JP7282637B2 - SIMULATOR OF CHROMATOGRAPHIC SEPARATION EQUIPMENT, OPERATING CONDITION SEARCHING METHOD FOR CHROMATOGRAPHIC SEPARATION EQUIPMENT, AND SIMULATION PROGRAM FOR CHROMATOGRAPHIC SEPARATION EQUIPMENT - Google Patents

Description

TECHNICAL FIELD The present invention relates to a simulator of a simulated moving bed chromatography separation apparatus, an operating condition search method, and a simulation program.

Conventionally, a chromatographic separation technique using a simulated moving bed system is known (for example, Patent Documents 1 and 2). Chromatographic separation by the simulated moving bed method uses a circulation system in which a plurality of unit packed towers (columns) packed with adsorbent are endlessly connected and a fluid flows in one direction. Then, a fluid to be treated (undiluted solution) containing strongly adsorbable components (=strongly adsorbable components) and weakly adsorbable components (=weakly adsorbable components) is introduced into this circulation system, fractionated by the filler, and weakly adsorbed. A soluble fraction (=weakly adsorbed fraction) and a strongly adsorbed fraction (=strongly adsorbed fraction) are obtained. For this purpose, a stock solution inlet, a weakly adsorbed fraction outlet, an eluent inlet, and a strongly adsorbed fraction outlet are provided in this order in the direction of fluid flow. The fraction outlet, the eluent inlet, and the strongly adsorbed fraction outlet are intermittently moved in the fluid flow direction while maintaining their relative positional relationship. As a result, chromatographic separation is performed with the same effect as moving the packing material for chromatographic separation in the direction opposite to the flow direction of the eluent.

Here, for separation in a chromatographic separation column, the reaction can be formulated (see, for example, Non-Patent Document 1). Therefore, in designing a chromatographic separation device, it is possible to obtain information for appropriate device design by performing simulations with various settings.

Also, in operation management, it would be useful if it were possible to obtain accurate operating conditions by simulating changes in conditions.

Japanese Patent Publication No. 60-55162 JP 2009-36536 A

Kenji Hashimoto, "Chromatographic Separation Engineering - From Batch to Simulated Moving Bed Operation," Baifukan (2005-09)

Here, in the case of batchwise chromatographic separation, since the number of parameters for the operating conditions is small, it is not difficult to search for the operating conditions by trial and error. However, in the simulated moving bed chromatographic separation, there are many operating condition parameters, and it takes a lot of labor and time to search for the operating conditions by trial and error. Accordingly, an object of the present invention is to easily obtain appropriate operating conditions for a chromatographic separation device.

The simulator according to the present invention circulates a fluid in one direction in a circular path in which three or more chromatographic columns are connected in a circular manner, and makes the inflow positions of the fluid to be treated and the elution fluid different from each other and sequentially shifted in the other direction. , A chromatographic separation device that simulates the operation of a pseudo-moving-bed chromatographic separation device in which the outflow positions of the strongly adsorbed component fraction and the weakly adsorbed component fraction are made different from each other and sequentially shifted in the other direction. wherein the annular path is provided with a strongly adsorbed component extraction port for mainly extracting the strongly adsorbed component and a weakly adsorbed component extraction port for mainly recovering the weakly adsorbed component, and from the strongly adsorbed component extraction port A first process of classifying the extracted weakly adsorbed components into a plurality of categories based on the migration path of the weakly adsorbed components; At least one of the second processes of classifying into a plurality of sections based on the above is executed, the annular path is provided with a stock solution inlet into which the stock solution is injected, and the weakly adsorbed component extraction port is downstream of the stock solution inlet. , the strongly adsorbed component extraction port is located upstream of the stock solution inlet, and the first process is performed by injecting the weakly adsorbed component into the stock solution inlet and then reaching the strongly adsorbed component extraction port. According to the number of circulations of the undiluted solution inlet, the weakly adsorbed component extraction port, and the strongly adsorbed component extraction port in the annular path and the number of circulations of the weakly adsorbed component in the annular path, the above The weakly adsorbed components are classified into a plurality of categories, and the second process is performed by dividing the stock solution injection accompanying the shift during the time from when the strongly adsorbed components are injected into the stock solution inlet to when the weakly adsorbed components are extracted. The strongly adsorbed components are classified into a plurality of categories according to the number of times the inlet, the weakly adsorbed component extraction port, and the strongly adsorbed component extraction port circulate in the annular path, and the number of circulations of the strongly adsorbed component in the annular path.

Further, the annular path has four or more chromatographic columns, is divided into four continuous sections 1 to 4 from the upstream side to the downstream side so that each section has at least one chromatographic column, and the annular path is provided with an eluent injection port into which an eluent is injected, the first section is a region from the eluent injection port to the strongly adsorbed component extraction port, and the fourth section is the weakly adsorbed component extraction port. The area from the outlet to the eluent inlet, and the first process includes counting the number of times the weakly adsorbed component moves from section 1 to section 4 across the boundary between section 1 and section 4, and counting the number of times the weakly adsorbed component moves The weakly adsorbed components are classified into a plurality of categories according to the number of times the strongly adsorbed components move across the boundary from section 4 to section 1, and the second process includes the number of times the strongly adsorbed components move from section 1 to section 4 and the It is preferable to classify the strongly adsorbed components into a plurality of categories according to the number of times the strongly adsorbed components move from section 4 to section 1 .

In the search method according to the present invention, a fluid is circulated in one direction in a circular path in which three or more chromatography columns are endlessly connected, and the inflow positions of the fluid to be treated and the elution fluid are made different from each other, and sequentially in the other direction. Operation of a chromatographic separation apparatus using a simulated moving bed type chromatographic separation apparatus simulator in which the outflow positions of the fraction for the strongly adsorbed component and the fraction for the weakly adsorbed component are shifted to differ from each other and sequentially shifted in the other direction. In the condition search method, the circular path has four or more chromatographic columns, and is divided into four continuous sections 1 to 4 from the upstream side to the downstream side so that each section has at least one chromatographic column. In the annular path, a stock solution inlet into which the stock solution is injected, a weakly adsorbed component outlet for mainly extracting the weakly adsorbed component, a strongly adsorbed component outlet for mainly extracting the strongly adsorbed component, and an eluent are injected. Section 1 is a region from the eluent inlet to the strongly adsorbed component outlet, Section 2 is a region from the strongly adsorbed component outlet to the stock solution inlet, Section 3 is a region from the stock solution inlet to the weakly adsorbed component extraction port, Section 4 is a region from the weakly adsorbed component extraction port to the eluent inlet, and at least Either one is performed, and in the first process, the weakly adsorbed component extracted from the strongly adsorbed component extraction port can be dealt with by changing the integrated flow rate in section 4, and the part can be dealt with by changing the integrated flow rate in section 2. and obtaining a first value specifying the amount of at least one of the two portions ; a second step of selecting either a condition change including a change in the integrated flow rate or a condition change including a change in the integrated flow rate in section 2, and adjusting the amount of the weakly adsorbed component extracted from the strongly adsorbed component extraction port; and the second processing includes two parts, that is, a part where the strongly adsorbed component extracted from the weakly adsorbed component extraction port can be dealt with by changing the integrated flow rate in section 1 and a part where the strongly adsorbed component can be dealt with by changing the integrated flow rate in section 3. a third step of dividing into a plurality of portions including a portion and obtaining a second value specifying the amount of at least one portion of the two portions; and a fourth step of selecting either a condition change including a change in or a condition change including a change in the integrated flow rate of the section 3, and adjusting the amount of the strongly adsorbed component extracted from the weakly adsorbed component extraction port. .

A program according to the present invention circulates a fluid in one direction in a circular path in which three or more chromatography columns are circularly connected, and makes the inflow positions of the fluid to be processed and the elution fluid different from each other and sequentially shifted in the other direction. , A chromatographic separation device that simulates the operation of a pseudo-moving-bed chromatographic separation device in which the outflow positions of the strongly adsorbed component fraction and the weakly adsorbed component fraction are made different from each other and sequentially shifted in the other direction. wherein the annular path is provided with a strongly adsorbed component outlet for mainly extracting strongly adsorbed components and a weakly adsorbed component outlet for mainly recovering weakly adsorbed components; A first process of classifying the weakly adsorbed components extracted from the component extraction port into a plurality of categories based on the movement path of the weakly adsorbed components, and classifying the strongly adsorbed components extracted from the weakly adsorbed component extraction port into the strongly adsorbed components. At least one of the second processes of classifying into a plurality of categories based on the movement path of is executed, the circular path is provided with a stock solution inlet into which the stock solution is injected, and the weakly adsorbed component extraction port is provided with the stock solution The strongly adsorbed component outlet is located downstream of the inlet, the strongly adsorbed component extraction port is located upstream of the stock solution inlet, and the first treatment is performed by injecting the weakly adsorbed component into the stock solution inlet and then the strongly adsorbed component. The number of circulations of the undiluted solution inlet, the weakly adsorbed component outlet, and the strongly adsorbed component outlet in the circular path, and the number of cycles of the weakly adsorbed component in the annular path during the time until the extraction port is reached. The weakly adsorbed components are classified into a plurality of categories according to the number of times, and the second process is performed on the shift in the time from when the strongly adsorbed components are injected into the stock solution inlet to when the weakly adsorbed components are extracted. The strongly adsorbed components are divided into a plurality of categories according to the number of revolutions of the undiluted solution inlet, the weakly adsorbed component outlet, and the strongly adsorbed component outlet in the circular path, and the number of revolutions of the strongly adsorbed components in the annular path. classified into

According to the simulator according to one embodiment, strongly adsorbed components (leak components) erroneously extracted from the weakly adsorbed component extracting port can be classified according to their movement paths. Similarly, weakly adsorbed components (leakage components) erroneously extracted from the strongly adsorbed component extraction port can be classified according to their movement paths. Therefore, since the cause of the leakage can be identified from the movement path of the adsorbed component, it is possible to easily determine the appropriate operating conditions for the chromatographic separation device.

Further, according to the operating condition search method of one embodiment, the operating conditions of the simulator are changed in accordance with the first value that identifies a portion of the strongly adsorbed component (leakage component) mistakenly extracted from the weakly adsorbed component extraction port. can. Similarly, the operating conditions of the simulator can be changed according to a second value that specifies a portion of the weakly adsorbed component (leakage component) mistakenly extracted from the strongly adsorbed component extraction port. Therefore, since the operating conditions of the simulator can be changed according to the values (first value, second value) related to the cause of leakage, it is possible to easily search for suitable operating conditions of the chromatographic separation apparatus.

In this way, suitable operating conditions can be easily determined, and the device can be designed and operated based on the determined operating conditions.

It is a figure explaining operation|movement of the chromatographic separation apparatus of a simulated moving bed system. It is a figure which shows the distribution of the weakly adsorbed component and the strongly adsorbed component in one step. It is a figure which shows the relationship between a column position, a position in each column, and time. It is a figure shown about each section. FIG. 4 is a diagram showing flows of weakly adsorbed components and strongly adsorbed components. It is a figure which shows a motion of a weak adsorption component. It is a figure which shows the structure of a simulator. FIG. 4 shows the labeling of weakly adsorbed components. FIG. 4 shows the labeling of weakly adsorbed components. FIG. 4 shows the labeling of strongly adsorbed components. In the example shown in the figure where the column is fixed and the inlet and outlet are moved, the weakly adsorbed component (indicated by ○ in the figure) is injected in step 1 and flowed downstream, and is recovered in fraction C in step 3 of the next cycle. It is a figure which shows the state performed. In the example shown by moving the column and fixing the inlet and outlet, weakly adsorbed components (indicated by ○ in the figure) are injected in step 1 and flowed downstream, and are recovered in fraction C in step 3 of the next cycle. It is a figure which shows the state performed. In the example shown in the figure where the column is fixed and the inlet and outlet are moved, the strongly adsorbed component (indicated by ○ in the figure) is injected in step 1 and flowed downstream, and is recovered in fraction A in step 4 of the next cycle. It is a figure which shows the state performed. In the example shown by moving the column and fixing the inlet and outlet, the strongly adsorbed component (indicated by ○ in the figure) is injected in step 1 and flowed downstream, and is recovered in fraction A in step 4 of the next cycle. It is a figure which shows the state performed. FIG. 4 is a diagram showing the leakage of weakly adsorbed components and strongly adsorbed components. FIG. 4 is a diagram showing causes of leakage of weakly adsorbed components. FIG. 4 is a diagram showing causes of leakage of strongly adsorbed components.

BEST MODE FOR CARRYING OUT THE INVENTION Hereinafter, embodiments of the present invention will be described based on the drawings. It should be noted that the invention is not limited to the embodiments described herein.

<Chromatographic separation device>
First, the simulated moving bed type chromatographic separation apparatus will be briefly described. FIG. 1 shows an example in which six chromatographic columns (columns) 1 to 6 are connected in series to form a circular path to constitute a circulatory chromatographic separation apparatus. Since there are six columns, six steps from Step 1 to Step 6 are processed as one cycle, and this cycle is repeated. In the actual apparatus, the position of the column is fixed, and the inflow position of the fluid to be treated (undiluted solution) and the extraction (outflow) position of the fraction are shifted. The inflow position of the fluid (stock solution) and the withdrawal (outflow) position of the fraction were set to be fixed. Therefore, in FIG. 1, the columns are shown to be shifted.

In the present embodiment, gas is also targeted as the fluid to be processed and the elution fluid, but liquids will be described below.

In Step 1, the columns are arranged in order of 4, 5, 6, 1, 2, and 3 from the right, and the fluid (liquid) flows in one direction, from right to left, from the leftmost column to the rightmost column. column. An inlet F for the stock solution is provided between the columns 6 and 1 and flows into the column 1 . An outlet A for fraction A mainly containing weakly adsorbed components (weakly adsorbed component outlet) is provided between columns 2 and 3, and part of the fluid flowing out from column 2 is discharged as fraction A. . An inlet D for an elution fluid (eluent) is provided between the columns 3 and 4 and flows into the column 4 . An outlet C for fraction C mainly containing strongly adsorbed components (strongly adsorbed component outlet) is provided between columns 4 and 5, and part of the effluent fluid from column 4 is extracted as fraction C. . The fraction A is also referred to as a weakly adsorbed fraction in the present specification, and is also referred to as a raffinate in the literature. Fraction C is also referred to herein as the strongly adsorbed fraction and is also referred to as an extract in the literature.

Here, at the beginning of the treatment, both the weakly adsorbed component and the strongly adsorbed component flowed in from the injection port F and were in the column 1, and chromatographic separation was not performed. The components are unevenly distributed in different columns, enabling separation. That is, the weakly adsorbed component moves to the left in the figure due to the flow of the eluent, while the strongly adsorbed component moves to the right in the figure due to the column shift (movement of the packing material). At a certain point in time, predetermined amounts of stock solution and eluate are introduced, and fractions A and C corresponding to the inflow are withdrawn from outlets A and C. Assuming this as 1 step, the inflow positions of the undiluted solution and the eluent and the outflow positions of the fractions A and C are shifted to move to the next step. This is repeated for the endlessly connected columns. In the example of FIG. 1, since there are six columns, six steps constitute one cycle. The inflow timing of the stock solution and the eluent may be different, and the outflow timing of the fractions A and C may also be different.

FIG. 2 schematically shows the concentration distribution of each column in Step 1 when the process is in a steady state. State (a) is reached by the inflow of the stock solution into column 1, the inflow of the eluent into column 4, the outflow of fraction A from column 2, and the outflow of fraction C from column 4, and then by circulation of the eluent. , weakly adsorbed components (concentration shown by solid lines), and strongly adsorbed components (concentrations shown by dashed lines) migrate (states (b) to (d)). Then, when the column has moved by one column, Step 1 ends, and the column moves one column in the direction opposite to the flow direction of the eluent, resulting in the state of Step 2 (e). As a result, the density distribution becomes the same as the state (a) of Step 1. FIG. In this manner, weakly adsorbed components and strongly adsorbed components can be separated from the undiluted solution.

FIG. 3 shows the position and time in the column when it is assumed that the eluent flows in the column from right to left, and the passage of time is assumed to flow from top to bottom. A column is of length L and the step time is Ts. 2(a) to 2(e) show weakly adsorbed components (solid lines) and strongly adsorbed components (dashed lines) at respective column positions at times (a) to (e) shown on the right side of FIG. It shows the concentration of

<Triangle theory>
Next, the triangle theory will be explained. In the chromatographic separation by the simulated moving bed method, conditions for inflow or outflow of the liquid necessary for realizing good separation and purification will be described with reference to the conceptual diagram of the simulated moving bed method shown in FIG. FIG. 4 shows section 1 between eluent supply port D and outlet C, section 2 between outlet C and stock solution inlet F, section 3 between stock solution inlet F and outlet A, Section 4 is shown between outlet A and eluent inlet D. FIG. Also, V1 to V4 mean the integrated flow rates of sections 1 to 4 per step, respectively. Arrows in FIG. 4 indicate the direction of fluid flow. Corresponding to the example in Figure 1, Section 1 is the rightmost column (Column 4 of Step 1), Section 2 is the 2nd and 3rd columns from the right (Columns 5 and 6 of Step 1), Section 3 is the right 2 columns from 4 to 5 (columns 1 and 2 of Step 1), and section 4 corresponds to the leftmost column (column 3 of Step 1). The following description assumes that each section uses the same column.

Focusing on sections 2 and 3 in FIG. 2, and at integrated flow rate V3, weakly adsorbed components migrate over the length of section 3, and strongly adsorbed components migrate less than section 3 distance. By doing so, when moving downstream with respect to the positions of the inlets and outlets D, C, F, and A, the weakly adsorbed components move further downstream, and the strongly adsorbed components move upstream. and relatively (pseudo) movement. That is, the weakly adsorbed component can be extracted from the extraction port A, and the strongly adsorbed component can be extracted from the extraction port C, respectively.

That is, V _j ^acc is the integrated flow rate of section j (j = any of 1 to 4) _{, and} V ^elution and the elution volume of the weakly adsorbed component A is the _{weakly adsorbed component} ^elution , the following equations (1) to (4) are established.
V ₂ ^acc > V _{weakly adsorbed component} ^elution formula (1)
V ₂ ^acc < V _{strongly adsorbed component} ^elution formula (2)
V ₃ ^acc > V _{weakly adsorbed component} ^elution formula (3)
V ₃ ^acc < V _{strongly adsorbed component} ^elution formula (4)

On the other hand, focusing on section 1, if the moving distance of the strongly adsorbed component is less than the length of section 1 at the integrated flow rate V1, the strongly adsorbed component (passed upstream without being extracted from outlet C) (strongly adsorbed components) cannot be extracted from the extraction port C. In this state, the strongly adsorbed components remain adsorbed on the packing material of the column and proceed to section 4 at the time of the next step change, resulting in contamination of fraction A. In other words, it can be said that the filler is not sufficiently regenerated. Therefore, in section 1, the moving distance of the strongly adsorbed component must exceed the length of section 1 at integrated flow rate V1.

That is, the following formula (5) holds.
V ₁ ^acc > V _{strongly adsorbed component} ^elution formula (5)

Focusing on section 4, when the moving distance of the weakly adsorbed component exceeds the length of section 4 at integrated flow rate V4, the weakly adsorbed component (the weakly adsorbed component that has passed through outlet A without being extracted) ) cannot be extracted from the extraction port A. In that situation, the weakly adsorbed components will have to go with the eluent to section 1, resulting in contamination of fraction C. It can be said that the eluent is not sufficiently regenerated. Therefore, in section 4, the travel distance of the weakly adsorbed component must be less than the length of section 4 at integrated flow rate V4.

That is, the following formula (6) is established.
V ₄ ^acc < V _{weakly adsorbed component} ^elution formula (6)

A relationship that satisfies each of the above expressions is called a so-called triangle theory. 709-738 and the like.

The above is an explanation of the conditions for establishing the triangle theory, and in the simulated moving bed system, the separation and purification of the target substance are carried out under the operating conditions that satisfy this triangle theory. Note that the triangle theory can also be applied when each section has two or more columns. In this case, the "length of the section" in the above description should be read as "the length of one column".

When considering the separation of the weakly adsorbed component and the strongly adsorbed component, according to the triangle theory, the direction of movement of the weakly adsorbed component and the strongly adsorbed component in each section should be as shown in FIG. That is, the weakly adsorbed component can be moved leftward in the figure according to the flow direction of the eluent in Sections 1 to 3, and rightward in Section 4 opposite to the flow direction of the eluent. The strongly adsorbed component may be moved rightward in the drawing in accordance with the direction of movement of the column (opposite to the direction of flow of the eluent) in Sections 1 to 4, and to the left in Section 1, which is opposite to the direction of movement of the column.

The "elution volume" of each adsorbed component used to derive the above triangle theory is based on the average migration speed of each adsorbed component. Here, in the case of an actual device, since the velocity of each adsorbed component due to diffusion has a distribution width, it is not sufficient to follow the above-mentioned triangle theory, and a certain margin must be provided. The larger the margin, the more reliable the separation, but the disadvantage is that the amount of stock solution to be processed is reduced, the amount of eluent used is increased, or both. Therefore, it is necessary to obtain the minimum necessary margin for economical operation. For that search, trial and error is performed through experiments or simulations, and through the simulations of this embodiment, appropriate design conditions and operating conditions can be obtained.

<About leakage>
An unacceptable amount of leakage will result if the above margins are not sufficient. Here, the weakly adsorbed components recovered in fraction C (strongly adsorbed fraction) are referred to as "leakage of weakly adsorbed components", and the strongly adsorbed components recovered in fraction A (weakly adsorbed fraction) are referred to as "strongly adsorbed components Let's call it "leakage".

If only weakly adsorbed components are considered, the following cases can be considered. This will be described with reference to FIG. In the figure, weakly adsorbed components are schematically indicated by gray circles.

First, in FIG. 6(a), after the weakly adsorbed component is introduced, it moves leftward from the inlet F along with the flow of the eluent. As shown in FIG. 6(b), it is pulled out from the extraction port A. Here, in the column at column position 6, conditions are set so that the direction of movement of the weakly adsorbed component is opposite to the direction of flow of the eluent. As for the unacceptable amount of weakly adsorbed components, some of them flow together with the eluent, as shown in FIG. 6(c). This is the "leakage due to the path through the circular progress" among the leaks of the weakly adsorbed components.

Also, in the simulated moving bed system, the column shifts in the direction opposite to the direction of flow of the eluent. Therefore, the weakly adsorbed component adsorbed on the packing material in FIG. 6(c) moves in the direction opposite to the flow direction of the eluent as the column shifts. Then, as shown in FIG. 6(d), part of the weakly adsorbed component is drawn out from the drawing port C. As shown in FIG. This is "leakage through a route that does not go around" among the leaks of weakly adsorbed components.

In this way, the migration path along which the weakly adsorbed component flows out from the extraction port C can be classified into at least two. The handling of path leakage is different. Therefore, it is necessary to distinguish between leaks due to the two travel paths. As for the strongly adsorbed components, there are at least two types of leaks, "leakage due to a circuit delay path" and "leakage due to a circuit without a circuit delay", depending on the movement path of the strongly adsorbed component.

<First Embodiment>
FIG. 7 shows the simulator 10 according to the first embodiment. Generally, a general-purpose computer is used, and a simulation program for a chromatographic separation device that simulates a simulated moving bed chromatographic separation device that separates components by calculation described below is installed as an application program. The simulator 10 has an input section 12 , a processing section 14 and an output section 16 . From the input unit 12, various conditions regarding the apparatus are input, and properties of the liquid to be treated are input. The processing unit 14 executes the simulation program to calculate weakly adsorbed components and strongly adsorbed components to be extracted as the fractions A and C, and outputs the results from the output unit 16 .

The simulator according to the first embodiment classifies each adsorbed component into a plurality of parts based on the movement path leading to the leak. The simulator uses the above-mentioned two migration routes ("leakage through the route that passes through the orbital advance" and "leakage through the route that does not undergo the orbital advance" for the leakage of weakly adsorbed components, Each adsorbed component is classified based on "leakage due to leakage" and "leakage due to route without round-trip delay").

The simulator according to the first embodiment labels each adsorption component according to the following rules.
・When injected as a stock solution, the label is set to (0).
・Provide a circuit advance/delay judgment unit in the part where the eluent flows.
・Increase the number of labels by 1 when passing from the upstream to the downstream of the circuit advance/delay determination section.
The number of labels is decremented by one when moving from downstream to upstream of the circuit advance/delay determination unit.
This is an event that occurs with respect to the components in the column when the advance/delay determination unit (that is, the portion into which the eluent flows) overtakes the column during step transition.
・Use the label at the time of recovery as the label for leakage of each adsorbed component.

With such labeling, the leakage through the route of each adsorbed component can be judged by the label at the time of recovery as follows.
・The label at the time of recovery will be positive for leaks due to the path through which the weakly adsorbed component has traveled.
・If the weakly adsorbed component leaks through a route that does not go around, the label at the time of recovery will be incorrect.
・Leakage due to a route that passes through the circulation delay of the strongly adsorbed component will have a negative label at the time of collection.
・A non-negative label will be displayed at the time of collection for leaks that do not pass through the circulation delay of strongly adsorbed components.

The table below summarizes leaks due to paths, labels, and their countermeasures. Note that countermeasures against leakage shown in the table will be described later in the second embodiment.

Next, labeling of weakly adsorbed components will be described with reference to FIGS. 8 and 9. FIG. In other words, one place (specific position) in the circulatory system is provided with a place for determining whether the circuit advances or lags. In the figure, a circuit advance/delay determination unit is provided at the portion where the eluent flows into the rightmost column. As a result, the states shown in FIGS. 8(a) to 8(d) are the weakly adsorbed component (0) because the weakly adsorbed component has not passed the orbit advance/delay determination unit since being injected as the undiluted solution. and labeled.

On the other hand, as shown in FIG. 9(a), when the weakly adsorbed component passes through the extraction port A without flowing out and passes through the orbit advance/delay determination unit, it is labeled as a weakly adsorbed component (+1). be. Then, as shown in FIG. 9B, the weakly adsorbed component discharged from the extraction port C is also labeled as a weakly adsorbed component (+1). Furthermore, as shown in FIG. 9(c), the weakly adsorbed component passed through the outlet C and extracted from the outlet A is also labeled as the weakly adsorbed component (+1). In this way, labels are given according to the number of times of passage.

Further, as shown in FIG. 9(d), with respect to a column having a weakly adsorbed component labeled as a weakly adsorbed component (+1) past the cycle advance/delay determination part, the cycle advance/delay determination is performed by column shift. When shifted to the front of the part, the label is returned to the weakly adsorbed component (0).

In FIG. 9, only the weakly adsorbed component (0) and the weakly adsorbed component (+1) are shown. ) or more, and also weakly adsorbed components (-1) or less.

In this way, by performing the round advance/delay determination, the recovered weakly adsorbed components are labeled according to the route. In this embodiment, the leakage of the weakly adsorbed component is assumed to be due to the path through which the circulation advances if the label at the time of recovery is non-positive, and that it is due to the path that does not undergo the circulation advance if the label is positive. can judge. Therefore, it is possible to check the amount of the same weakly adsorbed component for each route, and to consider countermeasures for reducing leakage.

The strongly adsorbed component can also be calculated in the same manner. That is, as shown in FIG. 10, the strongly adsorbed components mainly move along with the packing material due to the column shift, and basically move in the direction opposite to the moving direction of the eluent. Therefore, the strongly adsorbed component that has flowed in through the inlet F is labeled as strongly adsorbed component (0). and labeled. On the other hand, when it passes through the outlet C and passes the orbital advance/delay determination part, it is labeled as a strongly adsorbed component (-1), and if this is extracted from the outlet A, the component strength The adsorbed component is labeled as strongly adsorbed component (-1). Depending on the passage of the orbital advance/delay determination part, there are strongly adsorbed components (-2) or less, and there are strongly adsorbed components (+1) or more. 1) is mainly labeled. In the present embodiment, the leakage of the strongly adsorbed component is due to a route that passes through a round trip delay if the label value at the time of collection is non-negative, and through a route that does not pass through a round trip delay if the label value is negative. is.

As described above, in the first embodiment, the weakly adsorbed components and the strongly adsorbed components extracted from the extraction ports A and C are labeled by the circuit advance/delay determination, and their movement paths are specified. That is, the weakly adsorbed components and the strongly adsorbed components extracted from each of the extraction ports A and C are classified according to their movement paths. Therefore, the strongly adsorbed components erroneously extracted from the outlet A can be classified according to their movement paths (strongly adsorbed components (0) and strongly adsorbed components (-1)). Similarly, the weakly adsorbed components erroneously extracted from the extraction port C can be classified according to their movement paths (weakly adsorbed components (0) and weakly adsorbed components (+1)). Therefore, since the cause of the error can be identified from the movement path of the adsorbed component, it is possible to easily search for suitable operating conditions for the chromatographic separation apparatus.

In this way, in the first embodiment, it is possible to identify the movement path of the extracted component. Therefore, it is possible to adjust the operating conditions based on the specified movement route and perform appropriate operation.

In the first embodiment, weakly adsorbed components and strongly adsorbed components may be classified by the following calculation.

Circular path of stock solution inlet F, weakly adsorbed component outlet A, and strongly adsorbed component outlet C associated with a shift in the time from when the weakly adsorbed component is injected into the raw solution inlet F to when it reaches the strongly adsorbed component outlet C. Leakage of the weakly adsorbed component is classified into a plurality of categories according to the number of laps in the loop and the number of laps of the weakly adsorbed component in the annular path.

Specifically, the value Y1 based on the difference in the number of circulations is calculated using the formula "Y1 = (number of circulations of the weakly adsorbed component) - (number of circulations of the inlet/outlet) - X1", and classified based on the value Y1. do. Here, "the number of turns of the weakly adsorbed component" and "the number of turns of the injection port/extraction port" are calculated in an image in which the position of the column is fixed and the injection port and the extraction port are moved. In addition, X1 is a parameter for correcting Y1 to an integer. When the weakly adsorbed component is positioned upstream of the injection port D, it is taken to be positive. Y1 is equal to the value of the label above.

For example, in the example shown in FIG. 11 where the column is fixed and the inlet and outlet are moved, weakly adsorbed components (indicated by ◯ in the figure) are injected in step 1 and flowed downstream, and in step 3 of the next cycle When collected in fraction C, the "circulation number of weakly adsorbed components" is "12/6". Since the injection port F makes 8/6 turns during that time, the "number of turns of the injection port/withdrawal port" is "8/6". In this case, X1 is set to "-2/6" according to the relative position of the undiluted solution injection port of the weakly adsorbed component expressed in units of rotation. From these, Y1 at the time of recovery is "Y1=(12/6)-(8/6)-(-2/6)=6/6=1". FIG. 12 shows the same thing in a picture of moving the column with the inlet and outlet fixed, and when the above label is examined, the value of the label is Since it increases by 1 and goes from 0 to +1, the value of the label when collected in fraction C is +1 (equal to Y1 during collection above).

Circular path of stock solution inlet F, weakly adsorbed component outlet A, and strongly adsorbed component outlet C associated with a shift in the time from when the strongly and weakly adsorbed component is injected into the raw solution inlet F to when it reaches the weakly adsorbed component outlet A. Leakage of the strongly adsorbed components is classified into a plurality of categories according to the number of circulations in the loop and the number of cycles of the strongly adsorbed components in the annular path.

Specifically, the value Y2 based on the difference in the number of circulations is calculated using the formula "Y2 = (number of circulations of the strongly adsorbed component) - (number of circulations of the inlet/outlet) - X2", and classified based on the value Y2. do. Here, "the number of turns of the strongly adsorbed component" and "the number of turns of the injection port/extraction port" are calculated in an image in which the position of the column is fixed and the injection port and the extraction port are moved. In addition, X2 is a parameter for correcting Y2 to an integer. When the strongly adsorbed component is positioned upstream of the injection port D, it is taken to be positive. Y2 is equal to the value of the label above.

For example, in the example shown in FIG. 13 where the column is fixed and the inlet and outlet are moved, the strongly adsorbed component (indicated by ◯ in the figure) is injected in step 1 and flowed downstream, and in step 4 of the next cycle When collected in fraction A, the "circulation number of strongly adsorbed components" is "5/6". Since the injection port F makes 9/6 turns during that time, the "number of turns of the injection port/extraction port" is "9/6". In this case, X2 is set to "2/6" according to the relative position of the undiluted solution injection port of the weakly adsorbed component expressed in units of rotation. From these, Y2 at the time of recovery is "Y2=(5/6)-(9/6)-(2/6)=-6/6=-1". FIG. 14 shows the same thing in a diagram of moving the column with the inlet and outlet fixed, and when the above label is examined, the value of the label is changed when passing through the orbital advance/delay determination unit fixed to the eluent injection port D. Since it is decremented by 1 and goes from 0 to -1, the value of the label when collected in fraction A will be -1 (equal to Y2 during collection above).

Further, in the first embodiment, weakly adsorbed components and strongly adsorbed components may be classified by the following calculation.

The number of times the weakly adsorbed components moved from section 1 to section 4 shown in FIG. Classify weakly adsorbed component leaks into categories according to the number of times the aforementioned boundary is crossed in Section 1 (ie, the number of times the label value is incremented by one).

Specifically, the value Z1 based on the difference in the number of times the boundary is crossed is calculated using the formula "Z1 = (number of moves from section 4 to section 1 of the weakly adsorbed component) - (number of moves from section 1 to section 4 of the weakly adsorbed component)". and classify based on the value Z1. Z1 is equal to the value of the label above.

The number of times the strongly adsorbed components migrated from section 1 to section 4 across the boundary between sections 1 and 4 shown in FIG. Leakage of strongly adsorbed components is classified into multiple categories according to the number of times that section 1 crosses the aforementioned boundary (ie, the number of times the label value is incremented by 1).

Specifically, the value Z2 based on the difference in the number of times the boundary is crossed is expressed by the formula "Z2 = (number of moves from section 4 to section 1 of strongly adsorbed component) - (number of moves from section 1 to section 4 of strongly adsorbed component)". and classify based on the value Z2. Z2 is equal to the value of the label above.

[Specific calculation method of the first embodiment]
A specific calculation method of the first embodiment will be described below.

[1. Target to be calculated]
The object of calculation is the concentration distribution of each adsorbed component in the column. The liquid phase concentration C and the solid phase concentration C ^* must be calculated separately. More specifically, C _k ^* is the liquid phase concentration that is in equilibrium with the solid phase concentration. Let _Cck be the solid phase concentration, then _Ck ^* = _Cck / _Kk , where _Kk is the partition coefficient. Also, the two concentrations change depending on the time t and the position z in the column (position in the flow direction in the column). The subscript k represents the component species.

Therefore, the concentration distribution in the column shown below is expressed by (2 × the number of components) two-variable functions
_Ck (t,z) and _Ck ^* (t,z)
will be represented by

FIG. 2 mentioned above graphically shows the values of C _k (t,z) at several times t. As will be described later, C _k ^* is also obtained by simulation as a simultaneous differential equation, but is not shown in FIG. 2 .

[2. partial differential equation]
The above equations for C _k and C _k ^* can be the following partial differential equations in Non-Patent Document 1, for example.

Here, ε _b is the porosity, u ₀ is the superficial linear velocity of the mobile phase, K _fk is the overall mass transfer coefficient, and a _v is the surface area of the packing material per unit volume of the column. Also, K _fk a _v is the overall mass transfer capacity coefficient.

[3. Area where calculation is performed]
The above formulas (1) and (2) are for one column and one process. When the columns and processes are arranged, the whole is as shown in FIG. The distance of each square is 0≤z≤L, and the time is 0≤t≤Ts, and _Ck (t,z) and _Ck ^* (t,z) are obtained within this range.

It also introduces the number "column position". These are numbered according to the column position when the doorway is fixed as shown in FIG.

Regarding the differential equations (1) and (2) mentioned above, it is clarified that they are in each column and each step as follows.

Here, L is the length of one column, and Ts is the process time.

The graph in FIG. 2 is a "snapshot" of the distribution in the column at several times after the process has progressed to steady state. Therefore, the "sections" indicated by (a) to (e) in FIG. 3 are displayed.

[4. Initial conditions]
First, the concentration in the column is all zero at the start of the operation, and the formula is as follows.

[5. boundary condition]
[5.1. Boundary conditions for column connection]
The columns are connected endlessly as shown in FIG. That is, for column connections,
・For directly connected parts, the end of the column = the beginning of the next column ・For the parts where the stock solution or eluent is injected, the end of the column + injected material = the beginning of the next column・For extracted parts, the end of the column = the beginning of the next column (extraction does not affect the concentration in the column)
becomes.

In other words, z=L at the end of the column and z=0 at the beginning of the column. Specifically, it is as follows.

[5.2. Boundary conditions for the transition of the process (Step)]
For process transitions, column shifts are reflected. As shown in FIGS. 1 and 2, when Step 1 is completed, the process proceeds to Step 2, so the position of the column is shifted at this point.

Here, since "column position 1 to column position 6" are used to specify the columns, all the components are shifted each time the step is changed. Specifically, it is as follows.

The same applies to the solid phase side as follows. (7a) to (7f) differ only in that (6a) to (6f) are marked with ^* .

[6. Evaluation of Separation Results]
For example, the average of the 20th cycle is taken as the separation performance. Therefore, the liquid entering the outlet fraction is averaged for the 20th cycle (ie step number 115-120). Expressed as a formula, it is as follows.

Then, if the concentration of each component in each fraction is known, the purity/recovery rate can be calculated.

[7. Model for Simulator of First Embodiment]
The basic idea is to label each molecule. However, this would result in a huge amount of calculation. Therefore, the densities C _k and C _k ^* are classified by labels, and are written as C _k(i) and C _k(i) ^* with i as the label. Of course, summing to the original C _k and C _k ^* :

In this way, the components were classified by label, but the adsorbability of each component to the adsorbent did not change. Therefore, the partial differential equation can be expressed as follows by simply adding a subscript (i) to equations (1) and (2).

At the start of the operation, the concentrations in the column are all zero as well. This also differs from formulas (3) and (4) only in the addition of the subscript (i).

As for column connection, most of the numbers are simply suffixed with (i) based on the formulas (5a) to (5f), but the following points should be noted.
• Only those with a label value of 0 are injected as undiluted solutions. Therefore, it is necessary to classify the cases of formulas (16c1) and (16c2).
・The label value must be incremented by 1 for the component exceeding the round-trip advance/delay determination part. Therefore, the formula (16f) for the connection of column position 6 and column position 1 has the difference that the subscript becomes (i+1) on the right side.

Most of the process transitions also have only a suffix (i) attached, but when the process transitions from column position 1 to column position 6, the label value must be decremented by one. Therefore, in the equations (17f) and (18f), the subscript on the right side is (i-1).

Thus, concentrations for liquid phase, solid phase weakly adsorbed components, and strongly adsorbed components at all column positions 1-6 can be obtained on the label. Therefore, the concentration of each component in each fraction required for realizing the first embodiment can be obtained for each label. The concentration of the liquid phase at column position=5 is the concentration of fraction A, and the concentration of the liquid phase at column position=1 is the concentration of fraction C. For example, when k=weakly adsorbed component and k=strongly adsorbed component, by averaging the 20th cycle (Steps 115 to 120), each concentration can be obtained for each label.

[8. Limitation of label range in the model for the simulator of the first embodiment]
In the model method for simulators described above, the computational load is proportional to the number of labels to be computed. Since the range of label values is theoretically from -∞ to +∞, the amount of calculation becomes infinite. Therefore, in practice, it is preferable to set N=5 and limit to −N≦label value≦+N. Of course, N may be anywhere from about 2 to 10, or more. Then, if the label value needs to be further decreased by -1 from -N, the label value is shifted to +N. Also, if the label value needs to be increased by +1 from +N, the label value is shifted to -N.

Specifically, the boundary conditions in the model for the simulator of the first embodiment described above are changed as follows.

The boundary conditions for connecting columns are as follows.

Also, the boundary conditions for the column shift are as follows on the liquid phase side.

Then, the solid phase side is as follows.

Strictly speaking, the model is changed by restricting the labels from -∞ to +∞ to -N to +N in this way. However, under realistic operating conditions, N=5 is sufficient, and only a negligibly small error occurs when -∞ to +∞.

By limiting the range of labels, it is possible to prevent the computational complexity from becoming infinite. However, if the label range is from -5 to +5, 11 functions must be calculated, so the amount of calculation is approximately 11 times that of the simulator that does not consider labels.

[9. Solution of Partial Differential Equation]
Above, a general partial differential equation is obtained. After that, it is sufficient to solve the above equation by a general method. In Non-Patent Document 1, the finite difference method is used, and there is no problem in using the finite difference method in this embodiment as well.

<Second embodiment>
Next, the operating condition search method for the chromatographic separation apparatus according to the second embodiment will be described with reference to FIG. The operating condition search method according to the second embodiment uses part of the simulator of the chromatographic separation apparatus described above to perform at least one of the first process and the second process.

The first process has a first step and a second step. In the first step, as shown in FIG. 15, the leakage of the weakly adsorbed component (the weakly adsorbed component extracted from the extraction port C) is divided into a plurality of portions to obtain a first value specifying at least one portion. The second step selects either a condition change including a change in the integrated flow rate V4 in section 4 or a condition change including a change in the integrated flow rate V2 in section 2 based on the first value, and selects a weakly adsorbed component extraction port C Adjust the recovery amount to

The first value is, for example, a value related to the breakdown of each portion when classified into "leakage due to a route that has undergone a circuit advance" and "leakage due to a route that has not undergone a circuit advance" obtained by the simulator of the first embodiment. (see Table 1 of the first embodiment).

The second process has a third step and a fourth step. In the third step, as shown in FIG. 15, the leakage of the strongly adsorbed component (the strongly adsorbed component extracted from the extraction port A) is divided into a plurality of portions to obtain a second value specifying at least one portion. . A fourth step selects either a condition change including a change in the integrated flow rate V1 in section 1 or a condition change including a change in the integrated flow rate V3 in section 3 based on the second value, and the strongly adsorbed component outlet A Adjust the recovery amount to

The second value is, for example, a value related to the breakdown of each portion when classified into "leakage due to a route that passes through a round-trip delay" and "leakage due to a route that does not pass through a round-trip delay" obtained by the simulator of the first embodiment. , (see Table 1 of the first embodiment).

Next, leakage countermeasures will be specifically described. As mentioned above, strongly adsorbed components and weakly adsorbed components can be separated by following the triangle theory. That is, the directions of movement of strongly adsorbed components and weakly adsorbed components according to the triangle theory are as shown in FIG. Separation is ensured by observing this. However, the actual process does not follow theory and requires some margin, and a lack of margin will result in an unacceptable amount of leakage. 16 and 17 show examples of leakage when labels are limited to the weakly adsorbed component (0), the weakly adsorbed component (+1), the strongly adsorbed component (0), and the C strongly adsorbed component (-1). Then, countermeasures against leaks (changes in operating conditions) are shown below. By changing the operating conditions in this way and repeating the simulation under the changed operating conditions, it is possible to search for suitable operating conditions.

"1. When a weakly adsorbed component leaks into the component fraction C" and "2. When a strongly adsorbed component leaks into the component fraction A" will be described below.

(1) When the weakly adsorbed component leaks into the component fraction C The label is determined from the simulation results, and if the weakly adsorbed component with the label of the weakly adsorbed component (+1) leaks into the fraction C, in section 4 , some of the weakly adsorbed components are considered to be traveling around to section 1. That is, in Section 4, the direction of movement of the weakly adsorbed component is not the opposite direction (→) to the flow of the eluent, but the direction (←) of the flow of the eluent as shown in FIG. 16(a). be done. Therefore, the countermeasure in this case is to reduce the flow rate of section 4 ((iii) countermeasure).

On the other hand, from the simulation results, if the weakly adsorbed component with the label of weakly adsorbed component (0) is leaking into the fraction C, then in section 2, some of the weakly adsorbed component is going backward to the section 1 side. is judged to be That is, in section 2, the moving direction of the weakly adsorbed component is considered to be not "←" but "→" as shown in Fig. 16(b). Therefore, the countermeasure in this case is to increase the flow rate of section 2 ((iv) countermeasure).

(2) When the strongly adsorbed component leaks to fraction A If the strongly adsorbed component of the label strongly adsorbed component (-1) leaks to fraction A, in section 1, the strongly adsorbed component circulates to section 4 It is determined that it was delayed and passed through section 4 and leaked into fraction A. That is, as shown in FIG. 17(a), in section 1, the direction of movement of the strongly adsorbed component is considered to be the direction of the flow of the eluent, not the direction of the flow of the eluent, but the direction of the flow of the eluent. be done. Therefore, the countermeasure in this case is to increase the flow rate of section 1 ((i) countermeasure).

On the other hand, from the simulation results, if the strongly adsorbed component labeled as strongly adsorbed component (0) leaks into fraction A, as shown in FIG. It is determined that the component is "←" instead of "→" and passes through section 3. Therefore, the countermeasure in this case is to reduce the flow rate of section 3 ((ii) countermeasure).

Thus, in this embodiment, when weakly adsorbed components leak to fraction C, and when strongly adsorbed components leak to fraction A, the cause can be known. Appropriate remedial measures can then be selected. Then, based on the results of the simulation, it is possible to search for appropriate conditions by repeating the simulation after taking the above-described darkening countermeasures.

<Third Embodiment>
Similar to the simulator of the first embodiment, the simulator according to the third embodiment has the function of classifying into a plurality of fractions based on the migration path leading to leakage of each adsorbed component. Specifically, the simulator according to the third embodiment classifies based on two movement routes. The two migration routes are "leakage through the route that passes through the circulation advance" and "leakage through the route that does not undergo the circulation advance" for the leakage of the weakly adsorbed component, and "leakage through the route that passes through the delay in the circulation" for the leakage of the strongly adsorbed component. "leakage" and "leakage due to a route that does not undergo a round trip delay".

The simulator according to the third embodiment performs simplified calculations under certain assumptions to calculate information for estimating the above classification.

First, leakage of weakly adsorbed components will be described. Here, it is assumed that the amount of the weakly adsorbed components flowing out from the extraction port C is small under general operating conditions, that is, operating conditions in which separation is achieved to some extent satisfactorily. In that case, the weakly adsorbed components flowing out from the extraction port C can be considered to be limited to the weakly adsorbed component (0) and the weakly adsorbed component (+1). The weakly adsorbed components flowing out of the outlet A are referred to as the weakly adsorbed component (0) _A and the weakly adsorbed component (+1) _A , and the weakly adsorbed components flowing out of the outlet C are added as subscripts indicating that they flow out from the outlets A and C. The weakly adsorbed components are expressed as weakly adsorbed component (0) _C and weakly adsorbed component (+1) _C .

"(Amount of weakly adsorbed components flowing out from extraction port C and extraction port A after passing through orbit advance/delay determination unit) = Weakly adsorbed components (+1) _C + Weakly adsorbed components (+1) _A = (Circulation advance/delay determination unit) The amount of the weakly adsorbed component that has passed through the delay determination section while moving forward)−(the amount of the weakly adsorbed component that has passed through the orbital advance/delay determination section while retreating)".

In addition, since the concentration is the target in the simulation, when calculating the weakly adsorbed component (+1) _C + weakly adsorbed component (+1) _A in the above formula, it is converted to the amount by multiplying the set flow rate. do.

Such approximation can reduce the amount of computation. Note that the information obtained here is the weakly adsorbed component (+1) _C + the weakly adsorbed component (+1) _A , so information on the upper limit value of the leaked weakly adsorbed component (+1) _C that has passed through the circulation of the weakly adsorbed component. is obtained.

In addition, since the strongly adsorbed component mainly retreats with the shift of the column, the amount of the strongly adsorbed component flowing out from the outlet A = strongly adsorbed component (-1) _C + strongly adsorbed component (-1) About _A , and the amount of the strongly adsorbed component retreating in the orbital advance/delay determination section. The strongly adsorbed components flowing out from outlet A are denoted as strongly adsorbed component (0) _A and strongly adsorbed component (-1) _A , and the strongly adsorbed components flowing out from outlet C are denoted as strongly adsorbed component (0) _C and strongly adsorbed component ( -1) Represented by _C.

Next, leakage of strongly adsorbed components will be described. Here, it is assumed that the amount of the strongly adsorbed components flowing out from the extraction port A is small under general operating conditions, that is, operating conditions in which separation is achieved to some extent satisfactorily. In this case, the strongly adsorbed components flowing out from the extraction port A can be limited to strongly adsorbed components (0) and strongly adsorbed components (-1). The strongly adsorbed components flowing out from outlet C are denoted as strongly adsorbed component (0) C and strongly adsorbed component (-1) _C , and the strongly adsorbed components flowing out from outlet C are denoted as strongly adsorbed components (0) _C and outflow from outlet A. The strongly adsorbed components are expressed as strongly adsorbed component (0) _A and strongly adsorbed component (-1) _A.

"(Amount of strongly adsorbed components flowing out of extraction port A and extraction port C after passing through orbital advance/delay determination unit) = strongly adsorbed components (-1) _A + strongly adsorbed components (-1) _C = (rotation The amount of strongly adsorbed components that have passed through the advance/lag determination section while retreating)-(the amount of strongly adsorbed components that have passed through the orbital advance/delay determination section while moving forward).

In addition, since the concentration is the target in the simulation, when calculating the strongly adsorbed component (-1) _A + strongly adsorbed component (-1) _C in the above equation, the amount is calculated by multiplying the set flow rate. Convert to

Such an approximation can reduce the amount of calculation compared to the simulator of the first embodiment. This is because, as shown below, it can be realized by adding processing with a relatively small amount of calculation to the simulator that is the basis of the first embodiment (based on [1] to [7] of the first embodiment). Note that the obtained here is the strongly adsorbed component (-1) _A + strongly adsorbed component (-1) _C , so the upper limit for the strongly adsorbed component (-1) _A leaked through the circulation delay of the strongly adsorbed component Value information is obtained.

[Model for the simulator of the third embodiment]
In the third embodiment, the following calculations are performed after performing calculations using a model for the simulator that is the basis of the first embodiment.

In the 20th cycle, the amount of the component that advances and exceeds the circulating advance/delay determination part is the liquid phase concentration of the corresponding component in the column at column position 6 and the flow rate that flows out from column position 6 during Steps 115 to 120. It is multiplied and becomes as follows.

In the 20th cycle, the amount of the component that retreated and exceeded the lap lead/lag determination part is the total amount of the corresponding component in the solid phase and liquid phase in the column at column position 1 in Steps 115 to 120. , as follows:

Here, S is the cross-sectional area of the column and u is the linear velocity.

For the weakly adsorbed component, the amount obtained by subtracting the receding amount from the advance amount is set as the upper limit amount of the weakly adsorbed component (+1) _C leaked through the orbital advance of the weakly adsorbed component. is the upper limit of the leaked strongly adsorbed component (-1) _A after the circulation delay of the strongly adsorbed component.

<Fourth Embodiment>
The simulator according to the fourth embodiment multiplies the value (weakly adsorbed component (+1) _C +weakly adsorbed component (+1) _A ) obtained in Embodiment 3 in relation to leakage of the weakly adsorbed component by the parameter α. , the leakage through the round trip of the weakly adsorbed component, the weakly adsorbed component (+1) _C .

For the weakly adsorbed component, if the estimated value of α = weakly adsorbed component (+1) _C / (weakly adsorbed component (+1) _C + weakly adsorbed component (+1) _A ) is obtained, then from the amount calculated in the third embodiment The amount of leakage through the round trip of the weakly adsorbed component can be converted to the weakly adsorbed component (+1) _C .

This α is estimated as follows. α = {(integrated flow rate of section 1) - (integrated flow rate of section 2)} / {(integrated flow rate of section 1) - (elution volume of weakly adsorbed component)} where the integrated flow rate is for one step, elution The volume is that of one column.

In addition, the value (strongly adsorbed component (-1) _C + strongly adsorbed component (-1) _A ) obtained in the second embodiment in relation to the leakage of the strongly adsorbed component is multiplied by the parameter β to obtain the strongly adsorbed component Calculate an estimate for the strongly adsorbed component (-1) _A , which leaks through a round-trip delay of .

For strongly adsorbed components, if an estimated value of β = strongly adsorbed component (-1) _A / (strongly adsorbed component (-1) _C + strongly adsorbed component (-1) _A ) is obtained, it is calculated in the second embodiment. This amount can be converted to the strongly adsorbed component (-1) _A , which is the amount of leakage after the circulation delay of the strongly adsorbed component.

This β is estimated as follows. β = {(integrated flow rate of section 3) - (integrated flow rate of section 4)} / {(elution volume of strongly adsorbed component) - (integrated flow rate of section 4)} where the integrated flow rate is for one step, elution The volume is that of one column.

[Model for the simulator of the fourth embodiment]
In the fourth embodiment, after calculating the model for the simulator of the third embodiment, it is only necessary to multiply α or β, and the same calculation can be performed.

<Specific example>
Next, a specific example using the simulator of the above embodiment will be described. A simulation was performed using a calcium-substituted Y-type zeolite with fructose as a strongly adsorbed component and glucose as a weakly adsorbed component. Various parameters are as follows.

[1. Comparative Example 1]
In Comparative Example 1, one operating condition is selected from the range of operating conditions according to the triangle theory, and the search for the operating condition is started. Aim for a fraction C purity of 97% or higher. A simulator with a basic configuration is used.

[1.1. Trial 1]
The operating conditions were set as follows. u1-u4 are flow velocities in sections 1-4.

(Evaluation of results)
Fraction C currently has a purity of 95.55%, but in order to raise this to above 97%, the leakage of the weakly adsorbable component (glucose) (currently 4.12%) must be reduced. For this purpose, "countermeasure (iii)" and "countermeasure (iv)" can be considered.

[1.2. Trial 2 (take countermeasure (iii) for Trial 1)]
As countermeasure (iii), the flow rate in section 4 was reduced from trial 1.

Leakage of the weakly adsorbed component (glucose) was reduced to 3.36% recovery, but the reduction was still insufficient, so the purity of fraction C only increased to 96.14%.

[1.3. Trial 3 (take countermeasure (iv) for Trial 1)]
As countermeasure (iv), the flow rate in section 2 was increased from trial 1.

Leakage of the weakly adsorbed component (glucose) was reduced to 2.43% recovery, resulting in an increased purity of 97.25%, meeting the requirements.

[1.4. Summary of Comparative Example 1 (Trials 1 to 3)]
For trial 1, since it is not known which of the measures (iii) and (iv) should be selected as a measure to reduce the leakage of the weakly adsorbed component (glucose), trial and error of trying each measure. is necessary.

[2. Example 1]
In Example 1, as in Comparative Example 1, one operating condition is selected from the range of operating conditions according to the triangle theory, and a search for the operating condition is started. Starting from the operating conditions of Trial 1, we aim for a fraction C purity of 97%. The simulator of the first embodiment was used.

[2.1. Trial 1]
The same conditions as in Trial 1 of Comparative Example 1 were set.

In order to achieve a purity of 97% for fraction C, the leakage of the weakly adsorbed component (glucose) must be reduced to approximately 3% or less as a recovery rate. Here, the allowable amount of leakage depends on the recovery rate of the strongly adsorbed component (fructose) to fraction C, but if the recovery rate is 100%, the weakly adsorbed component (glucose ) is 3.09% as a recovery rate. As the recovery rate decreases from 100%, the allowance decreases from 3.09% accordingly. In the above result, it can be seen from the conventional simulator of Comparative Example 1 that the leakage of the weakly adsorbed component is 4.12%, but in the simulator of this example (first embodiment), it is further confirmed by the countermeasure (iii) 0.73% as a recovery rate of leaks to be dealt with (positive value on the label)
・Recovery rate of leaks to be addressed by measure (iv): 3.39% (non-positive label value)
Information was obtained that the breakdown was

Therefore, it can be seen that the measure (iii) alone cannot reduce the leakage of the weakly adsorbed component (glucose) below 3.39%, and therefore the purity cannot reach 97%. Therefore, measure (iv) is tried without trying measure (iii).

[2.2. Trial 2 (take countermeasure (iv) for Trial 1)]
The conditions were set to the same conditions as in Trial 3 of Comparative Example 1.

The requirement (purity of fraction C >97%) was met.

[2.3. Summary of Example 1]
For Trial 1, it is possible to choose to try measure (iv) without trying measure (iii) as a measure to reduce leakage of the weakly adsorbed component (glucose). Compared to Comparative Example 1, the number of trials and errors could be reduced.

[3. Example 2]
In Example 2, as in Comparative Example 1, starting from the operating conditions of Trial 1, the purity of Fraction C is aimed at 97%. The simulator of the third embodiment was used.

[3.1. Trial 1 (operating conditions selected only by triangle theory)]
The same conditions as in Trial 1 of Comparative Example 1 were set.

In order to achieve a purity of 97%, the leakage of the weakly adsorbed component (glucose) must be reduced to approximately 3% or less as a recovery rate. It can be seen from the conventional simulator of Comparative Example 1 that the leakage is currently 4.12%, but in the simulator of the present invention (second embodiment), the leakage that should be dealt with by the countermeasure (iii) is between 0% and 0.93% as a recovery rate. I also understand.
Since measure (iii) alone cannot reduce the leakage of the weakly adsorbed component (glucose) to less than 3.19%, it can be seen that the purity cannot reach 97%, so next measure will be taken without trying measure (iii) (iv) is tried.

[3.2. Trial 2. (Take measure (iv) for Trial 1)]
The conditions were set to the same conditions as in Trial 3 of Comparative Example 1.

The requirement (purity of fraction C >97%) was met.

[3.3. Summary of Example 2]
For Trial 1, it is possible to choose to try measure (iv) without trying measure (iii) as a measure to reduce leakage of the weakly adsorbed component (glucose). Compared to Comparative Example 1, the number of trials and errors could be reduced.

[4. Example 3]
In Example 3, as in Comparative Example 1, starting from the operating conditions of Trial 1, the purity of Fraction C is aimed at 97%. The simulator of the fourth embodiment was used.

[4.1. Trial 1. (Operating conditions selected only by triangle theory)]
The same conditions as in Trial 1 of Comparative Example 1 were set.

The approximate value for the amount of leakage to be addressed by measure (iii) obtained by the simulator of the fourth embodiment is 0.73%, which is the exact value ( This is a good approximation of the leakage to be addressed by measure (iii) obtained in Trial 1: 0.73%). Therefore, as in the first embodiment, countermeasure (iv) can be tried without trying countermeasure (iii).

[4.2. Trial 2 (take countermeasure (iv) for Trial 1)]
The conditions were set to the same conditions as in Trial 3 of Comparative Example 1.

The requirement (purity of fraction C >97%) was met.

[4.3. Summary of Example 3]
For Trial 1, it is possible to choose to try measure (iv) without trying measure (iii) as a measure to reduce leakage of the weakly adsorbed component (glucose). Compared to Comparative Example 1, the number of trials and errors could be reduced.

[5. Comparative Example 2]
In Comparative Example 2, one operating condition is selected from the range of operating conditions according to the triangle theory, and the search for the operating condition is started. Starting from the operating conditions of Trial 1, a recovery of 92% of the strongly adsorbed component (fructose) to fraction C is aimed. A simulator with a basic configuration is used.

[5.1. Trial 1 (Operating conditions appropriately selected only by the triangle theory)]
The same conditions as in Trial 1 of Comparative Example 1 were set.

Here, the strongly adsorbed component (fructose) recovery to fraction C is currently 88.40%, but in order to increase this to 92%, the amount of strongly adsorbed component (fructose) leakage (currently 11.4%) is required. 58%) must be reduced. For this purpose, "countermeasure (i)" and "countermeasure (ii)" can be considered.

[5.2. Trial 2 (take countermeasure (i) for Trial 1)]
As countermeasure (i), the flow rate in section 1 was increased from trial 1.

Leakage of the strongly adsorbed component (fructose) decreased to a recovery rate of 9.27%. only go up.

[5.3. Trial 3 (taking countermeasure (ii) for Trial 1)]
As countermeasure (ii), the flow rate in section 3 was reduced from trial 1.

Leakage of the strongly adsorbed component (fructose) decreased to a recovery of 7.58%, resulting in a recovery of the strongly adsorbed component (fructose) in fraction C increased to 92.61%, meeting the requirements.

[5.4. Summary of Comparative Example 2 (Trials 1 to 3)]
For Trial 1, since it is not known which of countermeasures (i) and (ii) should be selected as a countermeasure to reduce the leakage of the strongly adsorbed component (fructose), trial and error of trying each countermeasure. is necessary.

[6. Example 4]
In Example 4, as in Comparative Example 2, one operating condition is selected from the range of operating conditions according to the triangle theory, and a search for the operating condition is started. Starting from the operating conditions of Trial 1, a recovery of 92% of the strongly adsorbed component (fructose) to fraction C is aimed. The simulator of the first embodiment was used.

[6.1. Trial 1]
The same conditions as in Trial 1 of Comparative Example 2 were set.

In order to achieve a recovery rate of 92%, the leakage of the strongly adsorbed component (fructose) must be reduced to 8% or less as a recovery rate. It can be seen from the conventional simulator of Comparative Example 2 that the amount of leakage is 11.58%. as 2.16 + 0.02 = 2.18% (negative label value)
・Recovery rate of leaks to be addressed by measure (ii): 9.41% (non-negative label value)
Information was obtained that the breakdown was

Since measure (i) alone cannot reduce the leakage of the strongly adsorbed component (fructose) below 9.41%, it can be seen that the recovery rate of the strongly adsorbed component (fructose) in fraction C cannot reach 92%. . Therefore, the countermeasure (ii) is tried next without trying the countermeasure (i).

[6.2. Trial 2 (take countermeasure (ii) for Trial 1)]
The same conditions as in Trial 3 of Comparative Example 2 were set.

The requirement (recovery rate of 92% or more of strongly adsorbed component (fructose) to fraction C) was met.

[6.3. Summary of Example 4]
One can choose to try measure (ii) without trying measure (i). Compared to Comparative Example 2, the number of trials and errors could be reduced.
[7. Example 5]
In Example 5, as in Comparative Example 2, starting from the operating conditions of Trial 1, the recovery rate of the strongly adsorbed component (fructose) to fraction C was aimed at 92%. The simulator of the third embodiment was used.

[7.1. Trial 1 (Operating conditions appropriately selected only by the triangle theory)]
The same conditions as in Trial 1 of Comparative Example 2 were set.

In order to achieve a recovery rate of 92%, the leakage of the strongly adsorbed component (fructose) must be reduced to 8% or less as a recovery rate. It can be seen from the conventional simulator of Comparative Example 2 that the leakage of the strongly adsorbed component (fructose) is 11.58%, but the simulator of the present invention (second embodiment) further:
・Leakage to be dealt with by measure (i) is between 0% and 3.05% as recovery rate,
・Therefore, the leakage to be addressed by measure (ii) is between 11.58-3.05% = 8.53% and 11.58%,
It is also understood that

Since the leakage of the strongly adsorbed component (fructose) cannot be reduced below 8.53% only by reducing the contamination caused by measure (i), the recovery rate of the strongly adsorbed component (fructose) in fraction C reaches 92%. Since it is found that it is not possible, the countermeasure (ii) is tried next without trying the countermeasure (i).

[7.2. Trial 2 (take countermeasure (ii) for Trial 1)]
The same conditions as in Trial 3 of Comparative Example 2 were set.

The requirement (recovery rate of 92% or more of strongly adsorbed component (fructose) to fraction C) was met.
[7.3. Summary of Example 5]
For Trial 1, as a measure to reduce leakage of the strongly adsorbed component (fructose), it is possible to choose to try measure (ii) without trying measure (i). Compared to Comparative Example 2, the number of trials and errors could be reduced.

[8. Example 6]
In Example 6, as in Comparative Example 2, starting from the operating conditions obtained in Trial 1, the recovery rate of the strongly adsorbed component (fructose) to fraction C was aimed at 92%. The simulator of the fourth embodiment was used.

[8.1. Trial 1 (Operating conditions appropriately selected only by the triangle theory)]
The same conditions as in Trial 1 of Comparative Example 2 were set.

The approximate value of the amount of leakage to be countermeasured by measure (i) obtained by the simulator of the fourth embodiment is 2.24%, which is the exact value (the trial of Example 4 by the simulator of the first embodiment). 1: 2.18%). Therefore, the same discussion as in Example 4 can be made, and similarly, countermeasure (ii) is tried next without trying countermeasure (i).

[8.2. Trial 2 (take countermeasure (ii) for Trial 1)]
The same conditions as in Trial 3 of Comparative Example 2 were set.

The requirement (recovery rate of 92% or more of strongly adsorbed component (fructose) to fraction C) was met.

[8.3. Summary of Example 6]
Without trying measure (i), one can choose to try measure (ii) as a measure to reduce leakage of the strongly adsorbed component (fructose). Compared to Comparative Example 2, the number of trials and errors could be reduced.

<Effects of Embodiment>
As described above, according to the present embodiment, it is possible to reduce the number of trials and errors for examining which of a plurality of measures is effective as measures for reducing leakage of weakly adsorbed components. In addition, it is possible to reduce the number of times of trial and error for examining which of a plurality of measures is effective as measures for reducing leakage of strongly adsorbed components.

"others"
As described in Non-Patent Document 1, a simulated moving bed chromatographic separation device can be configured from three columns.

In addition, it is preferable to use the simulator according to the present embodiment in designing a simulated moving bed type chromatographic separation apparatus.

Furthermore, it is preferable that the simulated moving bed chromatographic separation apparatus is provided with the simulator according to the present invention, and the output of this simulator is used in the operation of the simulated moving bed chromatographic separation apparatus.

10 simulator, 12 input section, 14 processing section, 16 output section.

Claims (4)

A fluid is circulated in one direction in a circular path in which three or more chromatographic columns are connected in a circular fashion, and the inflow positions of the fluid to be treated and the elution fluid are made different from each other and sequentially shifted in the other direction to obtain an image of the strongly adsorbed component. A simulator for a chromatographic separation device that simulates the operation of a simulated moving bed type chromatographic separation device in which the outflow positions of the fractions for the fraction and the weakly adsorbed component are made different from each other and sequentially shifted in the other direction,
The annular path is provided with a strongly adsorbed component outlet for mainly extracting the strongly adsorbed component and a weakly adsorbed component outlet for mainly recovering the weakly adsorbed component,
a first process of classifying the weakly adsorbed components extracted from the strongly adsorbed component extraction port into a plurality of categories based on the movement path of the weakly adsorbed components; performing at least one of the second processing of classifying into a plurality of categories based on the migration path of the strongly adsorbed component,
The annular path is provided with a stock solution injection port into which the stock solution is injected,
The weakly adsorbed component extraction port is located downstream of the stock solution injection port,
The strongly adsorbed component extraction port is located upstream of the stock solution injection port,
The first process includes the stock solution injection port, the weakly adsorbed component extraction port, and the classifying the weakly adsorbed components into a plurality of categories according to the number of times the strongly adsorbed component outlet circulates in the annular path and the number of times the weakly adsorbed component circulates in the annular path;
The second process includes the stock solution injection port, the weakly adsorbed component extraction port, and the The strongly adsorbed components are classified into a plurality of categories according to the number of times the strongly adsorbed component outlet circulates in the annular path and the number of times the strongly adsorbed component circulates in the annular path.
A chromatographic separator simulator. The annular path has four or more chromatographic columns, and is divided into four continuous sections 1 to 4 from the upstream side to the downstream side so that each section has at least one chromatographic column,
The annular path is provided with an eluent injection port into which the eluent is injected,
The section 1 is a region from the eluent inlet to the strongly adsorbed component outlet,
The section 4 is a region from the weakly adsorbed component outlet to the eluent inlet,
The first treatment includes the number of times the weakly adsorbed component moves from section 1 to section 4 across the boundary between sections 1 and 4, and the number of times the weakly adsorbed component moves from section 4 to section 1 across the boundary. Classify the weakly adsorbed components into a plurality of categories according to the number of times,
The second processing classifies the strongly adsorbed components into a plurality of categories according to the number of times the strongly adsorbed components have moved from section 1 to section 4 and the number of times the strongly adsorbed components have moved from section 4 to section 1. A simulator of the chromatographic separation apparatus according to 1 . A fluid is circulated in one direction through a circular path in which three or more chromatographic columns are connected in an endless manner, and the inflow positions of the fluid to be treated and the elution fluid are made different from each other and sequentially shifted in the other direction, so that the strong adsorption component can be detected. A method of searching operating conditions for a chromatographic separation device using a simulator for a simulated moving bed chromatographic separation device in which the outflow positions of the fraction and the weakly adsorbed component are made different from each other and sequentially shifted in the other direction,
The annular path has four or more chromatographic columns, and is divided into four continuous sections 1 to 4 from the upstream side to the downstream side so that each section has at least one chromatographic column,
The annular path includes a stock solution inlet for injecting a stock solution, a weakly adsorbed component extraction port for mainly extracting the weakly adsorbed component, a strongly adsorbed component extraction port for mainly extracting the strongly adsorbed component, and an elution solution for introducing an eluent. A liquid inlet is provided,
Section 1 is a region from the eluent inlet to the strongly adsorbed component outlet,
Section 2 is a region from the strongly adsorbed component outlet to the stock solution inlet,
Section 3 is a region from the stock solution inlet to the weakly adsorbed component outlet,
Section 4 is a region from the weakly adsorbed component outlet to the eluent inlet,
Performing at least one of the first treatment and the second treatment,
The first processing is
The weakly adsorbed component extracted from the strongly adsorbed component extraction port is divided into a plurality of portions including two portions, a portion that can be counteracted by changing the integrated flow rate in section 4 and a portion that can be counteracted by changing the integrated flow rate in section 2. a first step of obtaining a first value specifying the amount of at least one portion of the two portions ;
Either a condition change including a change in the integrated flow rate in the section 4 or a condition change including a change in the integrated flow rate in the section 2 is selected based on the first value, and the weakly adsorbed component extracted from the strongly adsorbed component outlet is selected based on the first value. and a second step of adjusting the amount of the adsorbed component,
The second processing is
The strongly adsorbed component extracted from the weakly adsorbed component extraction port is divided into a plurality of portions including two portions, a portion that can be counteracted by changing the integrated flow rate in section 1 and a portion that can be counteracted by changing the integrated flow rate in section 3. a third step of obtaining a second value specifying the amount of at least one portion of the two portions;
Either a condition change including a change in the integrated flow rate in the section 1 or a condition change including a change in the integrated flow rate in the section 3 is selected based on the second value, and the strong adsorbent extracted from the weakly adsorbed component extraction port is selected. A method for searching operating conditions for a chromatographic separation device, comprising: a fourth step of adjusting the amount of adsorbed components. A fluid is circulated in one direction in a circular path in which three or more chromatographic columns are connected in a circular fashion, and the inflow positions of the fluid to be treated and the elution fluid are made different from each other and sequentially shifted in the other direction to obtain an image of the strongly adsorbed component. A simulation program for a chromatographic separation device for simulating the operation of a simulated moving bed type chromatographic separation device in which the outflow positions of the fractions for the fraction and the weakly adsorbed component are made different from each other and sequentially shifted in the other direction,
The annular path is provided with a strongly adsorbed component outlet for mainly extracting the strongly adsorbed component and a weakly adsorbed component outlet for mainly recovering the weakly adsorbed component,
to the simulator
a first process of classifying the weakly adsorbed components extracted from the strongly adsorbed component extraction port into a plurality of categories based on the movement path of the weakly adsorbed components; Execute at least one of the second processing of classifying into a plurality of categories based on the migration path of the strongly adsorbed component ,
The annular path is provided with a stock solution injection port into which the stock solution is injected,
The weakly adsorbed component extraction port is located downstream of the stock solution injection port,
The strongly adsorbed component extraction port is located upstream of the stock solution injection port,
The first process includes the stock solution injection port, the weakly adsorbed component extraction port, and the classifying the weakly adsorbed components into a plurality of categories according to the number of times the strongly adsorbed component outlet circulates in the annular path and the number of times the weakly adsorbed component circulates in the annular path;
The second process includes the stock solution injection port, the weakly adsorbed component extraction port, and the The strongly adsorbed components are classified into a plurality of categories according to the number of times the strongly adsorbed component outlet circulates in the annular path and the number of times the strongly adsorbed component circulates in the annular path.
A simulation program for a chromatographic separator.

Images