Determination of Undesirable Volatile Organic Compounds in Petroleum-Derived Products by Thermal Desorption and Gas Chromatography-Mass Spectrometry Technique ^†

Abstract

Petroleum remains the principal raw material for the manufacture of a wide variety of products with multiple uses. Within this framework, it is possible to find petroleum-derived products (PDPs), such as waxes, with applications in the agri-food industry. Due to the above-mentioned use, the adaptation of the quality control of these products to the requirements of this industry is a factor to be considered by its producers to comply with food safety standards. The present study is focused on establishing the starting conditions for the development and optimization of an analytical methodology based on TD-GC-MS (Thermal Desorption-Gas Chromatography-Mass Spectrometry) that allows for the determination and identification of volatile organic compounds (VOCs) that are undesirable not only because of their smell but also because they may pose a health risk due to prolonged exposure. The desorption temperature was studied, and 80 °C was selected as the initial sample heating temperature to guarantee the integrity of VOCs. A total of 10 compounds, relevant from a food safety point of view, were identified. The n-alkanes were the most represented chemical family in the studied samples.

1. Introduction

Petroleum-derived products (PDPs) are essential for manufacturing a wide range of products with multiple applications in different fields. Among PDPs’ numerous usages, it is possible to find some of them in the agri-food sector, mainly as food-packaging material such as paraffin waxes used to preserve the humidity and slow down the deterioration of some cheeses, for example [1].

Generically, petroleum waxes can be defined as a solid wax at room temperature that is obtained in the refining of lubricating oils. In addition to the use mentioned above, petroleum waxes are used in the manufacture of a wide variety of products, the most well-known use being in the manufacture of candles [2]. In recent years, consumers have become increasingly demanding of food quality due to a greater degree of knowledge about biotic and abiotic contaminants and their direct consequences on health. For this reason, companies in charge of manufacturing and distributing petroleum waxes for food-packaging applications must have quality control programs to ensure their products are in compliance with safety and hygiene standards. Because of this, food-grade waxes are subjected to a process of hydrotreatment at a high pressure and temperature [1,3] to meet the requirements of the U.S. Food and Drug Administration (FDA) 21 CFR 172.886 [4] for their use in food and 21 CFR 178.371 [5] for direct contact with food.

One of the principal parameters to be evaluated in quality control is smell. First, petroleum waxes with applications in the agri-food industry must be odorless to avoid interaction with the overall aroma of the food and, on the other hand, to prevent aroma-producing compounds from being transferred to the product [2,6]. Petroleum waxes’ aromatic and volatile profile was studied by Men et al. (2018) by using the global profile Headspace Electronic Nose (HS eNose) technique [6]. Nevertheless, global profile techniques only provide overall information on samples’ volatile composition [7].

It is interesting to study which VOCs are present in petroleum waxes because they could be transferred to the food during food processing and storage. In addition, the individual identification of VOCs is important for assessing possible health risks from consumption and/or prolonged exposure to these compounds. Within this context, Durrett (1966) [8] applied the headspace (HS) coupled to gas chromatography with a flame ionization detector (GC-FID) to identify the compounds involved in the odor of industrial petroleum waxes. Headspace is one of the most widely used VOC extraction techniques for evaluating the aromatic profile of samples. However, the headspace technique has certain disadvantages. For example, its sensitivity is limited to ppm concentrations, and it does not allow the detection of higher boiling volatiles and semi-volatile compounds. Thus, the direct thermal desorption (TD) technique coupled to GC-MS is currently gaining interest since it is able, with little sample preparation, to detect a wide range of volatiles and semi-volatile compounds with high sensitivity (in the order of ppb).

The present work is focused on establishing the heating temperature and time in the thermal desorption unit, as well as the chromatographic separation conditions, for the development of an analytical methodology based on TD-GC-MS (Thermal Desorption-Gas Chromatography-Mass Spectrometry) that allows for the characterization of VOCs in petroleum wax packaging. This type of analysis avoids sample handling and reduces waste generation. To the best of our knowledge, this is the first time that TD-GC-MS has been applied to the analysis of VOCs from paraffin-based packaging for foods.

2. Materials and Methods

2.1. Petroleum Wax-Based Packaging Samples

Different kinds of red petroleum wax used for food packaging were purchased at local markets available to consumers in Spain. Specifically, cheese and cheese for children’s petroleum wax-based packaging were used in this study. Before the analysis, each sample was wrapped in aluminum foil and stored individually in an odorless glass container, perfectly sealed, and labeled.

2.2. Analysis of VOCs in Petroleum Wax-Based Packaging by TD-GC-MS

The analysis of the VOCs of the petroleum wax-based samples was carried out by the TD-GC-MS technique. In total, 1.5 mg of each sample was placed together with glass wool in an empty glass sample tube of length 90 mm and outer diameter 1/4″ (Shimadzu Scientific Instruments, Kyoto, Japan) for thermal desorption. No organic solvent is required in sample preparation, so the thermal desorption was performed directly on the petroleum wax-based packaging. Sample tubes were thermally desorbed by an automated thermal desorption unit (TD-20; Shimadzu Scientific Instruments, Kyoto, Japan) coupled to a gas chromatograph with a triple quadrupole (Q3) mass spectrometry detector (GC-MS TQ8040; Shimadzu Scientific Instruments, Kyoto, Japan).

Thermal desorption of the sample tubes was performed in a two-step mode. First, the sample tube was heated to 80 °C for 10 min. VOCs were desorbed in a flow rate of 60 mL/min of He (5N grade), collected into a 50–60 mg Tenax^TM (Shimadzu Scientific Instruments, Kyoto, Japan) TA cryogenic trap, and cooled at −15 °C. In the second step, trap desorption was carried out at 280 °C for 3 min and desorbed VOCs were transferred (split 1:50) to a Silicosteel^® (Shimadzu Scientific Instruments, Kyoto, Japan) transfer line heated to 250 °C. The transfer line introduced the VOCs into the GC-MS system. The GC was equipped with a BPX5 capillary column (length 30 m; internal diameter 0.25 mm; film thickness 0.25 μm; SGE^TM Analytical Science, Melrose Park, NSW, Australia). Helium (5N grades) was the carrier gas at a linear velocity flow mode of 0.94 mL/min. The GC oven temperature program for VOCs analysis started a 40 °C (holding time 5 min), then it increased to 220 °C at 3 °C/min (holding time 15 min), and finally increased to 270 °C at 40 °C/min (holding time 2 min). The MS ion source and interface temperatures were 200 °C and 275 °C, respectively. In addition, electron ionization at 70 eV was the ionization mode, and MS was run in a Q3 Scan mode within a 50–600 m/z range. The GC-MS Postrun Analysis software (Shimadzu Scientific Instruments, Kyoto, Japan) was used for the chromatographic treatment of the data as well as to carry out compound identification.

2.3. VOCs Identification

VOCs were identified by comparing the mass spectrum from each chromatographic peak with the mass spectrum of the compounds in the NIST 2014 library version (National Institute of Standards and Technology, Gaithersburg, MD, USA) using the GC-MS Postrun Analysis software (Shimadzu Scientific Instruments, Kyoto, Japan). Furthermore, a saturated alkane standard (C₇–C₄₀; Sigma-Aldrich, St. Louis, MO, USA) was analyzed using the same method to calculate the retention index to identify the VOCs by comparison with the NIST library or the literature.

3. Results and Discussion

3.1. Sample Heating Temperature

The heating temperature and time are important variables since they determine the content of VOCs desorbed from the sample, then transferred from the sample to the Tenax^TM TA cryogenic trap. No references related to the determination of VOCs in paraffin-based packaging materials by TD-GC-MS were found in the literature. Thus, the heating temperature was evaluated to achieve the right conditions for the determination of VOCs in the samples. The following sample heating temperatures were investigated: 80 °C, 100 °C, and 150 °C. The sample heating time was set to 10 min. The results obtained indicated that the sample heating temperature directly affects the peak area along with the absolute intensity. The heating temperature clearly conditioned the resulting chromatograms; more intense chromatographic peaks were obtained at the highest temperature; however, no additional compounds were detected using higher temperatures. Therefore, 80 °C was selected as the initial sample heating temperature to guarantee the integrity of the VOCs.

3.2. Identification of VOCs in Petroleum Wax-Based Packaging

More than 30 VOCs were detected in the samples from the packaging materials from both cheese and cheese for children’s petroleum wax-based packaging. A total of 10 compounds were selected because they are relevant from a food safety point of view.

Table 1. List of the identified compounds in cheese and cheese for children’s petroleum wax-based packaging with their corresponding chemical family group, identifier (ID), retention time (RT), calculated retention index (Calc I), and theoretical retention index (Lit I).

			Cheese Petroleum Wax-Based Packing		Cheese for Children Petroleum Wax-Based Packing
Family	ID	Compound	RT (min)	Calc I	RT (min)	Calc I	Lit I
Aldehydes	1	Nonanal	21.943	1112	n.d. 4	n.d. 4	1104 5
n-Alkanes	2	Dodecane 1	26.290	-	26.270	-	-
n-Alkanes	3	Tridecane 1	31.009	-	30.990	-	-
n-Alkanes	4	Tetradecane 1	35.456	-	35.432	-	-
n-Alkanes	5	Pentadecane 1	39.641	-	39.613	-	-
Phenol derivatives	6	BHT 2	39.986	1506	39.965	1504	1512 6
n-Alkanes	7	Hexadecane 1	43.593	-	43.574	-	-
n-Alkanes	8	Heptadecane 1	47.351	-	47.324	-	-
n-Alkanes	9	Octadecane 1	50.931	-	50.894	-	-
Phthalates	10	DBP 3	n.d. 4	n.d. 4	56.620	1900	1973 7

¹ Compound identified by comparison with a commercially available standard. ² BHT—Butylated hydroxytoluene (2,6-di-tert-butyl-4-methylphenol). ³ DBP—Dibutyl phthalate. ⁴ n.d.—Not detected. ⁵ NIST Library (2014 versions). ⁶ Kotowska, et al. (2012) [9]. ⁷ Dickschat et al. (2004) [10].

shows the VOCs identified by using the retention index and MS library matching criteria and comparing the chromatographic and mass spectra with commercial standards.

As can be seen in Figure 1, the chromatographic profile of the studied petroleum wax-based packaging is very similar. Nevertheless, there are some differences, such as the height and area of the chromatographic peaks. In general, it was observed that the most represented chemical family in both chromatograms is the n-alkanes group. Specifically, the C₁₂-C₁₈ alkanes were positively identified in both samples. This result was within our expectations since petroleum waxes are solids of a complex mixture essentially constituted of paraffin hydrocarbons. On the other hand, BHT (butylated hydroxytoluene) was found in all the petroleum wax-based packaging samples. Undoubtedly, BHT is a widely used antioxidant in foods and food-related products, such as packaging, thus, the exposure of society, as well as the environment, to this substance is likely to happen [11]. Although it has not been listed as a cancerogenic compound, there are many studies related to its negative effects on animal models [12,13]. Nonetheless, the information about BHT exposure in humans is limited. Currently, an ADI (acceptable daily intake) of 0.25 mg/kg BW (body weight) was established by the EFSA (European Food Safety Authority) [14]. DBP (dibutyl phthalate) was also identified in the cheese for children’s petroleum wax-based packaging. This substance is commonly used as a plasticizer in a great variety of consumer products. To date, no information is available regarding the effects of exposure to DBP on humans. Therefore, it is not considered to be a carcinogen for humans and is classified by the EPA (U.S. Environmental Protection Agency) as a Group D (not classifiable as to human carcinogenicity) compound [15,16]. For its part, the EFSA has established a TDI (tolerable daily intake) of 10 µg/kg BW for DBP [17]. Conversely, nonanal was detected in the cheese’s petroleum wax-based packaging. Nonanal is a C9-saturated fatty aldehyde characterized by a rose-orange odor. The JECFA (Joint FAO/WHO Expert Committee on Food Additives) established an ADI of 0–0.1 mg/kg BW for this substance and declared no safety concern at the current levels of intake when used as a flavoring agent [18]. The results obtained here were compared with the literature. A concordance was observed with the results previously published by Durrett (1996) [8] which indicated the attribution of part of the odor to n-alkanes and aldehydes. However, Durrett (1996) [8] identified C₃ to C₅ alkanes and aldehydes, while in the study presented here, they were C₁₂-C₁₈ and C₉, respectively. In addition, the literature indicated that the main sources of odor in industrial waxes were toluene and methyl ethyl ketone. Both are solvents used in the dewaxing process at the refinery. The absence of these two compounds in the present study could be explained in relation to the fact that the waxes used for contact with food must be previously hydroprocessed to eliminate the dewaxing solvents as well as other aromatic compounds.

4. Conclusions

The identification of 10 volatile organic compounds in different petroleum wax-based packing was performed. The n-alkanes were the most represented chemical family in the studied samples. The results obtained have demonstrated the sensitivity of the DT-GC-MS technique for the determination of VOCs in this type of sample. Furthermore, the established instrumental parameters have allowed for obtaining a good chromatographic separation and will enable the future quantification of the identified VOCs after the optimization and validation of this analytical method.