Evaluation of center-cut separations applying simulated moving bed chromatography with 8 zones

doi:10.1016/j.chroma.2016.05.060

Journal of Chromatography A

Volume 1456, 22 July 2016, Pages 123-136

https://doi.org/10.1016/j.chroma.2016.05.060 Get rights and content

Highlights

•
Estimation of 8 zone SMB operating space using the equilibrium model.
•
Efficient scanning algorithm based on a recursive solution of the TMB stage model.
•
Novel strategy for the calculation of outlet concentrations in ideal 8 zone TMB.
•
Analytical performance optimization of 8 zone TMB based on a profit function.
•
Definition of decision criteria for the optimal split sequence in 8 zone SMB.

Abstract

Different multi-column options to perform continuous chromatographic separations of ternary mixtures have been proposed in order to overcome limitations of batch chromatography. One attractive option is given by simulated moving bed chromatography (SMB) with 8 zones, a process that offers uninterrupted production, and, potentially, improved economy. As in other established ternary separation processes, the separation sequence is crucial for the performance of the process. This problem is addressed here by computing and comparing optimal performances of the two possibilities assuming linear adsorption isotherms. The conclusions are presented in a decision tree which can be used to guide the selection of system configuration and operation.

Introduction

Simulated moving bed chromatography (SMB) was first patented by Broughton et al. in 1961 for application in the petrochemical industry [1]. It is based on the principle of the true moving bed (TMB), which is itself based on the concept of a continuous countercurrent extractor, as the notation commonly used for its description suggests. The idea behind this process is to maximize the separation driving force by moving the adsorbent and the solvent in opposite directions. However, in practice, moving the adsorbent is not a trivial task, since it is usually composed of small particles and pumping it implies in back-mixing of the solvent, which reduces resolving power. One practical solution found for this problem was to divide the adsorbent into columns and instead of moving the adsorbent in a continuous way, to move the columns upstream in discrete intervals (or by changing the valve connections, effectively changing the relative positions of the columns in the system), thus creating the SMB [2].

In recent years the application of the SMB concept has been mostly studied for the separation of petrochemical derivatives, sugars, biotechnological products, plant extracts and chiral molecules. These separations are typically done using chromatography due to the need to separate solutes with very similar physicochemical properties. Furthermore these separations are also characterized by small differences in adsorption isotherms and/or large scale production and/or relatively expensive mobile phases, a combination of factors typically best dealt with by SMB in comparison to batch chromatography [3], [4].

A major shortcoming of the classical SMB is that the solute being isolated must either be the one with the highest or the one with the lowest affinity to the adsorbent [5], [6], [7]. This restricts its application to more specific cases, usually simpler mixtures.

Taking as an example a mixture composed of 3 components eluting in sequence A, B and C, the most straightforward approach to employ the SMB process to isolate the intermediary eluting solute B is to use two SMBs in sequence in such a way that the first unit separates the solute in the elution series that stands between the intermediary eluting solute B and one of the extremes of the elution series. The feed entering the second unit then contains the intermediary eluting solute in either the first or the last position of the elution series, making it possible to separate it from the other solute. It is possible to merge these two SMB units into a larger integrated SMB unit with 8 zones [4], [5], [8], [9], [10], [11]. There are two ways of doing this coupling, either by recycling the extract from the first subunit into the second subunit (Fig. 1a)) or the raffinate (Fig. 1b)). The difference being which side of the elution series (the least or the most retained solute, respectively) is separated first [9], [11].

Many different SMB configurations have been developed for the separation of ternary mixtures. Nicolaos et al. [10] compared different options of cascades with 4 and 5 zone SMB units in the frame of the equilibrium theory, other authors have used dynamic calculations and optimization algorithms for more realistic comparisons between different SMB column arrangements [5], [6], [9], [12], [13]. Furthermore, many promising semi-continuous configurations, such as the JO (Japan Organo Corp.) and the MCSGP (Multicolumn Countercurrent Solvent Gradient Purification) processes have also been proposed and thoroughly analyzed for performing center-cut separations [9], [14].

An 8 zone SMB works, essentially, as a cascade of two SMBs [11], but it is one pump and one multi-port valve simpler and consumes less solvent [5]. Its main weakness is caused by the fact that all columns must be switched with the same period, limiting the maximal achievable recovery yield [10]. This limitation complicates the optimization process, since it creates a trade off between productivity and recovery yield, causing the optimal separation sequence in a SMB cascade to be different than that of an 8 zone SMB.

The purpose of this work is to use the TMB analogy with the SMB [15], [16], [17] to provide tools for the choice between its two configuration options (extract or raffinate recycle, shown in Fig. 1) and for the definition of optimal flow rates in each zone. This will be done in the three following steps:

First, the operating regions in the parameter space allowing recovery of pure intermediary eluting solute will be defined using the equilibrium theory [18], [19] and cross-validated using the stage model of TMB [8], [20], [21].

Second, a novel strategy for the calculation of ideal 8 zone TMB performance is presented and used, together with the economic description of the process components, to optimize the flow rates. The analytical optimization of 8 zone TMB operating conditions has not yet been done, since this system does not allow complete separation of three fractions, which makes a straightforward solute mass balance impossible [8], [10]. Also because of the maximal recovery yield limitation, the optimization of 8 zone TMB operating conditions is more complex than that of a cascade of 4 zone TMBs and cannot be done solely based on physicochemical data. It also requires economic information about the components involved in the process. This problem has often been dealt with using objective functions [5], [6], [9]. Here the objective function is defined as the profit of the process in order to compare the two configuration options on an economical basis.

Third, the comparison of the optimized performances of each configuration is used to derive explicit criteria for the choice between them, that is, which separation should be done in the first subunit [8], [9], [10], a problem also encountered in the field of continuous distillation [22], [23].

Section snippets

Optimization

We assume that the purpose of optimizing the operating conditions is to maximize the profit of the production process while respecting purity constrains. A definition for the profit (P) is: $P = \frac{(m o n e y e a r n e d - m o n e y i n v e s t e d)}{t i m e i n t e r v a l}$

According to Kawajiri and Biegler [24], the profit can be expressed as a function of various performance indicators as shown in Eq. (2): $P = {\dot{V}}_{S} c_{F e e d} \frac{ε_{T o t a l}}{1 - ε_{T o t a l}} \times P r \times P S P - S C \times S C o - {\dot{V}}_{O u t l e t} \times S R C o - c_{F e e d} {\dot{V}}_{F e e d} \times F C o$ where SC is the total solvent consumption, and, PSP, SCo, SRCo,

Operating regions: extract recycle

The operation of the 8 zone TMB with extract recycle is defined by the following solvent mass balances: ${\begin{array}{c} m_{I (2)} - m_{I I (2)} = m_{E 2} \\ m_{I I I (2)} - m_{I V (2)} = m_{R 2} \\ m_{I (1)} - m_{I V (2)} = m_{D 1} \\ m_{I I I (1)} - m_{I I (1)} = m_{F 1} \\ m_{I I I (1)} - m_{I V (1)} = m_{R 1} \\ m_{I (2)} - m_{I V (1)} = m_{D 2} \\ m_{I (1)} = m_{I I I (2)} + m_{I I (1)} - m_{I I (2)} \end{array}$

Because of the solvent mass balance in the recycled stream, zone II(2) cannot prevent solute B from leaking out in the second extract port without causing either zones I(1), II(1), or, III(2) to fail at their functions (which would compromise the purity of the

Discussion

The results presented in Fig. 7 show that for 8 zone TMB operated at the HP point, when crude feed costs have a higher impact on the economics of the process than solvent removal costs, the choice between raffinate and extract recycle is done on the basis of recovery yield (Fig. 6), while at the opposite situation, the choice between the two system configurations is done on the basis of product concentration. When these two cost contributions are the same, both configurations perform equally

Conclusions

In this work, the equilibrium theory, an equilibrium stage model, and, a profit function were used to optimize the flow rates of 8 zone TMB with extract and raffinate recycle for separating ternary mixtures. The performances at the respective optimum operating points were compared, which led to the derivation of simple criteria for the decision of which of the two separation sequences (easier one first, or harder one first) allows the highest profit (Fig. 7). Simple formulae are given

Acknowledgement

The authors acknowledge the financial support of Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) for the undertaking of this work.

References (27)

D.M. Ruthven et al.
Countercurrent and simulated countercurrent adsorption separation processes
Chem. Eng. Sci.
(1989)
A. Rajendran et al.
Simulated moving bed chromatography for the separation of enantiomers
J. Chromatogr. A
(2009)
G. Agrawal et al.
Comparison of various ternary simulated moving bed separation schemes by multi-objective optimization
J. Chromatogr. A
(2012)
G. Agrawal et al.
Systematic optimization and experimental validation of ternary simulated moving bed chromatography systems
J. Chromatogr. A
(2014)
J. Nowak et al.
Theoretical study of using simulated moving bed chromatography to separate intermediately eluting target compounds
J. Chromatogr. A
(2012)
A. Nicolaos
Application of equilibrium theory to ternary moving bed configurations (four plus four, five plus four, eight and nine zones) I. Linear case
J. Chromatogr. A
(2001)
G.M. Zhong et al.
Analytical solution for the linear ideal model of simulated moving bed chromatography
Chem. Eng. Sci.
(1996)
G.M. Zhong et al.
Steady-state analysis of simulated moving-bed chromatography using the linear, ideal model
Chem. Eng. Sci.
(1998)
C.B. Ching et al.
Experimental-study of a simulated countercurrent adsorption system. 3. SORBEX operation
Chem. Eng. Sci.
(1985)
D. Beltscheva et al.
Linear two-step gradient counter-current chromatography—analysis based on a recursive solution of an equilibrium stage model
J. Chromatogr. A
(2003)

Broughton, D.B. and C.G. Gerhold, Continuous sorption process employing fixed bed of sorbent and moving inlets and...

A. Seidel-Morgenstern et al.

New developments in simulated moving bed chromatography

Chem. Eng. Technol.

(2008)

H. Colin

Foreword—Preparative chromatography today

Analusis

(1998)

Cited by (9)

Simulated-moving-bed technology for purification of the prebiotics galacto-oligosaccharides
2021, Separation and Purification Technology
Citation Excerpt :
In the sense of an efficient separation, the target component should elute as the first or last component in order to be able to separate them. Intermediate target components in a ternary mixture of substances can also be obtained in so-called “center-cut” separations [31–33]. However, such a separation cannot be solved with a classic 4-zone SMB and will not be further discussed at this point.
A commercial galacto-oligosaccharide (GOS) syrup with prebiotic properties was purified with simulated moving bed (SMB) chromatography to separate undesired side products induced by synthesis reaction mechanism. A promising polystyrene-divinylbenzene resin stationary phase has been characterized with regard to its selective adsorption behavior. The chosen chromatographic system allows a pseudo binary separation of the multi-component GOS syrup mixture exploiting a multi-column continuous chromatographic process. Suitable SMB operation parameters were estimated by means of the equilibrium (triangle) theory, further optimized by modelling the separation in more detail and finally validated experimentally. Changes of temperature, concentration of the feed as well as in the instrumental configuration (open loop, closed loop and 3 zones) were examined with a classical SMB plant in lab scale size. Highly purified GOS products could be obtained with purities up to >99% in the extract and raffinate fractions. A quantitative description of the transfer from batch chromatography to the continuous separation process via SMB technology is provided, indicated significant potential for a scale-up.
Model-based process design of a ternary protein separation using multi-step gradient ion-exchange SMB chromatography
2020, Computers and Chemical Engineering
Citation Excerpt :
Cascade schemes require designing and operating two four-zone smb processes, which can be laborious and costly. Hence, integrated schemes have been developed using a single multi-port valve and fewer pumps than cascade schemes (Seidel-Morgenstern et al., 2008; da Silva and Seidel-Morgenstern, 2016). Moreover, process design is simplified due to simultaneous switching of all columns.
Model-based process design of ion-exchange simulated moving bed (iex-smb) chromatography for center-cut separation of proteins is studied. Use of nonlinear binding models that describe more accurate adsorption behaviors of macro-molecules could make it impossible to utilize triangle theory to obtain operating parameters. Moreover, triangle theory provides no rules to design salt profiles in iex-smb. In the modeling study here, proteins (i.e., ribonuclease, cytochrome and lysozyme) on chromatographic columns packed with strong cation-exchanger SP Sepharose FF is used as an example system. The general rate model with steric mass-action kinetics was used; two closed-loop iex-smb network schemes were investigated (i.e., cascade and eight-zone schemes). Performance of the iex-smb schemes was examined with respect to multi-objective indicators (i.e., purity and yield) and productivity, and compared to a single column batch system with the same amount of resin utilized. A multi-objective sampling algorithm, Markov Chain Monte Carlo (mcmc), was used to generate samples for constructing Pareto optimal fronts. mcmc serves on the sampling purpose, which is interested in sampling the Pareto optimal points as well as those near Pareto optimal. Pareto fronts of the three schemes provide full information on the trade-off between the conflicting indicators, purity and yield here. The results indicate that the cascade iex-smb scheme and the integrated eight-zone iex-smb scheme have similar performance and that both outperforms the single column batch system.
Biopharmaceutical development, production, and quality
2020, New and Future Developments in Microbial Biotechnology and Bioengineering: Microbial Biomolecules: Properties, Relevance, and Their Translational Applications
The present chapter continues the description of the pharmaceutical applications of biomolecules that was started in the previous one. However, here we shift our focus onto production and design strategies. We begin by presenting different biopharmaceutical processing technologies, highlighting the role of continuous manufacturing and molecular biology tools in improving production yield and quality. Finally, prompted by the recent advances of the pharmaceutical sector, we provide the reader with a road map for implementing “Quality-by-Design” in the development of biopharmaceuticals.
Biopharmaceutical molecules
2020, New and Future Developments in Microbial Biotechnology and Bioengineering: Microbial Biomolecules: Properties, Relevance, and Their Translational Applications
In this chapter the reader is invited to an in-depth discussion of biotechnology and its applications in the pharmaceutical industry. We begin with a brief overview of the expected health-care challenges for the near future in the light of current demographic trends, which will substantiate the recent technological breakthroughs described in the following sections. Afterward, we describe the importance of microbial molecules in the inception of the vaccine industry and proceed to comprehensively lay out its current tendencies. Next, we cover the main biopharmaceutical categories, including cytokines, hormones, blood factors, and enzymes, stressing production strategies that use microbial expression systems. The chapter ends with a discussion on the growing importance of gene and cell therapy in the future of medicine and lists some of the most exciting innovations in this field.
Efficient numerical simulation of simulated moving bed chromatography with a single-column solver
2018, Computers and Chemical Engineering
Citation Excerpt :
Moreover, process design is simplified due to simultaneous switching of all columns. The left panel of Fig. 3 shows an integrated eight-zone scheme, in which the most retained component is collected at the extract I port, while the raffinate I stream is reinjected across the feed port (Seidel-Morgenstern et al., 2008; da Silva and Seidel-Morgenstern, 2016). The working principle is very similar to the cascade scheme, as the columns are regenerated in zones I and V.
We present four different numerical methods for the numerical simulation of simulated moving bed chromatography. Two approaches use fixed-point iteration for computing cyclic steady states, and two other approaches use operator splitting for computing complete system trajectories. All methods are based on weak coupling of individual column models and can easily be implemented using any existing single-column solver. Simulation software is implemented based on the CADET project and published as open source code.
The numerical performance is compared using five case studies. For both fixed-point iteration and operator-splitting, an alternative approach is found to be more efficient than the standard approach. Namely, the one-column analog saves time in computing the cyclic steady state, while lag-aware operator-splitting yields more detailed information on the system trajectory. The presented methods can be combined with other models, for example to consider hold-up volumes, and have applications beyond simulated bed chromatography.
Research progress on simulated moving bed separation process and its optimization methods
2023, Huagong Jinzhan/Chemical Industry and Engineering Progress

View all citing articles on Scopus

View full text

Evaluation of center-cut separations applying simulated moving bed chromatography with 8 zones

Highlights

Abstract

Introduction

Section snippets

Optimization

Operating regions: extract recycle

Discussion

Conclusions

Acknowledgement

Chem. Eng. Sci.

J. Chromatogr. A

J. Chromatogr. A

J. Chromatogr. A

J. Chromatogr. A

J. Chromatogr. A

Chem. Eng. Sci.

Chem. Eng. Sci.

Chem. Eng. Sci.

J. Chromatogr. A

New developments in simulated moving bed chromatography

Chem. Eng. Technol.

Foreword—Preparative chromatography today

Analusis