Adyst, May, 1968, Vol. 93, pp. 311-318 311 Origanum Oils and their Investigation by Gas-chromatographic and Infrared Spectroscopic Analysis BY C. CALZOLARI, B. STANCHER AND G. PERTOLDI MARLETTA (Istituto di Merceologia, UniwersitG di Trieste) A gas-chromatographic study of the essential oil of origanum has been carried out with both capillary and packed columns. Furthermore, some new components of the essential oil of origanum were found by coupling preparative gas-chromatographic analysis and infrared spectroscopy. WE have frequently observed that spices of various geographic origin have different organo- leptic characteristics, indicating that the environmental conditions of the plant growth influence the composition of aromatic components of the spice. To determine the possible variations we examined the essential oil that is extracted from the plant.The present paper deals with the examination of oil of origanum obtained from samples of different origin and its characterisation by gas-chromatographic analysis. The composition of the essential oil of origanum has already been partially studied,lJ,* and the presence of some of the main components recorded. In order to define accurately the aromatic properties of the essential oil of origanum, it was examined by gas-chromatographic methods, and the substances separated were identified by infrared spectroscopy. By this procedure almost all of the constituents of the essential oil that were present in amounts greater than 0.1 per cent. were identified. The examination was carried out on five samples of commercial essential oil of origanum and four samples distilled in our laboratory from origanum grown in Greece (var.Heraclaeotium), Turkey (var. Onites), Spain (var. floribundum Munby) and Italy (Sicily) (var. Gracile). Gas-chromatographic analysis of the oils was carried out with packed columns and capillary columns of different diameters. SAMPLE PREPARATION- The essential oil of origanum was extracted by steam-distillation according to the A.0.A.C.4 method. The commercial samples were supplied by the following firms: Bertrand Freres, Grasse, France (stated Lebanese origin) ; Dragoco, Milano, Italy; Esperis, Milano, Italy (stated Italian origin) ; Valerio, Milano, Italy (stated Italian origin) ; Viansino, Milano, APPARATUS- Analytical and preparative gas chromatograph-Wilkens Instrument and Research Inc., Aerograph Moduline, Model 1521-1, dual column, equipped with a dual thermal conductivity detector, two flame-ionisation detectors and linear temperature programmer. Gas chromatogra$h-C. Erba (Milano, Italy), Fractovap, Model D, dual column, equipped with linear temperature programmer, flame-ionisation detector, constant flow of pressure and completely modular. Spectrophotometer-Perkin-Elmer, Infracord, Model 137, double beam. GAS-CHROMATOGRAPHIC ANALYSIS- The following columns were used. (A) Stainless-steel column, 2 metres long, 0.d. 4 inch, packed with 10 per cent. Carbowax 20M on silanised Chromosorb (80 to 100 mesh); carrier gas nitrogen. The Wilkens gas chromatograph was used with flame-ionisation detector. (B) Stainless-steel column, 7 metres long, i.d.1 mm, packed with 2 per cent. Carbowax 20M on silanised Chromosorb (100 to 120 mesh) ; carrier gas nitrogen. The Erba gas chroma- tograph was used with flame-ionisation detector. EXPERIMENTAL Italy. 0 SAC and the authors.312 CALZOLARI et al. : ORIGANUM OILS AND THEIR INVESTIGATION BY [Analyst, Vol. 93 (C) Stainless-steel column, 30 metres long, id. 0.25mm, coated with Carbowax 20M (C. Erba's capillary column); carrier gas nitrogen. The Erba gas chromatograph was used with flame-ionisation detector. (D) Stainless-steel column, 50 metres long, i.d. 0*25mm, coated with Carbowax 20M (C. Erba's capillary column); carrier gas nitrogen. The Erba gas chromatograph was used with flame-ionisation detector.(E) Aluminium preparative column, 2 metres long, 0.d. 4 inch, packed with 20 per cent. Carbowax 20M on Chromosorb AW (60 to 80 mesh); carrier gas helium. The Wilkens gas chromatograph was used with thermal conductivity detector. (F) Aluminium preparative column, 10 metres long, 0.d. Q inch, packed with 20 per cent. Carbowax 20M on Chromosorb AW (60 to 80 mesh); carrier gas helium. The Wilkens gas chromatograph was used with thermal conductivity detector. PROCEDURE- The preparative gas-chromatographic analysis of the essential oil of origanum was carried out by first fractionating on a 2-metre fractionating column that separated the oil into five main fractions. Thus the main component, carvacrol, which is present in amounts up to a possible maximum of 83 per cent., was eliminated.Each of the fractions then underwent preparative gas chromatography on a 10-metre column (2500 theoretical plates), which allowed a better separation. The best conditions for analysis were chosen for each fraction and the separated compounds were collected on dry ice - acetone at -70" C. The samples were then analysed by infrared spectroscopy.K RESULTS AND DISCUSSION For the qualitative and quantitative determination of the essential oil, the Kovats Indexes6 (Table I) and the areas for each of the peaks were calculated as percentages. Table I1 shows a more accurate percentage composition for the results previously reported (see Note I in Calzolari, Pertoldi Marletta and Stancher' and Note I1 in Pertoldi Marletta and Stancher*), which had been obtained with a shorter column. To compare the results obtained by the various columns, the peaks were numbered by reference to the number of peaks separated by capillary column (D).The 2-metre 10 per cent. Carbowax 20M column was used for these determinations, as less satisfactory results had been obtained with other stationary phases previously tried (Apiezon L, Apiezon M, methylsilicon polymer SE30, butanediol succinate, LAC3 R-728, polydiethyleneglycol succinate, LAC2 R-446, polydiethyleneglycol adipate) . As an example, Fig. 1 shows the chromatogram obtained from the essential oil extracted from Greek origanum. 3 In I Fig. 1. Greek origanum on column A, sample 0.2p1, programme temperature 65 to 220" C (rate 4" C per minute), nitrogen flow 9 ml per minute, injector temperature 270" C, detector temperature 230" C, sensitivity = 4 x 10Peak number 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 29b (Spain) 29b (Esfieris) TABLE I KOVATS INDEXES CALCULATED FOR EACH COLUMN Columns A + > + + + + + + + + + + > + + + + + + + + -t- + + + + + + + + + + + + + + May, 19681 GAS-CHROMATOGRAPHIC AND INFRARED SPECTROSCOPIC ANALYSIS Mark + indicates the resolved peaks in each column examined.B + > + + + + + + + + + 3- + + + + -4- + + + + + + + + + + + + { + -t + + + + + + + + + + + + + + C + > + + + + + + + + + + + + + + + + + + + + + { + + { + + ( + + + + + + + + + + + + + + i- + -1- + + + + + D + + + + + + + + + + + + + + + + + + + + + + + -+ { + + { + + -I- + + + + + + + 4- + + + 3- + + + + + + + + + + + Kovats Index A 1025 } 1042 1090 1135 1156 1175 1196 1212 1233 1268 1296 ) 1370 1438 1470 1534 1558 1570 1621 1645 1694 1715 1750 1763 1773 1836 1870 1946 1984 2021 2118 2152 2181 2264 1630 1645 Kovats Index D 1010 1023 1030 1072 1113 1129 1153 1162 1172 1178 1200 1212 1225 1239 1251 1281 1291 1338 1389 1437 1450 1466 1495 1538 - ) 1548 1564 { 1595 1610 1634 1647 1674 1692 1701 1715 1726 1741 1759 1771 1834 1867 1894 1929 1972 1985 2094 2118 2126 2137 2171 2252 1595 1595 313 Table I1 shows that both the commercial origanum oils and the oils we extracted contained the same components, although in different amounts.314 CALZOLARI et al.: ORIGANUM OILS AND THEIR INVESTIGATION BY [Analyst, Vol.93 TABLE I1 RESULTS OF THE QUANTITATIVE ANALYSIS FOR THE LABORATORY-EXTRACTED AND COMMERCIAL OILS OF ORIGANUM Laboratory-extracted oils Commercial oils f A A \ r > Greece Turkey Spain Italy Bertrand Dragoco Esperis Valerio Viansino per cent. per cent. per cent. per cent. per cent. Peak Areas, Areas, Areas, Areas, Areas, Areas, Areas, Areas, Areas, number per cent. per cent. per cent. per cent. 1 v 4 6 8 9 10 11 12 ;: } ;; } 2 } 16 16 21 22 25 26 27 28a 28b* 29a 29bi 32 36 37 39 40 41 42 43 44 E } 2 } 47 0.03 0-96 0.14 0.10 0.93 traces 0.86 0.11 0.22 traces 3-67 6.90 0.04 0.22 0-13 traces 0.19 0.10 traces 0.85 1.05 traces 0-50 0.74 1.53 traces 0.10 0.1 1 0.08 traces traces 0.24 0-12 - - 0.02 0.31 0.28 0.03 0.4 1 traces 0.47 0.17 0.23 traces 1.90 2-27 traces 0.16 traces traces 0.39 0.04 traces 1-44 0.82 traces 0.1 1 3.15 2.46 traces 0-36 0.27 0.03 traces traces 0.36 0.37 - - 0.03 1.69 0.30 0.32 2.13 traces 2.22 0.26 0.36 6.50 23.00 12.64 0.19 0.2 1 0.04 traces 0.93 0.58 traces 13.30 3.32 traces 0.09 1.42 4-63 traces 0.43 0.14 traces traces traces 0.78 1-56 - - traces 1.13 0.03 0.10 1.31 traces 1.66 0.1 1 0.77 7.82 6.01 0.02 0.26 0.30 traces 2.60 0.12 4-87 0-67 - - - - traces 0.61 0.77 1.60 traces 0.66 - - 0-23 0.14 - - 0.03 2.00 0-12 0.21 2-20 traces 2-22 0.32 0.20 traces 6.68 9.60 0.63 0.89 0.7 1 0.02 0-17 0.20 traces 0.63 2-41 traces 0-03 0.51 0.17 traces 0.09 0.12 0.03 traces 0.02 0.53 traces - - traces 1.86 0.24 0.27 1-36 traces 0.91 0.26 0.72 traces 1-66 14-12 0.01 0.30 0.33 traces 2.64 0.08 traces 1-35 3.46 traces 0.76 1.33 0.68 0.09 0.14 0.16 0.04 - - - - 0.64 - traces 0.66 0.04 0.08 0.28 traces 0.23 0.06 0.32 traces 0.46 2.14 0.04 0.14 0.06 0.47 0.38 0.02 0.40 1-64 13.64 traces 0.51 0.07 traces 0.06 0.04 0.02 - - - - - 0.06 - 0.04 1.94 0.10 0 21 2.16 traces 2.04 0.26 0.21 traces 6.03 8.59 0-46 0.76 0.7 1 traces 0.19 0.10 traces 0.80 2.26 traces traces 0.56 0.03 0.09 0.04 0.10 0-08 traces traces 0.14 traces - - 0.04 1-64 1-16 0.20 1.91 traces 2.06 0.36 0.45 traces 4.68 7.75 0.26 0.63 0.40 0.01 0.82 0.03 1.03 2-75 traces 0.02 1-62 0-23 0.19 0.23 0.04 traces traces 0-60 - - - - - 60 6-19 0.86 18.34 69.90 31.28 1.60 1-24 18-83 21.21 61 74-90 83.10 4-66 9.33 38.67 66.21 76-08 63-29 49.08 62 traces traces traces traces traces 0.22 0.08 traces 0.70 * Peak 28b is present only in the Spanish essential oil and was identified with No.17 (Spanish) (see Note I11 in Stancher and Pertoldi Marlettas). t Peak 29b is present only in the essential oil, Esperis, and was identified with No. 18 bis (see Note I11 in Stancher and Pertoldi Marlettab). of Table I11 lists the constituents we identified and the methods we used for the identification each comDonent. PreparGive gas chromatography was used in the analysis of all nine available essential oils in order to ascertain whether the same components were always present. Only in the essential oil of origanum, Esperis, was the component, peak number 28b, and in the Spanish essential oil the presence of the component, peak number 29b, found (Table 111). The possibility of changes in composition of the terpene compounds during gas-chromato- graphic analysis had previously been ~uggested,~ but pure terpene compounds that we subjected to successive preparative gas-chromatographic analysis and infrared analysis showed no alteration.1°May, 19681 GAS-CHROMATOGRAPHIC AND INFRARED SPECTROSCOPIC ANALYSIS TABLE I11 IDENTIFICATION OF COMPONENTS IN ESSENTIAL OIL OF ORIGANUM 315 Identification methods A I t Peak number 2 3 4 6 8 10 11 12 13 16 16 19 21 22 23 26 26 28a 28b 29a 29b 32 36 37 39 40 46 60 61 62 To Compound Bicyclic or tricyclic terpene .. .. Camphene .. .. .. .. 8-Pinene .. .. .. .. Myrcene . . .. .. .. .. a-Terpinene . . .. .. .. Limonene .. .. .. .. 1.8-Cineole . . .. .. .. Pentyl alcohol . . .. . . . . y-Terpinene .. .... .. p-C ymene .. .. .. Primary alcohol. . .. .. .. a-Pinene . . .. . . . . .. Most likely aliphatic secondary alcohol Secondary or tertiary terpene alcohol Chain-branched aliphatic alcohol . . Linalol . . .. .. .. .. Primary unsaturated alcohol . . . . Aromatic ether . . .. .. .. Secondary alcohol . . . . . . Borneo1 . . .. .. .. .. Carvone .. .. .. .. .. Secondary terpene alcohol . . .. Caryophyllene . . .. .. .. Sesquiterpene? . . .. .. .. Sesquiterpene . . .. .. .. Aromatic compound . . .. .. Polyterpene hydrocarbon . . .. Thymol .. .. .. .. .. Phenolic compound . . .. .. Carvacrol .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. Infrared spectra + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + Elemental analysis + + + + + + + + + + + + + + Me1 ting- nh0 point + + + + + + + + + + + + + + + + + + ascertain the relationship between the chromatographic resolution and the number of plates used, we carried out a gas-chromatographic analysis of a sample of essential oil of origanum with columns of different diameter each packed with Carbowax 20M.Columns B, C and D were used for this purpose and Table IV shows the number of plates and the standard effective peak number11 of all analytical columns used so far. TABLE IV COMPARISON OF EFFICIENCIES OF THE PACKED AND CAPILLARY COLUMNS Theoretical plates Standard effective Column number peak number A 2700 4-8 B 6600 7.1 C 9000 8 D 14,600 9 The standard effective peak number of a column is the effective peak number calculated for the consecutive alkanes, which are eluted at retardation ratios of about 5-0.11 Figs.2, 3 and 4 show the chromatograms obtained on columns B, C and D. The resolution obtained on our capillary columns did not improve the separation, especially for column B . Some new peaks of minor constituents were noted, and in some instances the separation between adjacent peaks improved. Moreover, it is interesting to note that this did not occur with peaks Nos. 28 and 29, which overlapped in the capillary column.316 CALZOLARI et al. : ORIGANUM OILS AND THEIR INVESTIGATION BY [Autalyst, Vol. 93 _1 Fig. 2. Greek origanum on column B, sample 0 . 2 4 , programme temperature 65 to 220" C (rate 4.5" C per minute), nitrogen flow 2.6 ml per minute, by-pass 26 ml per minute, injector temperature 270" C, detector temperature 230" C, sensitivity = 4 x 10 51 tli W 50 Fig.3. Greek origanum on column C, sample 0 . 1 ~ 1 , programme temperature 65 to 220" C (rate 4.5" C per minute), nitrogen flow 1 ml per minute, by-pass 45 ml per minute, injector temperature 270" C, detector temperature, 230" C, sensitivity = 4 x 1 2f 37 I 36 35 29 16 N m X Fig. 4. Greek origanum on column D, sample O-lpl, programme temperature 65 to 220" C (rate 4.5" C per minute), nitrogen flow 1 ml per minute, by-pass 50 ml per minute, injector temperature 270" C, detector temperature 230" C, sensitivity = 4 x 1May, 19681 GAS-CHROMATOGRAPHIC AND INFRARED SPECTROSCOPIC ANALYSIS 317 However, the 50-metre column gave the best resolution and Table I shows the new Kovats Indexes of the single peaks. CONCLUSIONS In the gas-chromatographic analysis of essential oil of origanum we used various columns of different diameters and lengths to study the relationship between the numbers of plates used and the resolution obtained.For this investigation we used two packed columns: (A), 2 metres long, 0.d. Q inch; and (B), 7 metres long, i.d. 1 mm, and two capillary columns, 30 and 50 metres long. In the four kinds of columns examined we retained Carbowax 20M as the stationary phase because previous results had shown it to be the best. The percentage areas were calculated only for packed column (A), as in the other three gas-chromatographic columns the use of a by-pass was necessary to inject a minimum amount of sample into the column.This by-pass can sometimes lead to the components injected into the column being selected according to their volatility and consequently to an erroneous quantitative evaluation. The chromatograms obtained with the packed column B were particularly satisfactory, as the resolution obtained with this column could be compared with the resolution with the capillary column. This study has shown that it is unnecessary to use capillary columns to obtain a good definition of the components of the essential oils, as a long packed column of small diameter containing a small percentage of stationary liquid phase on a silanised support can often yield satisfactory results. Thus the disadvantages of the capillary columns, vix., high cost, short life and difficulty of coating in the laboratory are eliminated.The preparative gas-chromatographic analysis of nine different kinds of essential oils of origanum (five commercial oils and four laboratory-extracted oils) increased our knowledge of the components of this oil and established the presence of nine constituents not previously identified in essential oil of origanum. Furthermore, we obtained the infrared spectra of fourteen other substances that are contained in this essential oil. These substances were only partially identified, as direct comparison with the pure compounds was not made and this alone could give more accurate information about them. However, the fourteen compounds were classified with satisfactory accuracy, and, whenever possible, the refractive index was determined and elemental analysis carried out, although frequently the amount available was only sufficient for an infrared spectrum to be obtained. Of the fourteen partially identified components, two belonged to different oils, vix., No.28b was present only in the Spanish origanum essential oil and No. 29b in the origanum essential oil, Esperis. Thirty components were totally or partially identified, and the presence confirmed of seven other components that had not been previously described in the literature. The possibility of isomerisation or changes in composition of the terpene compounds during the preparative gas chromatography was studied by analysing various pure terpenes, first by preparative gas-chromatographic analysis and then by infrared spectroscopy. This control did not show any alteration in our experimental conditions. The quantitative composition of the origanum oils examined varied according to their place of origin.Therefore, the quantitative area-percentage results obtained from chromato- grams could supply sufficient indication to establish the origin of the spice, although examination of a series of samples of the same origin would be necessary to form a more complete picture. The results obtained demonstrate the advantages of coupling preparative gas-chromato- graphic analysis and infrared spectroscopy for the isolation and identification of the com- ponents of mixtures such as essential oils. Results are more accurate by this method than by the method of comparison of retention times with those of known compounds, which is inaccurate when carried out on high-resolution capillary columns. REFERENCES 1. 2. 3. 4. Fenaroli, G., Sostanze aromatiche naturali,. 1963, I, 796. Blake, M. I., Analyt. Chem., 1958, 30, 400. Ikeda, R. M., Stanley, W. L., Vannier, S . H., and Spliter, E. M., Fd Res., 1962, 27, 455. Horwitz, W., Editor, “Official Methods of Analysis of the Association of Official Agricultural Chemists,” Tenth Edition, The Association of Official Agricultural Chemists, Washington, D.C., 1965, p. 472.318 6. 6. 7. 8. 9. 10. 11. CALZOLARI, STANCHER AND MARLETTA Stancher, B., and Pertoldi Marletta, G., Ann. Facoltb di Economia e Commercio, Universita Messina, Italy, Atti del V Convegno della Qualitb, 10-12 Settembre 1966, Anno IV, N. 2, p. 663. Kovats, E., Helv. Chim. Ada, 1968, 41, 1916. Calzolari, C., Pertoldi Marletta, G., and Stancher, B., Universitb degli Studi di Trieste, Facolta di Economia e Commercio, Istituto di Merceologia, Pubbl. N. 28, 1966, p. 1. Pertoldi Marletta. G., and Stancher, B., Universitb degli Studi di Trieste, Facoltb di Economiae Commercio, Istituto di Merceologia, Pubbl. N. 29, 1966, p. 1. Zubyk, W. J., and Conner, A. Z., Analyt. Chem., 1960, 32, 912. Stancher, B., and Pertoldi Marletta, G., Rass. Chim., 1967, 3, 99. Perry, S. G., and Hurrell, R. A., J. Gas Chromat., 1966, 3, 2. Received Azsgust lst, 1967