Method for separation with simulated moving bed

Abstract

The present invention relates to a method for separating one or more components from a liquid feed mixture in an EBA-SMB operating mode without the need of pumps at the outlets of the EBA columns. The present invention also relates to a simulated moving bed separation device with expanded bed adsorption columns which can be used in the method according to the invention.

Claims

1. A method for the separation of at least one component from a liquid feed mixture in a simulated moving bed separation device, wherein the separation device comprises a plurality of separation units each with a fixed inlet and a fixed outlet through which during operation of the simulated moving bed separation device a liquid flow can be established in each separation unit in an upward direction, resulting for each separation unit into an input stream and an output stream; wherein each of the plurality of separation units comprises an expanded bed of a separation matrix; wherein in each of the separation units the space above the separation matrix bed is completely filled with liquid without a gas void above the separation matrix bed; wherein each of said plurality of separation units is communicatively coupled to at least one inflow control unit and to at least one outflow control unit; wherein each inflow control unit comprises at least one pump and at least one valve; wherein each outflow control unit comprises at least one valve; wherein each outflow control unit does not comprise a pump; wherein the liquid feed mixture is fed to the inlet of each separation unit in sequence; wherein each of the separation units is subsequently subjected to one of a washing step, an elution step, a cleaning step, an equilibration step or a compensation step; and wherein a purified product outlet stream containing the at least one component is collected.

2. A method according to claim 1, wherein each separation unit comprises one or more columns.

3. A method according to claim 2, wherein a separation unit comprises two or more columns, which are operated in series.

4. A method according to claim 2, wherein a separation unit comprises two or more columns, which are operated in parallel.

5. A method according to claim 1, wherein the device comprises a plurality of in-flow detectors which can provide an output signal corresponding to the detection of a chemical or physical parameter.

6. A method according to claim 5, wherein the device further comprises a processor capable to process the output signal from the in-flow detectors and to regulate the inflow and outflow control units while maintaining a predetermined level of the expanded bed of the separation matrix in each of the separation units.

7. A method according to claim 1, wherein the liquid feed mixture is a biological broth.

8. A method according to claim 1, wherein the fixed outlet of each separation unit is provided with a fixed top adapter.

Description

BRIEF DESCRIPTION OF THE FIGURES

(1) FIG. 1: EBA-SMB setup is further explained in EXAMPLE 1. The meaning of the numbering in FIG. 1 is given in the following table:

(2) TABLE-US-00001 Number Meaning 2 Output port 1 3 Output port 2 4 Output port 3 5 Output port 4 6 Output port 5 7 Input port 1 8 Input port 2 9 Input port 3 10 Input port 4 11 Input port 5 12 pH 13 μS/cm 15 QI/001 16 QI/002 17 QI/003 19 V1-001 20 V2-001 21 V3-001 22 V4-001 23 V5-001 24 V6-001 25 V7-001 26 V8-001 27 V1-002 28 V2-002 29 V3-002 30 V4-002 31 V5-002 32 V6-002 33 V7-002 34 V8-002 35 V1-003 36 V2-003 37 V3-003 38 V4-003 39 V5-003 40 V6-003 41 V7-003 42 V8-003 43 V1-004 44 V2-004 45 V3-004 46 V4-004 47 V5-004 48 V6-004 49 V7-004 50 V8-004 51 V1-005 52 V2-005 53 V3-005 54 V4-005 55 V5-005 56 V6-005 57 V7-005 58 V8-005 59 V1-006 60 V2-006 61 V3-006 62 V4-006 63 V5-006 64 V6-006 65 V7-006 66 V8-006 67 V1-007 68 V2-007 69 V3-007 70 V4-007 71 V5-007 72 V6-007 73 V7-007 74 V8-007 75 V1-008 76 V2-008 77 V3-008 78 V4-008 79 V5-008 80 V6-008 81 V7-008 82 V8-008 83 V1-009 84 V2-009 85 V3-009 86 V4-009 87 V5-009 88 V6-009 89 V7-009 90 V8-009 91 V1-010 92 V2-010 93 V3-010 94 V4-010 95 V5-010 96 V6-010 97 V7-010 98 V8-010 99 V1-011 100 V2-011 101 V3-011 102 V4-011 103 V5-011 104 V6-011 105 V7-011 106 V8-011 108 LI-001 109 LI-002 110 LI-003 111 LI-004 112 LI-005 113 LI-006 114 LI-007 115 LI-008 116 C-001 117 C-002 118 C-003 119 C-004 120 C-005 121 C-006 122 C-007 123 C-008 124 Abs 125 Pump 1 126 Pump 2 127 Pump 3 128 Pump 4 129 Pump 5 Items No 116-123 represent eight columns (C-001-C008). Items No 108-115 (LI/001-LI/008) represent level sensors which each monitor the level (expressed in mm) of the separation matrix in the respective columns. Items No 19-106 all represent solenoid valves with internal diameter of 2 mm. The separation device comprises eight columns (numbered 116 through 123) with inlet and outlet through which a liquid flow of the mixture can be established in each column in an upward direction. Each of the columns comprises an expanded bed of a separation matrix. Each of columns is communicatively coupled to an inflow control unit and to an outflow control unit. Each inflow control unit comprises a pump (P) and a valve (one or more of items numbered 67 through 106). Each outflow control unit consists of a valve (one or more of items numbered 19 through 58) without a pump. The in-flow detectors included are item 15 for pH measurement, item 16 for measuring conductivity, and item 17 for measurement of UV light absorbance.

(3) FIG. 2: Schematic representation EBA-SMB control, with different sensors including level sensor (LI-001), pH (QI-001), conductivity (QI-002) and UV sensors (QI-003) sending signals to controller C, which further sends the signal to the active column inlet pump (P-001).

(4) FIG. 3: Cycle time vs pH

EXAMPLES

Example 1

(5) EBA-SMB Setup

(6) EBA-SMB technology as described in the claims consists of multiple EBA columns with every column connected to an input control unit consisting of at least one valve and one pump and output control unit consisting of at least one valve. The specifications of the EBA columns used for testing in EBA-SMB mode are described in Table 1.

(7) TABLE-US-00002 TABLE 1 expanded bed column dimension Total column height 65 cm Expanded bed height Varied with flow rate Column diameter  2 cm Settled bed height 32 cm Flow distribution mechanism 3 cm ceramic bead bed Expansion factor Varied with flow rate Bed level detection Using ultrasound sensor mounted to fixed top adapter at the column outlet

(8) TABLE-US-00003 TABLE 2 The EBA-SMB set-up as described in FIG. 1 includes Number of columns 8 Number of inlet valves/column 5 Number of outlet valves/column 5 Number of series valves/column 1 Number of inlet pumps for 8 5 (specific inlet valve of every column columns is connected to a common pump, for example inlet valve 1 of 5 of every column is connected to pump 1) Number of input ports for pumps 5 Number of output ports for 5 fractionation EBA bed level measuring 8 (1 attached to every column outlet) ultrasound sensors Control unit (C) described in FIG. 2 describes that when a column is FIG. 2 actively controlled, the controller receives signal from the sensors attached to the column or to the column inlet/outlet stream lines and sends signal to the inlet pump based on desired set-point.

(9) EBA-SMB Software

(10) The EBA-SMB software for use according to the present invention runs a recipe, which is an adaptation of the time-based switching of column positions within a SMB cycle. As the EBA-SMB recipe might require executing bed level control depending on the column position, it is important to note that the constant time per position will result in variable bed volumes of the input stream per position. Therefore, constant volume is used as the set point for switching, in case of positions with active bed level control. However, this further results in inconsistent switch times for such positions, which can lead to accumulation or unavailability of columns in certain positions. To avoid this, an approach has been derived with compensation positions, which account for the inconsistent switch times in case of positions with active bed level control. These compensation positions can also be utilized for optimization of EBA-SMB, depending on the process flexibility.

(11) TABLE-US-00004 TABLE 3 Gel type matrix properties Composition 50-55% sulphonated polymer of styrene, divinylbenzene and ethylstyrene Na.sup.+ form 45-50% water Porosity gel type Density/specific gravity 1150-1200 g/l Operating pH 0-14 Maximum operating temperature 120° C. Particle diameter 300 μm Particle size distribution ≥95% Ion-exchange capacity 1.7-1.8 eq/l

(12) TABLE-US-00005 TABLE 4 Feed composition Liquid feed mixture fermentation broth containing GABA GABA concentration 70-110 g/l Biomass dry weight concentration 1-16 g/l Sugars 0.1-2 g/l Organic acids 0.5-5 g/l Glycerol 0.1-1.4 g/l Density 1010-1100 g/l Viscosity 1.1-1.3 mPa .Math. s pH 4-6.5

(13) Test Conditions:

(14) The results were obtained by testing various process conditions using the EBA-SMB set-up which involved uniform and variable zone switching times during an SMB cycle. The matrix properties and feed composition are described in Table 3 and Table 4.

(15) As described in the table below (Table 5), the experiments involved: 1. Change in the amount of feed fed/ml settled bed matrix volume (SBV) 2. Variation of the NaOH concentration in the elution buffer 3. Variation of the number of columns in feed and elution steps 4. Entrainment rejection (ER), where the liquid void in the column moving to a new step is replaced with the input liquid stream of that particular step 5. Extended elution to prevent non-ideal flow distribution of a single pump feeding more than 1 step 6. Fraction of the product-rich elution stream is to collect the elution peak sample using a specific output port

(16) TABLE-US-00006 TABLE 5 Pump flow rate, input stream, inlet number, outlet number and switch time vs SMB step for experiments 002, 003 and 005 No. Total Number of Output Pump Inlet Flow rate of number of columns port No. No. Step Input stream (ml/min) SBV columns in series No. 1 1 Regeneration 4 wt % H.sub.2SO.sub.4 20 2 1 NA 5 2 2 Elution 5-8 wt % NaOH 20 2 2 2 2 and 4 3 3 Adsorption Unclarified 15 1 1 NA 5 fermentation broth 4 4 Equilibration Demineralized 20 2 1 NA 5 water 4 4 Elution wash Demineralized 20 2 1 NA 3 and 4 water 5 5 Adsorption Demineralized 20 2 2 2 1 wash water

(17) TABLE-US-00007 TABLE 6 Pump flowrate, input stream, inlet number, outlet number and switch time vs SMB step for exp 006 and 007 No. Total Number of Output Pump Inlet Flow rate of number of columns port No. No. Step Input stream (ml/min) SBV columns in series No. 1 1 Regeneration 4 wt % H.sub.2SO.sub.4 20 2 1 NA 5 2 2 Elution 8 wt % NaOH 20 2 2 2 2 and 4 3 3 Adsorption Unclarified 15 0.7 2 2 5 fermentation broth 4 4 Equilibration Demineralized 20 2 1 NA 5 water 4 4 Elution wash Demineralized 20 2 1 NA 3 and 4 water 5 5 Adsorption Demineralized 20 2 1 NA 1 wash water

(18) Test Results:

(19) The pH profile during the cycle time of several switches is described in FIG. 3. It is observed that the system exhibits a cyclic steady state performance under the conditions described in Table 4.

(20) TABLE-US-00008 TABLE 7 Experimental results from the tests performed using the conditions described in table 5 and table 6 Binding Product Bound GABA Yield in Overall Exp Experimental capacity titer recovery in feed zone GABA Yield No. description (g GABA/L SBV) (g GABA/L) elution (%) (%) (%) 002 1 column feed 65.4 33 55.02 53 29.29 without ER, 2 column 5 wt % NaOH elution with selective product collection 003 1 column feed 63 47 78.46 50 39.56 without ER, 2 column 8 wt % NaOH elution with selective product collection 005 1 column feed 82 47 84.45 51 43 without ER, 2 column 8 wt % NaOH elution modified fractionation compared to 003 006 2 column feed with 50 50 60 74 45 ER, increased SBH, 2 column 8 wt % NaOH elution with fractionation 007 2 column feed with 50 50 84 74 64 ER, increased SBH, 2 column 8 wt % NaOH extended elution with fractionation

(21) Conclusions:

(22) From the results described in Table 7, the following conclusions are derived, 1. Decreasing the amount of feed/column and increasing the number of feed zone columns from 1 to 2 along with ER, resulted in a feed zone GABA recovery increase from about 50% (EXP002, 003, 005) to about 74% (EXP006, 007), but at a lower binding capacity of 50 g/L SBV. However, considering the 8-column configuration and critical adsorption wash and elution zones, the maximum number of columns that can be accommodated for the feed zone is only 2. Therefore, to achieve a further increase in yield at higher binding capacities and to capture the break through GABA, it is required to configure more than 2 columns in the feed zone. 2. On increasing the NaOH concentration from 5 to 8 wt % in the elution buffer, the GABA recovery in feed zone increased from about 55% (EXP002) to about 84% (EXP005). This change also resulted in a product titer increase from 33 g/L to 47 g/L. 3. In addition to the above factors, it was observed that the part of the elution product fraction was collected during ER in elution wash zone (EXP006). During this recipe, the elution wash zone employed the same pump as the regeneration wash. As there has been no flow distribution mechanism to ensure an equal flow to the two different zones, a non-ideal flow of the elution product stream resulted during the course of the SMB cycle, which reduced elution recovery to 60%.”. This was avoided by performing a run with extended elution (EXP007), where the elution zone duration was increased to collect the complete product fraction before the column proceeds to the elution wash zone. This approach resulted in an elution recovery of 84% with a product titer of 50 g/L when the binding capacity was 50 g/SBV.

(23) The overall conclusion of these EBA-SMB experimental studies is that the system can be further optimized to improve GABA recovery in the feed zone and to enhance the overall yield. As a result, the product titer can be enhanced due to higher binding capacity when more columns are available in feed zone, without compromising on yield. Individual inlet pump/column can avoid the need for an extended elution zone. The EBA-SMB technology itself performed consistently during the optimization studies with defined bed level control mechanism. Based on the impurity analysis, the EBA-SMB process achieved a purity of >92% GABA from unclarified fermentation broth, comparable to >93% GABA purity in case of purification of GABA from clarified broth using packed bed adsorption. From the results so far, under the most optimal conditions, the productivity has been improved by 2-fold compared to SMB packed bed. Thus, building the case to eliminate clarification steps and increase the productivity.