The Selective Extraction of Natural Sesquiterpenic Acids in Complex Matrices: A Novel Strategy for Isolating Zizanoic Acid in Vetiver Essential Oil

Abstract

Essential oils are complex mixtures of apolar components, mainly phenylpropanoids, monoterpenes, and sesquiterpenes. Vetiver (Vetiveria zizanioides (L.) Nash) is a non-endemic grass in several tropical regions, widely used for slope stabilization and erosion control because of its long and deep roots that help to bind the soil together, preventing landslides and soil loss. From these roots, vetiver essential oil is obtained, which is extracted and produced worldwide and highly valued for its diverse range of bioactive substances used by the cosmetics and perfume industries. These substances, present in a very complex mixture, are difficult to isolate. Zizanoic acid is a very rare substance in nature and also very interesting because of the biological properties already described. In the present study, zizanoic acid was selectively isolated with 84–87% purity from vetiver commercial essential oils, in which it was present at less than 10%, using KOH-impregnated silica gel column chromatography alone. The experiments were monitored using GC-MS and UHPLC-HRMS, and the isolated substances (zizanoic and valerenic acids) were further determined by NMR experiments. The whole methodology and analytical approach proved to be very efficient for natural product complex mixture analysis and also very selective, allowing for a distinct capacity to recover carboxylic acids from complex biological samples.

1. Introduction

The roots of the grass Vetiveria zizanioides, also known as Chrysopogon zizanioides (L.) Roberty (family Poaceae), are used to extract vetiver essential oil [1]. It is a highly complex oil with several terpenoid derivatives and diverse chemical skeletons [2]. A recently published paper reviewed several aspects of vetiver essential oil’s uses, market, and chemistry [3]. Together with Haiti, Indonesia, China, Madagascar, and Japan, Brazil is one of the major producers of vetiver, which is currently readily available worldwide [4]. With a compound annual growth rate of 9.7%, the global vetiver oil market is expected to reach USD 1560.2 million by 2032 from an anticipated USD 815.6 million in 2025 [5]. One of the primary natural ingredients in the cosmetics business, vetiver essential oil is frequently utilized to create high-end scents [6]. One of the primary uses of vetiver cultivation is to reduce soil erosion, while it is also produced to purify soils that contain heavy metals, herbicides, nitrates, and other contaminants [7,8]. Vetiver oil also presents anti-inflammatory [9], antibacterial [10], and antifungal [11] properties, which are frequently used in traditional medicine.

The chemical composition of this essential oil is very complex. Some of its unique substances are present at such a low concentration that pharmacological studies are rarely performed. The substances present are usually hydrocarbons, sometimes alcohols and ketones, and uncommonly carboxylic acids. Even when they are present, these acids are rarely observed due to the gas chromatographic methods commonly used to analyze essential oils; the methyl-silicone stationary phase does not present a good resolution in this chemical class. Indeed, carboxylic acids are very reactive and active pharmacologically, and developing a way to selectively extract them from complex matrices is crucial to expand the knowledge of natural products’ bioactivities. Zizanoic acid is one such substance, a sesquiterpenoid found mainly in vetiver essential oils in very small quantities [12,13,14]. Not many properties of this substance have been described, except for those analyzed through traditional acid–base separation approaches described below, such as strong antimycobacterial activity, especially against drug-resistant strains of Mycobacterium TB; moreover, it lacks sensory properties for perfumery [15]. Additionally, other pharmacological properties are described, such as anti-inflammatory properties due to its capacity to modulate mediators in the arachidonic acid pathway and reduce oxidative stress [16] and antioxidant properties that can scavenge free radicals and reduce oxidative stress, potentially protecting against cellular damage [17].

Because zizanoic acid is present in vetiver essential oil (VEO) but produced in tiny quantities, it is highly prized [18]. Reports of acidic components in vetiver oils have been observed since the early 20th century [12]. Zizanoic acid was first separated using an acid–base procedure with an aqueous alkaline solution. It was then converted into a methyl ester using diazomethane, purified using fractional distillation, and then analyzed using silica gel chromatography [13]. The methyl ester was then subjected to alkaline hydrolysis to generate the pure acid, which yielded about 0.4%. After chemical degradation revealed the acid’s structure, the molecule was extracted using aqueous Na₂CO₃, esterified, and separated by fractional distillation and crystallization [14]. This approach can lead to several degradations of the molecule. Diazomethane is also a very difficult reagent to produce today since diazald salts have been banned due to their explosive characteristics and toxicity. However, other esterification techniques are available.

Another interesting substance detected during gas chromatographic analysis focused on carboxylic acids in vetiver essential oil was valerenic acid. This is a sesquiterpene with significant pharmacological properties, such as anti-inflammatory and anti-convulsant effects, as well as important actions on the central nervous system (CNS), such as anxiolytic, sedative, and anesthetic effects, even in cases of psychological stress [19]. Its beneficial effects on brain-derived neutrophil factor (BDNF), a vital CNS protein that may aid in neuronal illnesses like Alzheimer’s, have also been investigated, since the modulation of depression and other neurogenic diseases is strongly influenced by BDNF [20,21]. It is usually described as a minor constituent of one of the most important medicinal plants worldwide, Valeriana officinalis, which is extensively marketed and studied. The availability of valerenic acid in other species allows for the isolation and application of more selective and effective extractive methodologies.

The main focus of this study was to develop a selective and efficient way to extract carboxylic acids present naturally in very low concentrations in complex matrices, such as essential oils, as a way to allow their study in pharmacological models and make their medicinal use feasible. To achieve this goal, commercial vetiver essential oils were acquired and subjected to elution using lab-produced silica gel modified with KOH, acting as an ion exchange column in open-air column chromatography. Only three solvents were applied: the first two to separate the non-acid substances, and then one to obtain the pure sesquiterpene carboxylic acid fraction. To verify the entire process, in addition to applying gas chromatography coupled with mass spectrometry to identify the substances at these acid fractions, several experiments using nuclear magnetic resonance (NMR) and ultra-high-resolution liquid chromatography coupled with high-resolution mass spectrometry (UHPLC-HRMS) were carried out.

2. Materials and Methods

2.1. Chemicals and Reagents

Hexane (Hex, 99% PA ACS), methyl alcohol (MeOH 99.89% PA), dichloromethane (DCM), and ethyl acetate (EtOAc, 99.98% PA ACS) were the solvents used in the experiments, and they were purchased from Sigma Aldrich, Inc., São Paulo, Brazil. Commercially available vetiver oils were used in all experiments. Vetiveria zizanioides root oils (5 mL (g)) were stored at 4 °C until sample preparation. Silica gel (0.04–0.063 mm/230–400 mesh, pH 10% in 20 °C) was used to prepare the filtration column and was purchased from Sigma Aldrich, Inc. Hydrochloric acid (HCl, 37% ACS grade) was purchased from Sigma Aldrich, Inc. The solution was prepared in water at a 0.1 M concentration. Sodium chloride (NaCl), ACS Reagent Grade (≥99.0%), was purchased from Sigma Aldrich, Inc. Deionized (DI) water was produced using a Milli-Q system from Millipore (Bedford, MA, USA). Sigma Aldrich, Inc. provided the pH strips.

Preparation of the KOH Solution and Impregnation in Silica Gel

A 5% KOH solution (10 g KOH/200 mL H₂O) was added to 100 g of silica gel G60. The mixture was placed in an oven at 120 °C for 24 h. After this period, the KOH-impregnated silica was cooled and sieved using a 0.25 mesh sieve.

2.2. Filtration Column (Anion Exchange)

The ion exchange column was used to extract essential oils from the root of Vetiveria zizanioides using KOH-impregnated silica gel (Figure 1). A glass open column (44 cm in length; 11.5 cm in diameter) was packed with KOH-impregnated silica gel following the classic procedure using 200 mL of n-hexane. After filling the glass column with silica, 5 g of oil was solubilized in 3 mL of n-hexane and applied to the top. The eluents used in this system were n-hexane, dichloromethane, and methanol.

The obtained fractions were dried with anhydrous sodium sulfate, filtered, and evaporated to dryness using a rotary evaporator. A ratio of 1:4 was used in the filtration column. The sample (5 mL) was adsorbed on a chromatographic column containing 20 g of silica impregnated with KOH. The mobile phases used were hexane for the elution of hydrocarbons, dichloromethane for the elution of molecules of medium polarity, and methyl alcohol for the desorption of acids adsorbed on the column as their respective potassium salts. The following was used in the column: 1 L of hexane + 200 mL of dichloromethane + 300 mL of methanol. Soon after, a liquid–liquid partition was made with the methanolic fraction to recover the acids in potassium salts. A separatory funnel was used for separation by partition. The fraction was acidified with a 10% HCl solution until pH = 4.0. Figure 1 provides a brief illustration of this procedure.

2.3. Percentage Yield Calculation

The extraction yield percentage was based on the calculation of the dry extract of each fraction of the filtration column after the repair process using the funnel. The percentage yield was calculated using a conventional method and determined using the following formula:

$$\mathrm{Y} \mathrm{pec} \left(\%\right)={\displaystyle \frac{\mathrm{Mass} \mathrm{of} \mathrm{dry} \mathrm{fraction} \times 100}{\mathrm{Mass} \mathrm{of} \mathrm{sample} \left(\mathrm{vetiver} \mathrm{oil}\right) \mathrm{used} \mathrm{for} \mathrm{extraction} \left(\mathrm{g}\right) }}$$

2.4. Instrumental Analysis

2.4.1. UHPLC-HRMS Experiments

A Dionex Ultimate 3000 ultra-high-performance liquid chromatography (UHPLC) system coupled to a QExactive hybrid quadrupole Orbitrap mass spectrometer (Thermo Fisher Scientific, Bremen, Germany) equipped with an electrospray ionization (ESI) source was used. Separation was performed in a reversed-phase column (Syncronis 1.7 µm C18, 50 mm; 2.1 mm; 1.7 µm) at 40 °C, with a constant flow rate of 300 μL/min and an injection volume of 5 µL. A gradient chromatographic run started at 5% of mobile phase B (methanol with 0.1% formic acid) and 95% of mobile phase A (water with 5 mM ammonium formate and 0.1% formic acid). Mobile phase B was increased to 10% at 1.0 min, 25% at 2 min, and 90% at 10 min. After reaching 100% of B at 14 min and maintaining this ratio until 16 min, the initial chromatographic condition was restored from 16.1 to 20.0 min.

The LC effluent was pumped to the mass spectrometer operating in a positive and negative ESI mode, calibrated daily with a manufacturer’s calibration solution (Thermo Fisher Scientific, Bremen, Germany). The ESI parameters were further optimized with the final setup: spray voltage of 2.9 kV, S-lens voltage of 80 V, capillary temperature of 380 °C, and auxiliary gas heater temperature of 350 °C. The nitrogen sheath, auxiliary, and sweep gas were set at 30, 10, and 1 arbitrary units, respectively. Full-scan data were acquired in a range of m/z 70–1050 at a resolution of 70.000 full widths at half-maximum (FWHM), automatic gain control (AGC) of 1 × 10⁶, and maximum injection time (IT) of 100 ms. In the full MS approach, in the negative mode, the target exact mass was m/z 233.1547 ([M − H]⁻), and in the positive mode, it was m/z 235.1693 ([M + H]⁺), to C₁₅H₂₂O₂.

Data dependent analysis (DDA) was carried out. Each full-scan mass spectrum was recorded in positive and negative ion mode in profile mode at a resolution of 70.000. The AGC target was 3 × 10⁶ with a maximum injection time of 50 ms. The most intense peaks were selected for HCD fragmentation. Tandem spectra were collected at a resolution of 17.500 with an AGC target of 1 × 10⁵ and a maximum injection time of 60 ms. An isolation window of 1.5 and a normalized collision energy of 30 and 60 were used when triggering a fragmentation event. Data was acquired and processed using Thermo ScientificTM TraceFinderTM 4.1 software (Thermo Fisher Scientific, Austin, TX, USA), with a ±5 ppm mass tolerance.

2.4.2. GC-MS and GC-MS-MS Experiments

A Thermo Scientific 1300 Trace Gas Chromatograph coupled with an ISQ LT single quadrupole MS in a DB-5ms column of 30 m; 0.250 mm ID and 0.25 µm film thickness, 5–phenyl-methylpolysiloxane. The pulsed split injection mode was selected to inject 3 µL of sample into the Ultra Inert Liner (Agilent, 4 mm; 900 µL) at 280 °C, using helium as the carrier gas at a flow rate of 1.0 mL/min and a split ratio of 100:1. The injection pulse pressure was set to 50.0 psi (3.960 mL/min) for 0.30 min, at a rate of 1 mL/min, whereas the purge flow was set to 0.800 mL/min. The initial temperature ramp of the oven started at 60.0 °C for 3 min, and increased to 200.0 °C at a rate of 3.0 °C/min, followed by 290.0 °C at 10.0 °C/min. The total running time was 58.67 min.

For the GC-MS/MS experiments, a Thermo Scientific 1300 Trace Gas Chromatograph coupled with a TSQ 8000 Evo Triple Quadrupole MS was used, equipped with a VF-17ms column of 30 m (0.250 mm ID and 0.25 µm film thickness; 50% phenyl/50% dimethylpolysiloxane). The pulsed split injection mode was selected to inject 1 µL of the sample into the Ultra Inert Liner (Agilent, 4 mm; 900 µL) at 280 °C, using helium as the carrier gas at a flow rate of 1.0 mL/min and a split ratio of 10:1. The injection pulse pressure was set to 50.0 psi (3.960 mL/min) for 0.30 min, at a rate of 1 mL/min, whereas the purge flow was set to 0.800 mL/min. The initial temperature ramp of the oven started at 60.0 °C for 3 min, and increased to 200.0 °C at a rate of 3.0 °C/min, followed by 290.0 °C at 10.0 °C/min. The total running time was 58.67 min.

In both experiments, the MS transfer line temperature was set at 280 °C, and the ion source temperature was kept at 280 °C. The system was operated in EI mode at an energy level of 70 eV. The chromatogram was scanned in SCAN mode, and the mass range was m/z 30 to m/z 700. The identification of compounds was confirmed using the National Institute of Standards and Technology Library, 2017 (NIST17).

2.4.3. NMR Experiments

NMR experiments were carried out using an Agilent 600 MHz spectrometer operating at 599.87 MHz for ¹H and 150.85 MHz for ¹³C, at 298 K, using a 5 mm NMR tube in CD₃OD. The spectra were performed applying the standard pulse sequence. The ¹H NMR spectrum was acquired with a single 45^o radiofrequency excitation pulse, a spectral width of 5.38 kHz, an acquisition time of 1.52 s, 8 k data points, and 16 scans. The ¹³C NMR spectrum was acquired with a single 45° radiofrequency pulse, a spectral width of 30.5 kHz, an acquisition time of 1.05 s, a relaxation delay of 2 s, and 32 k data points. The DEPT 135 pulse sequence was performed using a spectral width of 30.5 kHz. All spectra were processed and analyzed using MestreNova 15 software.

3. Results and Discussion

3.1. Ion Exchange Column Yields

The methodology used to isolate bioactive acids was the ion exchange column, since this type of chromatography is faster and more efficient for this type of separation. The acidic diterpenes were adsorbed onto the silica, while the non-acidic ones were eluted with equal amounts of hexane and dichloromethane. Adding methanol at the end of the column enabled the elution of the terpenic acids.

A particular ion exchange mechanism was used to perform filtration in a column impregnated with KOH. In this instance, a certain ion exchanger had to be connected to the stationary phase. Neutral and apolar substances elute the stationary phase more rapidly, whereas neutral and medium polar substances tend to interact with the silica, resulting in more retention. Finally, in methanol, the initially absorbed ions from the carboxylic acids were liberated as their respective potassium salts. This technique was previously used to remove mixtures of resinoid acids from oil resins, but, as far as we know, it has never been used to selectively extract very rare substances from essential oils [22]. Based on the sample quantity available, the optimal width and height of the column to be utilized in the procedure had to be established. The solvent flow had to remain constant to prevent the sample and stationary phase from drying out in open-column chromatography. These solvents selectively passed along the column as the essential oil components were separated based on their respective affinities. Very apolar substances interacted with hexane through London or Van der Waals forces. Medium polar substances were eluted with dichloromethane. Additionally, intermediate nonpolar compounds with hydroxyls (-OH) or carbonyls (=O) that set them apart from hydrocarbons were extracted by dichloromethane. Methanol was used to remove salts, primarily from the acidic parts of silica impregnated with KOH and complexed with the ion exchange.

After these solvents had passed through, a separatory funnel was employed for partition separation. Since this acidic part was still in its ionized form, the pH of this fraction of the acid salts was 12. Dichloromethane had to be supplied along with water for the initial liquid–liquid separation. The acidic fraction tended to migrate to the aqueous phase since it was in the ionized state, but the organic phase displayed the acidic fraction’s medium polarity remnants. Since the organic phase was physically on the bottom and the aqueous phase was on top of the funnel, because chlorinated solvents have a higher weight, the aqueous phase was on top of the funnel at the end of the first partition separation (Figure 2).

These sesquiterpenes were eluted from the column in the form of potassium salts and then acidified with hydrochloric acid to pH 4 and extracted with dichloromethane.

Table 1. Extraction yields of each fraction of the filtration column.

Vetiveria zizanioides Root Oil	F. Hexane (g)	F. Dichloromethane (g)	F. Methanol (g)
V1	3.83	0.24	0.22
V2	2.99	0.65	0.12

F = Fraction.

presents the masses of the yields of each fraction obtained.

Three commercial vetiver oil samples were coded V1 and V2. The methanolic fractions, which comprise the acidic parts of vetiver essential oil, showed a recovery percentage of 2.4–4.4% when compared to the initial samples weighing 5 g.

3.2. Identification in GC-MS

After being extracted from the separatory funnel, the acidified fractions of commercial vetiver oils were examined. In Figure 3A, the chromatogram from the essential oil shows a red arrow at the zizanoic acid. As seen in Figure 3B, it was noteworthy that a major peak was present at the same retention time, which was between 43.20 and 43.90 min. It was feasible to detect the presence of an acid compound with the same property by using the mass spectrometer and knowing that the mass of zizanoic acid that we searched for matched m/z 234.

GC-MS confirmed the presence of zizanoic acid based on its characteristic fragments: m/z 219 (42.57%), m/z 164 (80.65%), m/z 145 (98.99%—base peak), m/z 119 (100%), and m/z 91 (60.05%). Fragments with an intensity above 10% of the spectrum were considered. The molecular ion at m/z 234 was detected but showed low intensity. By comparing the obtained spectrum with that from the NIST library, 90% similarity was found, which supports the presence of zizanoic acid in the samples. Another peak, less intense than the first one, was detected between 46.40 and 46.60 min, with m/z 234 as the molecular ion and fragments at m/z 161 (base peak), m/z 133 (36.10%), m/z 105 (80.00%), and m/z 91 (60%). The fragmentation profile and the NIST library confirm the presence of valerenic acid [23].

Table 2. Percentages of zizanoic and valerenic acid in the acid fraction.

	Zizanoic Acid			Valerenic Acid
Sample	Rt (min.)	Area	%	Rt (min.)	Area	%
V1	43.92	580,366,466	84.18	46.42	57,105,897	8.28
V2	43.64	203,248,113	87.07	46.36	15,083,369	6.46

displays the values of the regions that correlate to the peak at the retention time, which fell between 43.60 and 43.92 min for zizanoic acid and another main acid peak. This other substance, present at 46.36–46.42 min, was identified as valerenic acid. Table 2 shows the percentage areas of zizanoic and valerenic acid after extraction. These percentages were obtained by area normalization.

The extraction yield was determined based on the percentage of peaks in the total ion chromatogram, both before and after extraction. Gas chromatography coupled with mass spectrometry was widely employed for analyzing volatile compounds, such as those found in essential oil samples. Electron ionization, utilized as the ionization method, has the advantage of producing mass spectra with numerous fragments that facilitate identification through comparison with spectral libraries, as well as structural elucidation based on the fragmentation pattern of the compound. The extraction yields for samples V1 and V2 were 92.46% and 93,53%, respectively.

3.3. Identification in GC-MS/MS

The presence of zizanoic and valerenic acid was confirmed using GC-MS/MS with a medium polarity stationary phase. On the VF-17ms column, valerenic acid was eluted at 42.66 min and zizanoic acid at 48.70 min in all three samples, V1, and V2. The following fragments characterized valerenic acid: m/z 234 (27.00%), m/z 161 (100%), m/z 133 (36.24%), m/z 105 (82.06%), and m/z 91 (63.00%). The similarity percentage was greater than 80%. Zizanoic acid was identified by its fragments at m/z 234 with less than 10% abundance, m/z 219 (65.98%), m/z 164 (80.93%), m/z 145 (100%) as the base peak, m/z 119 (82.10%), and m/z 91 (49.51%). The match percentage was greater than 90% for all three samples compared to the NIST library spectrum.

Methyl zizanoate was detected, as E. Belhassen et al. previously described [6], in both samples. The characteristic fragments resulting from the electron ionization of this ester were m/z 248 (molecular ion), m/z 233 (52.99%), m/z 188 (91.70%), m/z 145 (100%), and m/z 119 (61.27%). The similarity percentage was greater than 90% for both samples.

3.4. Identification in UHPLC-HRMS

The presence of zizanoic and valerenic acid was confirmed by ultra-high-performance liquid chromatography coupled with high-resolution mass spectrometry (UHPLC-HRMS), an orthogonal technique to GC-MS. The high-resolution analysis detected both acids as pseudomolecular ions in the three samples, with a mass error of 0.01 ppm. In the negative ESI mode, zizanoic acid was identified; it was detected as the most abundant peak in the total ion chromatogram, appearing as a single peak upon ion extraction and identified by DDA set to 30. Even when the DDA experiment was set to 60, there was no fragmentation of zizanoic acid. DDA in mass spectrometry is a technique where the mass spectrometer dynamically selects and fragments ions based on their abundance in a survey scan. This means that the most abundant precursor ions are chosen for subsequent fragmentation, resulting in MS/MS spectra for those specific ions.

The extraction of the zizanoic acid ion from the TIC derived from the commercial sample and then from the acid fraction of the essential oil is shown in Figure 4. It is evident that the target compound-related peak in both situations was selective, exhibited a somewhat narrow band, and showed the filtration column’s high efficiency. Due to a solubility incompatibility caused by using dichloromethane as the solvent to prepare the samples injected into the HPLC and the presence of methanol in the mobile phase, very broad bands were formed, and the TIC baseline was raised. To correct this problem, the samples of the acid fraction were subsequently solubilized in methanol, ensuring better compatibility with the mobile phase and avoiding a new elevation of the baseline.

In the positive ESI mode, valerenic acid was identified. According to Shukla et al., valerenic acid could be identified by the pseudomolecular ion m/z 235.1693, fragment ions m/z 217.1595, m/z 189.1637, and m/z 123.1174 [24]. Valerenic acid was identified in both samples.

Table 3. Valerenic acid identification data.

Shukla et. al. [24]	V1 Sample		V2 Sample
Fragments Ions (m/z)	Fragments Ions (m/z)	Error (Δppm)	Fragments Ions (m/z)	Error (Δppm)
235.1693	235.1697	1.62	235.1696	1.15
217.1595	217.1591	1.70	217.1590	2.30
189.1637	189.1642	2.80	189.1642	2.54
123.1174	123.1173	0.81	123.1172	1.54

presents the exact mass of the pseudomolecular ion and the four most intense fragments, along with their associated mass errors.

The results of area normalization applied to UHPLC-HRMS confirm those obtained by GC-MS, despite the differences in the ionization mode principles. In the acidic fractions of samples V1 and V2, the amounts of zizanoic acid were 90% and 92%, respectively. For valerenic acid, the values obtained were 7% in sample V1 and 5.8% in V2.

The results from both analytical methods demonstrate the selectivity provided by the extraction approach used on the crude sample. Gas chromatography, used for analyzing volatile compounds detected by electron ionization (EI), and liquid chromatography, applied to polar compounds detected by high-resolution electrospray ionization (ESI), revealed complementary profiles. A comparison between the total ion chromatograms (TICs) of the crude samples and those obtained after extraction demonstrates the efficiency of the extraction method.

3.5. Identification in NMR

NMR spectroscopy was used to achieve the structural elucidation of zizanoic acid. The results were compared with data from the literature [13,14,25]. The limited sensitivity of NMR compared to analytical techniques utilizing chromatography coupled with mass spectrometry permitted only the identification of zizanoic acid, the major component, in the acid portion of vetiver essential oil. Figure 5 shows the ¹H NMR spectrum of sample V2, where the zizanoic acid was identified, and its structure numbered according to IUPAC.

Figure 6 shows the ¹³C NMR spectrum confirming the structure of zizanoic acid, the major compound in sample V2. The carbon of the carboxylic group presented a chemical shift at 180.09 ppm. The olefinic carbons appeared at 106.29 and 157.29 ppm. Six methylene groups (36.69; 32.35; 25.62; 25,60; and 24.92 ppm) and three CH (51.03; 48.89; and 48.18 ppm) were identified by the DEPT 135 experiment. Two methyl groups were observed at 24.94 ppm and 27.50 ppm.

The most deshielded signals in the ¹H NMR spectrum were attributed to geminal protons from C13: 4.60 and 4.74 ppm. The signals at 2.70 ppm and 2.61 ppm were attributed to H2 and H10, respectively. The signals on the region from 2.0 to 1.0 ppm were assigned to five methylene diasterotopic groups. The two methyl groups were detected at 1.04 and 1.05 ppm. These values are in accordance with those in the literature [13,14,25].

4. Conclusions

The ion exchange system operating with two cleaning solvents (hexane and dichloromethane), and further use of methanol to elute the acid fraction in an open-column chromatograph filled with silica modified with KOH, was successfully applied and found to be a selective and efficient way to separate zizanoic and valerenic acids from the complex matrix of Vetiveria zizanioides essential oil, in which they are very minor substances. The isolation process was achieved in only one step with very high efficiency. These bioactive acids are of great medical interest because of their many effects on the central nervous system. The identification was fully performed with the high-resolution methods of NMR and mass spectrometry. This development opens several possibilities to isolate and study minor constituents from several complex matrices from the oil and food industries.