HUNGARIAN JOURNAL OF INDUSTRIAL CHEMISTRY VESZPRÉM Vol. 33(1-2). pp. 23-30. (2005) SIMULATED MOVING BED (SMB) SEPARATION OF PHARMACEUTICAL ENANTIOMERS G. GÁL1*, L. HANÁK1, J. ARGYELÁN1, A. STRBKA1 and T. SZÁNYA1 A. ARANYI2, K. TEMESVÁRI2 1University of Veszprém, Department of Chemical Engineering Processes 8201 Veszprém, P.O.Box 158, HUNGARY, *E-mail: gaborgal@freemail.hu 2Gedeon Richter LTD., 1475 Budapest P.O.Box. 27, HUNGARY Authors investigated the separation of chiral racemic mixture - pharmaceutical enantiomers - in a laboratory scale SMB equipment. They chose the proper chiral chromatographic packing and eluent for the separation of SMB with the help of previous analytical HPLC investigation. This was followed by frontal adsorption- elution measurements on the chosen Chiralcel OD packing (20µm particle size) in n-hexane-IPA eluent at 20°C. The measurements of adsorption equilibrium (k’ values) and NTP, HETP data were carried out with laboratory scale preparative HPLC equipment (L=25 cm, ID=1 cm). The parameter planning, previous estimation of SMB operation was carried out with KROM-N and SMB-KROM-N simulation programs. This was followed by the experimental study of the SMB preparative liquid chromatographic operation (laboratory scale open loop eluent circle 4 column (L=25 cm, ID=1 cm) equipment), and the comparison of the mathematical simulation with the results achieved in practice. The prescribed 99,9 % m/m purity for the „S” component of raffinate can be reached. At the optimum experiment the yield for „S” was over 99 %, the productivity was 62 mg S g-1 packing day-1 and the eluent consumption was 5,4 mL eluent mg-1 S. Keywords: preparative liquid chromatography, simulated moving bed chromatography, pharmaceutical enantiomers, chiral chromatographic packing Introduction Nowadays almost half of the registered pharmaceutical products have chiral structure, accordingly they are of importance in pharmaceutical industry [1]. Pharmacologically, most often only one optical isomer has proper activity, while the other one is inactive, possibly toxic. As a consequence the optical purity of enantiomers have got significant importance. As the enantiomers have the same physicochemical features and they show different characters only in optically active surroundings, the separation of them is unachievable without chiral interactions. Chiral stationary phases can be produced by chiral selectors modifying the next natural materials (proteins, cyclodextrins, saccharides, antibiotics), synthetic polimers, and small, completely synthetic chiral compounds. Nowadays the production and separation of enantiomer can be done by asymmetric catalysis, biotransformation, liquid-liquid extraction, capillary electrophoresis, membrane separation, crystallization, chromatography, and within chromatography: with capillary electrochromatography (CEC), supercritical fluid chromatography (SCF), gas chromatography (GC) and liquid chromatography (LC). That is thin-layer chromatography (TLC), countercurrent chromatography (CCC) and simulated moving bed chromatography (SMB). The advantage of the SMB method is that the procedure can be made continuously, the columns are completely used, the productivity and yield is higher, and the consumption of the eluent is lower compared to the batch chromatographic process. The disadvantage is the high investment cost, and the considerable sensitivity of the operational parameters. In the traditional batch elution chromatography the sample is injected on the top of the column and the components get separated after a certain time by moving through the column forced by the mobile phase [2]. However, this process is not very efficient as during the chromatography only a small part of the whole stationary phase is used for separation. A possibility to improve the packing utilization, is given by the true moving bed chromatography (TMB) principle. This chromatographic process was first introduced in the late 1960s by the Universal Oil Product Company [3-5] and mailto:gaborgal@freemail.hu 24 was intensively investigated by Ching and co-workers [6-11]. According to this concept, not only the liquid phase is moving but the solid phase as well (TMB, Fig. 1). For example, at a given counter current liquid phase and solid phase velocities, the faster eluting compound (Raffinate, „B”) moves forward in the liquid direction and the slower eluting compound (Extract, „A”) moves in the opposite direction together with the solid phase. Under this condition, the full mass of the solid phase contributing to the separation is continuosly used, thus improving considerably the productivity of the system. Obviously, this principle is particularly suitable for a binary mixture, especially for racemates (optical isomer mixtures). However, it is technically difficult to move a solid phase, thus the solid phase movement is simulated (SMB, Fig 2). In fact, it is a continuous process, the system consists of a number of small columns arranged in series in practice, as shown schematically in Fig 2. As the inlet and outlet points are regularly changed (switching time), the net result is the same as it would be if the stationary phase were moving. In the last decades simulated moving bed processes were widely spread in preparative chromatography on the field of high-purity materials production in the pharmaceutical, fine chemical and biotechnology industries. If materials are difficult to be separated (α ≈ 1) the process efficiency is disadvantegous. Otherwise if a component of a mixture was strongly adsorbed, the process would become uneconomical because a large volume of eluent should be used. Experimental Examination of Adsorbent Packings During examination Chiralcel OD-H, Chiralcel OJ, Chiralpak AD, Chiralpak IA, Chiralpak AS chiral packings were compared. Selection of chiral chromatographic packings and eluents The selection of the chiral stationary packings (CSP) and eluents carried out by a reference book published by DAICEL. These materials are coated or immobilised polysaccharide-derived CSPs. The chiral stationary packings tested by analytical HPLC are the follows (Fig. 3): • Chiralcel OD-H: Cellulose tris(3,5-dimethyl- phenylcarbamate), coated on a 5 µm silica support, particle size 5 µm, • Chiralcel OD: Cellulose tris(3,5-dimethyl- phenylcarbamate), coated on a 20 µm silica support, particle size 5 µm, • Chiralcel OJ: Cellulose tris(4-methylbenzoate), coated on a 20 µm silica support, particle size 20 µm, • Chiralpak AD: Amylose tris(3,5-dimethyl- phenylcarbamate) coated on a 20 µm silica support, particle size 20 µm, • Chiralpak IA: Amylose tris(3,5-dimethyl- phenylcarbamate) immobilized on a 5 µm silica support, particle size 5 µm, • Chiralpak AS: Amylose tris[(S)-phenylethyl- carbamate] coated on a 5 µm silica support, particle size 5 µm. Fig. 1: Column disposition in the true moving bed chromatography (TMB) Fig. 2: Column disposition in the simulated moving bed chromatography (SMB), short columns in series 25 Fig. 3: Chiral packings tested by analitycal HPLC The measurements were done by GILSON type analytical HPLC. Detection was carried out at 254 nm UV wavelength, at 20°C. As can be seen in Table 1, the n-hexane-IPA eluent and the Chiralcel-OD chiral stationary packing proved to be the most appropriate according to the selectivity ( ), so this system was chosen for further examination. R Sαα = Determination of the Number of Theoretical Plates (NTP) – Height of Equivalent Theoretical Plate (HETP) and Selectivity ( ) R Sα The definition of the equilibrium data of „S”(B) and „R”(A) optical isomers was carried out on SUPELCO product preparative HPLC column (I.D. = 1 cm, L = 25 cm), filled with Chiralcel OD (particle size: 20µm) packing, with the aid of a column-packing vibrator, at 20°C. Air was removed from the column by means of 95:5 (v/v) n-hexane-IPA and an LMIM D-167 pump. A Rheodyne injection valve with 100µL loop was connected to the column inlet and a Waters UV detector to the outlet, where we monitored the signal of optical isomers. The elution residence-time curve was recorded by UV spectrophotometer (λ = 254 nm). The sample was 541/BK chiral racemic mixture soluted in eluent in 50 mg mL-1 concentration, from which 100µL was injected into the column. The eluent was n-hexane-IPA used in three different volumetric ratio with five different volume flow rates. The results of the measurements are shown in Table 2. The residence time curve was evaluated by the triangulation method determining σ and tR values: 0 0' t tt k R − = S R R RR S k k tt tt S R ' ' 0 0 = − − =α 2 R σ t ⎟ ⎠ ⎞ ⎜ ⎝ ⎛=NTP NTP L =HETP Frontal Adsorption - Elution Measurements The Langmuir constants and the data of the adsorption isotherm were calculated from the k’ values: ε ε − = 1 ' )()( SBSB kK ε ε − = 1 ' )()( RARA kK columnmL liquidmL 67.0=ε columnmL packingg BULK 6.0=ρ BULK SBSB Ka ρ ε− = 1 )()( BULK RARA Ka ρ ε− = 1 )()( )()( RASB bb = (computative data) packingg volumefreeliquidmL a 3366.5*B = componentBmg volumefreeliquidmL b 016.0B = packingg volumefreeliquidmL a 4405.6*A = componentAmg volumefreeliquidmL b 016.0A = Thus enough information was assembled to do the computer simulations (with KROM-N Software) of the frontal adsorption-elution. The data input to the software is shown on Table 3. Hereby there was a possibility to compare the simulation and the laboratory measurements. The laboratory measurements were performed with the column used before. Air was removed from the column by means of 95:5 (v/v) n- hexane-IPA at 20°C, directed downwards by means of an LMIM-D167 piston pump. During the frontal adsorption a 10 mL of 50:50 (m/m) mixture of chiral mixtures S+R (total conc. 2-5-10 g L-1) was applied downwards to the column at a flow rate 2-2,5-5 mL min-1 at 20°C. Feeding of the mixture S and R was stopped after 10 mL and pure eluent (95:5 and 93:7 n- hexane-IPA) was pumped into the column at a flow rate of 2-2,5-5 mL min-1 (Table 4). The eluent was collected in sample collectors and concentration analysed on-line by UV spectrophotometer. The concentration of the given samples were measured by analytical HPLC. The frontal adsorption-desorption simulation and laboratory measurement results are shown in (Fig. 4.). It’s remarkable that the separation of “S” and “R” optical isomers are favourable at small total contrantrations compared to the high ones. 26 The estimation of the SMB measurements with computer simulations The simulations were calculated by SMB-KROM-N software. The model of the software uses the physical and chemical data of chemicals, the number of theoretical plates (NTP), volumetric flow rates, adsorption equilibrium data, switching time, etc. published by Morbidelli [12] and his partners. The software of the simulation solves the differential equations by the numerical method of finite differencies [13]. Determination of Morbidelli Parameters On the bases of a theoretical method assuming independent adsorption and linear isotherms, published by Massimo Morbidelli [12] and his team, the right values of volumetric stream can be well estimated. Values of distribution quotient: silicagelsolidmL liquidvolumefreemL 703.9 )(1 H * B B =− ⋅ = ε ρa K silicagelmL liquidvolumefreemL 3366.5* =Ba silicagelsolidmL liquidvolumefreemL 71.11 )1( H * A = − ⋅ = ε ρa K A silicagelmL liquidvolumefreemL 4405.6* =Aa By the Morbidelli criteria the next relations must be true for producing pure „S” (B component) and „R” (A component) isomers: 11.71 = KA < mI 9.703 = KB < mII < KA = 11.71 9.703 = KB < mIII < KA = 11.71 mIV < KB = 9.703 The following parameters were chosen for the purpose of measurement, because these parameters accomplish Morbidelli criteria providing adequate results during simulations. Further data can be seen in the next chapter. F = 0.3 mL min-1 E = 4.4 mL min-1 R = 4.4 mL min-1 S = 12 mL min-1 LR OUT = 3.5 mL min-1 D = S + REC = 12 mL min-1 T = 10 min (switching time) L = 25 cm (column length) 4 πD A 2 f = Values of Morbidelli parameters: 7108.114989.16 )1( >= − − = ε ε L LT A D m fI 7108.117049.9 )1( 7028.9 <= − − − =< ε ε L LT A ED m fII 7108.111681.10 )1( 7028.9 <= − − +− =< ε ε L LT A FED m fIII 7108.111681.10 )1( 7028.9 <= − − +− =< ε ε L LT A FED m fIII Data input to the SMB-KROM-N software Number of components: k = 2 Column inner diameter: I.D. = 1 cm Column length: L = 25 cm Number of columns: N = 4 Free volume coefficient: EPS = 0.67 mL liquid free volume mL-1 column Bulk density: ρBULK=0.6 g packing mL -1 column Feed: F = 0.3 mL min-1 Fresh eluent: S = 12 mL min-1 Extract: E = 4.4 mL min-1 Raffinate: R = 4.4 mL min-1 Recycling: REC = 0 mL min-1 Langmuir constants: as given above Feed concentration: mg B component mL 5.2FB =c -1 liquid 5.2FA =c mg A component mL-1 liquid Number of Theoretical Plates: NTP = 200 per 25 cm column Switching time: 10 min Calculation time: 400 min SMB Equipment Planning, Construction and Installation The SMB preparative liquid chromatographic equipment with four columns, four sectors and open eluent circle was constructed in the Central Mechanical Workshop of the University of Veszprém (Figs. 5 and 6). During installation the four preparative liquid chromatographic columns (I.D. = 10 mm, L = 250 mm) were filled with Chiralcel-OD packing by the vibration method (~30 min filling time). The column packing density was 0.6 g mL-1, the free volume factor was 0.67. Each column was filled with approximately 11.77 g packing. Stainless steel frits (2µm) were placed at the top and the bottom of each column. Before measurements air was removed with eluent. 27 Fig. 5: Photograph of the SMB equipment Fig. 6: The block diagram of the SMB equipment with four columns, four sectors, and open eluent cycle SMB measurements Out of the 22 executed simulations the best ones were chosen, according to which four measurements were done with the laboratory scale 1:1:1:1 column configuration, open eluent circle SMB equipment of the Department of Chemical Engineering Processes. The conditions and the results of the SMB measurements are included in Table 5 and in the graphs (Fig.7). Results Measurement Results of SMB Compared to Simulation Results Columns of the four-column SMB equipment, previously equilibrated at 20°C with 95:5 (v/v) n- hexane-IPA as eluent, were used to separate a racemic mixture of 2.5 g L-1 isomer A (R) and 2.5 g L-1 isomer B (S) in the same solvent mixture. The mixture to be separated was fed at the top of column III. up to 10 min at a flow rate of 0.3 mL min-1. (The other volumetric ratios: E = 4.4 mL min-1, R = 4,4 mL min-1, fresh eluent: 12 mL min-1). It was followed by switching columns were exchanged according to the SMB process and the eluent was not recirculated. Flow rates were controlled by digital balances with the help of computer during the 10 min switching time. Fig. 7 shows the eluent consumption, productivity, purity and yield for component “S” in the raffinate fraction at quasi-stationary state. Our conclusion on the bases of laboratory measurements is that the SMB-KROM-N software is very adaptable to optimalize SMB operation. The prescribed 99,9 % m/m purity for the „S” component of raffinate can be reached. At the optimum experiment the yield for „S” was over 99 %, the productivity was 62 mg „S” g-1 packing day-1 and the eluent consumption was 5,4 mL eluent mg-1 „S”. With the help of experimental and theoretical optimization of SMB process (switching time decrease, feed concentration increase, volumetric flow-rate change, gradient methods application, column number increase, column configuration change etc.) markedly can be improve the specific values of the SMB process. The investigation is in progressive state at the University of Veszprém and at the Gedeon Richter LTD. Acknowledgement The authors express their gratitude to Chemical Engineering Institute Cooperative Research Center of the University of Veszprém and the Gedeon Richter LTD. for financial support of this research study. 28 RG OD FR 01 0,0 0,5 1,0 1,5 2,0 2,5 0 20 40 60 80 100 120 140 V (m L) c (m g m L -1 ) S measured R measured S calculated R calculated Eluent: n-hexane:IPA=95:5 (v/v ) csample= 5 mg mL -1 F= 5 mL min-1 RG OD FR 02 0,0 0,5 1,0 1,5 2,0 2,5 0 20 40 60 80 100 120 140 V (m L) c (m g m L -1 ) S measured R measured Eluent: n-hexane:IPA=95:5 (v/v ) csample= 5 mg mL -1 F= 2 mL min-1 RG OD FR 05 0,0 0,5 1,0 1,5 2,0 2,5 0 20 40 60 80 100 120 140 V (m L) c (m g m L -1 ) S measured R measured Eluent: n-hexane:IPA=95:5 (v/v ) csample= 2 mg mL -1 F= 2.5 mL min-1 RG OD FR 07 0,0 1,0 2,0 3,0 4,0 5,0 0 20 40 60 80 100 120 140 V (m L) c (m g m L -1 ) S measured R measured Eluent: n-hexane:IPA=95:5 (v/v ) csample= 10 mg mL -1 F = 2.5 mL min-1 RG OD FR 03 0 1 2 3 4 5 6 7 0 20 40 60 80 100 120 140 V (m L) c (m g m L -1 ) S measured R measured S calculated R calculated Eluent: n-hexane:IPA=93:7 (v/v ) csample= 10 mg mL -1 F= 2.5 mL min-1 RG OD FR 04 0 1 2 3 4 5 6 7 0 20 40 60 80 100 120 14 V (m L) c (m g m L -1 ) 0 S measured R measured Eluent: n-hexane:IPA=93:7 (v/v ) cs ample= 5 mg mL -1 F= 5 mL min-1 RG OD FR 06 0,0 0,5 1,0 1,5 2,0 0 20 40 60 80 100 120 140 V (m L) c (m g m L -1 ) S measured R measured Eluent: n-hexane:IPA=93:7 (v/v) csample= 2 mg mL -1 F= 2.5 mL min-1 Fig. 4: The results of frontal adsorption-desorption simulation and laboratory measurements 29 0 1 2 3 4 5 6 7 8 9 10 RG OD SMB 01 RG OD SMB 02 RG OD SMB 03 RG OD SMB 04 E lu en t c on su m pt io n (m L el ue nt m g- 1 S ) R calculated R measured 0 10 20 30 40 50 60 70 RG OD SMB 01 RG OD SMB 02 RG OD SMB 03 RG OD SMB 04 P ro du ct iv ity (m g S g -1 p ac ki ng d ay -1 ) R calculated R measured 0 10 20 30 40 50 60 70 80 90 100 RG OD SMB 01 RG OD SMB 02 RG OD SMB 03 RG OD SMB 04 P ur ity (% ) R calculated R measured 0 10 20 30 40 50 60 70 80 90 100 RG OD SMB 01 RG OD SMB 02 RG OD SMB 03 RG OD SMB 04 Y ie ld (% ) R calculated R measured Fig. 7: Results of 4 measurements and 1 simulation: eluent consumption, productivity, purity and yield for component “S” in the raffinate Concentration [% v/v ] Chiralcel OD-H Chiralcel OJ Chiralpak AS Chiralpak AD Chiralpak IA n-hexane:IPA 70:30 α=1 n-hexane:IPA 80:20 α=1.17 α=1.196 n-hexane:IPA 90:10 α=1.173 α=1 α=1 n-hexane:IPA 97.5:2.5 α=1.06 n-hexane:IPA 95:5 α=1.19 α=1 α=1.03 n-hexane:Et-OH 95:5 α=1.122 n-hexane:IPA:AcN 80:10:10 α=1 AcN 100 α=1.31 α=1 AcN-Me-OH 80:20 α=1 α=1 Ethanol 100 α=1 n-hexane:MTBÉ 80:20 k'1,2>20 n-hexane:MTBÉ 60:40 k'1,2=6.10 n-hexane:MTBE:Et-OH (60:40)+5% Et-OH k'1,2=1.086 n-hexane:IPA α=1.08 n-hexane:Met-OH 99:1 α=1.12 n-hexane:Et-OH 99:1 α=1.11 n-hexane:DKM 75:25 α=1.03 n-hexane:IPA:EtOH 95:2.5:2.5 α = 1.15 n-hexane:IPA:MetOH 95:2.5:2.5 α = 1.05 n-hexane:EtOH:MetOH 95:2.5:2.5 α = 1.05 PackingEluent Table 1: HPLC measurements on different charges Eluent: n-hexane:IPA=95:5 [% v/v] B t0 4σ HETP 4σ HETP Pressure Pressure [mL min-1] [sec] tR,1 k'1 [s] [mm] tR,2 k'2 [s] [mm] [psi] [bar] 2.5 316 1902 5.027 378 405 0.617 2 532 7.023 510 394 0.634 1.397 42 3.0 5 158 912 4.779 192 361 0.693 1068 5.768 264 262 0.955 1.207 66 4.6 10 79 468 4.932 108 300 0.832 576 6.300 150 236 1.060 1.278 110 7.7 15 53 306 4.817 78 246 1.015 378 6.186 108 196 1.276 1.284 168 11.8 20 39 228 4.779 66 191 1.309 282 6.148 90 157 1.592 1.286 240 16.9 30 26 168 5.388 54 155 1.614 210 6.985 90 87 2.870 1.296 336 23.6 Eluent: n-hexane:IPA=90:10 [% v/v] B t0 4σ HETP 4σ HETP Pressure Pressure [mL min-1] [sec] tR,1 k'1 [s] [mm] tR,2 k'2 [s] [mm] [psi] [bar] 2.5 316 1248 2.954 258 374 0.668 1 530 3.848 336 332 0.754 1.302 30 2.1 5 158 612 2.878 126 377 0.662 732 3.639 192 233 1.075 1.264 48 3.4 10 79 303 2.840 72 283 0.882 366 3.639 102 206 1.214 1.281 100 7.0 15 53 204 2.878 56 212 1.177 242 3.601 80 146 1.708 1.251 144 10.1 20 39 152 2.853 48 160 1.558 182 3.613 72 102 2.445 1.267 210 14.8 30 26 102 2.878 40 104 2.403 120 3.563 72 44 5.625 1.238 336 23.6 Eluent: n-hexane:IPA=80:20 [% v/v] B t0 4σ HETP 4σ HETP Pressure Pressure [mL min-1] [sec] tR,1 k'1 [s] [mm] tR,2 k'2 [s] [mm] [psi] [bar] 2.5 316 1008 2.194 210 369 0.678 1 218 2.859 249 383 0.653 1.303 0 0.0 5 158 492 2.118 102 372 0.672 588 2.726 138 290 0.861 1.287 0 0.0 10 79 242 2.067 58 279 0.898 288 2.650 78 218 1.146 1.282 54 3.8 15 53 162 2.080 46 198 1.260 192 2.650 58 175 1.426 1.274 108 7.6 20 39 120 2.042 40 144 1.736 142 2.599 58 96 2.607 1.273 168 11.8 30 26 81 2.080 35 88 2.835 94 2.574 40 88 2.829 1.238 306 21.5 NTP S (-) R (+) α (k'+/k'-) α (k'+/k'-) α (k'+/k'-)NTP NTP NTP NTP NTP S (-) R (+) S (-) R (+) Table 2: The parameters and the results of the elution measurements 30 Number of components: k = 2 Column inner diameter: I.D. = 1 cm Column length: L = 25 cm Free volume coefficient: EPS = 0.67 mL liquid free volume mL-1 column Bulk density: ρB= 0.6 g packing mL -1 column Feed: F = 2-2.5-5 mL min-1 Langmuir constants: as given above Sample feeding volume: 10 mL Sample concentration: mg B component mL 1052FB −−=c -1 liquid 1052FA −−=c mg A component mL-1 liquid Number of Theoretical Plates: NTP = 200 Calculation time: 70 min Table 3: Data input for the KROM-N software Identifier Feed [mL min -1] csample [mg mL -1] Eluent, (v/v) n-hexane-IPA RG OD FR 05 2.5 2 95:5 RG OD FR 02 2 5 95:5 RG OD FR 01 5 5 95:5 RG OD FR 07 2.5 10 95:5 RG OD FR 06 2.5 2 93:7 RG OD FR 04 5 5 93:7 RG OD FR 03 2.5 10 93:7 Table 4: The parameters of the frontal adsorption- elution measurements Identifier Eluent Sample Switching time (min) D E F R LR OUT simulation Raffinate cS (mg mL-1) 0,0468 0,5702 Purity (%) 99,66 100,00 Yield (%) 99,95 100,00 4,22 2,94 7,5 15,76 6,59 0,50 6,01 3,64 10 11,93 "S" isomer Productivity (mg S g-1 packing day-1) (mL min-1) Eluant consumption (mL eluent mg-1 S) RG OD SMB 01 10 11,55 4,29 0,23 3,59 3,72 RG OD SMB 02 RG OD SMB 03 RG OD SMB 04 RG OD SMB 01-04 measurements n-hexane:IPA = 93:7 (v/v) 5,23 5,39 0,3442 10 12,01 4,29 0,30 5,06 0,33 57,50 100,00 100,00 5 g racemic mL-1 in n-hexane:IPA = 95:5 (v/v) 5 g racemic mL-1 in n-hexane:IPA = 93:7 (v/v) 59,00 61,96 34,98 3,91 3,72 RaffinateRaffinate 0,2983 9,48 7,62 100,00 100,00 Raffinate 0,3477 9,25 34,69 100,00 100,00 Table 5: The conditions and the results of the SMB measurements SYMBOLS ε – free volume coefficient [mL liquid free volume mL-1 column] F – volume flow rate of eluent [mL min-1] t0 – dead time [min] tR – retention time [min] k’ – retention factor S(–) – the „S”, L-isomer, bonds weakly to adsorbent (B) R(+) – az „R”, D-isomer, bonds stronger to adsor- bent (A) R Sα – separation factor NTP – number of theoretical plates HETP – height of theoretical plates [mm] REFERENCES 1. ILONA HORVATH, SANDOR KOVACS: Days of Technical Chemistry (2004) 91-94 2. ERIC R. FRANCOTTE, PAUL RICHERT: J.Chromatogr A, (1997) 769:101-107 3. D.B. BRAUGHTON AND C.G GERHALD: US Pat, (1961) 2985589 4. D.B. BRAUGHTON AND S.A. GEMBICKI: A.I.Ch.E. Symp.Ser. 5. P.E. BAKER AND G.GANESTOS: Sep. Sci. Technol., 22 (1987) 2011 6. C.B. CHING, B.G. LIM, E.J.D. LEE AND S.C. NG: J. Chromatogr, 634 (1993) 215 7. B.G. LIM, C.B. CHING, R.B.H. TAN AND S.C. NG: Chem. Eng. Sci., 50 (1995) 2289 8. C.B. CHING, D.M. RUTHVENAND K. HIDAJAT: Chem. Eng. Sci., 40 (1985) 141 9. C.B. CHING, C. HO AND D.M. RUTHVEN: Chem. Eng. Sci., 43 (1988) 703 10. C.B. CHING, K.H. CHU.,K. HIDAJAT AND D.M. RUTHVEN: Chem. Eng. Sci., 48 (1993) 1343 11. C.B. CHING AND D.M. RUTHVEN: A.I.Ch.E. Symp.Ser. 38 (1992) 1744 12. MIGLIORINI C, MAZZOTTI M, MORBIDELLI M.: J Chromatogr A, (1998) 827:2:161-173 13. M. MIRON, T. SZÁNYA, L. HANÁK, J. ARGYELÁN AND GY. MARTON: Hung. J. Ind. Chem. (1995) 23:293-298