Applications Using GC×GC

Introduction to Applications
 

The GC×GC system is a powerful tool for the analysis of natural substances with complex matrices that are difficult to analyze by conventional GC or GCMS. Some applications are introduced below.

 

 

Data on this page was provided by the group run by Professor Luigi Mondello, University of Messina, Italy. An interview with Professor Mondello is featured in our section, A bridge with our customers.

 

 

Analysis of pesticides in drinking water by using SPME-GCxGC-quadMS
 

In this application the performance of the system was evaluated in the analysis of high-molecular weight compounds, namely pesticides. For such a scope, a wide mass range was applied (50-440 m/z) and an in-sample SPME-GCxGC-quadMS method was developed, then validated, for the analysis of 28 drinking-water pesticides. Chromatographic conditions were optimized by tuning the operational parameters in the analysis of a standard solution at the 50 μg/L level. The head pressure generated gas linear velocities of circa 20 cm/s and 215 cm/s in the first and second dimension, respectively. The negative offset (10°C) used in the second dimension enabled a more homogeneous occupation of the 2D space. A TIC SPME-GCxGC-quadMS chromatogram is illustrated in Figure 2.1. Peak identification was achieved by using a dedicated MS Pesticide library, characterized by the presence of LRI data used to filter out incorrect matches.

 

Figure 2.1 TIC SPME-GCxGC-quadMS chromatogram of a pesticide standard solution at the 10 ppb level. Refer to Table 1 for peak assignment.

 

The GC-MS and GCxGC-MS softwares used enable the application of a twin-filter: (i) spectral similarity and (ii) LRI range. The primary filter eliminates matches with a spectral similarity (expressed in %) lower than a minimum value set by the analyst (90% in this case); the other filter eliminates library spectra, characterized by an LRI value outside a predefined range (±20 in this case), and with respect to the LRI value of the unknown pesticide. The LRI range was determined by analyzing first the C7–C30 alkane mixture and then the 28 pesticides.
An example of the validity of using a double filtered search, in a GCxGC-quadMS analysis, is shown in Figure 2.2: with only the % similarity filter, the database search generated acceptable matches for vernolate and pebulate, eliminating butylate. The application of the LRI filter greatly simplified the search procedure, eliminating pebulate from the "hit" list, leaving only the correct match (vernolate).

 


ChromSquare window showing the LRI window applied for the library search of vernolate. Spectra of butylate (I), vernolate (II), and pebulate (III).

 

Apart from retention time precision, a series of parameters regarding SPME-GCxGC-quadMS method validation were considered (Table 2.1): the calibration curves, estimated using the least squares method, for the 28 pesticides were linear up to 100 mg/L for 21 pesticides, whereas eight were linear up to 50 mg/L. The lack of linearity in the latter cases was clearly due to two factors, namely, analytes competition and fiber saturation. In general, LOQ values were distributed over a rather wide concentration range (0.003–0.084 mg/L). Peak area precision (n=5) was measured for each pesticide at 10 mg/L: CV% values were slightly high, generally in the range of 5–15% and were due to the fact that the SPME process was carried out manually. Accuracy (n=5) was also determined for each pesticide at the 10 mg/L level: relative % error values were acceptable, namely within the range -11.3 to +12.7%.

 

Table 2.1 With regard to the MS parameters a scan frequency of 33 Hz was applied with a scan speed of 20 000 amu/s and a mass range of m/z 50–450, to embrace all the MW values of the pesticides analyzed. The consistency of mass spectra profiles across the upper peak half was evaluated by measuring ratios for significant ions relative to 6 pesticides; CV% were lower than 15% (Table 2.2). The narrowest peak in the chromatogram was heptachlor (base width (4σ) of circa 300 msec) and it was re-constructed with 10 data points. A series of parameters regarding SPME-GCxGC-quadMS method validation were considered and are reported in Table 2.1 and 2.2.

 

 

Table 2.2. Peak widths at half height, number of peak data points (half height), averaged library match (MS%) values (MS% range in parenthesis refer to values derived at each data point across the peak widths at half height), ratios for significant ions (CV% values reported in parenthesis refer to ratio values derived at each data point across the peak widths at half height), for six pesticides.

 

Analysis of Mineral Oil Aromatic Hydrocarbons (MOAH)
 

Mineral oil products derive from crude oil, through distillation processes and various refining steps, and contain proportions of mineral oil saturated hydrocarbons (MOSH) and mineral oil aromatic hydrocarbons (MOAH). The occurrence and danger of mineral oil products in foods has been discussed in recent years. It is said that contamination derives from the printing inks applied directly to the packaging and/or from the ink used in newspapers, employed to produce recycled fiber. This article presents the GCxGC analysis of MOAH fractions of pasta.
When analyzing mineral oil, MOSH and MOAH fractions are separated, and then each component is quantified. The fractions separated with off-line SPE using an Ag silica-gel SPE cartridge contain the peaks in addition to the target components in the chromatogram. These peaks can include information on the source of contamination.
The figure below shows the GCxGC chromatogram of a MOAH fraction for pasta purchased in a supermarket. The series of peaks in the MOAH fraction were identified as esterified fatty acids. The tentatively identified peaks are listed in the table below. MOAH fractions were quantified to be 1.6mg/Kg (< C25) with GCxGC -FID; attention was paid, during integration, to eliminate the "unknown" peaks. The esterified fatty acids derived from the paperboard packaging. In fact, in a sample of pasta analyzed prior to box packaging, no sign of MOAH contamination was observed.

 

GCxGC-qMS chromatogram of MOAH fraction for pasta (1st column: SLB-5msiL=30m, i.d.=0.25mm, df=0.25μm), 2nd column:F Supelcowax-10 iL=1m, i.d.=0.1mm, df=0.10μm), Modulation time: 6 sec )

 

No Compound No Compound No Compound
1 Isopropyldodecanoate 2 Dioctylether 3 2-Ethylhexyl octanoate
5 Isopropyltetradecanoate 9 Methylhexadecanoate 10 Ethylhexadecanoate
11 Isopropylhexadecanoate 12 Abietatriene 13 Octyldodecanoate
15 Methyloctadecanoate 16 Dodecyloctanoate 17 n-Butylhexadecanoate
18 Octyltetradecanoate 19 Tetradecyloctanoate 20 n-Butyloctadecanoate
22 Octylhexadecanoate 23 Di(ethylhexyl) phthalate 25 Squalene
26 1-Hexacosanol @ @ @ @

 

GCxGC-MS/MS Analysis of Mandarin Essential Oil
 

Shimadzu’s triple quadrupole GCMS-TQ8030 is capable of operating under high-speed conditions in both scan and MRM modes. Moreover, this instrument can generate simultaneous scan/MRM data, also in a rapid manner.
A GCxGC-MS/MS method was developed for simultaneous scan qualitative analysis of untargeted essential oil and MRM qual/quantitative analysis of target ones. SLB-5ms was used as the 1st column and IL-60 as the 2nd column.
The figure below is the scan GCxGC-MS/MS 2D chromatogram expansion for mandarin essential oil. The spectral matched 16 mono/sesquiterpenes are reported in the table.

 

The scan GCxGC-MS/MS 2D chromatogram for mandarin essential oil and the peak identification (modulation time: 5 sec)

 

 

 

 

No Compound No Compound No Compound
1 citronellal 2 terpinen-4-ol 3 α-terpineol
4 decanal 5 neral 6 geranial
7 perillaldehyde 8 thymol 9 linalool isobutyrate
10 α-copaene 11 dodecanal 12 methyl, N-methyl anthranilate
13 (E,E), α-farnesene 14 δ-cadinene 15 caryophyllene oxide
16 δ-sinensal @ @ @ @

 

MRM quantitative analysis of three preservatives, such as o-phenyl phenol (OPP), butylated hydroxytoluene (BHT) and butylated hydroxyanisole (BHA), was performed. A calibration curve was constructed over the 0.1 ppm-100 ppm range for OPP, while another covered the 0.5 ppm-200 ppm range for BHA and BHT. The figure below is the MRM GCxGC-MS/MS 2D chromatogram at the 1ppm level. An average OPP concentration corresponding to 57.0 ppm was found and no trace of either BHT or BHA was found in this mandarin oil (LODs of BHT and BHA were 3ppb and 11ppb, respectively).

 

MRM GC~GC-MS/MS 2D chromatogram for standard solution at the 1ppm level IS (1,4-dibromobenzene), BHT, OPP and BHA.

 

GCxGC-MS/MS has the potential to chromatographically separate the overlapped MRM peaks even in conventional GC/MS/MS analysis.

 

Analysis of Mate Tea
 

Mate tea is widely consumed in the countries of South America as a tonic and stimulating beverage to overcome fatigue. We performed GC×GC analysis of the volatile components in a mate tea beverage (Ilex paraguariensis leaves and twigs) purchased in the Brazilian market.
GC×GC-qMS Chromatogram of Mate Tea
(First column: SLB-5ms (L=30 m, I.D.=0.25 mm, df=0.25 μm); Second column: Equity 1701(L=1.5 m, I.D.=0.1 mm, df=0.1 μm), modulation time 6 s)

 

 

The first column is a micropolar column and the second column is a mid-polarity column with dimensions suitable for fast analysis. An extremely large number of components were detected in the 2D chromatogram obtained. Hydrocarbon group components were detected in the lower part of the 2D diagram (low-polarity region). Caffeine was also detected as a prominent compound.

 

Comparison of GC×GC and Single GC Analyses
 

  Detected peaks Identified peaks
GC×GC-MS 1000 or more 241
Single GC-MS 200 70

 

A library search using the mass spectrum identified 241 of the over 1000 peaks detected. It is apparent that GC×GC is an effective means of analyzing complex samples.

 

GC×GC-qMS Chromatogram of Mate Tea and Identification Results

 

 

No Compound Name No Compound Name No Compound Name
20 4-hydroxy-2-butanone 30 5-methyl-3-methylene-5-hexen-2-one 40 alpha-pinene
5 Isopropyltetradecanoate 9 Methylhexadecanoate 10 Ethylhexadecanoate
21 methylpyrazine 31 2-heptanone 41 2-octanone
15 Methyloctadecanoate 16 Dodecyloctanoate 17 n-Butylhexadecanoate
22 furfural 32 nonane 42 2-heptenal
23 isovaleric acid 33 4-heptenal 43 2,2-dimethyl-3-heptanone
24 (2E)-hexenal 34 2-butoxyethanol 44 5-ethyl-2(5H)-furanone
25 2-allylfuran 35 2,4-hexadienal 45 5-methyl furfural
26 (2Z)-hexenal 36 2(5H)-furanone 46 benzaldehyde
27 furfuryl alcohol 37 gamma-butyrolactone 47 hexanoic acid
28 hexanol 38 pyrazine 38 pyrazine, 2,5-dimethyl- 48 3-methyl-2(5H)-furanone
29 pentanoic acid 39 2,7-dimethyloxepine 49 1-octen-3-ol

 

This application data was provided by the group run by Professor Luigi Mondello, ( University of Messina, Italy, Chromareleont S.r.l. )

 

Fatty Acids in Blood Plasma

Fats in foods are attracting attention due to their deep relationship to a series of diseases including hypertension, heart disease, obesity, and hypercholesterolemia. They have been widely researched using chromatography in recent years. However, conventional methods suffer from several problems: 1) inability to identify the double-bond position in fatty acid isomers due to the similarity of the mass spectra, 2) poor GC resolution, and 3) inability to detect trace peaks due to poor sensitivity. In this example,a high-resolution, high-sensitivity GC×GC method was applied to the determination of fatty acid methyl esters in human blood plasma.

2D Chromatogram of Fatty Acid Methyl Esters in Blood Plasma

(First column:SLB-5ms(L=30m, i.d.=0.25mm, df=0.25μm), Second column:Supercowax-10(L=0.95m, i.d.=0.1mm, df=0.1μm), modulation time:6sec)

 

Peak FAME Peak FAME Peak FAME Peak FAME
1 C8:0 18 a-C19:0 35 C18:2ω6 (st) 52 C22:4ω6
2 C9:0 19 C19:0 36 C20:2 53 C22:4ω3
3 C10:0 (st) 20 C20:0 (st) 37 C20:2ω6 (st) 54 C24:4ω6
4 C11:0 (st) 21 C21:0 (st) 38 C22:2ω6 (st) 55 C20:5ω3 (st)
5 C12:0 (st) 22 C22:0 (st) 39 C24:2ω6 56 C20:5ω1
6 i-C14:0 23 C23:0 (st) 40 C18:3ω6 (st) 57 C21:5
7 C14:0 (st) 24 C24:0 (st) 41 C18:3ω3 (st) 58 C22:5ω6
8 i-C15:0 (st) 25 C14:1ω5 (st) 42 C18:3 59 C22:5ω3 (st)
9 a-C15:0 (st) 26 C16:1ω7 (st) 43 C19:3 60 C24:5ω3
10 C15:0 (st) 27 C17:1ω7 (st) 44 C19:3ω6 61 C24:5
11 i-C16:0 (st) 28 C18:1ω9 (st) 45 C20:3ω6 (st) 62 C20:6ω1
12 C16:0 (st) 29 C19:1 46 C20:3ω3 (st) 63 C22:6ω3 (st)
13 i-C17:0 (st) 30 C20:1ω9 (st) 47 C22:3ω6 64 C23:6
14 a-C17:0 31 C22:1ω9 (st) 48 C18:4ω3 65 C24:6ω3
15 C17:0 (st) 32 C24:1ω9 (st) 49 C20:4ω6 (st)    
16 i-C18:0 33 C16:2ω6 50 C20:4ω3 (st);    
17 C18:0 (st) 34 C17:2 51 C21:4    

 

It is apparent that the FAME peaks corresponding to the carbon number (C), double bond number (DB), and double bond position (ω) are regularly distributed on the 2D chromatogram. This spatial distribution arrangement is extremely effective for the identification of compounds. Of the 65 peaks, 29 could be identified based on this arrangement. (The peaks labeled (st) in the diagram result from standard samples.) In addition, low levels of fatty acids with odd carbon numbers were also detected.

 

This application data was provided by the group run by Professor Luigi Mondello, ( University of Messina, Italy, Chromareleont S.r.l. )  

 

 

Analysis of Coffee

The aroma of roasted coffee is characterized by the presence of several thousand types of volatile compounds, mainly belonging to the pyran, pyrazine, and pyrrole group. The olfactory sensitivity, concentration, and chemical characteristics differ from type to type and the mutual interactions between them determine the characteristic aroma of the coffee. We used GCGC-MS to analyze the volatile components in coffee beans that are hard to analyze by conventional GC due to their extremely complex compositions.

2D Chromatogram of the Volatile Components in Arabica Coffee

(First column:Supercowax-10 (L=30m, i.d.=0.25mm, df=0.25μm), Second column:SPB-5ms(L=1.0m, i.d.=0.1mm, df=0.1μm), modulation time:6sec)

A polar-nonpolar column pair was used for this analysis. Several thousand peaks on the two-dimensional plane,and we obtained a good plot of the extremely complex coffee aroma.

 

2D Chromatogram of the Volatile Components in Arabica Coffee

It is apparent that 14 types from the pyrazine group form according to the side-chain carbon number and are aligned as horizontal bands.

This application data was provided by the group run by Professor Luigi Mondello, ( University of Messina, Italy, Chromareleont S.r.l. )

 

 

No Compound No Compound No Compound
1 Isopropyldodecanoate 2 Dioctylether 3 2-Ethylhexyl octanoate
5 Isopropyltetradecanoate 9 Methylhexadecanoate 10 Ethylhexadecanoate
11 Isopropylhexadecanoate 12 Abietatriene 13 Octyldodecanoate
15 Methyloctadecanoate 16 Dodecyloctanoate 17 n-Butylhexadecanoate
18 Octyltetradecanoate 19 Tetradecyloctanoate 20 n-Butyloctadecanoate
22 Octylhexadecanoate 23 Di(ethylhexyl) phthalate 25 Squalene
26 1-Hexacosanol @ @ @ @

 

 

For Research Use Only. Not for use in diagnostic procedures.

This page may contain references to products that are not available in your country.
Please contact us to check the availability of these products in your country.