Introduction

The preservation of working conditions and employees’ health must be among the main concerns of any employer. Work locations where employees are daily or frequently exposed to hazardous scenarios, like factories and production lines, must have strict regulations and precise systems for continuously controlling and monitoring the work conditions.1

One of the main contributing factors to the degradation of working conditions is the lack of air quality in both indoor and outdoor environments. The existence of toxic compounds in the air, emitted by all types of chemical products and coatings used in factories, represents a health hazard and may lead to the development of a significant number of health conditions and pathologies in employees.2,3

An example of the aforementioned scenario is the painting line of a car factory. Here, all the coatings used throughout the entire process of coating a car are abundant sources of compounds potentially hazardous to the health of the employees. The coatings used in the line have compositions often rich in organic compounds that can be effortlessly emitted to the indoor air of the factory during the drying stages. In addition, even though they are not directly exposed to these compounds, the employees can still be affected by the toxicity of the overall atmosphere of the painting line. In this way, a gas chromatography-ion mobility spectrometry (GC-IMS) device was employed, for the first time, to detect and identify the VOCs emitted by three types of coatings, specifically, primers, basecoats (color coatings) and clearcoat (varnish), used during the coating processes of the painting line of a world-renowned automotive brand and, consequently, contribute to develop further studies to characterize the work conditions in the interior of the line. Further details regarding the goals of the study are addressed in due time.

Volatile organic compounds

A volatile organic compound is defined as “… any organic compound …, having, at 293.15 K, a vapor pressure of 0.01 kPa or more, or having a corresponding volatility …”, and in this way, any organic compound that volatilizes at around room temperature (20°C) can be categorized as a VOC.4,5

The existence of VOCs in both indoor and outdoor environments is directly caused by daily-use objects and activities. Cooking, smoking and cleaning are among the main activities responsible for the emission of VOCs into the air. Detergents, pesticides, personal care products, creams, perfumes, tobacco, food, building materials, paints, furniture and many other ordinary objects are equally abundant sources of VOCs.6,7,8 Consequently, the air of both public locations, such as hospitals, schools, shops or public transportation, and private locations, like houses or personal cars, is frequently rich in this type of analyte.9,10,11

Due to their idiosyncrasies, VOCs can effortlessly cross biological membranes, such as alveolar or pulmonary tissues, ocular tissues and even cutaneous tissues.12,13 Their interaction with humans, even at trace concentrations, leads to the development of several health conditions and pathologies. Although not health-threatening, cutaneous pruritus and ocular allergies are the most common reactions in short-term exposure scenarios. For example, ethanol, methyl ethyl ketone and isopropanol are known for causing such reactions in the skin and eyes.14,15 The cellular oxidative stress caused by the exposure to VOCs, however, can lead to more serious diseases. Asthma and chronic obstructive pulmonary disease are examples of such illnesses.16,17 Carcinogenic conditions have equally been linked to exposure of the organism to hazardous compounds. Formaldehyde, chloroform, benzene and trichloroethylene are examples of analytes known for causing severe forms of lung, breast and some other forms of cancer.18,19

In the specific scenario of a car factory painting line, coatings like primers, basecoats and clearcoats, are the key sources responsible for the emission of VOCs and, consequently, for the degradation of the air quality.20 In fact, a scientific study estimated that the total amount of VOCs emitted by a car assembly facility ranges from 1.18 to 4.30 kg per produced vehicle,21 and a significative portion of these emissions occur during the coating and drying processes. These values show how mandatory it is to characterize the main sources of VOCs and identify the main analytes emitted, in order to define the list of potential VOCs existent in the indoor air of the factory and, then, assess the eventual consequences for the employees in further studies.

Car factory painting line

The complete coating of a car chassis is a dense and intricate process that involves the utilization of several products.22 Due to being irrelevant to the scope of the present work, a complete description of the coating process was not included; nonetheless, it is important to mention that among all the steps occur several immersions in chemical baths with various purposes that range from protecting the metal against future corrosions to enabling a better application of the coatings. Then, several steps in which robotic arms are used to apply coatings like the primers, basecoats or clearcoats, are also a part of the extensive process of coating.23 All of the coatings are maintained in large reservoirs stored in a temperature- and humidity-controlled warehouse, where they are kept under constant blending to prevent eventual substrate deposition. Between the painting line and the warehouse, a complex and exclusive piping system is utilized for the transportation of each type of coating in order to prevent contaminations or undesired mixings.23,24 Between some of the stages, industrial size ovens are used for drying purposes. All the chassis are carefully and continuously verified by the employees of the painting line aiming to detect eventual imperfections.23 As mentioned, the coating process is intricate and can significantly change between different factories; in this way, the generic steps of the coating process were superficially addressed. Besides the differences, several coatings, like the primers, basecoats and clearcoats, are employed throughout the process.25,26

The majority of coatings are divided into two categories: solvent- and water-based. Solvent-based coatings contain several organic compounds that act as solvents in their composition.27 The main VOCs are toluene, xylene, benzene, ethylbenzene, 1,2,4-trimethyl benzene, naphthalene, acetone, methyl butyl ketone, methyl isobutyl ketone, methyl amyl ketone, di-isobutyl ketone, ethyl acetate, butyl acetate, ethyl propionate, methanol, ethanol, propanol and butanol, among many others.28 All the solvent analytes ease the application, drying and durability of the paint; however, their volatilization and consequent presence in the indoor air of the line represent a serious hazard for the exposed employees and the environment.25,27,29 Due to the growing concerns about the toxicological and environmental effects, restrictive laws have been implemented aiming to limit the application of solvent-based coatings and boost the use of water-based coatings. Besides these legislations, they are still used in an approximately equal number of cases and scenarios, nowadays, due to the idiosyncrasies of both categories.

Water-based coatings have a considerably lower ratio of organic solvents in their composition, which leads, consequently, to lower emissions of VOCs during the coating process.30,31 The most common analytes existent in water-based coatings are toluene, xylenes, ethyl acetate, butyl acetate, benzene, ethylbenzene and methyl ethyl ketone, among others. Besides the advantage of being less hazardous to both the employees’ health and the environment, water-based coatings present some disadvantages if compared with solvent-based versions.32,33,34 Due to the lower amount of solvent analytes in their composition, they are considerably more viscous and require the utilization of higher temperatures during longer periods to dry completely. In this way, when compared to the performance of solvent-based coatings, larger amounts of water-based coatings are required to coat the same area and larger amounts of energy are equally required during the drying process.30,35

Independently of their composition and the number of scenarios in which each one is employed, both types of coating are mainly responsible for the emission of VOCs in the painting line. In fact, it was even estimated that the concentration of VOCs in the composition of generic automotive solvent-based coatings ranges between 650 and 800 grams per liter and, in the case of generic automotive water-based coatings, the concentration reaches up to 420 grams per liter.21 A significative portion of these quantities is emitted to the indoor air during the application and drying of the coatings; in this way, it is relevant to fully address these emissions for further considerations regarding their eventual impacts.

Gas chromatography-ion mobility spectrometry

Spectrometric and chromatographic techniques, and sensors array-based electronic noses have often been employed for detecting VOCs.36,37 Among all these techniques, ion mobility spectrometry (IMS) has gained relevance and played an important role in several scientific fields.38 These applications include air quality control,39 health assessment,40 fraud detection,41 food quality and spoilage,42 security purposes,43 and many others.

Outstanding sensitivity, portability, analytical flexibility, instrumental simplicity and quasi-real-time monitoring capability are among the main characteristics of the IMS.44 The analytical flexibility enables the combination of IMS with several additional techniques, namely gas chromatography (GC) and even mass spectrometry (MS).45,46 As previously mentioned, a GC-IMS device was used in the present work. This device merges the high selectivity, wide dynamic concentration range and good precision of the gas chromatography with the aforementioned characteristics of the ion mobility spectrometer, creating an analytical technique with improved capacities for in situ detection, identification and quantification of VOCs based on their shape, weight and size, without requiring additional chemicals or complex sample preparation.44,47

A generic GC-IMS analysis starts with the injection of the volatile sample into the spectrometer. Once injected, the analytes constituents of the sample are separated by their capacity of adsorbing and desorbing to the inner surface of the chromatographic column.48 The time required for each compound to completely elude from the column is defined as retention time, rt. This temporal value corresponds to one of the three coordinates represented in the three-dimensional spectrum plotted after the GC-IMS analysis.

After the chromatographic pre-separation, the analytes proceed to the IMS section of the device. The operating principle of the IMS is based on the fact that an organic molecule, once ionized, can be accelerated if exposed to an electric field.44 Thus, the compounds enter an ionization chamber where they are ionized. An X-ray source was used in this work; nonetheless, tritium (3H) and nickel (63Ni) are equally often utilized sources.49

The previously formed primary ions participate in new reactions with molecules of NO, NH3 or H2O existent in the interior of the IMS tube. From these reactions, new ions are formed. Known as reactant ions, they can present the form of (H2O)nNO+, \(\left( {{\text{H}}_{{2}} {\text{O}}} \right)_{n} {\text{NH}}_{4}^{ + }\) or (H2O)nH+ and are visible in the final three-dimensional spectrum; they form an intense peak, named reactant ion peak (RIP), throughout the entire spectrum.44,50 Finally, these ions react with the organic analytes constituent of the sample under analysis. Considering a sample composed of an arbitrary compound identified as M, the reaction occurs as below:

$$M + \left( {{\text{H}}_{{2}} {\text{O}}} \right)_{n} {\text{H}}^{ + } \leftrightarrow {\text{M}}\left( {{\text{H}}_{{2}} {\text{O}}} \right)_{n - x} {\text{H}}^{ + } + x{\text{H}}_{{2}} {\text{O}}$$
(1)

The product ions formed during this reaction correspond to protonated monomers of the compound M. If the concentration of M is sufficiently elevated, the monomers can experience a second reaction with the remaining molecules of M, leading to the formation of dimers (equation 2). Larger clusters are also possible but uncommon.44,50

$$M + M\left( {{\text{H}}_{{2}} {\text{O}}} \right)_{n - x} {\text{H}}^{ + } \leftrightarrow M_{2} \left( {{\text{H}}_{{2}} {\text{O}}} \right)_{{n + \left( {x + i} \right)}} {\text{H}}^{ + } + i{\text{H}}_{{2}} {\text{O}}$$
(2)

An IMS device can operate in both positive and negative polarities. The process described above summarizes the IMS positive operating mode. For the negative operating mode, the reactant ions have the form of \(\left( {{\text{H}}_{{2}} {\text{O}}} \right)_{n} {\text{O}}_{2}^{ - }\) and are formed from the reaction between the primary ions and molecules of O2.44,50 Protonated monomers and dimers of an arbitrary compound named MX are formed, in the IMS negative working mode, through dissociative and associative electron attachments, as, respectively, represented below.

$$MX + \left( {{\text{H}}_{{2}} {\text{O}}} \right)_{n} {\text{O}}_{2}^{ - } \leftrightarrow M + X^{ - } + n{\text{H}}_{{2}} {\text{O}} + {\text{O}}_{2}$$
(3)
$$MX + \left( {{\text{H}}_{2} {\text{O}}} \right)_{n} {\text{O}}_{2}^{ - } \leftrightarrow \left( {MX} \right){\text{O}}_{2}^{ - } + n{\text{H}}_{2} {\text{O}}$$
(4)

Once formed, the product ions are exposed to a homogeneous and weak electric field which accelerates them across the IMS drift tube. Here, numerous collisions occur between the ions and the molecules of the drift gas existent in the interior of the tube. After the collision and since the ions are still under the effect of the electric field, they are accelerated again. This continuous acceleration-collision process leads the velocity of the ions to tend to a constant value, defined as drift velocity, vd. This analyte-specific velocity can be related to the electric field, E, by the ion mobility constant, K.44,51

$$K = \frac{{v_{{\text{d}}} }}{E}$$
(5)

When the ions reach the end of the drift tube, they are detected by a Faraday plate that registers the time required for each ion to traverse the complete length, L, of the tube. This temporal value is often defined as drift time. Since the drift velocity corresponds to the quotient between a length and a time, a new formulation of K arises (equation 6). Here, L corresponds to the length of the drift tube and dt to the time required by each ion to traverse the entire tube, generally known as drift time.44,52

$$K = \frac{L}{{E \cdot d_{t} }}$$
(6)

The ion mobility constant is an analyte-specific value but it depends on temperature, T and pressure, P, conditions, so it is useful to normalize K to standard environmental values of temperature, T0 = 273.15 K, and pressure, P0 = 760 Torr.44,51 The normalized ion mobility constant, K0, is also a compound-specific value and can be written as:

$$K_{0} = K\left( {\frac{P}{{P_{0} }}} \right)\left( {\frac{{T_{0} }}{T}} \right)$$
(7)

A three-dimensional spectrum is produced as outcome of each GC-IMS measurement. The spectrum includes both the retention (s) and drift (ms) times of all the detected VOCs, as well as their intensity (V). Figure 1 exemplifies an IMS spectrum in two- and three-dimensional views. The long red column across the spectrum corresponds to the reactant ion peak, RIP. Five peaks are identified for illustrative purposes (1: monomer of ethanol, 2: monomer of isopropanol, 3: dimer of ethanol, 4: monomer of acetone and 5: dimer of isopropanol). As addressed, each volatile compound has its specific drift (x-axis) and retention (y-axis) times, so each peak corresponds to a particular volatile organic compound. These two variables can be used for identification purposes. A color scheme is used to represent the different levels of intensity (z-axis). The intensity of each analyte can be used for quantification purposes.

Fig. 1
figure 1

Standard three-dimensional GC-IMS spectrum (1: monomer of ethanol, 2: monomer of isopropanol, 3: dimer of ethanol, 4: monomer of acetone and 5: dimer of isopropanol)

Full size image

Considering all the information addressed, this paper aims to fully characterize the emissions of all the coatings used in the painting line of the factory under study, namely their content in VOCs, in order to define the main organic compounds that can be present in the indoor air of the factory.

It is worth saying that this study is a part of a larger project developed by a car factory, whose goals are wider than the goals of this study. The project aims to completely evaluate the work conditions of the employees of the painting line. To do so, air analyses are expected to be performed at all the critical locations of the factory and once the potentially dangerous VOCs are identified and compared with the VOCs emitted by the coatings, further studies will be developed to assess the eventual consequences to the employees’ health, if any.

Concluding, this study stresses the results achieved during the evaluation of the emissions of all the coatings used in a car factory painting line, as a way of deepening the knowledge on the coatings-borne VOCs that might be present in the indoor air and for the development of further studies for later assessment of their real impact on the employees’ health. The focus of the study was given to the emissions of the coating and not to their original composition since some of the compounds used in the manufacturing of the coatings are not emitted to the atmosphere, i.e. they are not a direct threat to the employees. To do so, a GC-IMS device was employed, allowing the authors to accurately detect and identify all the volatile organic compounds in the headspace emitted by the coatings. A database of analytes was exclusively developed for the purpose of identifying all the detected analytes. Overall theoretical considerations regarding the identified VOCs were included in the manuscript and, as mentioned, these results will allow the development of further studies regarding the composition of the indoor air and the eventual consequences to the employees’ health, if that is the case.

Methodology

Analyzed samples

A total of 22 distinct samples, among primers, basecoats and clearcoat, were considered for this work. Four different water-based primers are used in the painting line. They are commercially named after their color, i.e. anthracite, light grey, light red and mid-grey. The used primer is selected accordingly with the color of the paint to be later applied. The factory has access to a pallet of 17 colors of paints. Like the primers, all these 17 water-based coatings were analyzed during this work. Finally, a twenty-second coating, commercially known as Lumeera, is used by the employees to varnish the already-painted metal surface. The clearcoat Lumeera corresponds to the only solvent-based coating considered for this study.

Table 1 summarizes the analyzed samples. The information here contained was fully provided by the car factory involved in the study. Regarding the samples, the names and respective colors of both primers and basecoats are included as provided by the factory and following the information publicly available on the commercial website of the automobile brand. Information regarding the solid or metallic nature of the coating and the primer color to be used for each basecoat is also provided. Finally, the ratio of VOCs in the composition of each coating is included in the form of a percentage, as provided by the producer. No additional information regarding the composition of the samples was provided by the factory or by the producer.

Table 1 Samples of primers (water-based coatings), basecoats (water-based coatings) and clearcoat (solvent-based coatings) analyzed during the study
Full size table

Sample preparation

As mentioned, the 22 types of samples were fully provided by the car factory. To do so, 100 mL volume samples were collected directly from the factory reservoirs into glass bottles. The bottles were then closed with metal caps and properly isolated from contaminations and radiation, aiming to avoid eventual degradation of the samples during the transportation process.

Once in the laboratory, a total of 10 replicas were prepared and analyzed for each one of the 22 samples. To do so, 2 μL of each sample was pipetted into 10-mL glass vials. Then, the vials were closed with a septum cap and were stored for 30 min at room temperature in order to create headspace. Once reached the equilibrium between the liquid and the gaseous portions of the sample, 2 mL of the headspace was collected with a syringe and two needles and injected into the GC-IMS device. Figure 2 schematizes the formation and collection of the headspace.

Fig. 2
figure 2

Formation (left) and collection (right) of the sample headspace

Full size image

Gas chromatography-ion mobility spectrometry

As mentioned, a GC-IMS device was used during this study. The chromatographic column assembled in the device was an MXT-200 model with 30 m of length and 0.53 mm of internal diameter coated with stainless steel with a mid-polar stationary phase of trifluoro propyl methyl polysiloxane with a thickness of 1 μm. The chromatographic column was operated at 70°C. For ionization purposes, an X-ray ionization source was used. Since the spectrometer can simultaneously operate both positive and negative working modes, two IMS tubes with a drift length of 8.23 cm were assembled in the spectrometer. The IMS was operated at room temperature. Purified air was used as drift and carrier gases; their flows were 300 mL/min and 10 mL/min, respectively. Table 2 summarizes the operating parameters of the GC-IMS.

Table 2 Operating parameters of the GC-IMS
Full size table

Identification of VOCs

Aiming to identify the analytes detected during the measurements of the coatings, a database of volatile organic compounds was developed. To do so, pure samples of analytes were analyzed under the same conditions used during the studies of the coatings. A predefined standard volume (2 μL) of pure analyte was pipetted into a glass vial, which was then sealed with a septum cap and isolated from potential contaminants. After 30 min, a portion of the created headspace was collected with a syringe and two needles and injected into the spectrometer, as schematized in Fig. 2. The characteristic drift and retention times, ion mobility constant and normalized ion mobility constants were then registered in the database for further purposes. Figure 3 illustrates an IMS spectrum for a VOC-free air sample, a), and a spectrum utilized for the identification of both monomer and dimer of a randomly selected analyte, specifically, 3-pentanol, b). This same process was repeated for a total of 130 potentially relevant volatile organic compounds. Once completed the database of VOCs, the identification of the analytes emitted by the analyzed primers, basecoats and varnish were attained by crosschecking the drift times, retention times and the normalized ion mobility constant of each analyte with the corresponding quantities previously registered in the developed database. The adopted procedure enabled an accurate and rapid identification of most of the detected analytes.

Fig. 3
figure 3

IMS spectra for a VOCs-free sample (a) and for the detection and identification of 3-pentanol (b), with the reactant ion peak (RIP), and the monomer and dimer of the analyte properly indicated

Full size image

Results and discussion

Spectra

Since a dual-polarity GC-IMS was employed during the study, a total of 440 spectra, 220 positive polarity spectra and 220 negative polarity spectra, were collected during the measurements. The negative polarity spectra, however, did not include any relevant data regarding the existence of analytes; in this way, they were disregarded from the study. Nonetheless, this fact leads to an evident conclusion; none of the coatings employed in the car factory painting line emits or has in its constitution analytes that originate negative ions once ionized. Figure 4 illustrates three-dimensional positive polarity spectra for three arbitrarily selected samples, in specific, a spectrum of (a) primer, (b) basecoat and (c) clearcoat. Figure 5 illustrates a magnified version of Fig. 4.

Fig. 4
figure 4

Three-dimensional positive polarity spectra for arbitrary samples of (a) primer, (b) basecoat and (c) clearcoat

Full size image
Fig. 5
figure 5

Magnified view of the three-dimensional positive polarity spectra for arbitrary samples of (a) primer, (b) basecoat and (c) clearcoat

Full size image

Repeatability

Aiming to assess the lack of variability of the data collected with the GC-IMS device, repeatability graphs were plotted. To do so, the intensity of all the detected analytes was summed for each one of the 10 repetitions, resulting in the total intensity of each spectrum. These values were, then, normalized and plotted in a line chart for further analyses. Figure 6 illustrates the repeatability of the data throughout the 10 spectra of an arbitrarily selected coating. The average value for the normalized intensity and the respective average variation for this data set are (0.87 ± 0.01), which proves the repeatability of the data and, consequently, the suitability of GC-IMS for such studies. The repeatability graphs of the remaining 21 coatings were not included in this article but similar variability results were attained for all of them.

Fig. 6
figure 6

Variability line chart for an arbitrarily selected coating. The average value for the normalized intensity and the respective average variation for the data set here included are (0.87 ± 0.01)

Full size image

Identification of VOCs

It is relevant to reinforce that, for all the data processing included in this chapter, 10 replicas were performed per type of coating. It is equally important to mention that only the analytes common to all those 10 spectra were considered for the study, and the remaining peaks were interpreted as sporadic contaminations and disregarded. The identification of the detected analytes was accomplished, as addressed, through the database of compounds exclusively developed for this study.

Primers

As previously mentioned, four different primers are used during the coating process of the car factory painting line. From the 10 repetitions performed with the GC-IMS for each type of primer, a total of 84, 99, 95 and 86 analytes were detected, respectively, for the primers light red, light grey, mid-grey and anthracite. Among monomers, dimers and even trimers, it was possible to use the exclusively developed database to, respectively, identify 29, 38, 28 and 29 analytes. These analytes correspond to a total of 26 volatile organic compounds. Table 3 summarizes all the VOCs identified among the analytes emitted by each kind of primer. Further information regarding the identification data is given in due time but it is relevant to state that the VOCs are listed in ascending order of retention time.

Table 3 Identified volatile organic compounds per type of primer, listed by ascending order of retention time
Full size table

By consulting the previous table, several facts can be noticed. The four primers contain a total of nine common volatile organic compounds, namely ethanol, isopropanol, methyl acetate, 2-butanol, acetone, ethyl acetate, 2-butanone, butyl acetate and 1-octanol. On the other hand, six VOCs are present in the composition of just one of the primers, namely tert-butyl methyl ether, sec-butylamine, 3-methyl-2-butanol and hexane for the primer light grey, tetrahydrofuran for anthracite and 3-methyl-1-butanol for mid-grey. Due to their scarcity, they can be seen as sporadic contaminations rather than constituent compounds. As expected, alcohol-based compounds, with a total of 10 identified analytes, represent the largest group of VOCs. Further considerations regarding the identified VOCs are addressed in due time.

Basecoats

Since the negative mode spectra did not contain relevant information regarding the emission of the coatings, a total of 170 positive mode spectra were considered during the analysis of this data set. A significantly elevated number of analytes were detected throughout all the samples. Table 4 summarizes the number of analytes detected throughout the 17 colors of coatings, the number of identified analytes among the detected ones and the corresponding number of identified VOCs considering monomers, dimers and even some trimers.

Table 4 Number of analytes detected throughout the 17 colors of coatings, number of identified analytes among the detected ones and the corresponding number of identified VOCs considering monomers, dimers and even some trimers
Full size table

Considering all 17 types of coatings, 37 volatile organic compounds were identified overall. Table 5 summarizes all the VOCs identified among the analytes detected from the emissions of each coating. For aesthetical reasons, the names of the coatings were replaced by numbers; nonetheless, the correspondence between the names and numbers is provided in the caption of the table.

Table 5 Identified volatile organic compounds per type of coating, listed by ascending order of retention time (1: ascot grey, 2: atlantic blue, 3: beech grey, 4: black oak, 5: deep black, 6: indium grey, 7: kings red, 8: lapiz blue, 9: moonstone silver, 10: petroleum blue, 11: pure white, 12: pyrite silver, 13: ravenna blue, 14: reflex silver, 15: romance red, 16: urano grey, 17: white silver)
Full size table

Further considerations regarding the identified analytes are addressed below; nonetheless, some evident conclusions can be drawn by analysing Table 5. Six volatile organic compounds are present in the constitution of all the 17 types of coatings, namely 4-fluoroaniline, ethanol, isopropanol, acetone, 1-butanol and 2-butanone meaning that they belong to the base mixture used for manufacturing all inks. Another six are solely present in the samples of one coating, namely 2-heptanone for moonstone silver, butylamine and 3-methyl-2-butanol for romance red, propyl acetate for beech grey, toluene for urano grey and 3-pentanol for atlantic blue. As for the results attained during the identification of VOCs in the samples of primers, alcohol-based VOCs are the most common analytes (12 VOCs). Additional remarks will be promptly addressed.

Clearcoat

Among monomers, dimers and even some trimers of each compound, 87 analytes were detected throughout the 10 analyses performed with varnish samples. From these, 30 analytes were accurately identified with the developed database, totaling 16 VOCs. The VOCs identified in the varnish samples are condensed in Table 6.

Table 6 Volatile organic compounds identified among the analytes emitted by the clearcoat
Full size table

Most of the analytes identified during the analyses of varnish samples are common to both primers and basecoats, nonetheless, three VOCs seem to be varnish-exclusive. They are ethylbenzene, p-xylene and o-xylene. Considering the 16 compounds, five belong to the family of alcohols, three to the esters family and two to the ketones family. Additional considerations are addressed promptly.

Identification data

As addressed herein, 26, 37 and 16 volatile organic compounds were, respectively, identified during the analyses of primers, basecoats and clearcoat. Since some of these are common to several samples, a total of 45 compounds were accurately identified throughout the entire study using the exclusively developed database. Table 7 condenses the identification data for all VOCs, namely their retention time, drift time, ion mobility constant (K) and normalized ion mobility constant (K0). The VOCs are listed in ascending order of retention time.

Table 7 Identification data of every volatile organic compound identified during the entire study, namely their retention time, relative drift time, ion mobility constant (K) and normalized ion mobility constant (K0)
Full size table

Considering all the compounds detected in the emissions analyzed from the samples of coatings, a total of 45 volatile organic compounds were accurately identified with the database exclusively developed for this study. The identified analytes include 14 alcohol-based analytes, nine ketones, six esters, five aromatic hydrocarbons, four amines, three ethers, two alkanes and two fatty acids.

As addressed in the introductory section of this article, analytes from the family of alcohols have been largely used in the composition of all types of coatings, inks and paints, primers and varnish, as solvents of the mixture. It was equally mentioned that solvent-based and water-based coatings are still used in comparable number of applications, despite the efforts to replace the first by the second type. The composition of water-based coatings is considerably less threatening to the environment and the human organism, nonetheless, they are still considerably rich in alcohols and other organic compounds.25,27 In fact, even in reduced quantities, the utilization of alcohols in the composition of the coatings is essential for reducing the coatings’ viscosity and time of drying;30 in this way, the existence of a significantly high number of alcoholic analytes in the composition of the coatings is justified.

Ketones and aromatic hydrocarbons are among the predominant compounds in the overall results, which may be justifiable by the fact that both these families of analytes have played a relevant role in the manufacturing of coatings.33 Ketones like acetone, butanone, or 4-methyl-2-pentanone and aromatic hydrocarbons like benzene, toluene, ethylbenzene and xylenes, commonly known as BTEX compounds, have been largely used as oxygenated solvents in the manufacturing of coatings for metal surfaces in detriment of the alcohol-based solvents. This substitution aims to reduce the hazardousness to the human organism and the environmental impacts of the alcoholic VOCs.30,34 In addition, the employment of BTEX VOCs, specifically xylenes, contributes to the decrease of the coating's viscosity.53 Besides that, it is relevant to reinforce that these analytes, and BTEX compounds in particular, are also considerably dangerous to human health.54 All the aforementioned VOCs were detected and identified during this study, meaning that these results corroborate the information available in the literature.

Due to their relevant role in the properties of the coatings, ester-based analytes have been largely employed in their manufacturing for many years. The inclusion of these compounds in the composition of coatings enables the improvement of their performance, in specific, their evaporation rate, blush resistance, leveling properties and viscosity.27,34 All these factors are of extreme relevance for the protection of the metal surface and for the durability of the coatings, and ester-based VOCs have ensured such factors. Some of these VOCs, namely methyl, ethyl, butyl, propyl and octyl acetate, were successfully detected and accurately identified during this study, which agrees with the results anticipated by the literature.55

The identification of several amine-based VOCs among the analytes emitted by the coatings was expected. Due to its large applicability, mechanical properties and low cost, the car body is mostly made of metal; however, metal is susceptible to corrosion if exposed to natural conditions. Aiming to prevent or delay the corrosion process, amines like triethylamine and 4-fluoroaniline have been included as corrosion inhibitors in the composition of modern coatings specifically used for metal-made surfaces. Several scientific studies have been developed concerning the suitability of amines for the development of corrosion-resistant coatings, so the detection of these analytes in the coatings used in the car factory painting line was anticipated.56,57,58

Three ethers were identified during the study here described. They are tert-butyl methyl ether, tetrahydrofuran and 1,4-dioxane. Ether-based compounds are commonly employed in the preparation of coatings due to their contribution to minimizing the drying time and to the solubility properties of the dyes.59,60 As with many other VOCs, their emission and consequent presence in the environment represent a potential hazard, so several studies have been developed aiming to eliminate or, at least, reduce the impact of such compounds in nature.61,62

Alkane-base analytes, like hexane and 2,4-dimethylpentane,28,63 and fatty acids, like butyric and propionic acids,64 have been reported in a few studies regarding the emissions of automotive coatings and overall dyes; nonetheless, their role in the composition of these mixtures is not as deeply addressed as for the previously referred compounds.

Conclusions

As previously addressed, this study is a significant part of a larger project which aims to fully characterize the work conditions of a car factory painting line for the employees chronically exposed to this environment. To do so, analyzing the coatings used in the painting line and identifying all the emitted compounds are mandatory tasks due to the well-known hazardousness to both the human organism and the environment of the VOCs. In this way, the study described in this article aimed to assess the volatile organic compounds existent in the composition and consequently emitted into the indoor air, by the coatings used in the painting line of a car factory. These results are of interest to deepen the knowledge on the content of the coatings and to develop further studies and draw further considerations regarding their impacts on employees’ health.

Among primers, basecoats and clearcoat, a total of 22 types of coatings were collected at the factory and explored in laboratory facilities. To do so, the headspace emitted by each coating was analyzed through gas chromatography-ion mobility spectrometry. Considering the 10 analyses performed per coating, the average variation of the normalized intensity was estimated in the 10−2 range (0.87 ± 0.01) in specific for the coating illustrated in Fig. 5. Similar results were attained for the remaining 21 coatings. These values ensure the repeatability of the collected data and, consequently, the suitability of GC-IMS for such type of study.

Aiming to identify the analytes emitted by all the 22 analyzed coatings, a database of VOCs was dedicatedly developed. This library enabled the accurate and rapid identification of 26, 37 and 16 compounds, respectively, emitted by the primers, basecoats and clearcoat. Overall, 45 volatile organic compounds were identified, including 14 alcohols, nine ketones, six esters and five aromatics. From the list of identified analytes, it was possible to conclude that most of them match the expected results and corroborate the scarce information available in the literature.

The results achieved during this study expand the knowledge regarding the emissions of the coatings employed in the painting line of the car factory and, since these products represent the largest and most relevant VOCs-emitting sources, define the compounds that must be targeted during further studies on the indoor air content and, later, on the potential risks to the employees. Further studies are mandatory to evaluate if the emitted VOCs here identified are, in fact, present in the indoor air of the factory and assess in which concentration levels that presence occurs. Consequently, the overall consequences of long-term exposure to the employees’ health, if any, can then be fully understood and described in order to comply with the goals of the overall project developed by the car manufacturer.