The identification of phenolic compounds by a gas chromatographic method on three capillary columns with the same non-polar stationary phase

https://doi.org/10.1016/j.aca.2005.03.024Get rights and content

Abstract

The reproducibility and repeatability of retention indices values of phenolic compounds studied on several capillary columns with the same methyl–5% phenylsiloxane stationary phase indicated that various mode of such columns could be of use for the confirmative analysis. The application of three columns with manufactures’ symbols of PE-5ms, PE-5ht and DB-5 has made it possible to identify 16 of free phenolic compounds present in pine needles.

Introduction

For the purpose of identification, the retention indices (RIs) are frequently used in the gas chromatography method (i.e. RI = 100[z + (tx − tz)/(tz+1 − tz)], where “t” is a total retention time defined as follows: tx for an unknown substance; tz and tz+1 are the retention times of the first and second n-alkanes, respectively, between which the unknown compound appeared). For the identification of compounds present in the sample under examination, first the mixture of standards with the addition of the n-alkanes and then the sample under investigation are being chromatographed to calculate the RI values. If the RI value of an unknown compound and that of standard are one and the same, the chances of some probability are assumed that these substances are identical.

A way of confirmation that this identification is apt is to make an analysis at the same temperature programme, but on the column with the stationary phase of other polarity [1]. In some cases, however, a little change of stationary phase polarity may make a separation process of the chromatographed mixture difficult, and as a result, the overlapped peaks on the chromatogram make the identification practically impossible. And that was the case when considering the data of RI found for the stationary phase of OV 1 and OV 17 [2]. According to McReynolds, the said phases are ranked among the medium-polar ones, though at the same time, the latter (i.e. methyl–50% phenylsiloxane phase) is more polar than the former (i.e. methylsiloxane phase). The mixture of eight biological acids (i.e. glyoxylic(oxime), 3-hydroxypropionic, 3-hydroxybutyric, 3-hydroxyisobutyric, succinic, fumaric, 2-methyl-3-ketovaleric and 2-propyl-3-hydroxypentanoic) is quite easily separable on the column with the OV 1 stationary phase (RIOV 1 value: 1121, 1140, 1160, 1163, 1308, 1349, 1354 and 1392, respectively). In the case of the OV 17 phase, there is no separation but four or maybe two peaks on the chromatogram instead of eight ones (RIOV 17 value: 1196, 1194, 1194, 1196, 1402, 1403, 1403 and 1403, respectively).

A similar situation can happen when the phytogenic samples are analysed. The extracts obtained from green or deciduous leaves may contain numerous compounds of the same chemical class so that there are no chances to match such two stationary phases, in which every compound present in a sample can be separated just as good. In other words, just only of the few stationary phases are of any use for analysis of phytogenic samples by the GC method. The capillary columns with the methylsiloxane [3], [4], [5] or methyl–5% phenylsiloxane [6], [7], [8], [9], [10] phase were most frequently applied for analysis of phenolic acids as trimethylsilyl derivatives (i.e. TMS-derivatives).

The authors have observed that when chromatographed these compounds on two capillary columns with the methyl–5% phenylsiloxane phase made by different producers, distinctly different values of RI for one and the same compound were received. Consequently, it was decided to check out, if it is possible to perform the confirmation analysis on several capillary columns equipped with the same stationary phase. Practically, the findings are expected to be of substantial importance for analysis by the gas chromatography methods.

Section snippets

Materials and sample

Nineteen simple phenolic compounds that are frequently found in natural samples of urine [2], leaves and tissues of trees [3], [4], [11], [12], [13], juices [14] and wines [15] have been chosen to be studied. Table 1 displays these compounds. They all, n-alkane standards: C10, C14, C17, C20, C23, anhydrous pyridine, bis(trimethylsilyl)trifluoroacetamide (BSTFA) (with 1% trimethylchlorosilane, TMCS) and adsorbent Florisil 60/100 mesh were purchased from Sigma–Aldrich Co. (Poznan, Poland).

Repeatability and reproducibility of RI value

The first part of Table 1 gives the RI values obtained on the PE-5ht, PE-5ms, and DB-5 columns. It was observed that three sets of the RI values were obtained on three columns with the same stationary phase. It is also worthy to note that the peaks of some compounds can be moved in relation to each other. For example, the hydroquinone appears earlier than C14 on the PE-5ms column (RIhydroq. = 1397), but it eluted later in the PE-5ht and DB-5 columns (RIhydroq. = 1408, 1410, respectively). The

References (19)

  • Ch. Kraus et al.

    Phytochemistry

    (1997)
  • T.-Y. Chu et al.

    Talanta

    (2001)
  • K. Głowniak et al.

    J. Chromatogr. A

    (1996)
  • D. Ryan et al.

    Anal. Chim. Acta

    (2001)
  • P. Swatsitang et al.

    Anal. Chem. Acta

    (2000)
  • F. Buiarelli et al.

    J. Chromatogr. A

    (1995)
  • R. Causon

    J. Chromatogr. A

    (1997)
  • D. Ryan et al.

    Trends Anal. Chem.

    (1999)
  • L.M. Weaver et al.

    Trends Plant Sci.

    (1997)
There are more references available in the full text version of this article.
View full text
Copyright © 2005 Elsevier B.V. All rights reserved.