A

Emwsp©

K

1

hsccolewua[timelth

0d

Journal of Chromatography A, 1151 (2007) 193–196

Separation of solanesol in tobacco leaves extract by slow rotarycounter-current chromatography using a novel non-aqueous

two-phase solvent system

Yang Zhao, Qizhen Du ∗Institute of Food and Biological Engineering, Zhejiang Gongshang University, Hangzhou 310035, China

Available online 7 January 2007

bstract

Non-aqueous solvent system composed of sunflower oil–ethanol was found to be suitable for the separation of lower polarity components.mploying this solvent system, 3 g of tobacco leaves extract containing 15% solanesol was separated using a slow rotary counter-current chro-atograph equipped with an 1100 mL column made of 5.7 mm I.D. convoluted PTFE tube. The separation yielded 1.5 g of yellow oily product in

hich solanesol occupied 26.8% determined by HPLC. The tobacco tar and other main impurities in the tobacco leaves extract were removed after

low rotary counter-current chromatographic separation. The separation processing may be developed as a technique for producing a solanesolroduct with food grade.

2007 Elsevier B.V. All rights reserved.

graph

v(ltbHpttenowabg

2

eywords: Non-aqueous solvent system; Slow rotary counter-current chromato

. Introduction

The very low polar solanesol is a non-cyclic terpene alco-ol possessing nine isoprene units, which is the key material ofynthesizing coenzyme Q and vitamin K [1]. Solanesol itselfan be used as a food additive for preventing cardiac and can-er [2,3]. Solanesol exists in tobacco leaves with an amountf 0.3–3% [4]. Commonly, the content of solanesol in ethano-ic extract of tobacco leaves is about 15%. But the ethanolicxtract contains pigments, tobacco tar and other impuritieshich hinder it to be used as a food additive. Traditional col-mn chromatography is the main method for the separationnd purification of solanesol with industrial organic solvents5,6]. But for food grade products, everybody hope no indus-rial organic solvents are used in the processing. Therefore, its significant to establish a processing that can remove the pig-

ents, tobacco tar and other main impurities in the ethanolicxtract of tobacco leaves. Counter-current chromatography is a

iquid–liquid chromatographic technology. Du et al. [7] reportedhat solanesol in tobacco leaves extract could be separated byigh-speed counter-current chromatography (HSCCC) with sol-

∗ Corresponding author. Tel.: +86 571 8821 8710; fax: +86 571 8882 3509.E-mail address: qizhendu@163.com (Q. Du).

2

cPm

021-9673/$ – see front matter © 2007 Elsevier B.V. All rights reserved.oi:10.1016/j.chroma.2007.01.009

y; Solanesol; Tobacco leaves

ent system composed of light petroleum–ethanol–methanol200:1:100, v/v/v). But light petroleum and methanol are dis-iked in food industry. In addition, to scale up HSCCC columno more than 10 L for industrial separation is not so easyecause of the high rotation speed and planetary run mode ofSCCC. However, slow rotary counter-current chromatogra-hy (SRCCC) column may be scaled up to the level to meethe requirement of industrial separation [8]. Thus, it is possibleo achieve industrial separation of solanesol in tobacco leavesxtract by using SRCCC. In this paper, we introduce a novelon-aqueous two-phase solvent system composed of sunfloweril and ethanol to separate solanesol in tobacco leaves extractith SRCCC [9,10]. The method may be used for industrial sep-

ration of food grade solanesol after the instrument is scaled upecause the solvents for the solvent system of SRCCC are foodrade.

. Experimental

.1. Materials

The tobacco leaves extract used for separation of solanesolontained about 15% of solanesol, which was provided by Threeower Group, Weifang, China. The extract was produced byacerating tobacco leaves with ethanol. Sunflower oil was of

1 atogr

esC

2

(ag

2

osbaltct

2

iC45hw5lsftIsltsptmmTf

2

p(Ut

a2

3

mpfetertboflrflsbfT62a3tjawl

efacistfaboHSt9pitae

94 Y. Zhao, Q. Du / J. Chrom

dible grade, and ethanol was of analytical grade. Solanesoltandard was purchased from Sigma–Aldrich, Shanghai,hina.

.2. TLC analysis

TLC analysis [11] was performed on GF254 aluminum platesMerck, Darmstadt, Germany), developed with n-hexane–ethylcetate (4:1), and colorized with sulfuric acid–anisaldehyde–lacial acetic acid (5:5:90, v/v/v) at 110 ◦C.

.3. Selection of solvent system

The partition effect estimate of solanesol between two phasesf a solvent system was used for initial selection of a solventystem that was used for SRCCC separation. The estimate wasased on TLC analysis. First, a little tobacco leaves extract wasdded into a tube containing equal volumes of upper phase andower phase of a given solvent system. Then the tube was shakeno make a thorough mixture of the two phases. The solanesoloncentration in the two phases was analyzed by TLC to estimatehe partition coefficient after layering the two phases.

.4. Isolation of solansesol with SRCCC

A seal-free SRCCC instrument (Institute of Food and Biolog-cal Engineering, Zhejiang Gongshang University, Hangzhou,hina) [9] equipped with one layer coil column made of4 m convoluted polytetrafluoroethylene (PTFE) tubing with a.7 mm average I.D. was used in the present study. The coil hubolder was 60 cm wide and 17 cm O.D. The column capacityas 1100 mL. The mobile phase was delivered using a Waters90 HPLC pump (Millipore, Milford, MA, USA). An injectionoop was used for sample loading. The selected non-aqueousolvent system composed of sunflower oil–ethanol was appliedor the SRCCC separation with the oil upper phase as the sta-ionary phase and the ethanolic lower phase as mobile phase.n each separation, the SRCCC column was first filled withtationary phase. Then the sample solution, 3 g of the tobaccoeaves extract in 50 mL of the upper phase, was injected intohe column through the injection loop when the column rotationpeed was adjusted to 100 rpm. After the injection was com-leted, the lower phase was delivered into the column to elutehe components. The effluent was collected with a FCS-300

ode fraction collector (Shanghai Kezhe Biochemical Instru-ent, Shanghai, China), collecting 50 mL in each fraction tube.he fractions were analyzed by TLC to identify the solanesol

ractions.

.5. HPLC analysis

The HPLC system was composed of a Waters 510

ump, a manual injector, a YMC-Pack ODS-AQ column5 �m, 250 mm × 4.6 mm I.D., Kyoto, Japan), a Waters 480V variable-wavelength monitor and a data processing sys-

em. The separation was performed at 30 ◦C eluting with

trgt

. A 1151 (2007) 193–196

cetonitrile–isopropanol (65:35). The detection wavelength was13 nm [12].

. Results and discussion

The solvent system is crucial to counter-current chro-atographic separation. Solvent system composed of light

etroleum–ethanol–methanol (200:1:200, v/v/v) was success-ully used for the separation of solanesol in tobacco leavesxtract by HSCCC [7]. But that solvent system showed lesshan 30% retention of stationary phase on SRCCC. With ourxperience, higher viscosity of stationary phase is good foretention of stationary phase on SRCCC column. Therefore,he present study tested the partition coefficient of solanesoletween the two phases of three solvent systems composedf sunflower oil–methanol, sunflower oil–ethanol and sun-ower oil–acetonitrile, in which sunflower oil is viscous. Theesult showed the partition coefficients of solanesol in sun-ower oil–methanol and sunflower oil–acetonitrile were toomall while sunflower oil–ethanol was about 0.5–1, estimatedy TLC analysis. Finally, sunflower oil–ethanol was selectedor SRCCC separation of solanesol in tobacco leaves extract.he retention rate of stationary phase without sample load were2% and 38% when the flow rate of the mobile phase was atmL/min and 4 mL/min, respectively, when the column rotatedt the optimum rotational speed of 100 rpm. The separation of.0 g sample was performed at a flow rate of 2 mL/min. Forhe separation of the sample, sample loading was conductedust after filling column with stationary phase. After the sep-ration of 3.0 g sample the retention rate of stationary phaseas 55%, which showed a smaller loss caused of the sample

oading.Fig. 1A shows the HPLC chromatogram of the tobacco leaves

xtract-a dark adhensive plaster which was not able to be usedor food additives because the extract contained tobacco tarnd other impurities. Under the given HPLC conditions, thehromatogram of the tobacco leave extract showed 10 peaksn which peak 6 was solanesol comparing to the retention oftandard solanesol. Fig. 2 showed the TLC analysis of 33 frac-ions from the separation of 3.0 g tobacco leaves extract. TLC ofractions 18–28 yielded two spots, one possibly for solanesolnd the other for sunflower oil. Fractions 18–28 were com-ined and evaporated under vacuum to yield 1.5 g of yellowily product in which solanesol occupied 26.8% determined byPLC (Fig. 1B). The HPLC chromatogram also indicated thatRCCC separation removed most of the impurities including

he components corresponding to HPLC peaks 1, 2, 3, 4, 5,and 10 of the tobacco leaves extract. Thus, the yellow oily

roduct may possibly be used as a food additive because ofts higher purity and non-industrial organic solvent processing,hough the safety of the components corresponding peaks 2nd 3 in Fig. 1B is needed for further estimation. The recov-ry of solanesol was 89% since a small part of solanesol went

o the adjacent fractions of solanesol fractions, which can beepeatedly separated after collecting them. As far as knowledgeoes, SRCCC apparatus can be easily scaled up to an indus-rial level [8]. Therefore, the present method may possibly be

Y. Zhao, Q. Du / J. Chromatogr. A 1151 (2007) 193–196 195

Fig. 1. HPLC chromatograms of the tobacco leaves extract for SRCCC separation and the oily solanesol product from SRCCC separation. Column: YMC-PackODS-AQ column (5 �m, 250 mm × 4.6 mm I.D.); mobile phase: acetonitrile–isopropanol (65:35, v/v); flow rate: 1.0 mL/min; detection wavelength: 213 nm. (A)The tobacco leaves extract for SRCCC separation; (B) the oily solanesol product from SRCCC separation.

Fig. 2. The TLC analysis of the fractions from SRCCC separation of 3.0 g tobacco leaves extract. TEE: tobacco leaves extract; S: solanesol standard. SRCCC columnc per p5

dp

4

lpew1

wpsl

R

apacity: 1100 mL; solvent system: sunflower oil–ethanol; stationary phase: up0 mL. 10–11: stationary phase; 12–33: mobile phase.

eveloped for separation technology of food grade solanesolroduct.

. Conclusion

SRCCC can be used for separation of solanesol in tobaccoeaves extract using slow rotary counter-current chromatogra-

hy with a novel non-aqueous solvent system composed ofthanol–sunflower oil. Three grams of tobacco leaves extractas separated with a SRCCC instrument equipped with an100 mL capacity column to yield 1.5 g of yellow oily product in

hase; flow rate: 2 mL/min; rotary speed of column: 100 rpm; fraction volume:

hich solanesol occupied 26.8%. The separation method mayossibly be developed for industrial separation of food gradeolanesol product by using SRCCC apparatus equipped with aarge capacity column.

eferences

[1] R.A. Morton, Biochem. J. 128 (1972) 11.[2] M. Wang, Integ. Utiliz. Chin. Res. 2 (2003) 16.[3] N.K. Khidyrova, K.M. Shakhidoyatov, Chem. Nat. Compd. 38 (2002) 107.[4] W. Xia, D.H. Yu, X.Z. Chen, Chin. J. Pharm. 34 (2003) 328.

1 atogr

[9] Q. Du, P. Wu, Y. Ito, Anal. Chem. 72 (2001) 3363.

96 Y. Zhao, Q. Du / J. Chrom

[5] M. Keca, S. Gross, I. Malnar, Farm. Glas. 53 (1997) 173.

[6] R.L. Rowland, P.H. Latimer, J.A. Giles, J. Am. Chem. Soc. 78 (1956) 4680.[7] Q. Du, J. Wang, Y. Ito, J. Liq. Chromatogr. Related Technol. 29 (2006)

2587.[8] Q. Du, G. Jerz, Y. He, L. Li, Y. Xu, Q. Zhang, Q. Zheng, P. Winterhalter,

Y. Ito, J. Chromatogr. A 1045 (2005) 59.

[

[[

. A 1151 (2007) 193–196

10] Q. Du, P. Winterhalter, Y. Ito, J. Liq. Chromatogr. Related Technol. 26(2003) 1827.

11] B.H. Woolen, D.H. Jones, J. Chromatogr. 61 (1971) 180.12] J. Zhao, C. Wang, X. Sun, Chin. J. Chromatogr. 15 (1997) 544.