Accepted Manuscript Title: Tubing modifications for countercurrent chromatography (CCC): Stationary phase retention and separation efficiency Author: Michael Englert Walter Vetter PII: DOI: Reference:
S0003-2670(15)00582-6 http://dx.doi.org/doi:10.1016/j.aca.2015.04.055 ACA 233899
To appear in:
Analytica Chimica Acta
Received date: Revised date: Accepted date:
30-1-2015 23-4-2015 25-4-2015
Please cite this article as: Michael Englert, Walter Vetter, Tubing modifications for countercurrent chromatography (CCC): Stationary phase retention and separation efficiency, Analytica Chimica Acta http://dx.doi.org/10.1016/j.aca.2015.04.055 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Tubing modifications for countercurrent chromatography (CCC): Stationary phase retention and separation efficiency
Michael Englert, Walter Vetter*
University of Hohenheim, Institute of Food Chemistry, Garbenstrasse 28, D-70599 Stuttgart, Germany * Corresponding author: Walter Vetter Phone: +49 711 459 24016 Fax: +49 711 459 24377 e-mail: [email protected] KEYWORDS Countercurrent chromatography; Instrumentation; Separation Efficiency; Tubing Modification; Highlights ► Column; Multilayer Coil; Highlights
• A crimping tool was developed for the production of tubing modifications in CCC • Six different tubing modifications were produced and tested • Sf in a polar, intermediate and non-polar solvent system was determined • Separation efficiency of tubing modifications was evaluated with standards
• Most tubing modification could improve the separation power of the CCC system
Abstract Countercurrent chromatography (CCC) is a separation technique in which two immiscible liquid phases are used for the preparative purification of synthetic and natural products. In CCC the number of repetitive mixing and de-mixing processes, the retention of the stationary phase and the mass transfer between the liquid phases are significant parameters that influence the resolution and separation efficiency. Limited mass transfer is the main reason for peak broadening and a low number of theoretical plates and impaired peak resolution in CCC. Hence, technical improvements with regard to column design and tubing modifications is an important aspect to enhance mixing and mass transfer. In this study we constructed a crimping tool which allowed us to make reproducible, semi-automated modifications of conventional round-shaped tubing. Six crimped tubing modifications were prepared, mounted onto multilayer coils which were subsequently installed in the CCC system. The stationary phase retention of the tubing modifications were compared to the conventional system with unmodified tubing in a hydrophobic, an intermediate and a hydrophilic two-phase solvent system. Generally, the tubing modifications provided higher capabilities to retain the stationary phase depending on the solvent system and flow rates. In the intermediate solvent system the separation efficiency was evaluated with a mixture of six alkyl p-hydroxybenzoates. The peak resolution could be increased up to 50% with one of the tubing modifications compared to the unmodified tubing. Using the most convincing tubing modification at fixed values for the stationary phase retention, a reasonable comparison to the unmodified tubing was achieved. The peak width could be reduced up to 49% and a strong positive impact at increased flow rates regarding peak resolution and
theoretical plate number was observed compared to unmodified tubing. It could be concluded that the tubing modification enhanced the interphase mixing and mass transfer of the two phases by additional and more vigorous agitation.
1. Introduction
Countercurrent chromatography (CCC), a liquid-liquid separation technique introduced by Ito [1], is based on the respective partitioning of solutes in a two-phase solvent system [2,3]. Nowadays, CCC is a well-established tool for the purification of various natural [4-6] and synthetic products [7] as well as for analytical purposes [8-11]. A characteristic difference of CCC from conventional liquid chromatography techniques is the absence of solid media inside the separation column which is usually necessary for the retention of the stationary phase [12,13]. Unfortunately, CCC suffers from (i) long separation times, (ii) a low number of theoretical plates N and (iii) a low peak resolution Rs compared with high-performance liquid chromatography (HPLC) [14-17]. This is due to the limited number of repetitive mixing and de-mixing processes and mass transfer between the liquid phases which impairs the resolution and separation efficiency and leads to peak broadening [18].
In this regard, the design and nature of the separation column is of utmost importance as it can be considered as the heart of a chromatographic system. In CCC the separation column usually consists of a single, round-shaped polytetrafluoroethylene (PTFE) tubing wrapped directly around a holder hub in multiple layers, referred to as conventional multilayer coils [12,19]. Surprisingly, only few studies and small scale tests on modification and improvement of tubing shapes have been conducted so far. A comparison of the separation performance of CCC multilayer coils equipped with different tubing geometries was made by Degenhardt et
al. in the separation of a standard anthocyanin mixture [20]. It was suspected that the fluid dynamics occurring inside the tubing were influenced in such a manner that undesirable longitudinal flow of the mobile phase could be interrupted [20]. More recently, a similar approach was followed for the improvement of the spiral tube support assembly for CCC (specially developed for the separation of extremely polar compounds with polar solvent systems) by changing the shape of the tubing through compressions perpendicularly along the length of the tubing [21-23]. Convoluted tubing was introduced to improve the retention of the stationary phase during the scale-up of slow-rotary CCC by Du and Winterhalter and it has been successfully applied in the purification of natural products [24-27]. In addition, Yang et al. investigated novel geometric designs of tubing in the rarely used and noncommercialized toroidal zigzag CCC apparatus and observed improved peak resolution [28,29]. Yet, all reported tubing modifications were handmade and a direct comparison to the unmodified tubing was not feasible because the capacity or length of the CCC column changed. Moreover, all these approaches have been one-time attempts and none of them have been converted into standard instrumentation.
The goal of this study was to examine different tubing modifications for CCC by a systematic, reliable and reproducible approach. These considerations were taken up by introducing a crimping tool which allowed us to generate reproducible, semi-automated modifications on conventional round-shaped tubing. Performances of six different crimped tubing modification geometries were directly compared to the unmodified, round-shaped tubing in a hydrophobic, an intermediate and a hydrophilic two-phase solvent system [30]. 2. Material and methods
2.1 Solvents and reagents
All organic solvents used for CCC experiments were of analytical grade or distilled before use. Methanol and tert-butylmethylether (tBME) were from Fisher Scientific (Leicestershire, United Kingdom). Acetonitrile and n-hexane were from Th. Geyer (Renningen, Germany). n-Butanol was from Merck (Darmstadt, Germany). Methyl-, ethyland propyl p-hydroxybenzoate (purity 99%) and acetic acid (purity 99.7%) were purchased from Sigma-Aldrich (Steinheim, Germany). Butyl p-hydroxybenzoate (purity 99%) was from Fluka (Taufkirchen, Germany) and isopropyl- and isobutyl p-hydroxybenzoate (purity of both 98%) were from Alfa Aesar (Karlsruhe, Germany).
2.2 Design of the crimping tool
For the reproducible production of different tubing modifications (TM), a crimping tool was designed and constructed (Fig. 1). The PTFE tubing (1.6 mm internal diameter (i.d.), 2.5 mm outer diameter, Zeus Industrial Products, Orangeburg, SC, USA) was first inserted into a pair of stainless steel guide rollers followed by insertion into two crimping rollers. The distance between superimposed rollers can be regulated in a range from 0 to 4 cm through an eccentric screw which allows the insertion of different tubing diameters and the generation of different depths of the crimping. Four crimping rollers were built for getting equipped with up to eight protuberances of 0.5 cm length at the outer rim, two of them in a right angle (90°) and two of them at an angle of 45° relative to the central axis of the rollers. Different TM can be produced by means of different settings of the protuberances on the upper and lower crimping rollers. Depending of the number of protuberances the distance between two crimpings on the tubing can be regulated between 0.5 (eight protuberances) and 9.5 cm (one protuberance). Moreover, onesided crimping was obtained if one of the crimping rollers remained unfitted and depending of the starting points of the two crimping rollers, parallel (P) or staggered (S) crimping could
be achieved. These variants, along with the use of right angle and/or 45° angled crimping relative to the tubing allowed to generate multiple TM. The tubing was crimped by drawing it slowly and evenly through the guide and crimping rollers. Afterwards, the modified tubing was visually controlled for correct crimping depths and distances. The crimping tool also served as a column holder for the subsequent winding of the modified tubing onto the holder hub, making multiple layers of tubing in a reproducible manner. The holder was adapted to the commercially available CCC system (section 2.4).
2.3 Tubing modifications and denotation
Six different types of TM (Fig. 2, Fig. S1) were made with the crimping tool by exchanging the protuberances of the crimping rollers (section 2.2). The tubing was crimped at intervals of 1.0 cm either parallel (P) or staggered (S) at depths of 0.3 mm. Two TM were made in each case at 90 ° (| |) or at 45 ° in a parallel (//) or anti-parallel (/ \) manner. This resulted in the tubing modifications |P| (Fig. 2a), |S| (Fig. 2b), /P/ (Fig. 2c), /P\ (Fig. 2d), /S\ (Fig. 2e) and \S/ (Fig. 2f). The capacity and length of the tubing was the same for all TM to obtain identical experimental conditions.
2.4 Apparatus
In the present study a CCC-1000 instrument (Pharma-Tech Research, Baltimore, MD, USA) was used with the peripheral equipment as previously described [31] and a total column volume of Vc = 65 mL. The original semi-preparative coils from the manufacturer where replaced with self-constructed coils which were adapted to match the design of the apparatus and consisted of a polyamide holder hub, a stainless steel column axis fitted with two
bearings and a spur gear (Kress, Neuenstadt, Germany) with 60 teeth and a module (gear diameter/number of teeth) of 1.25 with a slight profile shift (± 0.5 mm) to provide high stability and low wear under rotation of the centrifuge. Each coil was equipped with 10 m PTFE tubing of 1.65 mm i.d. (Zeus Industrial Products, Orangeburg, SC, USA) which was either unmodified tubing or undertaken modification with the crimping tool (section 2.2). The tubing was wound onto the holder hubs tightly, forming six multiple layers. The distance R between the central axis of the centrifuge and the column axis was 7.6 cm, resulting in a βvalue (β = r/R; r is the distance between the coil holder shaft and the coil axis) ranging from 0.50 at the internal terminal to 0.62 at the external terminal. 2.5 2.6 Preparation of solvent systems and sample solutions
The stationary phase retention Sf is mainly influenced by the mean viscosity ηM, density differences Δρ and interfacial tension σ between the two phases [32]. Preferentially, a solvent system should possess a low mean viscosity ηM, a high interfacial tension σ and density difference Δρ to achieve high Sf [32,33]. Taking into account the different behavior of solvent systems in CCC [26], three different and routinely used two-phase solvent systems were selected (one hydrophobic, one intermediate and one hydrophilic). The non-aqueous hydrophobic system (HACN) was made of n-hexane/acetonitrile (1:1, v/v). The intermediate solvent system (HTMW) consisted of n-hexane/tBME/methanol/water (5:2:5:3, v/v/v/v) and the polar solvent system (BAW) was prepared from n-butanol/acetic acid/water (5:1:4, v/v/v). Densities ρ, viscosities η and settling times ts of the solvent systems (Table 1) were determined according to Englert and Vetter [31]. A SITE100 spinning-drop tensiometer (Krüss, Hamburg Germany) was used to measure the interfacial tension σ (Table 1). The intermediate two-phase solvent system HTMW was used for the separation of a homologous series of alkyl p-hydroxybenzoates. Sample solutions for the separation of six
alkyl p-hydroxybenzoates (methyl-, ethyl-, isopropyl-, propyl-, isobutyl- and butyl phydroxybenzoate) were prepared by dissolving 25 mg of each compound in 100 mL mobile phase. Partition coefficients K with selectivity factors α of the respective alkyl phydroxybenzoates are given in Table 2. Solvent systems were prepared by adding the corresponding volume ratios of solvents (Table 1) in a 2 L separatory funnel. The mixtures were shaken repeatedly and allowed to equilibrate overnight. The two phases were separated shortly before use and degassed for 10 min in an ultrasonic bath.
2.6 Separation procedure
CCC experiments regarding the stationary phase retention Sf of the different TM were initiated by filling the CCC column entirely with the stationary phase of the respective twophase solvent system without rotation. Then the centrifuge was rotated at 1000 ± 10 rpm and the mobile phase of the solvent system was pumped into the CCC column at flow rates of 0.5, 1, 2, 3 or 4 mL min-1 in separate experiments. A 100 mL graduated cylinder was used to measure the volume of the displaced stationary phase from the CCC column. After the appearance of the mobile phase front at the CCC outlet, indicating that a hydrodynamic equilibrium was reached inside the CCC column, the rotation of the centrifuge was stopped. For the separations of the alkyl p-hydroxybenzoate homologous series the centrifuge was continued to rotate at 1000 ± 10 rpm and 2 mL of the sample solution was injected through a 10 mL sample loop after the displacement of the stationary phase was finished. For the fixed stationary phase retention studies the mobile phase flow was increased during the equilibration process up to 10 mL min-1, until sufficient stationary phase was displaced to reach the Sf value 0.56.
Each CCC experiment was performed in duplicate with only slight deviations (below 5%) regarding Sf values, elution volumes VR, retention times tR and peak widths wb of the alkyl p-hydroxybenzoates and therefore mean values are listed and used for calculations.
2.7 Evaluation of stationary phase retention Sf and separation efficiency
With the knowledge of the total CCC column volume Vc and the mobile phase volume Vm, the stationary phase volume Vs can be expressed by equation (1):
Vs = Vc – Vm
Equation (1)
A unique property of CCC is that Vm and Vs are dependent on each other and that the ratio between them is not necessarily constant during a separation [34]. This relationship can be described by the stationary phase retention Sf with equation (2) relating them to each other:
Sf = Vs/(Vm+Vs) = Vs/Vc
Equation (2)
For each CCC separation of the alkyl p-hydroxybenzoate homologous series, the chromatographic data was recorded and the experimental retention times tR, retention volumes VR and the baseline peak widths wb were obtained for each peak. The resolution Rs between adjacent peaks was calculated using the classical equation (3):
Rs = 2 (VR2 - VR1)/(wb1 + wb2)
Equation (3)
The theoretical plate number N was calculated from VR and wb according to equation (4):
N = (4VR/wb)²
Equation (4)
3. Results and discussion
3.1 Stability tests and reproducibility
Initial stability tests with TM |P|, |S|, /P/, /P\, /S\ and \S/ by applying external hydrostatic pressures of 300 psi (2.1 MPa) and temperatures of 40 °C for 3 h did not indicate changes in shape or capacity of the tubing material. These tests verified robust CCC conditions and confirmed that the TM were well-suited for practical applications. Also the life limits of the TM were not exceeded during the investigations as no leaks were detected and no ruptures were observed. The reproducibility of the tubing winding and formation of multiple layers on the holder hub was also ensured in order to exclude effects on the separation efficiency (Fig. S2). The discrepancy of the separation parameters was 25 s the fluid behavior is reversed and the lower phase should be introduced from the tail and the upper phase from the head of the coil) [38]. Next to |P|, /S\ and \S/ also provided considerably high Sf values for all flow rates. For |S|, /P/ and /P\, the Sf values were slightly worse and dropped faster than for |P|, \P/ and \S/.
4. Conclusions The tubing modifications used in this study provided an equal or greater overall capability to retain the stationary phase in a hydrophobic, an intermediate and a hydrophilic solvent system than the unmodified tubing which is commonly used in commercial CCC apparatus. Although the effects were varied in dependence of the solvent systems, the flow rates and the partitioning coefficient of the analytes, the staggered tubing modification \S/ was usually the best among the systems tested. The latter was found to provide high retention of stationary phase and a unique separation efficiency in the intermediate solvent system during the separation of alkyl p-hydroxybenzoates which was improved with increasing flow rates. Our results suggest that there is room for improvements with regard to the depth of the TM and the distance between crimpings. Both, studying the mass transfer rate and theoretical simulations by means of computational fluid dynamics would help to confirm and explain the occurring effects. The present tubing modifications are amenable to be scaled-up for large-scale industrial applications.
Conflict of interest The authors declare no conflict of interest.
Acknowledgements
We are grateful to the state of Baden-Württemberg for providing a stipend grant to Michael Englert. The authors owe a debt of gratitude to the Machine Instrumentation Facility of the University of Hohenheim for their help in the fabrication of the crimping tool.
Appendix A. Supplementary data Supplementary data associated with this article can be found in the online version.
References [1] Y. Ito, R.L. Bowman, Countercurrent Chromatography: Liquid-liquid partition chromatography without solid support, J. Chromatogr. Sci. 8 (1970) 315-323. [2] Y. Ito, Countercurrent chromatography, J. Biochem. Biophys. Methods 5 (1981) 105-129 [3] Y. Ito, Golden rules and pitfalls in selecting optimum conditions for high-speed countercurrent chromatography, J. Chromatogr. A 1065 (2005) 145-168. [4] A. Marston, K. Hostettmann, Counter-current chromatography as a preparative tool — applications and perspectives, J. Chromatogr. A 658 (1994) 315-341. [5] G.G. Leitao, P.A. de Souza, A.A. Moraes, L. Brown, Step‐gradient CCC separation of phenylpropanoid and iridoid glycosides from roots of Stachytarpheta cayennensis (Rich.) Vahl, J. Liq. Chromatogr. Rel. Technol. 28 (2005) 2053-2060. [6] H. Adhami, M. Zehl, C. Dangl, D. Dorfmeister, M. Stadler, E. Urban, P. Hewitson, S. Ignatova, L. Krenn, Preparative isolation of oleocanthal, tyrosol, and hydroxytyrosol from olive oil by HPCCC, Food Chem. 170 (2015) 154-159. [7] R.G. Whiteside, D.E. Games, C. Strawson, A.S. Graham, L. Brown, Application of high speed CCC for the purification of lisinopril-diketopiperazine diastereomers, J. Liq. Chromatogr. Rel. Technol. 24 (2001) 1801-1810.
[8] T. Kapp, W. Vetter, Offline coupling of high-speed counter-current chromatography and gas chromatography/mass spectrometry generates a two-dimensional plot of toxaphene components, J. Chromatogr. A 1216 (2009) 8391-8397. [9] M.H. Grace, C.W. Warlick, S.A. Neff, M.A. Lila, Efficient preparative isolation and identification of walnut bioactive components using high-speed counter-current chromatography and LC-ESI-IT-TOF-MS, Food Chem. 158 (2014) 229-238. [10] M.P. Rodríguez-Rivera, E. Lugo-Cervantes, P. Winterhalter, G. Jerz, Metabolite profiling of polyphenols in peels of Citrus limetta Risso by combination of preparative highspeed countercurrent chromatography and LC–ESI–MS/MS, Food Chem. 158 (2014) 139152. [11] M. Schröder, H. Abdurahman, T. Ruoff, K. Lehnert, W. Vetter, Identification of aromatic fatty acids in butter fat, J. Am. Oil Chem. Soc. 91 (2014) 1695-1702. [12] Y. Ito, J. Sandlin, W.G. Bowers, High-speed preparative counter-current chromatography with a coil planet centrifuge, J. Chromatogr. A 244 (1982) 247-258. [13] W.D. Conway, Counter-current chromatography, J. Chromatogr. A 538 (1991) 27-35. [14] S. Li, W. Wang, H. Tang, K. Chen, J. Yang, L. He, H. Ye, A. Peng, L. Chen, Comparison of counter-current chromatography and preparative high performance liquid chromatography applied to separating minor impurities in drug preparations, J. Chromatogr. A 1344 (2014) 51-58. [15] S. Tong, Y. Zheng, J. Yan, Application and comparison of high performance liquid chromatography and high speed counter-current chromatography in enantioseparation of (±)2-phenylpropionic acid, J. Chromatogr. A 1281 (2013) 79-86. [16] J. He, R. Tsao, R. Yang, T. Zhou, Purification of patulin from Penicillium expansum culture: high-speed counter-current chromatography (HSCCC) versus preparative highperformance liquid chromatography (prep-HPLC), Food Addit. Contam. A 26 (2009) 101107.
[17] M.-C. Menet, D. Thiebaut, Preparative purification of antibiotics for comparing hydrostatic and hydrodynamic mode counter-current chromatography and preparative highperformance liquid chromatography, J. Chromatogr. A 831 (1999) 203-216. [18] C. Zhao, C. He, Mass transfer and separation criteria for high-speed countercurrent chromatography, AIChE J. 57 (2011) 359-372. [19] T.A. Maryutina, S.N. Ignatova, I.A. Sutherland, Organic phase retention in CCC: Effect of temperature, tubing material, tubing bore, mobile phase flow, and the addition of an extraction agent, J. Liq. Chromatogr. Rel. Technol. 26 (2003) 1537-1550. [20] A. Degenhardt, M. Schwarz, P. Winterhalter, Y. Ito, Evaluation of different tubing geometries for high-speed counter-current chromatography, J. Chromatogr. A 922 (2001) 355-358. [21] Y. Ito, R. Clary, J. Powell, M. Knight, T.M. Finn, Improved spiral tube assembly for high-speed counter-current chromatography, J. Chromatogr. A 1216 (2009) 4193-4200. [22] Y. Yang, H.A. Aisa, Y. Ito, Flat-twisted tubing: Novel column design for spiral highspeed counter-current chromatography, J. Chromatogr. A 1216 (2009) 5265-5271. [23] Y. Ito, M. Knight, T.M. Finn, Spiral countercurrent chromatography, J. Chromatogr. Sci. 51 (2013) 726-738. [24] Q. Du, Z. Li, Y. Ito, Preparative separation of isoflavone components in soybeans using high-speed counter-current chromatography, J. Chromatogr. A, 923 (2001) 271-274. [25] Q. Du, P. Winterhalter, Y. Ito, Large convoluted tubing for scale‐up of slow rotary countercurrent chromatograph, J. Liq. Chromatogr. Rel. Technol., 26 (2003) 1991-2002. [26] Q. Du, P. Wu, Y. Ito, Low-speed rotary countercurrent chromatography using a convoluted multilayer helical tube for industrial separation, Anal. Chem., 72 (2000) 33633365.
[27] T. Esatbeyoglu, V. Wray, Isolation of dimeric, trimeric, tetrameric and pentameric procyanidins from unroasted cocoa beans (Theobroma cacao L.) using countercurrent chromatography, P. Winterhalter, Food Chem., 179 (2015) 278-289. [28] Y. Yang, H.A. Aisa, Y. Ito, Novel design for centrifugal countercurrent chromatography: I. Zigzag toroidal column, J. Liq. Chromatogr. Rel. Technol. 32 (2009) 2030-2042. [29] Y. Yang, H.A. Aisa, Y. Ito, Novel design for centrifugal countercurrent chromatography: II. Studies on novel geometries of zigzag toroidal tubing, J. Liq. Chromatogr. Rel. Technol. 33 (2010) 336-348. [30] Y. Ito, Experimental observations of the hydrodynamic behavior of solvent systems in high-speed counter-current chromatography, J. Chromatogr. A 301 (1984) 387-403. [31] M. Englert, W. Vetter, Solvent systems with n-hexane and/or cyclohexane in countercurrent chromatography - Physico-chemical parameters and their impact on the separation of alkyl hydroxybenzoates, J. Chromatogr. A 1342 (2014) 54-62. [32] S.N. Ignatova, T.A. Maryutina, B.Y. Spivakov, Effect of physicochemical properties of two-phase liquid systems on the retention of stationary phase in a CCC column, J. Liq. Chromatogr. Rel. Technol. 24 (2001) 1655-1668. [33] A. Berthod, Practical approach to high-speed counter-current chromatography, J. Chromatog. A 550 (1991) 677-693. [34] A. Berthod, T. Maryutina, B. Spivakov, O. Shpigun, I.A. Sutherland, Countercurrent chromatography in analytical chemistry (IUPAC technical report), Pure Appl. Chem. 81 (2009) 355-387. [35] Q. Du, C. Wu, G. Qian, P. Wu, Y. Ito, Relationship between the flow-rate of the mobile phase and retention of the stationary phase in counter-current chromatography, J. Chromator. A 835 (1999) 231-235. [36] J.B. Friesen, G.F. Pauli, G.U.E.S.S. - A generally useful estimate of solvent systems in CCC, J. Liq. Chromatogr. Relat. Technol. 28 (2005) 2777-2806.
[37] O. Bousquet, F. Le Goffic, Counter-current chromatographic separation of polyunsaturated fatty acids, J. Chromatogr. A 704 (1995) 211-216. [38] M. Liu, F. Yang, H. Shi, C.C. Akoh, L. Yu, Preparative separation of triterpene alcohol ferulates from rice bran oil using a high performance counter-current chromatography, Food Chem. 139 (2013) 919-924.
[39] W. Vetter, J. Kirres, P. Bendig, Bromination of 2-methoxydiphenyl ether to an average of tetrabrominated 2-methoxydiphenyl ethers, Chemosphere 84 (2011) 1117-1124. [40] D. Gu, Y. Yang, X. Xin, H.A. Aisa, Y. Ito, Novel design for centrifugal counter-current chromatography: VI. Ellipsoid column, J. Liq. Chromatogr. Rel. Technol. 38 (2015) 68-73 [41] K. Shinomiya, Y. Kabasawa, Y. Ito, Countercurrent chromatographic separation of sugars and their p-nitrophenyl derivatives by cross-axis coil planet centrifuge, J. Liq. Chromatogr. Rel. Technol. 22 (1999) 579-592. [42] Y. Ito, W.D. Conway, Experimental observations of the hydrodynamic behavior of solvent systems in high-speed counter-current chromatography: III. Effects of physical properties of the solvent systems and operating temperature on the distribution of two-phase solvent systems, J. Chromatogr. A 301 (1984) 405-414. [43] P.L. Wood, B. Jaber, I.A. Sutherland, A New hypothesis on the hydrodynamic distribution of the upper and lower phases in CCC, J. Liq. Chromatogr. Rel. Technol. 24 (2001) 1629-1654.
Captions to Figures
Fig. 1. a) Schematic design of the crimping tool for tubing modifications with two guide rollers and two crimping rollers. The crimping rollers are equipped with up to eight protuberances with 0.5 cm distance between them. By changing the protuberances different
tubing modifications can be realized. After insertion, the tubing is drawn through the guide and crimping rollers manually and is then crimped in a reproducible manner. The crimping tool also serves as a column holder for the subsequent winding of the modified tubing onto the holder hub. b) A photograph of the crimping tool without installed coil.
Fig. 2. Schematic images of the six different tubing modifications denoted as a) |P|, b) |S|, c) /P/, d) /P\, e) /S\ and f) \S/ investigated in CCC experiments in side view. The black arrows indicate the direction of the crimping to distinguish between P: parallel and S: staggered crimping | |: at 90 °; //: at 45 ° parallel and / \: at 45 ° anti-parallel.
Fig. 3. Stationary phase retention Sf at flow rates of 0.5, 1, 2, 3 and 4 mL min-1 determined for unmodified tubing U and tubing modifications |P|, |S|, /P/, /P\, /S\ and \S/ in the intermediate polarity n-hexane/tBME/methanol/water (5:2:5:3, v/v/v/v, HTMW) solvent Determination was conducted in head-to-tail (a) and tail-to-head mode (b).
system.
Fig. 4. CCC separation of six alkyl p-hydroxybenzoates (∼5 mg of each compound) in headto-tail mode with (a) unmodified tubing U and tubing modifications (b) |P|, (c) |S|, (d) /P/, (e) /P\, (f) /S\, and (g) \S/ with n-hexane/tBME/methanol/water (5:2:5:3, v/v/v/v) at a flow rate of 1 mL min-1. The rotation speed was 1000 ± 10 rpm and ELSD detection was conducted with nitrogen gas stream flow set to 1.2 mL min-1 and nebulizer temperature set to 40 °C. The alkyl hydroxybenzoates detected are (1) methyl p-hydroxybenzoate, (2) ethyl p-hydroxybenzoate, (3) isopropyl p-hydroxybenzoate, (4) propyl p-hydroxybenzoate, (5) isobutyl phydroxybenzoate and (6) butyl p-hydroxybenzoate. Changes in peak resolution and peak width are particularly noticeable for the first three eluting compounds (1-3) and are indicated by a blue background.
Fig. 5. Evaluation of peak widths wb obtained during the CCC separation of six alkyl phydroxybenzoates with a) methyl p-hydroxybenzoate, b) ethyl p-hydroxybenzoate, c) isopropyl p-hydroxybenzoate, d) propyl p-hydroxybenzoate, e) isobutyl p-hydroxybenzoate and
f)
butyl
p-hydroxybenzoate
using
the
intermediate
polarity
n-
hexane/tBME/methanol/water (5:2:5:3, v/v/v/v, HTMW) solvent system with the unmodified
tubing U and six tubing modifications |P|, |S|, /P/, /P\, /S\ and \S/ in head-to-tail mode at five different flow rates 0.5, 1, 2, 3 and 4 mL min-1.
Fig. 6. Resolution Rs obtained between the peaks of a) methyl p-hydroxybenzoate and ethyl phydroxybenzoate, b) ethyl p-hydroxybenzoate and isopropyl p-hydroxybenzoate, c) isopropyl p-hydroxybenzoate and propyl p-hydroxybenzoate, d) propyl p-hydroxybenzoate and isobutyl p-hydroxybenzoate and e) isobutyl p-hydroxybenzoate and butyl p-hydroxybenzoate using the intermediate polarity n-hexane/tBME/methanol/water (5:2:5:3, v/v/v/v, HTMW) solvent system with the unmodified tubing U and six tubing modifications |P|, |S|, /P/, /P\, /S\ and \S/ in head-to-tail mode at five different flow rates 0.5, 1, 2, 3 and 4 mL min-1.
Fig. 7. Tubing modification \S/ utilized for the CCC separation of six alkyl phydroxybenzoates (∼5 mg of each compound) in head-to-tail mode at five different flow rates 0.5, 1, 2, 3 and 4 mL min-1. For conditions and compound numbering see Fig. 4.
Fig. 8. Performance of the alkyl p-hydroxybenzoate separation in terms of peak width wb and number of theoretical plates N with unmodified tubing U (a,c) and tubing modification \S/
(b,d) at five different flow rates 0.5, 1, 2, 3 and 4 mL min-1 and a fixed stationary phase retention Sf=0.56.
Fig. 9. Stationary phase retention Sf at five different flow rates 0.5, 1, 2, 3 and 4 mL min-1 in head-to-tail and tail-to-head mode determined for unmodified tubing U and tubing modifications |P|, |S|, /P/, /P\, /S\ and \S/ (a,b) in solvent systems HACN which consisted of nhexane/acetonitrile (1:1, v/v) and (c,d) BAW which consisted of n-butanol/acetic acid/water (5:1:4, v/v/v).
Table 1. Physico-chemical properties of the two-phase solvent systems used for measuring the stationary phase retention of different tubing modifications.
Solvent
Viscosity
Mean
Density
Density
Interfacial Settling
system
(c.p.)
viscosity
(g cm-3)
difference
Tension
time
ηUP/ηLP
(c.p.)
ρUP/ρUP
(g cm-3)
(mN m-1)
(s)
Δρ
σ
ts
ηM HACNa
0.32/0.30
0.31
0.67/0.77
0.10
1.10
9
HTMWb
0.30/0.68
0.49
0.66/0.91
0.25
1.14
8
BAWc
2.49/1.52
2.01
0.89/0.99
0.10
1.22
29
a
HACN: n-hexane/acetonitrile (1:1, v/v)
b
HTMW: n-hexane/tBME/methanol/water (5:2:5:3, v/v/v/v)
c
BAW: n-butanol/acetic acid/water (4:1:5, v/v)
Table 2. Partitioning coefficients K with selectivity factor α of the six alkyl phydroxybenzoates in the intermediate solvent system (HTMW) nhexane/tBME/methanol/water (5:2:5:3, v/v/v/v) and concentration of the compounds in the sample solution used in CCC.
Compound
Abbreviation
Methyl p-
Partitioning
Selectivity
Sample solution
coefficient
factor
concentration
K [27]
α
[mg L-1]
0.08
-
MePHB
254
hydroxybenzoate Ethyl p-
0.17
2.13
EtPHB
252
hydroxybenzoate Isopropyl p-
0.28
1.65
iPrPHB
251
hydroxybenzoate Propyl p-
0.50
1.79
PrPHB
252
hydroxybenzoate Isobutyl p-
0.73
1.46
iBuPHB
251
hydroxybenzoate Butyl p-
0.99 BuPHB
hydroxybenzoate
1.36 251
Table 3. Percentage of decrease of peak widths wb calculated between the unmodified tubing U and tubing modification \S/of six alkyl p-hydroxybenzoates using the intermediate polarity n-hexane/tBME/methanol/water (5:2:5:3, v/v/v/v, HTMW) solvent system.
Compound Flow rate MePHB
EtPHB
iPrPHB
PrPHB
iBuPHB
BuPHB
0.5
6.21
9.70
17.5
20.9
25.1
28.5
1
9.73
19.1
24.0
29.9
34.8
39.0
2
17.4
28.6
32.3
39.2
41.8
41.4
3
22.2
33.3
36.9
41.9
47.2
44.0
4
26.6
35.4
43.9
44.4
48.0
49.3
-1
[mL min ]
S0003-2670(15)00582-6 http://dx.doi.org/doi:10.1016/j.aca.2015.04.055 ACA 233899
To appear in:
Analytica Chimica Acta
Received date: Revised date: Accepted date:
30-1-2015 23-4-2015 25-4-2015
Please cite this article as: Michael Englert, Walter Vetter, Tubing modifications for countercurrent chromatography (CCC): Stationary phase retention and separation efficiency, Analytica Chimica Acta http://dx.doi.org/10.1016/j.aca.2015.04.055 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Tubing modifications for countercurrent chromatography (CCC): Stationary phase retention and separation efficiency
Michael Englert, Walter Vetter*
University of Hohenheim, Institute of Food Chemistry, Garbenstrasse 28, D-70599 Stuttgart, Germany * Corresponding author: Walter Vetter Phone: +49 711 459 24016 Fax: +49 711 459 24377 e-mail: [email protected] KEYWORDS Countercurrent chromatography; Instrumentation; Separation Efficiency; Tubing Modification; Highlights ► Column; Multilayer Coil; Highlights
• A crimping tool was developed for the production of tubing modifications in CCC • Six different tubing modifications were produced and tested • Sf in a polar, intermediate and non-polar solvent system was determined • Separation efficiency of tubing modifications was evaluated with standards
• Most tubing modification could improve the separation power of the CCC system
Abstract Countercurrent chromatography (CCC) is a separation technique in which two immiscible liquid phases are used for the preparative purification of synthetic and natural products. In CCC the number of repetitive mixing and de-mixing processes, the retention of the stationary phase and the mass transfer between the liquid phases are significant parameters that influence the resolution and separation efficiency. Limited mass transfer is the main reason for peak broadening and a low number of theoretical plates and impaired peak resolution in CCC. Hence, technical improvements with regard to column design and tubing modifications is an important aspect to enhance mixing and mass transfer. In this study we constructed a crimping tool which allowed us to make reproducible, semi-automated modifications of conventional round-shaped tubing. Six crimped tubing modifications were prepared, mounted onto multilayer coils which were subsequently installed in the CCC system. The stationary phase retention of the tubing modifications were compared to the conventional system with unmodified tubing in a hydrophobic, an intermediate and a hydrophilic two-phase solvent system. Generally, the tubing modifications provided higher capabilities to retain the stationary phase depending on the solvent system and flow rates. In the intermediate solvent system the separation efficiency was evaluated with a mixture of six alkyl p-hydroxybenzoates. The peak resolution could be increased up to 50% with one of the tubing modifications compared to the unmodified tubing. Using the most convincing tubing modification at fixed values for the stationary phase retention, a reasonable comparison to the unmodified tubing was achieved. The peak width could be reduced up to 49% and a strong positive impact at increased flow rates regarding peak resolution and
theoretical plate number was observed compared to unmodified tubing. It could be concluded that the tubing modification enhanced the interphase mixing and mass transfer of the two phases by additional and more vigorous agitation.
1. Introduction
Countercurrent chromatography (CCC), a liquid-liquid separation technique introduced by Ito [1], is based on the respective partitioning of solutes in a two-phase solvent system [2,3]. Nowadays, CCC is a well-established tool for the purification of various natural [4-6] and synthetic products [7] as well as for analytical purposes [8-11]. A characteristic difference of CCC from conventional liquid chromatography techniques is the absence of solid media inside the separation column which is usually necessary for the retention of the stationary phase [12,13]. Unfortunately, CCC suffers from (i) long separation times, (ii) a low number of theoretical plates N and (iii) a low peak resolution Rs compared with high-performance liquid chromatography (HPLC) [14-17]. This is due to the limited number of repetitive mixing and de-mixing processes and mass transfer between the liquid phases which impairs the resolution and separation efficiency and leads to peak broadening [18].
In this regard, the design and nature of the separation column is of utmost importance as it can be considered as the heart of a chromatographic system. In CCC the separation column usually consists of a single, round-shaped polytetrafluoroethylene (PTFE) tubing wrapped directly around a holder hub in multiple layers, referred to as conventional multilayer coils [12,19]. Surprisingly, only few studies and small scale tests on modification and improvement of tubing shapes have been conducted so far. A comparison of the separation performance of CCC multilayer coils equipped with different tubing geometries was made by Degenhardt et
al. in the separation of a standard anthocyanin mixture [20]. It was suspected that the fluid dynamics occurring inside the tubing were influenced in such a manner that undesirable longitudinal flow of the mobile phase could be interrupted [20]. More recently, a similar approach was followed for the improvement of the spiral tube support assembly for CCC (specially developed for the separation of extremely polar compounds with polar solvent systems) by changing the shape of the tubing through compressions perpendicularly along the length of the tubing [21-23]. Convoluted tubing was introduced to improve the retention of the stationary phase during the scale-up of slow-rotary CCC by Du and Winterhalter and it has been successfully applied in the purification of natural products [24-27]. In addition, Yang et al. investigated novel geometric designs of tubing in the rarely used and noncommercialized toroidal zigzag CCC apparatus and observed improved peak resolution [28,29]. Yet, all reported tubing modifications were handmade and a direct comparison to the unmodified tubing was not feasible because the capacity or length of the CCC column changed. Moreover, all these approaches have been one-time attempts and none of them have been converted into standard instrumentation.
The goal of this study was to examine different tubing modifications for CCC by a systematic, reliable and reproducible approach. These considerations were taken up by introducing a crimping tool which allowed us to generate reproducible, semi-automated modifications on conventional round-shaped tubing. Performances of six different crimped tubing modification geometries were directly compared to the unmodified, round-shaped tubing in a hydrophobic, an intermediate and a hydrophilic two-phase solvent system [30]. 2. Material and methods
2.1 Solvents and reagents
All organic solvents used for CCC experiments were of analytical grade or distilled before use. Methanol and tert-butylmethylether (tBME) were from Fisher Scientific (Leicestershire, United Kingdom). Acetonitrile and n-hexane were from Th. Geyer (Renningen, Germany). n-Butanol was from Merck (Darmstadt, Germany). Methyl-, ethyland propyl p-hydroxybenzoate (purity 99%) and acetic acid (purity 99.7%) were purchased from Sigma-Aldrich (Steinheim, Germany). Butyl p-hydroxybenzoate (purity 99%) was from Fluka (Taufkirchen, Germany) and isopropyl- and isobutyl p-hydroxybenzoate (purity of both 98%) were from Alfa Aesar (Karlsruhe, Germany).
2.2 Design of the crimping tool
For the reproducible production of different tubing modifications (TM), a crimping tool was designed and constructed (Fig. 1). The PTFE tubing (1.6 mm internal diameter (i.d.), 2.5 mm outer diameter, Zeus Industrial Products, Orangeburg, SC, USA) was first inserted into a pair of stainless steel guide rollers followed by insertion into two crimping rollers. The distance between superimposed rollers can be regulated in a range from 0 to 4 cm through an eccentric screw which allows the insertion of different tubing diameters and the generation of different depths of the crimping. Four crimping rollers were built for getting equipped with up to eight protuberances of 0.5 cm length at the outer rim, two of them in a right angle (90°) and two of them at an angle of 45° relative to the central axis of the rollers. Different TM can be produced by means of different settings of the protuberances on the upper and lower crimping rollers. Depending of the number of protuberances the distance between two crimpings on the tubing can be regulated between 0.5 (eight protuberances) and 9.5 cm (one protuberance). Moreover, onesided crimping was obtained if one of the crimping rollers remained unfitted and depending of the starting points of the two crimping rollers, parallel (P) or staggered (S) crimping could
be achieved. These variants, along with the use of right angle and/or 45° angled crimping relative to the tubing allowed to generate multiple TM. The tubing was crimped by drawing it slowly and evenly through the guide and crimping rollers. Afterwards, the modified tubing was visually controlled for correct crimping depths and distances. The crimping tool also served as a column holder for the subsequent winding of the modified tubing onto the holder hub, making multiple layers of tubing in a reproducible manner. The holder was adapted to the commercially available CCC system (section 2.4).
2.3 Tubing modifications and denotation
Six different types of TM (Fig. 2, Fig. S1) were made with the crimping tool by exchanging the protuberances of the crimping rollers (section 2.2). The tubing was crimped at intervals of 1.0 cm either parallel (P) or staggered (S) at depths of 0.3 mm. Two TM were made in each case at 90 ° (| |) or at 45 ° in a parallel (//) or anti-parallel (/ \) manner. This resulted in the tubing modifications |P| (Fig. 2a), |S| (Fig. 2b), /P/ (Fig. 2c), /P\ (Fig. 2d), /S\ (Fig. 2e) and \S/ (Fig. 2f). The capacity and length of the tubing was the same for all TM to obtain identical experimental conditions.
2.4 Apparatus
In the present study a CCC-1000 instrument (Pharma-Tech Research, Baltimore, MD, USA) was used with the peripheral equipment as previously described [31] and a total column volume of Vc = 65 mL. The original semi-preparative coils from the manufacturer where replaced with self-constructed coils which were adapted to match the design of the apparatus and consisted of a polyamide holder hub, a stainless steel column axis fitted with two
bearings and a spur gear (Kress, Neuenstadt, Germany) with 60 teeth and a module (gear diameter/number of teeth) of 1.25 with a slight profile shift (± 0.5 mm) to provide high stability and low wear under rotation of the centrifuge. Each coil was equipped with 10 m PTFE tubing of 1.65 mm i.d. (Zeus Industrial Products, Orangeburg, SC, USA) which was either unmodified tubing or undertaken modification with the crimping tool (section 2.2). The tubing was wound onto the holder hubs tightly, forming six multiple layers. The distance R between the central axis of the centrifuge and the column axis was 7.6 cm, resulting in a βvalue (β = r/R; r is the distance between the coil holder shaft and the coil axis) ranging from 0.50 at the internal terminal to 0.62 at the external terminal. 2.5 2.6 Preparation of solvent systems and sample solutions
The stationary phase retention Sf is mainly influenced by the mean viscosity ηM, density differences Δρ and interfacial tension σ between the two phases [32]. Preferentially, a solvent system should possess a low mean viscosity ηM, a high interfacial tension σ and density difference Δρ to achieve high Sf [32,33]. Taking into account the different behavior of solvent systems in CCC [26], three different and routinely used two-phase solvent systems were selected (one hydrophobic, one intermediate and one hydrophilic). The non-aqueous hydrophobic system (HACN) was made of n-hexane/acetonitrile (1:1, v/v). The intermediate solvent system (HTMW) consisted of n-hexane/tBME/methanol/water (5:2:5:3, v/v/v/v) and the polar solvent system (BAW) was prepared from n-butanol/acetic acid/water (5:1:4, v/v/v). Densities ρ, viscosities η and settling times ts of the solvent systems (Table 1) were determined according to Englert and Vetter [31]. A SITE100 spinning-drop tensiometer (Krüss, Hamburg Germany) was used to measure the interfacial tension σ (Table 1). The intermediate two-phase solvent system HTMW was used for the separation of a homologous series of alkyl p-hydroxybenzoates. Sample solutions for the separation of six
alkyl p-hydroxybenzoates (methyl-, ethyl-, isopropyl-, propyl-, isobutyl- and butyl phydroxybenzoate) were prepared by dissolving 25 mg of each compound in 100 mL mobile phase. Partition coefficients K with selectivity factors α of the respective alkyl phydroxybenzoates are given in Table 2. Solvent systems were prepared by adding the corresponding volume ratios of solvents (Table 1) in a 2 L separatory funnel. The mixtures were shaken repeatedly and allowed to equilibrate overnight. The two phases were separated shortly before use and degassed for 10 min in an ultrasonic bath.
2.6 Separation procedure
CCC experiments regarding the stationary phase retention Sf of the different TM were initiated by filling the CCC column entirely with the stationary phase of the respective twophase solvent system without rotation. Then the centrifuge was rotated at 1000 ± 10 rpm and the mobile phase of the solvent system was pumped into the CCC column at flow rates of 0.5, 1, 2, 3 or 4 mL min-1 in separate experiments. A 100 mL graduated cylinder was used to measure the volume of the displaced stationary phase from the CCC column. After the appearance of the mobile phase front at the CCC outlet, indicating that a hydrodynamic equilibrium was reached inside the CCC column, the rotation of the centrifuge was stopped. For the separations of the alkyl p-hydroxybenzoate homologous series the centrifuge was continued to rotate at 1000 ± 10 rpm and 2 mL of the sample solution was injected through a 10 mL sample loop after the displacement of the stationary phase was finished. For the fixed stationary phase retention studies the mobile phase flow was increased during the equilibration process up to 10 mL min-1, until sufficient stationary phase was displaced to reach the Sf value 0.56.
Each CCC experiment was performed in duplicate with only slight deviations (below 5%) regarding Sf values, elution volumes VR, retention times tR and peak widths wb of the alkyl p-hydroxybenzoates and therefore mean values are listed and used for calculations.
2.7 Evaluation of stationary phase retention Sf and separation efficiency
With the knowledge of the total CCC column volume Vc and the mobile phase volume Vm, the stationary phase volume Vs can be expressed by equation (1):
Vs = Vc – Vm
Equation (1)
A unique property of CCC is that Vm and Vs are dependent on each other and that the ratio between them is not necessarily constant during a separation [34]. This relationship can be described by the stationary phase retention Sf with equation (2) relating them to each other:
Sf = Vs/(Vm+Vs) = Vs/Vc
Equation (2)
For each CCC separation of the alkyl p-hydroxybenzoate homologous series, the chromatographic data was recorded and the experimental retention times tR, retention volumes VR and the baseline peak widths wb were obtained for each peak. The resolution Rs between adjacent peaks was calculated using the classical equation (3):
Rs = 2 (VR2 - VR1)/(wb1 + wb2)
Equation (3)
The theoretical plate number N was calculated from VR and wb according to equation (4):
N = (4VR/wb)²
Equation (4)
3. Results and discussion
3.1 Stability tests and reproducibility
Initial stability tests with TM |P|, |S|, /P/, /P\, /S\ and \S/ by applying external hydrostatic pressures of 300 psi (2.1 MPa) and temperatures of 40 °C for 3 h did not indicate changes in shape or capacity of the tubing material. These tests verified robust CCC conditions and confirmed that the TM were well-suited for practical applications. Also the life limits of the TM were not exceeded during the investigations as no leaks were detected and no ruptures were observed. The reproducibility of the tubing winding and formation of multiple layers on the holder hub was also ensured in order to exclude effects on the separation efficiency (Fig. S2). The discrepancy of the separation parameters was 25 s the fluid behavior is reversed and the lower phase should be introduced from the tail and the upper phase from the head of the coil) [38]. Next to |P|, /S\ and \S/ also provided considerably high Sf values for all flow rates. For |S|, /P/ and /P\, the Sf values were slightly worse and dropped faster than for |P|, \P/ and \S/.
4. Conclusions The tubing modifications used in this study provided an equal or greater overall capability to retain the stationary phase in a hydrophobic, an intermediate and a hydrophilic solvent system than the unmodified tubing which is commonly used in commercial CCC apparatus. Although the effects were varied in dependence of the solvent systems, the flow rates and the partitioning coefficient of the analytes, the staggered tubing modification \S/ was usually the best among the systems tested. The latter was found to provide high retention of stationary phase and a unique separation efficiency in the intermediate solvent system during the separation of alkyl p-hydroxybenzoates which was improved with increasing flow rates. Our results suggest that there is room for improvements with regard to the depth of the TM and the distance between crimpings. Both, studying the mass transfer rate and theoretical simulations by means of computational fluid dynamics would help to confirm and explain the occurring effects. The present tubing modifications are amenable to be scaled-up for large-scale industrial applications.
Conflict of interest The authors declare no conflict of interest.
Acknowledgements
We are grateful to the state of Baden-Württemberg for providing a stipend grant to Michael Englert. The authors owe a debt of gratitude to the Machine Instrumentation Facility of the University of Hohenheim for their help in the fabrication of the crimping tool.
Appendix A. Supplementary data Supplementary data associated with this article can be found in the online version.
References [1] Y. Ito, R.L. Bowman, Countercurrent Chromatography: Liquid-liquid partition chromatography without solid support, J. Chromatogr. Sci. 8 (1970) 315-323. [2] Y. Ito, Countercurrent chromatography, J. Biochem. Biophys. Methods 5 (1981) 105-129 [3] Y. Ito, Golden rules and pitfalls in selecting optimum conditions for high-speed countercurrent chromatography, J. Chromatogr. A 1065 (2005) 145-168. [4] A. Marston, K. Hostettmann, Counter-current chromatography as a preparative tool — applications and perspectives, J. Chromatogr. A 658 (1994) 315-341. [5] G.G. Leitao, P.A. de Souza, A.A. Moraes, L. Brown, Step‐gradient CCC separation of phenylpropanoid and iridoid glycosides from roots of Stachytarpheta cayennensis (Rich.) Vahl, J. Liq. Chromatogr. Rel. Technol. 28 (2005) 2053-2060. [6] H. Adhami, M. Zehl, C. Dangl, D. Dorfmeister, M. Stadler, E. Urban, P. Hewitson, S. Ignatova, L. Krenn, Preparative isolation of oleocanthal, tyrosol, and hydroxytyrosol from olive oil by HPCCC, Food Chem. 170 (2015) 154-159. [7] R.G. Whiteside, D.E. Games, C. Strawson, A.S. Graham, L. Brown, Application of high speed CCC for the purification of lisinopril-diketopiperazine diastereomers, J. Liq. Chromatogr. Rel. Technol. 24 (2001) 1801-1810.
[8] T. Kapp, W. Vetter, Offline coupling of high-speed counter-current chromatography and gas chromatography/mass spectrometry generates a two-dimensional plot of toxaphene components, J. Chromatogr. A 1216 (2009) 8391-8397. [9] M.H. Grace, C.W. Warlick, S.A. Neff, M.A. Lila, Efficient preparative isolation and identification of walnut bioactive components using high-speed counter-current chromatography and LC-ESI-IT-TOF-MS, Food Chem. 158 (2014) 229-238. [10] M.P. Rodríguez-Rivera, E. Lugo-Cervantes, P. Winterhalter, G. Jerz, Metabolite profiling of polyphenols in peels of Citrus limetta Risso by combination of preparative highspeed countercurrent chromatography and LC–ESI–MS/MS, Food Chem. 158 (2014) 139152. [11] M. Schröder, H. Abdurahman, T. Ruoff, K. Lehnert, W. Vetter, Identification of aromatic fatty acids in butter fat, J. Am. Oil Chem. Soc. 91 (2014) 1695-1702. [12] Y. Ito, J. Sandlin, W.G. Bowers, High-speed preparative counter-current chromatography with a coil planet centrifuge, J. Chromatogr. A 244 (1982) 247-258. [13] W.D. Conway, Counter-current chromatography, J. Chromatogr. A 538 (1991) 27-35. [14] S. Li, W. Wang, H. Tang, K. Chen, J. Yang, L. He, H. Ye, A. Peng, L. Chen, Comparison of counter-current chromatography and preparative high performance liquid chromatography applied to separating minor impurities in drug preparations, J. Chromatogr. A 1344 (2014) 51-58. [15] S. Tong, Y. Zheng, J. Yan, Application and comparison of high performance liquid chromatography and high speed counter-current chromatography in enantioseparation of (±)2-phenylpropionic acid, J. Chromatogr. A 1281 (2013) 79-86. [16] J. He, R. Tsao, R. Yang, T. Zhou, Purification of patulin from Penicillium expansum culture: high-speed counter-current chromatography (HSCCC) versus preparative highperformance liquid chromatography (prep-HPLC), Food Addit. Contam. A 26 (2009) 101107.
[17] M.-C. Menet, D. Thiebaut, Preparative purification of antibiotics for comparing hydrostatic and hydrodynamic mode counter-current chromatography and preparative highperformance liquid chromatography, J. Chromatogr. A 831 (1999) 203-216. [18] C. Zhao, C. He, Mass transfer and separation criteria for high-speed countercurrent chromatography, AIChE J. 57 (2011) 359-372. [19] T.A. Maryutina, S.N. Ignatova, I.A. Sutherland, Organic phase retention in CCC: Effect of temperature, tubing material, tubing bore, mobile phase flow, and the addition of an extraction agent, J. Liq. Chromatogr. Rel. Technol. 26 (2003) 1537-1550. [20] A. Degenhardt, M. Schwarz, P. Winterhalter, Y. Ito, Evaluation of different tubing geometries for high-speed counter-current chromatography, J. Chromatogr. A 922 (2001) 355-358. [21] Y. Ito, R. Clary, J. Powell, M. Knight, T.M. Finn, Improved spiral tube assembly for high-speed counter-current chromatography, J. Chromatogr. A 1216 (2009) 4193-4200. [22] Y. Yang, H.A. Aisa, Y. Ito, Flat-twisted tubing: Novel column design for spiral highspeed counter-current chromatography, J. Chromatogr. A 1216 (2009) 5265-5271. [23] Y. Ito, M. Knight, T.M. Finn, Spiral countercurrent chromatography, J. Chromatogr. Sci. 51 (2013) 726-738. [24] Q. Du, Z. Li, Y. Ito, Preparative separation of isoflavone components in soybeans using high-speed counter-current chromatography, J. Chromatogr. A, 923 (2001) 271-274. [25] Q. Du, P. Winterhalter, Y. Ito, Large convoluted tubing for scale‐up of slow rotary countercurrent chromatograph, J. Liq. Chromatogr. Rel. Technol., 26 (2003) 1991-2002. [26] Q. Du, P. Wu, Y. Ito, Low-speed rotary countercurrent chromatography using a convoluted multilayer helical tube for industrial separation, Anal. Chem., 72 (2000) 33633365.
[27] T. Esatbeyoglu, V. Wray, Isolation of dimeric, trimeric, tetrameric and pentameric procyanidins from unroasted cocoa beans (Theobroma cacao L.) using countercurrent chromatography, P. Winterhalter, Food Chem., 179 (2015) 278-289. [28] Y. Yang, H.A. Aisa, Y. Ito, Novel design for centrifugal countercurrent chromatography: I. Zigzag toroidal column, J. Liq. Chromatogr. Rel. Technol. 32 (2009) 2030-2042. [29] Y. Yang, H.A. Aisa, Y. Ito, Novel design for centrifugal countercurrent chromatography: II. Studies on novel geometries of zigzag toroidal tubing, J. Liq. Chromatogr. Rel. Technol. 33 (2010) 336-348. [30] Y. Ito, Experimental observations of the hydrodynamic behavior of solvent systems in high-speed counter-current chromatography, J. Chromatogr. A 301 (1984) 387-403. [31] M. Englert, W. Vetter, Solvent systems with n-hexane and/or cyclohexane in countercurrent chromatography - Physico-chemical parameters and their impact on the separation of alkyl hydroxybenzoates, J. Chromatogr. A 1342 (2014) 54-62. [32] S.N. Ignatova, T.A. Maryutina, B.Y. Spivakov, Effect of physicochemical properties of two-phase liquid systems on the retention of stationary phase in a CCC column, J. Liq. Chromatogr. Rel. Technol. 24 (2001) 1655-1668. [33] A. Berthod, Practical approach to high-speed counter-current chromatography, J. Chromatog. A 550 (1991) 677-693. [34] A. Berthod, T. Maryutina, B. Spivakov, O. Shpigun, I.A. Sutherland, Countercurrent chromatography in analytical chemistry (IUPAC technical report), Pure Appl. Chem. 81 (2009) 355-387. [35] Q. Du, C. Wu, G. Qian, P. Wu, Y. Ito, Relationship between the flow-rate of the mobile phase and retention of the stationary phase in counter-current chromatography, J. Chromator. A 835 (1999) 231-235. [36] J.B. Friesen, G.F. Pauli, G.U.E.S.S. - A generally useful estimate of solvent systems in CCC, J. Liq. Chromatogr. Relat. Technol. 28 (2005) 2777-2806.
[37] O. Bousquet, F. Le Goffic, Counter-current chromatographic separation of polyunsaturated fatty acids, J. Chromatogr. A 704 (1995) 211-216. [38] M. Liu, F. Yang, H. Shi, C.C. Akoh, L. Yu, Preparative separation of triterpene alcohol ferulates from rice bran oil using a high performance counter-current chromatography, Food Chem. 139 (2013) 919-924.
[39] W. Vetter, J. Kirres, P. Bendig, Bromination of 2-methoxydiphenyl ether to an average of tetrabrominated 2-methoxydiphenyl ethers, Chemosphere 84 (2011) 1117-1124. [40] D. Gu, Y. Yang, X. Xin, H.A. Aisa, Y. Ito, Novel design for centrifugal counter-current chromatography: VI. Ellipsoid column, J. Liq. Chromatogr. Rel. Technol. 38 (2015) 68-73 [41] K. Shinomiya, Y. Kabasawa, Y. Ito, Countercurrent chromatographic separation of sugars and their p-nitrophenyl derivatives by cross-axis coil planet centrifuge, J. Liq. Chromatogr. Rel. Technol. 22 (1999) 579-592. [42] Y. Ito, W.D. Conway, Experimental observations of the hydrodynamic behavior of solvent systems in high-speed counter-current chromatography: III. Effects of physical properties of the solvent systems and operating temperature on the distribution of two-phase solvent systems, J. Chromatogr. A 301 (1984) 405-414. [43] P.L. Wood, B. Jaber, I.A. Sutherland, A New hypothesis on the hydrodynamic distribution of the upper and lower phases in CCC, J. Liq. Chromatogr. Rel. Technol. 24 (2001) 1629-1654.
Captions to Figures
Fig. 1. a) Schematic design of the crimping tool for tubing modifications with two guide rollers and two crimping rollers. The crimping rollers are equipped with up to eight protuberances with 0.5 cm distance between them. By changing the protuberances different
tubing modifications can be realized. After insertion, the tubing is drawn through the guide and crimping rollers manually and is then crimped in a reproducible manner. The crimping tool also serves as a column holder for the subsequent winding of the modified tubing onto the holder hub. b) A photograph of the crimping tool without installed coil.
Fig. 2. Schematic images of the six different tubing modifications denoted as a) |P|, b) |S|, c) /P/, d) /P\, e) /S\ and f) \S/ investigated in CCC experiments in side view. The black arrows indicate the direction of the crimping to distinguish between P: parallel and S: staggered crimping | |: at 90 °; //: at 45 ° parallel and / \: at 45 ° anti-parallel.
Fig. 3. Stationary phase retention Sf at flow rates of 0.5, 1, 2, 3 and 4 mL min-1 determined for unmodified tubing U and tubing modifications |P|, |S|, /P/, /P\, /S\ and \S/ in the intermediate polarity n-hexane/tBME/methanol/water (5:2:5:3, v/v/v/v, HTMW) solvent Determination was conducted in head-to-tail (a) and tail-to-head mode (b).
system.
Fig. 4. CCC separation of six alkyl p-hydroxybenzoates (∼5 mg of each compound) in headto-tail mode with (a) unmodified tubing U and tubing modifications (b) |P|, (c) |S|, (d) /P/, (e) /P\, (f) /S\, and (g) \S/ with n-hexane/tBME/methanol/water (5:2:5:3, v/v/v/v) at a flow rate of 1 mL min-1. The rotation speed was 1000 ± 10 rpm and ELSD detection was conducted with nitrogen gas stream flow set to 1.2 mL min-1 and nebulizer temperature set to 40 °C. The alkyl hydroxybenzoates detected are (1) methyl p-hydroxybenzoate, (2) ethyl p-hydroxybenzoate, (3) isopropyl p-hydroxybenzoate, (4) propyl p-hydroxybenzoate, (5) isobutyl phydroxybenzoate and (6) butyl p-hydroxybenzoate. Changes in peak resolution and peak width are particularly noticeable for the first three eluting compounds (1-3) and are indicated by a blue background.
Fig. 5. Evaluation of peak widths wb obtained during the CCC separation of six alkyl phydroxybenzoates with a) methyl p-hydroxybenzoate, b) ethyl p-hydroxybenzoate, c) isopropyl p-hydroxybenzoate, d) propyl p-hydroxybenzoate, e) isobutyl p-hydroxybenzoate and
f)
butyl
p-hydroxybenzoate
using
the
intermediate
polarity
n-
hexane/tBME/methanol/water (5:2:5:3, v/v/v/v, HTMW) solvent system with the unmodified
tubing U and six tubing modifications |P|, |S|, /P/, /P\, /S\ and \S/ in head-to-tail mode at five different flow rates 0.5, 1, 2, 3 and 4 mL min-1.
Fig. 6. Resolution Rs obtained between the peaks of a) methyl p-hydroxybenzoate and ethyl phydroxybenzoate, b) ethyl p-hydroxybenzoate and isopropyl p-hydroxybenzoate, c) isopropyl p-hydroxybenzoate and propyl p-hydroxybenzoate, d) propyl p-hydroxybenzoate and isobutyl p-hydroxybenzoate and e) isobutyl p-hydroxybenzoate and butyl p-hydroxybenzoate using the intermediate polarity n-hexane/tBME/methanol/water (5:2:5:3, v/v/v/v, HTMW) solvent system with the unmodified tubing U and six tubing modifications |P|, |S|, /P/, /P\, /S\ and \S/ in head-to-tail mode at five different flow rates 0.5, 1, 2, 3 and 4 mL min-1.
Fig. 7. Tubing modification \S/ utilized for the CCC separation of six alkyl phydroxybenzoates (∼5 mg of each compound) in head-to-tail mode at five different flow rates 0.5, 1, 2, 3 and 4 mL min-1. For conditions and compound numbering see Fig. 4.
Fig. 8. Performance of the alkyl p-hydroxybenzoate separation in terms of peak width wb and number of theoretical plates N with unmodified tubing U (a,c) and tubing modification \S/
(b,d) at five different flow rates 0.5, 1, 2, 3 and 4 mL min-1 and a fixed stationary phase retention Sf=0.56.
Fig. 9. Stationary phase retention Sf at five different flow rates 0.5, 1, 2, 3 and 4 mL min-1 in head-to-tail and tail-to-head mode determined for unmodified tubing U and tubing modifications |P|, |S|, /P/, /P\, /S\ and \S/ (a,b) in solvent systems HACN which consisted of nhexane/acetonitrile (1:1, v/v) and (c,d) BAW which consisted of n-butanol/acetic acid/water (5:1:4, v/v/v).
Table 1. Physico-chemical properties of the two-phase solvent systems used for measuring the stationary phase retention of different tubing modifications.
Solvent
Viscosity
Mean
Density
Density
Interfacial Settling
system
(c.p.)
viscosity
(g cm-3)
difference
Tension
time
ηUP/ηLP
(c.p.)
ρUP/ρUP
(g cm-3)
(mN m-1)
(s)
Δρ
σ
ts
ηM HACNa
0.32/0.30
0.31
0.67/0.77
0.10
1.10
9
HTMWb
0.30/0.68
0.49
0.66/0.91
0.25
1.14
8
BAWc
2.49/1.52
2.01
0.89/0.99
0.10
1.22
29
a
HACN: n-hexane/acetonitrile (1:1, v/v)
b
HTMW: n-hexane/tBME/methanol/water (5:2:5:3, v/v/v/v)
c
BAW: n-butanol/acetic acid/water (4:1:5, v/v)
Table 2. Partitioning coefficients K with selectivity factor α of the six alkyl phydroxybenzoates in the intermediate solvent system (HTMW) nhexane/tBME/methanol/water (5:2:5:3, v/v/v/v) and concentration of the compounds in the sample solution used in CCC.
Compound
Abbreviation
Methyl p-
Partitioning
Selectivity
Sample solution
coefficient
factor
concentration
K [27]
α
[mg L-1]
0.08
-
MePHB
254
hydroxybenzoate Ethyl p-
0.17
2.13
EtPHB
252
hydroxybenzoate Isopropyl p-
0.28
1.65
iPrPHB
251
hydroxybenzoate Propyl p-
0.50
1.79
PrPHB
252
hydroxybenzoate Isobutyl p-
0.73
1.46
iBuPHB
251
hydroxybenzoate Butyl p-
0.99 BuPHB
hydroxybenzoate
1.36 251
Table 3. Percentage of decrease of peak widths wb calculated between the unmodified tubing U and tubing modification \S/of six alkyl p-hydroxybenzoates using the intermediate polarity n-hexane/tBME/methanol/water (5:2:5:3, v/v/v/v, HTMW) solvent system.
Compound Flow rate MePHB
EtPHB
iPrPHB
PrPHB
iBuPHB
BuPHB
0.5
6.21
9.70
17.5
20.9
25.1
28.5
1
9.73
19.1
24.0
29.9
34.8
39.0
2
17.4
28.6
32.3
39.2
41.8
41.4
3
22.2
33.3
36.9
41.9
47.2
44.0
4
26.6
35.4
43.9
44.4
48.0
49.3
-1
[mL min ]
