Gas Chromatographic Separation of Oxygen-Containing Terpene Compounds on Low Temperature Columns SIR: James and Martin ( 2 ) have shown the fundamental relationship between retention volume and the amount of liquid phase present in a gas chromatographic column. Ring (5) has demonstrated on a conventional Celite column t h a t retention times decrease without loss of resolution, and fraction peaks become sharper with a decreasing amount of liquid phase. Hishta, hlesserly, and Reschke ( I ) employed
columns of 60- to 80-mesh glass beads with less than 0.3% liquid phase to separate high boiling ketones and hydrocarbons. They obtained good resolution and short retention times a t temperatures as much as 150' C. belo\T the boiling points of the samples. Littlewood (3) and Pollard and Hardy (4) obtained good results n i t h similar columns. I n the present study columns packed with glass beads containing low concentrations of stationary phases were explored for the separation of various oxygen-containing terpenes which are frequently found in spice oils. Low temperatures and retention times and inert solid supports are desirable because of the tendency of terpenes to decompose or isomerize a t elevated temperatures, particularly when in contact with foreign substances such as acids or alkalies.
L C S I.
HYPROSE
S P 80 I
I
I3
Figure 2. Determination of nature of oxygen-containing terpenes by comparing their behavior on two different columns
m
+J
EXPERIMENTAL
0
-
> 2
J
-1
a5
-
W
0 ~
v)
z 0 Q
v)
W
a a
8
w
4 3 ' 2
r"
Q
'
i
a
0 V W
a 4
8
6
TI ME,
4
2
0
MINUTES
Figure 1. G a s chromatograms of terpene alcohols 2. Linaloz, 1 9 8 ' C.; 3. Menthol, 2 1 6 ' C.; 4. a-Terpineol, 21 8 ' C.; 5. Citronellol 2 2 2 ' C.; 6. Nerol, 2 2 5 ' C.; 7. Geraniol, 2 2 9 ' C. (Peak 1 was the ether, Cineole, 177' C.) A. 0.20% Dow Corning 71 0 on glass beads at 90' . - c. _. 8. 0.1 25% Hyprose S.P. 8 0 on glass beads at 90' C.
1028
ANALYTICAL CHEMISTRY
Equipment. A Research Specialty Company Model 601-1 oven served as the heating unit, and a Cow-Mac tno-filament thermal conductivity cell served as the detector with a constant voltage regulator to control t h e heating of filaments. T h e recorder was a 10-mv. Leeds and Northrup, type G, instrument with 1-second response and a chart speed of 0.5 inches per minute. Emerging components were collected with a Model CH6-H fraction collector equipped with type D cells of 0.1-mm. thickness and manufactured by Connecticut Instrument Co. A Perkin Elmer Model 21 infrared spectrophotometer was used for sample identification. Preparation of Columns Stainless steel tubes, 6 feet in length and l / 4 inch o.d., containing either a polar or a nonpolar stationary phase on 60- t o 80-mesh glass beads were used. T h e concentration of the polar stationary phase, Hyprose S.P. 80, was 0.125Oj, and the nonpolar phase, Dow Corning 710, was 0.2%. Procedure. Three terpene mixtures rrere chromatographed on a Don, Corning 710 and on a Hyprose S.P. SO column, under t h e conditions listed i n Table I. T h e terpene mixtures were six alcohols, eight aldehydes, and three esters. Emerging components were trapped in a cooled fraction collector which deposits the compounds directly in an ultramicro type D infrared absorption cell. After the fraction had been collected, the inner cell of 0.1-mm. thickness was filled with carbon tetra-
Table 1. Composition of Mixtures and Relative Retention Times of Components The experimental conditions were: oven temperature, 90' + 1" C.; detector ternperature, 100" zk 1" C. for the Silicone 710 column and 110" 1" C. for the Hyprose S.P.80 column; injection block temperature, 210" it 2' C.; carrier gas flow a t 5 p.s.i., SO" F., 40 ml. per minute; column length, 6 feet; inert support, glass beads of mesh size 66-80; sample size, 1 pl. Relative Retention Times DOW Hyprose Corning S.P. Compound no 80 Alcohol Mixture" 1.39 1.78 Linalool Menthol 1.78 2.35 a-Terpineol 2.72 3.54 Citronellol 3.16 4.78 Nerol 3.711 5 78 Geraniol 4.28 7.26 hldehydes Ilixture Benzaldehyde 0.39 0.56 CaDroic aldehyde 0.62 0.85 M6thyl heptanone 1.23 1.61 Caprylic aldehyde 1.91 2.44 Citronellol 2.24 3.11 Capric aldehyde 2.41 2.67 Citral 3.81 4.50 Cinnammic aldehyde 5.02 5.83 Esters Mixture Linolyl acetate 0.50 0.44 Geranyl acetate 1.05 0.72 Xeryl acetate 3.28 1.61 Reference Standard (Tetraline) 1.00 1.00 a Each mixture contained approximately equal amounts of each component.
chloride, and the solute was identified by comparison of the infrared spectrum with spectra of authentic reference materials. RESULTS AND DISCUSSION
The ob$erred relative retention times of the three types of materials are listed in Table I. All terpenes were eluted from both columns in less than 8 minutes. The retention time of tetralin (taken as 1) was the reference standard. Under the experimental conditions described, the specific retention volume per gram of liquid phase was 264 ml. per gram for Dow Corning 710 and 164 ml. per gram for Hyprose S.P. 80. As shown in Figure 1, the complete separation of alcohols was achieved with Hyprose S.P. SO but not with Dow Corning 710 column. The aldehydes and esters examined, however, were completely separated in either column. Peak 7 in Figure 1B (geraniol) is skewed toward peak 6 (nerol, the cis isomer of geraniol). The infrared curve for peak 7 is qualitatively very similar to t h a t for pule geraniol. Both
isomers exhibit medium strength absorption bands at 1670 em.-‘ but only the trans form, geraniol, has a band a t The ratio of about 890 em.-’ &SO cm.-l to AI6i0 cm.-1 for pure geraniol was approximately 0.8 whereas this ratio for fraction 7 was lower, indicating the presence of some nerol in fraction 7 . I n general, the behavior of three types of compounds was quite different on the tm-o columns. For example, with the exception of the esters shown in Table I, all compounds had longer retention times on the Hyprose column. W t h the Don. Corning 710 column, all the terpenes within one mixture were eluted in the order of their boiling points, but with the Hyprose S.P. 80 column this was not the case. I n Figure 2 the logarithms of the relative retention times observed on the two columns are plotted against each other. The resulting points for the esters, aldehydes, and alcohols lie approximately in three different straight lines. Determination of the chemical nature of oxygen-containing terpenes can be facilitated in many
cases by comparing their behavior on a polar and a nonpolar column. LITERATURE CITED
(1) Hishta, C., Messerly, J. P., Reschke, R. F., ANAL.CHEX.32,1730 (1960). (2) James, A. T., Martin, A. J. P., Baochem. J . 50, 679 (1952). (3) LittleiTood, A. B., in “Gas Chromatography,” D. H. Desty, ed., p. 23,
Butterworth, London, 1958.
( 4 ) Pollard, F. H., Hardy, C. J;: in
“F‘apour Phase Chromatography, D. H. Desty, ed., p. 115, Academic Press, Sew York, 1957. (5) Ring, R. D., in “Gas Chromatogranhv.” V. J . Coates. H. J. Noebels. I: rST Fagerson, eds., p. 195, Academic Press, New York, 1958. HEIXOS r s ~ Eastern Utilization Research and Develoament Division U. S.Departhent of Agriculture Philadelphia 18,Pa. 1 Senior Research Fellow, American Spice Trade Association. Work was supported in part by funds from the American Spice Trade Association. Mention of a specific product does not constitute endorsement of that product over similar ones not mentioned.
Silicic Acid Chromatography of Methoxypiperonylic Acids SIR: 9 neiv lignan, possessing one unsubstituted and one methoxy-substituted methylenedioxyphenyl group, upon oxidation with potassium perrnanganate in acetone, gave piperonylic acid as expected, but only a minute amount of the niethoxypiperonylic acid. -4means of separating and identifying micro amounts of the methoxypiperonylic acids in the presence of piperonylic acid n-as needed. This can be done by thin-layer chrornatography, using a spray of chromotropic acid in sulfuric acid as the chroniogenic reagent. This reagent gives colors v i t h methylenedioxyphenyl compounds, which are found in many natural products. It will also respond to methylenedioxyl and other groupings that give formaldehyde on treatiiient with acid (g). The methosypiperonylic acid in niilligram amounts n-as isolated by silicic acid column chromatography, and its identity n-as confirmed as 6-methosypiperonylic acid by the ultraviolet and infrared spectral data reported below.
SILICAGEL G. Brinkmann Instruments, Inc., Great Neck, N. Y . , or Microchemical Specialties Co., Berkeley 3, Calif. SILICICACIDwas a specially prepared grade for chromatographing lipides (Bio-Rad Laboratories, Richmond, Calif.), used without drying. It lost 17.8y0 water on drying overnight a t 110’ c. SOLVENTSwere reagent grade and distilled before use. 3,4-llethylenediouyphenyl Acids. Melting points were: piperonylic, 228-0’ C.; 2-methoxypiperonylicJ 1567’ C.; 5rnethouypiperonylic, 2067’ C.; 6-methouypiperonylic, 150-1’ C. Solutions in acetone, 1% (w./v.), were used for thin-layer chromatography. CHROMOTROPIC-SULFURIC ACIDSPRAY RE.~GEST.One volume of freshly made 10% (w./v.) aqueous solution of sodium 1,8 - dihydroxynaphthalene - 3,6 - disulfonate (Eastman Kodak P230) was
Table I.
PROCEDURES AND RESULTS
Thin-layer Chromatography. A silica gel layer, 250 microns thick, applied on a glass plate 200 mni. square as described by Stahl ( 3 ) , was activated for 45 minutes at 105’ C., spotted 2.5 em. from one side with 1 to 3 pl. (10 to 30 kg.) of the acids, and allowed to develop at room temperature in a closed chamber (with a solventsaturated paper lining) containing a 1-em. depth of ethyl acetate-hexaneacetic acid (50:50:0.5) until the solvent front moved 11 to 14 em. past the starting line. The solvent front was marked, and the plate was removed. The sol-
Chromatographic Data on Piperonylic and Methoxy-Substituted Piperonylic Acids
EXPERIMENTAL
Apparatus and Reagents. THIKLAYER C H R O M A T O G R B P H T . The apparatus of Stahl(5) was used. COLVMP;CHROMATOGRAPHY. A smaller version of the all-glass apparatus of Gordon and Beroza (4) was used. (Tube was 1 cm. i.d. b y 30 cm. long; a plug of glass wool replaced the sintered glass disk.)
added to 5 volumes of sulfuric acid, prepared by adding carefully 5 volumes of concentrated sulfuric acid to 3 volumes of water. Renen-ed reekly.
Acid Piperonylic 5-Methoxypiperonylic 2-Methoxypiperon ylic
6-Methoxypiperonylic
Thin-laver Rj val;e 0.64 0.54 0.47 0.37
Column ______--_ Eluent Peak elution 280/300 mp (yoethyl volume absorbance acetate in ratio iso-octane) (ml.1 c0.6 10 13 3.1 10 110 1.1 20 35 0.3 20 140
VOL. 34, NO. 8, JULY 1962
1029
