METHOD FOR OBTAINING CRYSTALS FROM A MOTHER SOLUTION, AND CRYSTALLIZATION DEVICE SUITABLE FOR THIS PURPOSE

Abstract

A method for obtaining crystals from a mother solution operates such that mother solution is fed into a crystallisation device. Supersaturation of the mother solution is brought about by open and/or closed-loop control of the temperature (). Seed crystals are added to the mother solution at a seeding point (Sp). The seed crystals are grown as nuclei by continuous closed-loop process control and are finally removed from the method as crystals. They form the product yield. The formation of crystal nuclei in the mother solution is countered by steps in the closed-loop process control. During the closed-loop process control and the crystallisation procedure, the position of the limit (M) of the formation of secondary nuclei is determined using sensors that detect data currently, wherein the position of the limit (M) is established as a value of the concentration (c) and the temperature () from these data. This determined position of the limit (M) of the formation of crystal nuclei is used as the basis for the closed-loop process control of the crystallisation.

Claims

1. A method for obtaining crystals from a mother solution, in which mother the solution is fed into a crystallisation device, in which supersaturation of the mother solution is brought about by open and/or closed-loop control of the temperature (), in which seed crystals are added to the mother solution at a seeding point (Sp), in which the seed crystals are grown as crystal nuclei by continuous closed-loop process control and are finally removed from the method as crystals and form the product yield, in which the formation of crystal nuclei in the mother solution is countered by steps in the closed-loop process control, characterised in that, during the continuous closed-loop process control and while the method is running, in that the mother solution data are currently determined by sensors, in that from these data the position of the limit (M) of the formation of crystal nuclei is determined while the method is running, as a value of the concentration (c) and the temperature (), and in that this current determined position of the limit (M) of the formation of crystal nuclei is used as the basis for the closed-loop process control of the crystallisation.

2. A method according to claim 1, characterised in that the crystals are carbohydrates or sugar alcohols.

3. A method according to claim 1, characterised in that the mother solution is fed into the crystallisation device discontinuously in batches or continuously.

4. A method according to claim 3, characterised in that optical signals and/or signals relating to ultrasound, radar and/or microwaves are utilised for determining the position of the limit (M) of the formation of crystal nuclei, wherein in respect of the formation of crystal nuclei these signals a) are evaluated in a discontinuous operation during a batch, wherein the result of evaluation has an effect on the closed-loop control variables during processing of the batch or on the subsequent crystallisation batch, or alternatively b) are evaluated in a continuous operation over a period of time, wherein the result of evaluation has an effect on the closed-loop control variables for a subsequent period of time.

5. A method according to claim 4, characterised in that the optical signals and/or the signals relating to ultrasound, radar and/or microwaves are captured and evaluated as a function of time.

6. A method according to claim 4, characterised in that the type of optical signals and/or signals relating to ultrasound, radar and/or microwaves is selected such that they are characteristic of the formation of crystal nuclei of carbohydrates or sugar alcohols.

7. A method according to claim 4, characterised in that evaluation of the optical signals and/or the signals relating to ultrasound, radar and/or microwaves, including where appropriate their course over time, allows a conclusion to be drawn about the time of nucleation and/or the rate of nucleation.

8. A method according to claim 4, characterised in that the position of the limit (M) of nucleation is determined using iterative approximation.

9. A method according to claim 4, characterised in that the optical signals are generated by an interaction of electromagnetic radiation with the crystal nuclei and are in particular detected as turbidity.

10. A method according to claim 1, characterised in that the position of the limit (M) of the formation of crystal nuclei is determined using a light scattering measurement as turbidity, a reflectance measurement, in particular by laser in the faint of a focused beam reflectance measurement (abbreviated to FBRM) or using imaging methods, in particular using microscopes, online in the crystallisation device.

11. A method according to claim 10, characterised in that the light scattering measurement is performed by means of a turbidity meter or nephelometer, and/or in that the light scattering measurement is performed as turbidity at different angles.

12. A method according to claim 3, characterised in that the result of evaluation relating to the closed-loop control variables from parallel crystallisation processes operating in batches or continuously is in addition applied where appropriate to other crystallisation devices.

13. A method according to claim 1 characterised in that the results of evaluation are utilised in a self-training system for the purpose of online closed-loop control.

14. A method according to claim 4, characterised in that, from the determined optical signals and/or the signals relating to ultrasound, radar and/or microwaves, the fact that crystals and/or crystal nuclei have diminished and/or disappeared is determined.

15. A method according to claim 1, characterised in that, from the determined values, an evaluation of the cleaning required for the crystallisation device is performed, in particular establishing that a cleaning process is required and/or establishing that a cleaning process can be considered to have been successfully completed.

16. A crystallisation device for carrying out a method according to claim 1, having sensors for determining the position of the limit (M) of the formation of crystal nuclei.

17. A crystallisation device according to claim 16, characterised in that the sensor or sensors is or are arranged in a region of the crystallisation device that is free of bubbles.

18. A crystallisation device according to claim 16, characterised in that a heating chamber is provided, and in that the sensor or sensors is or are arranged below the heating chamber.

19. A crystallisation device according to claim 16, characterised in that the sensors are optical sensors and/or sensors for ultrasound, radar and/or microwaves.

Description

[0060] The invention is explained below in more detail with reference to the drawing, in which:

[0061] FIG. 1 shows a schematic illustration of the limit curves for seeded crystallisation processes in a crystallisation device;

[0062] FIG. 2 shows a schematic illustration of the conditions during crystallisation by evaporation, in a one-batch process;

[0063] FIG. 3 shows a schematic illustration of the conditions during crystallisation by evaporation, in a multi-batch process;

[0064] FIG. 4 shows a schematic illustration of the conditions during crystallisation by cooling, in a one-batch process; and

[0065] FIG. 5 shows a schematic illustration of the conditions during crystallisation by cooling, in a multi-batch process.

[0066] In a crystallisation process in a crystallisation device, crystal nuclei from which crystals of the desired kind are to grow are introduced into a mother solution in a targeted manner at a seeding point.

[0067] FIG. 1 represents in principle how the values for the mother solution for a crystallisation process in a crystallisation device behave. For this purpose, a temperature for the mother solution is represented on the x axis and a concentration c on the y axis. It can be seen that at high temperatures and low concentration an undersaturation is formed in the region U, in which crystals that are present are dissolved. Here, the mother solution is therefore liquid and no formation of solid sugar crystals is possible.

[0068] At a high concentration c and relatively low temperatures , by contrast, a supersaturated region U is produced. Here, sugar crystals that are present grow as a result of the high concentration in their environment, and moreover the formation of fines is also possible. This region is unstable since the mother solution has a tendency to crystallise.

[0069] Between the undersaturated region U and the supersaturated region U there is a saturation curve S forming the boundary.

[0070] Moreover, markedly above the saturation curve S there is a metastable limit M for the formation of crystal nuclei. In the region between the saturation curve S and the metastable limit M, seed crystals that are already present as a result of the supersaturation present can grow, but no new sugar crystals can be formed because the constraints required for nucleation are not met.

[0071] In the prior art, there was a tendency to prevent the formation of crystal nuclei as far as possible in that, although it was necessary to keep above the saturation curve S in order to enable any growth of sugar crystals at all, this saturation curve S was kept to as closely as possible. It was not permitted to overshoot an empirically determined operating point A, or in any case it had to be observed as closely as possible.

[0072] Also on this operating curve A there is for example an operating point .sub.1 for crystallisation by evaporation. Here, the seeding point is not shown.

[0073] Further indicated is a typical operating curve A.sub.T for crystallisation by cooling. Lowering the temperature, indicated by a horizontal line, brings the crystallisation device above the saturation curve S and the starting point close thereto, at a starting temperature .sub.s, and this is estimated by continuously reducing the temperature until just before the operating curve A, obtained from empirically determined values, is reached. Then the concentration c is also reduced so that this operating curve A is not overshot. This reduction in the concentration c is produced by depleting the mother solution as a result of crystallisation. The method is then terminated at a temperature .sub.e.

[0074] The operating curve A lies markedly below the metastable limit M in order to reliably prevent crystal nuclei from arising or fines from resulting.

[0075] FIG. 2 now represents the conditions in a similar form, wherein in this case time t is indicated on the x axis and concentration c is again indicated on the y axis.

[0076] When one considers the course of the method A.sub.T, it is initially of identical values and runs on the left, as seen in this representation, over the saturation curve S, of concentration c.sub.0, to the seeding point Sp. From there, the concentration increases further with time, as can also be seen in FIG. 1, until the operating curve A, of concentration c.sub.1, has been entirely reached. Then the curve A.sub.T bends away in conventional manner and runs just below the operating curve A, which consequently in practice forms the limit concentration B in the supersaturation of the prior art.

[0077] In a method according to the invention, this does not happen, but rather the curve rises further, above the operating curve A of concentration c.sub.1. Over time, the concentration c.sub.2 is finally exceeded. Now, sensors (not illustrated) establish that crystal nuclei are produced, and determine from this that the metastable limit M for the formation of crystal nuclei has undoubtedly been overshot. This recognition is immediately utilised now, in a next step, to reduce the concentration until it has again fallen far enough for no more formation of crystal nuclei to be observable. This is the case between the metastable limit M and the moment on the operating curve A that is marked 2. From a control engineering point of view, the reduction in concentration is ended at this moment designated 2. The method can now proceed until the concentration c.sub.2 is overshot again.

[0078] In this embodiment of the method according to the invention, this procedure then continues to be monitored iteratively within a single batch until the respective measurements quite clearly show where the metastable limit M for the formation of crystal nuclei actually lies.

[0079] In the representation, the assumption is made that this recognition has been reached at the iterative step 6, and from that point this value, recognised as more or less the optimum, is maintained for the concentration c, with the result that an optimised limit concentration N for the supersaturation is formed as the method continues to run.

[0080] FIG. 3 shows, in a similar form, an iterative sequence when a plurality of batches are used one after the other. The method of the prior art would once again be that having the limit concentration B of the prior art, just below the operating curve A of concentration c.sub.1.

[0081] In this case, a certain value is assumed in each case within a batch. Thus, the first batch is not reduced to the metastable limit M again as in FIG. 2, but remains constant once it reaches a certain value and it is recognised that the metastable limit M of concentration c.sub.2 has been overshot.

[0082] In a second batch this termination is carried out at a markedly lower value, and the recognition that no crystal nucleus at all has yet been formed is utilised for a higher value again for the third batch, and so on.

[0083] FIG. 4 now represents schematically an iterative sequence of process steps in a single batch, in this case for crystallisation by cooling.

[0084] Unlike the representation in FIGS. 2 and 3, over time t the values of the temperature thus also decrease. Thus, temperature falls to the right. Here, the concentration c does not remain constant over time; however, the same curves and values also appear herethat is to say the undersaturation region U, the supersaturation region , the saturation curve S that separates these two regions, and the metastable limit M for the formation of crystal nuclei, which is located at some distance above the saturation curve S.

[0085] Here too, an operating curve is indicated, which is iteratively approximated to this metastable limit M for the formation of crystal nuclei, within a single batch.

[0086] In FIG. 5, crystallisation by cooling takes place in a multi-batch process and, in a similar manner to that in FIG. 3, knowledge is acquired in each case from one batch to the next.

[0087] For this purpose, the cooling rate of the mother solution may be set to be successively higher from one batch to the next until the nucleation limit M is reached or overshot. The cooling rate of the mother solution may be reduced again somewhat once the nucleation limit M has been overshot and this overshooting has been recognised, with the result that in the next batch nucleation is only just no longer observed. Using this iterative process, it is possible to approximate to the upper limit of the metastable zone as the operating point. This is of particular interest at the seeding point Sp, since here there is a risk from a control engineering point of view that the process will move into the unstable region, and it is possible to counter this risk in a targeted manner.

[0088] The cooling rate of the mother solution that is associated with this operating point serves to specify a target value for the process. It should be monitored as the batch is run.

[0089] For determination of the position of the metastable limit M, various sufficiently sensitive process-analytical tools may be used, for example turbidity meters, a focused beam reflectance measurement (FBRM) or imaging methods, such as those using process microscopes.

[0090] Using one or more sensors for determining the nucleation limit M, in the sugar crystallisation sector it is possible to construct a self-training and self-optimising closed-loop control that improves the known closed-loop control concepts for a crystallisation batch.

[0091] In practice, in the sugar house of a refinery, for a product stationfor example for white sugarhaving a plurality of discontinuous crystallisation devices, only one device would be equipped in this way. If the crystallisation devices operating in parallel are of the same construction, the operating point that is identified can be applied to these devices. Closed-loop control there may then be performed using the known concepts, for example with the aid of mother solution dry substance values.

[0092] The dry substance values of the mother solution may for example be monitored using a process refractometer. In this context, this iterative process for determining the metastable limit M should be repeated periodically.

[0093] The principle is applicable to both continuous and discontinuous crystallisation by evaporation, and to continuous and discontinuous crystallisation by cooling.

LIST OF REFERENCE NUMERALS

[0094] A Operating curve [0095] A.sub.T Typical operating curve for crystallisation by cooling [0096] A.sub. Typical operating curve for crystallisation by evaporation for a constant temperature 8 [0097] B Limit concentration for supersaturation in the prior art [0098] c Concentration [0099] c.sub.0 Concentration at the saturation point [0100] c.sub.1 Concentration at the operating point for crystallisation by evaporation [0101] c.sub.2 Concentration at the metastable limit for the formation of crystal nuclei [0102] M Metastable limit for the formation of crystal nuclei [0103] N New, optimised operating curve [0104] S Saturation curve [0105] Sp Seeding point [0106] t Time [0107] U Region of undersaturation [0108] Unstable region of supersaturation [0109] Temperature [0110] .sub.1 Constant temperature value [0111] .sub.s Starting temperature of A.sub.T [0112] .sub.e End temperature of A.sub.T