sec] [Brix] [sec] [Brix] [sec] 1 (20 Brix) 27.3 360 26.2 43 25.5 52 2 (15 Brix) 20.1 92 19.1 152 19.2 77 3 (10 Brix) 13.7 331 14.0 49 14.9 74 4 (5 Brix) 8.1 89 7.0 40 10.2 65 5 (water) 2.0 73 1.5 115 4.3 130 Retentate 0.93 0.36 0.32

(34) The results in this table demonstrate an increased amount of substances dissolved in the extract with a starting material with a relatively small particle size, but also demonstrates that particles with a size smaller than 0.20 mm result in a filtration with a duration less suitable for practical implementation.

Experiment 6: Comparison of Sliced Roots with Particulate Matter Based Thereon

(35) Cichorium intybus plant roots were sliced, and oven dried overnight at 105 C., up to a dry matter content of 97.5% w (as confirmed using ISO 6496 as specified above). Part of the dried material was ground to a particulate starting material and sieved with sieves with a specific pore size. The samples are summarized in Table 11.

(36) TABLE-US-00009 TABLE 11 Set-up Sieve range Sample Shape of powder Comparative example K Dried sliced roots Not applicable Example 19 Particulate <1.0 mm Example 20 Particulate 1.0-2.0 mm

(37) The particle size distribution of sample 19 was measured and is summarized in Table 12. This measurement also demonstrated that the sample had a D.sub.5 of 63 m, a D.sub.50 of 619 m and a D.sub.95 of 1269 m.

(38) TABLE-US-00010 TABLE 12 Particle size distribution Size Volume over [m] [%] 1 100.00 5 100.00 10 99.47 15 98.89 20 98.32 25 97.79 30 97.32 40 96.53 50 95.85 75 94.26 100 92.74 125 91.46 150 90.46 175 89.67 200 88.97 250 87.33 300 84.82 350 81.18 400 76.49 450 70.98 500 64.95 600 52.38 700 40.46 800 30.12 900 21.70 1000 15.17 1100 10.30 1200 6.79 1300 4.34 1400 2.69 1500 1.60 1600 0.90 1700 0.48 1800 0.23 1900 0.09 2000 0.00

(39) 15 grams of each of the samples were brought in contact with 100 grams of water. During extraction, the water was kept at a temperature of 65 C. and a refractometer, calibrated with a plurality of aqueous solutions with a known carbohydrate content, was used in order to determine the concentration of dissolved substances in the juice after 5 minutes, which are summarized in Table 13.

(40) The filtrate of the extraction was analyzed by HPLC, for which the measured amounts of DP5+, expressed in grams per 100 g Brix are also listed in Table 13.

(41) TABLE-US-00011 TABLE 13 Solute concentration [solute] DP5+ Sample [ Brix] [g/100 g Brix] Comparative example K 3.8 84.3 Example 19 11.0 87.5 Example 20 7.7 85.6

(42) These results demonstrate that it is possible to arrive at a higher solute concentration within the same period of time for particles with a size below 2.0 mm, and especially below 1.0 mm, in comparison to sliced material from the same plant.

Experiment 7: Example 21, Dandelion Extraction

(43) Roots of Taraxacum officinale (dandelion) were harvested in October and oven dried for 12 hours at 60 C. after which they were ground to a particulate material. The particulate material was sieved with a sieve with a pore size of 1.0 mm, in order to obtain a starting material out of material which passed through the sieve.

(44) 15 grams of the sieved sample was brought in contact with 100 grams of water. During extraction, the water was kept at a temperature of 65 C. and a refractometer was used in order to determine the concentration of dissolved substances in the juice over continued 5-minute intervals until reaching plateau (i.e. no different value in three consecutive measurements). FIG. 7 shows a graph of the increase in carbohydrate content measured over time.

(45) The filtrate of the extraction was analyzed by HPLC. In a first aspect, this analysis discloses that the quantity of inulin in the filtrate was 89.8 grams per 100 grams carbohydrate, demonstrating that it is possible to extract high amounts of inulin from other materials than Cichorium intybus in a particulate form as well. In a second aspect, the inulin extracted has a DP5+ fraction of 69.2 grams per 100 grams Brix, which comes down to 79.0 grams per 100 grams carbohydrate. The average DP of the filtrate was determined to be 10.4 by AOAC Method 997.08.

(46) These results demonstrate that other particulate starting materials than Cichorium intybus, such as for instance Taraxacum officinale, can favourably be used in the method of the present invention as evidenced by the swift achievement of a high yield and by the high degree of polymerization (DP) of the inulin obtained.

Experiment 8

(47) Cichorium intybus plant roots were sliced and dried to a dry matter content, according to ISO 6496 (1999) as specified above, of 88.0%.sub.w. The sliced and dried material was ground to a particulate starting material and sieved with sieves with a specific pore size.

(48) With the fractions obtained, mixtures were prepared according to Table 14, which lists the weight percentage of each of the fractions in the compositions.

(49) TABLE-US-00012 TABLE 14 Set-up Size [mm] 0-0.2 0.2-1.0 1.0-2.0 2.0-4.0 2.0-3.15 3.15-4.0 4.0-5.6 Example 22 10% 90% Example 23 10% 70% 20% Example 24 10% 70% 20% Example 25 10% 70% 20% Example 26 10% 70% 20% Comparative 10% 70% 20% example L Comparative 10% 50% 40% example M

(50) 45 grams of each of the mixtures according to the Examples listed were extracted in 300 grams of water. A refractometer was used in order to determine the concentration of dissolved substances in the juice after 2.5 minutes and continued 2.5-minute intervals until reaching 20 minutes, and a last measurement after 1 h.

(51) FIG. 8 is a plot of the concentration of dissolved substances in the juice over time of each of the Examples of this Experiment. FIG. 9 shows the same data but differently expressed, namely, each time as the difference to Example 22, the difference being called extraction loss. This demonstrates that compositions with a significant amount of particles with a size above 4 mm result in a slower extraction, in particular during the initial 10 minutes of extraction, and moreover lead to a lower ultimate degree of extraction. Within the range with particles with a size below 4 mm, compositions with smaller particles exhibit increased extraction while surprisingly still showing good filterability.

Experiment 9

(52) Cichorium intybus plant roots were sliced and dried to a dry matter amount suitable for conservation, corresponding to a dry matter content according to ISO 6496 (1999) as specified above of 88.0% w. The sliced and dried material was ground to a particulate starting material. Using a sieve with a pore size of 100 microns, a fines fraction was separated off from the (now fines-depleted) main fraction. Subsequently, portions of the fines fraction and of the main fraction were recombined to prepare two compositions: a composition having 20% by weight of particles having a size lower than 100 microns (Example 27), and a composition having 30% by weight of particles having a size lower than 100 microns (Comparative Example N). Particle size distribution measurements on these Example 27 and Comparative Example N have been carried out according to the above-mentioned procedures, from which the results are shown in FIG. 10 (cumulative) and 11.

(53) A multi-step filtration experiment was carried out according to FIG. 12. The experiment comprises a number of cycles 1, 2, . . . , n, and each of these cycles comprises 5 steps 11-15; 21-25; n1-n5. In cycle 1, 45 grams of composition (111) was contacted for 150 seconds with a bath of 300 grams of warm water (112) at a temperature of 75 C. in step 11, and filtered by vacuum filtration to obtain a retentate or cake (113) on the filter with no cracks and a juice or filtrate (114). The filtrate 114 was collected. In each of the subsequent steps 12-15 of this cycle, 200 grams of warm water (75 C.) was poured over the filter with the cake from the previous step, and the filtrates were collected. In the last step 15, the filtration was carried out until cracks became visible in the cake, and the filtrate was furthermore pressed by hand. The subsequent cycles were carried out in a similar way, however instead of using warm water, a mixture of the obtained filtrates of the previous cycle was used in all of the steps but the last step. In the first step of the cycles subsequent to the first cycle, the bath consists of the filtrate of the second step of the previous cycle (i.e. step 12) as 212a, supplemented to a total of 300 grams with filtrate of the first step of the previous cycle (i.e. step 11), as 212b, which was, were necessary, diluted with the filtrate obtained from the last step of the previous cycle (i.e. step 15), as 212c, in order to obtain a maximum calculated solute concentration at the inlet of 16 Brix, in order to avoid crystallization during extraction. In the second to the fourth steps of the subsequent cycles, instead of using warm water, a mixture of the filtrate obtained from the next step of the previous cycle and the filtrate from the last step of the previous cycle, totaling to 200 grams and at a temperature of 75 C. was used.

(54) With Example 27, two experiments could be completed with 10 cycles, wherein the total filtration time of the 10th cycle amounted to 1292 and 986 seconds, hence 228 seconds per step on average. With Comparative Example N, two experiments were carried out, which each were aborted during the third cycle, since the filtration time in the 3rd cycle took more than 900 seconds per step.

(55) In this experiment, the processing of material having a fraction of particles with particle size smaller than 100 m of 20% by volume of the total composition went favorably; by contrast, the processing of material having a fraction of particles with particle size smaller than 100 m of 30% by volume of the total composition presented significant problems as the filtration time was unacceptably long.

Experiment 10: Different Freshly Ground Plant Roots

(56) Cichorium intybus plant roots were sliced, and freshly ground to a particulate starting material and sieved with sieves with a specific pore size. The particle size distribution of the sample of Example 28 was measured via image analysis. The sample had a D.sub.10 of 454 m, a D.sub.50 of 1405 m and a D.sub.90 of 2778 m. The sample had only 2 vol. % of particles smaller than 0.15 mm and no particles bigger than 4.0 mm.

(57) Jerusalem Artichoke roots were sliced, and freshly ground to a particulate starting material and sieved with sieves with a specific pore size. The particle size distribution of the sample of Example 29 was measured via image analysis. The sample had a D.sub.10 of 907 m, a D.sub.50 of 1913 m and a D.sub.90 of 3406 m. The sample had only 0.3 vol. % of particles smaller than 0.15 mm and no particles bigger than 4.0 mm.

(58) Freshly sliced Cichorium intybus plant roots were provided in cossette form as Comparative Example P.

(59) 55 grams of the samples were brought in contact with 60 grams of water. During extraction, the water was kept at a temperature T.sub.extr of 65 C. and a refractometer was used in order to determine the concentration of dissolved substances over a time frame of about 1 h (3600 s).

(60) FIG. 14 demonstrates a faster rate of extraction and higher final solute concentration rates after 1 h when using ground powders from Cichorium intybus and Jerusalem Artichoke plants roots which are sieved with sieves to obtain a particle size distribution as claimed, compared to cossettes of fresh Cichorium intybus roots.

Experiment 11: Freshly Ground Plant Roots Centrifuged Before Extraction

(61) Cichorium intybus plant roots were sliced, and freshly ground to a particulate starting material and sieved with sieves with a specific pore size. The particle size distribution of the sample of Example 30 was measured via image analysis. The sample had a D.sub.10 of 885 m, a D.sub.50 of 1909 m and a D.sub.90 of 3216 m. The sample had only 0.3 vol. % of particles smaller than 0.15 mm and 6 vol. % of particles bigger than 4.0 mm. The particulate plant roots had a solid content of 21.8 wt. %.

(62) The particulate plant roots were then centrifuged in a prior separation step to yield a juice separated from pulp. The juice contains a Brix of 23.5 wt. %. The pulp has a solid content of 27.3 wt. % from where solubles is 18.6 Bx.

(63) The pulp was then subjected to an extraction during 2.5 min with a 5 Bx solution (ratio 55 g pulp/60 g water) at 65 C. and filtrating the juice from the solids yielded a juice having a Brix of 11.5%

(64) The retentate was then further subjected to a second extraction (38.8 g retentate/76.2 g water) during 2.5 min at a temperature of 65 C. The filtrated juice had a Brix of 3.8. The solid part had a solid content of 18.7 wt. % from where solubles is 3.8 Bx.

(65) When comparing Experiment 11i.e. including the prior separation stepwith Experiment 10i.e. without a pror separation stepit was established that implementation of the prior separation step in case the inulin-containing plant material is not dried before providing the inulin-containing plant material in particulate form: significantly facilitated reaching the desired temperature for the extraction; significantly improved the functioning of the filtration after the extraction; significantly reduced the need for evaporating water from the filtrated juice; reduced the risk of degradation of inulin.