Studiu Comparativ Priv. Liofilicitatea Unor Coloranti Sintetici Prin Cromatografie

8/6/2019 Studiu Comparativ Priv. Liofilicitatea Unor Coloranti Sintetici Prin Cromatografie

1/6

REV. CHIM. (Bucureti) 61 Nr. 3 2010 http://www.revistadechimie.ro 229

A Comparative Study Concerning the Lipophilicity of SomeSynthetic Dyes Estimated by Thin Layer Chromatography

and Different Computation Methods

DORINA CASONI, CODRUTA SIMONA COBZAC, COSTEL SARBU*

Babe-Bolyai University, Faculty of Chemistry and Chemical Engineering, 11 Arany Janos Str., 400028, Cluj-Napoca, Romnia

The retention behaviour of some selected synthetic dyes have been investigated on RP-18F254S

, RP-18W/UV

254and CNF

254Splates using methanolwater mixture in different volume proportions as mobile phases.

The RMvalues of the compounds decreased linearly with increasing concentration of methanol in the mobile

phase in all cases. On the basis of the linear relationship between retention (RM) values and volume fraction

of methanol, values of RM0

corresponding to 100% water were obtained by extrapolation using five isocraticR

Mvalues. Theregression determination coefficients obtainedfor all stationary phases were excellent (higher

than 0.99 in all cases). The good regularities of profiles of retention indices (RM

and PC1/RM) indicated that the

same mechanism (lipophilic interactions) is dominant in all cases andall types of stationary phases (RP-18,RP-18W and CN) appear to be suitable for estimating the lipophilicity of selected synthetic dyes. Thechromatographic lipophilicity values (expressed by R

M0) estimated on three different stationary phases were

not highly correlated between them and also with computed log P values. The best correlations were

founded with lipophilicity parameters calculated by Dragon 5.4 software. By using PC1/RM or0values asestimators for lipophilic character of synthetic dyes, the correlation between these values obtained on that

three stationary phases were significantly improved, correlation coefficient being higher than 0.92 in somecases.

Keywords:lipophilicity,synthetic dyes, computed log P,RP-HPTLC,PCA

During the past decades an increasing number ofquantitative structure-activity (QSAR) models have beenusing different molecular descriptors for predictingtoxicological [1, 2] properties of chemicals.

Lipophilicity, defined as tendency of a chemical

compound to distribute between an immiscible non-polarorganic solvent and water, is one of the molecularparameters most frequently used for the assessment ofthe relationship between biological activity andphysicochemical characteristics[3].The most widely usedmeasure of lipophilicity is the partition coefficient in the 1-octanol/water system denoted in few different ways (logP, log Kow). The reference procedure to measure log P is theclassical shake-flask method, which however is time-consuming, limited in range and limited to extremely purecompounds [4].Measurement by use of equilibrationmethods is frequently made difficult, or even impossible,by the instability of compounds in some cases, bypreference of a compound for one of the two phases of thesystem, by the formation of stable emulsions after shakingor by difficulties in solubility limits. For this reason thesuitability of experimental and theoretical approaches fordetermination and estimation of the lipophilicity of organiccompounds remains a focus of scientific interest.

Due to their advantageous application parameters,various reversed phase lichid chromatographic methodshave been used for determination of lipophilicity [5, 6].These techniques require only a small amount ofcompounds and they do not need to be very pure becausetheir impurities are readily separated during thechromatographic process. Moreover, the lipophilicity valuesestimated by high-performance liquid chromatography

(HPLC) and high-performance thin-layer chromatography(HPTLC) are generally well correlated [7]. In recent years,HPTLC technique has been used to estimate lipophilicity

* email: [email protected]

of a considerable number of chemicalsand the relationshipof this property with biological activity has been assessed[8-13].The use of RM0values obtained from different typesof reversed-phase thin-layer chromatography (RP-TLC) isbased on the assumed linear relationship between RMand

the concentration of organic modifier in the mobile phase:R

M= R

M0+ bC (1)

RM

= log (1/RF-1) (2)

where RM values are derived by the retention factor (RF)according to the Eq. (2), RM0 is the RM value extrapolated to100% water, Cis the concentration of organic modifier andb is the slope of regression function associated withspecific hydrophobic surface area and it is considered analternative measure of lipophilicity [14-19]. RF is theretention factor calculated on the basis of migrationdistance of compound/migration distance of solvent front.

Recently, another retention related parameter (theisocratic chromatographic lipophilicity index), 0, havebeen introduced. It is defined as the volume fraction of theorganic solvent in the mobile phase for which the amountof solute in the mobile phase is equal to that in thestationary phase when the retention factor (k) is 1 [20]. Inthin layer chromatography0 is described by equation (3):

0

= RM0

/b (3)

In addition, a new lipophilicity scale may be obtained byapplying Principal Component Analysis (PCA) directly tothe RF and/or RM values corresponding to different fractions

of organic modifier and, as a direct consequence, thescores corresponding to the first principal component canbe used as an alternative lipophilicity scale. Much more,


2/6

REV. CHIM. (Bucureti) 61 Nr.3 2010http://www.revistadechimie.ro230

the scatter plots of scores onto the plan described by thefirst two principal components offers the possibility to geta lipophilcity chart [21, 22].

A wide variety of synthetic dyes have been used in foodprocessing and in a variety of modern industrial processes.Over the last few years a lot of concern has been expressedin the media over the safety of food additives generally andthe synthetic food dyes in particular. Many synthetic fooddyes used today have been proven to cause cancer,

hyperactivity, inattentiveness, asthma, and in rare casesdeath [23]. Moreover, many of the water-soluble azo dyesare known to be degraded by intestinal microorganismsand some of their degradation products may havemutagenic properties [24-26]. Because of theircommercial importance many analytical procedures hasbeen established and used for quality control of dyes andfor evaluation of their impact on human health orenvironmental pollution. The most relevant internationallyagreed testing methods used by government, industry andindependent laboratories to assess the safety of chemicalproducts takes in consideration its lipophilicity parameters.Unfortunately, experimental lipophilicity data are notavailable in literature for the compounds investigated inthis paper. The classical experimental procedure usingshake-flask method seems to be difficult for some ofstructurally dyes because of the large difference betweenwater solubility and anticipated solubility in octanol.

In the light of the above considerations, we found itinteresting to carry out a comparative study concerningthe chromatographic lipophilicity of several synthetic dyeson different stationary phases. Because most of theconsidered compounds are easily ionizing, the purpose ofthis paper was also the elucidation of the mechanism ofretention on different types of stationary phase and theuse of RP- TLC technique to assess the lipophilicity of thiskind of compounds.

Experimental partThin-Layer Chromatography

The chromatographic behaviour of sixteen syntheticdyes listed in table 1was studied on different stationaryphases: RP-18 F

254s(20 X 20cm, Merck Darmstadt-

Germany), RP-18W/UV254 (20 X 10cm, Macherey-Nagel)and CN F254s (10 X 10cm, Merck, Darmstadt-Germany).The standards of analytical grade were purchased fromMerck or Fluka.

Analytical-grade methanol was purchased fromChimopar (Bucharest, Romania). The standard solutions(1 mg mL-1) of dyes were separately prepared in methanoland 2 L from them were separately applied manually onthe plates by means of a 10 L Hamilton (Switzerland)microliter syringe. Chromatography was performed in anormal developing chamber (saturated for 15 min withsolvent vapors) at room temperature (~22oC), usingdifferent proportion mixtures of methanol-water as mobilephase (from 20 to 60% methanol in steps of 10% for alltypes of stationary phases). The developing distance was8 cm in all cases. After development the plates were driedin air at room temperature and the spots of dyes wereapparent from their colours.

Theoretical Partition CoefficientsIt is well known that many software and internet

modules are able to calculate lipophilicity values bydifferent algorithms. All of them require a previouslymolecule drawing that is usually performed by Hyperchemand optimized using the MM+ molecular mechanics forcefield [27]. On the basis of obtained geometry, differentsoftware allowing theoretical calculations of variouslipophilicity descriptors. One of the log P values wascalculated using fragmental and atomistic methods byChem3D Ultra 8.0 [28], (ClogPCD), four values (MLOGP,MLOGP2, ALOGP, ALOGP2) were calculated using

topological descriptors by the software Dragon 5.4 [29]and six Log P values (ALOGPs, AC logP,, miLogP, logPKOWWIN

,

Table 1

THE COMMON AND IUPAC NAME OF THE SYNTHETIC DYES


3/6

REV. CHIM. (Bucureti) 61 Nr. 3 2010 http://www.revistadechimie.ro 231

XLOGP2, XLOGP3) based on different computationmethods (electrotopological state descriptors, fragmentalmethod and atom additive model) were obtained by usingALOGPS 2.1-vcclab free internet module [30]. All thecalculated values are listed in table 2.

Results and discussionsThe retention behaviour of several synthetic dyes (fig.

1) was investigated by systematic changing the amountof organic modifier in mobile phases. The Rfvalues obtainedshowed that the retention of the studied dyes regularlyincreased as the methanol content of the mobile phasewas increased. Linear relationships characterized by high

correlation coefficients were obtained between RM valuesand volume fraction of methanol.

Because most of the studied compounds are easilyionizable, they could be susceptible to the secondarymechanisms of interactions with more or less non-polarstationary phase. The regular changes of the retention ofdyes with changing water-methanol ratio can be observedby the profiles of retention indices (RM) (fig. 2a-c).

These systematic regularities of retention observed forall three types of stationary phases might indicate that thesame mechanism (lipophilic interactions) is dominant inall cases and no secondary mechanisms were highlighted.The extrapolated RM0 values and other experimental

Table 2

Log P VALUES CALCULATED BY DIFFERENT COMPUTING PROGRAMS (CHEMDRAW ULTRA 8.0, DRAGON 5.0 ANDALOGPS 2.1 [25-27])

Fig. 1. The structures of thedyes investigated


4/6

REV. CHIM. (Bucureti) 61 Nr.3 2010http://www.revistadechimie.ro232

Fig. 2. The profiles of RM values for all fraction of methanol: (a) RP-18; (b) RP-18W and (c) CN

Table 3EXPERIMENTAL LIPHOPHILICITY INDICES ESTIMATED BY DIFFERENT RP-HPTLC STATIONARY PHASES

lipophilicity indices obtained on the three stationary phasesare summarized in table 3.

The similarity and/or differences between thelipophilicity parameters estimated from retention dataobtained on the three stationary phases can be illustratedby representation of the profiles of RM0 values (fig.3.a) andprofiles of scores obtained by applying PCA directly to thematrix of R

Mvalues (PC1/R

M, fig. 3.b). Given the large

number of existing stationary phases and their different

properties, the problem is the choice of the most suitablestationary phases so that chromatographic results obtainedto be comparable with the theoretical calculated values orobtained by established methods. In order to assess theexperimental lipophilicity of the investigated compounds(estimated by RM0, b, 0, scores corresponding to firstprincipal component of R f (PC1/Rf) and scorescorresponding to first principal component of R

M(PC1/R

M)),

the obtained results were compared with the differentcomputed Log P values. The correlation matrix (table 4)illustrates a low compatibility between chromatographicindices of lipophilicity and computed Log P values for theinvestigated dyes.

The best correlations were obtained with the valuescalculated by using topological descriptors (Dragon 5.4software). The weak correlation may be attributed to thefact that many computer programs do not recognize thepotentially ionic character of molecules. Unfortunately, forthe compounds considered in this study, the experimentallipophilicity determined by the classical shake-flaskmethod are missing from the literature, probably, becauseof the large difference between water solubility andanticipated solubility in octanol.

By using PC1/RM or 0values as estimators for lipophiliccharacter of synthetic dyes, the correlation between these values obtained on the all three stationary phases weresignificantly improved, correlation coefficient being higher

than 0.92 in some cases. These fairly high correlationbetween 0parameters on three stationary phases maybe further evidence that secondary retention mechanismsare absent in all cases.

Most of statistical algorithms do not require normalizedinput data but in some cases (linear regression)normalization is a recommended step. In statistics,normalization refers to the division of multiple sets of databy a common variable in order to allow data on differentscales to be compared, by bringing them to a comparablescale. Lipophilicity charts (fig. 4a-c) obtained by scatterplots of normalized data for experimental lipophilicity

(a)

(c)

(b)


5/6

REV. CHIM. (Bucureti) 61 Nr. 3 2010 http://www.revistadechimie.ro

(b)

Fig. 3. The profiles of lipophilicity indices RM0 (a) and PC1/RM (b) of the investigated dyes

Fig. 4. Scatterplot of normalized data forexperimental lipophilicity indices:

(a) RM0; (b) 0; (c) PC1/RM

(a)

Table 4

THE CORRELATION CONCERNING LIPOPHILICITY PARAMETERSOBTAINED ON DIFFERENT RP-TLC PLATES AND SOME COMPUTED log PVALUES (THE HIGHEST STATISTICAL SIGNIFICANT VALUES ARE BOLDED)

(a)

(b)

(c)

233


6/6

REV. CHIM. (Bucureti) 61 Nr.3 2010http://www.revistadechimie.ro

indices for three different stationary phases (table 5) aremore easily to use for looking of trends, groupings or outliers.

By careful visual examination of the lipophilicity space(fig. 4a-c), it is apparent that the chromatographic behaviorof the investigated dyes is firstly in agreement with thenumber of ionic groups;the normalized values of0 andPC1/RM parameters associate the compounds in better waythan RM0values.

ConclusionsRetention behaviour of sixteen synthetic dyes was

investigated by reverse phase thin-layer chromatographyon RP-18F254S, RP-18W/UV254 and CNF254S plates usingmethanolwater mixtures as mobile phases. Systematicregularities of retention revealed by the profiles of retention

indices (RM) no highlighted secondary mechanisms ofinteraction with studied stationary phases. In addition, fairlyhigh correlation between0parameters on three stationaryphases may be further evidence that secondary retentionmechanisms are absent. The low correlation betweenchromatographic indices of lipophilicity and computed LogP values might be attributed to the strong ionic characterof investigated dyes. The results obtained indicate thatdescriptors for molecular properties computed by Dragonsoftware are the best parameters for estimating thelipophilicity of this kind of compounds. The 0 and PC1/RMvalues of synthetic dyes estimated on the three stationaryphases seem to be the most suitable and appropriateestimators.

Acknowledgement: The financial support of the Ministry of Educationand Research of Romania (CNCSIS, IDEI 560/2007) is gratefullyacknowledged.

References1. VLAIA, V., OLARIU, T., CIUBOTARIU, C., MEDELEANU, M., VLAIA,L., CIUBOTARIU., D., Rev. Chim. (Bucuresti), 60, 2009, p. 13572. DONE, R., MNDRILA, G., TARKO, L., Rev. Chim. (Bucuresti), 60,2009, p. 9923. SELA SSIE, C. D., GARG, R., KAPUR, S., KURUP, A., VERMA, R. P.,MEKAPATI, S. B., HANSCH,C., Chem. Rev., 102, 2002, p. 25854. LEO, A., Chem. Rev. 93, 1993, p. 12815. SRBU, C., MALAWSKA, B., J. Liq. Chromatogr. Related Technol .,23, 2000, p. 2143

Table 5

THE NORMALIZED DATA FOR EXPERIMENTAL LIPHOPHILICITY INDICES ESTIMATED BY DIFFERENTRP-HPTLC STATIONARY PHASES

6. CASONI, D., KOT-WASIK, A., NAMISNIK, J., SRBU, C., J.Chromatogr. A, 1216, 2009, p. 2456

7. RAVETTI, S., GUALDESI, M. S., BRINON, M. C., J. Liq. Chromatogr.Related Technol., 31, 2008, p. 10148. PERISIC-JANJIC, N. U., LUCIC, B., JANJJICN, N. J., AGBABA, D., J.Planar Chromatogr., 16, 2003, p. 3479. PERISIC-JANJIC, N. U., JAVANOVIC, B. ,. JANJJIC , N. J., RAJKOVIC,O. S., ANTONEVIC, D. G., J. Planar Chromatogr.,16, 2003, p. 42510. JANICKA, M.,J. Planar Chromatogr., 20, 2007, p. 26711. PYKA, A., GURAK, D., J. Planar Chromatogr., 20, 2007, p.37312. MORAK-MLADOWSKA, B., PLUTA, K., J. Liq. Chromatogr. RelatedTechnol., 31 , 2008, p. 61113. ONISOR, C., KOVALA-DEMERTZI, D., DEMERTZIS, M. A., SRBU,C., Rev. Chim. (Bucuresti), 60, 2009, p. 103914. SRBU, C., KUHAJDA, K., KEVRESAN, S., J. Chromatogr. A., 917,2001, p. 36115. BUTTE, W., FOOKEN, C., KLUSSMANN, R., SCHULLER, D., J.Chromatogr., 214, 1981, p. 5916. ROZYLO, J. R., MATYSIAK, J., NIEWIADOMY, A., ZABINSKA, A., J.Planar Chromatogr., 13, 2000, p.17617. SRBU, C., TODOR, S., J. Planar Chromatogr. 11, 1998, p. 12318. SRBU, C., TODOR, S., J. Chromatogr. A, 822, 1998, p. 26319. MALAWSKA, B., KULIG, K., WISNIEWSKA, M., J. Planar Chromatogr.,13, 2000, p. 18720. JZWIAK, K., SZUMILO, H., SOCZEVINSKI, E., Wiad. Chem., 55,2001, p. 104721. NASAL, A., BUCISKI, A., BOBER, L., KALISZAN, R., Int. J. Pharm.,159, 1997, p. 4322. KALISZAN, R., Chem. Rev. 107, 2007, p. 3212

23.*** http://healthfieldmedicare.suite101.com/article.cfm/food_coloring#ixzz0DbWaF4Wd&A24. JONES, R. RYAN, A. J., WRIGHT, S. E., Food Cosmet. Toxicol., 2,1964, p. 44725. ROXON, J. J., RYAN, A. J., WRIGHT, S. E., Food Cosmet. Toxicol.,5, 1969, p. 36726. CHUNG, K. T., FULK, G. E., ANDREWS, A., W. Appl. Environ.Microbiol., 42, 1981, p. 64127. *** HyperChem, release 7.5 for Windows, Molecular ModelingSystem; Hypercube, Inc. and Autodesk, Inc.28. *** Chemical Structure Drawing Standard, ChemDraw Ultra 8.0.3(2003) http://www.cambridgesoft.com29. *** http://www.talete.mi.it30.*** Virtual Computational Chemistry Laboratory. http://

www.vcclab.org/lab/alogps/

Manuscript received. 28.10.2009

234

Studiu Comparativ Priv. Liofilicitatea Unor Coloranti Sintetici Prin Cromatografie

Documents

Transcript of Studiu Comparativ Priv. Liofilicitatea Unor Coloranti Sintetici Prin Cromatografie