The European Union requires that benzene in the air is continuously measured due
to its toxicity and widespread presence in the population nuclei, mainly
motivated by vehicle emissions. The reference measuring technique is gas
chromatography (GC). Automatic chromatographs used in monitoring stations
must verify the operating conditions established in Standard EN 14662 part 3,
which includes a type approval section with a number of tests that analysers
must pass. Among these tests, the potential interference of a number of
compounds is evaluated. The 2005 version of the mentioned standard requires
the evaluation of the potential interference of tetrachloromethane (TCM). The
2015 version eliminates TCM as a potential interferent. Although most
consumer uses of TCM have been banned, recent studies have measured
significant concentrations of TCM in the air. In this paper, the potential
interference of TCM in benzene measurements obtained with gas chromatography
coupled to a photoionisation detector (GC-PID) has been investigated. Our study
shows that the simultaneous presence of benzene and TCM causes a significant
decrease in benzene readings. For TCM concentrations of 0.7
Benzene is a volatile organic compound (VOC) (Tisserand and Young, 2014).
Directive 2008/50/EC (European Commission, 2008) defines VOCs as organic
compounds from anthropogenic and biogenic sources, other than methane, that
are capable of producing photochemical oxidants by reactions with nitrogen
oxides in the presence of sunlight. Benzene sources include natural
emissions from vegetation and oceans (Misztal et al., 2015), microbial
decomposition (Neves et al., 2005), wildfires (Wentworth et al., 2018), and
volcanoes (Tassi et al., 2015); anthropogenic emissions mainly from
vehicles that use fossil fuels (von Schneidemesser et al., 2010); and, in
central and northern European countries, emissions from the combustion of wood used
for domestic heating (Hellén et al., 2008). It is also present in
tobacco smoke (Darrall et al., 1998) and in a wide range of industrial and
household products (solvents, adhesives, paints and cleaning products) and
is also a raw material for the synthesis of other products, such as dyes,
detergents, plastics and explosives (Guenther et al., 1995). Its content in
gasoline is regulated by Directive 2009/33/CE, and it has to be < 1 % (
Due to the chemical stability of benzene compared with most VOCs (with a half-life of 9.4 days; Atkinson, 2000), its permanence in the atmosphere is high. Consequently, it can be transported over long distances. It is degraded by OH radicals in the troposphere, forming phenol and glyoxal, among other compounds (Atkinson, 2000; Volkamer et al., 2001).
Benzene is a recognised inducer of leukaemia (D'Andrea and Reddy, 2016) and
also affects the central nervous and immune systems and damages genetic
material (Bahadar et al., 2014). It is the only VOC in Europe whose
concentrations in the air are regulated. Its annual limit value is 5
The 2005 version of the Standard EN 14662-3 included a list of paraffinic, cyclic and halogenated organic compounds (including tetrachloromethane – TCM) that had to be tested as potential interferents. In the 2015 version, all hydrocarbons are maintained and isooctane (2,2,4-trimethylpentane) and 1-butanol have been added but TCM has been removed. Table 1 shows all the common and specific components of each version.
Organic compounds used to assess interferences in the measurement of benzene in the air, in accordance with standards EN 14662-3 (2005 and 2015 versions).
The synthesis of TCM for emissive uses was controlled and practically banned
by the Montreal Protocol because it is an ozone-depleting substance (Sherry
et al., 2018). However, its use as a raw material for the synthesis of other
substances such as hydrofluorocarbons, pyrethroid pesticides or
perchloroethylene is still allowed (Graziosi et al., 2016). Diffuse
emissions may occur in its manufacture or during its use in the
aforementioned syntheses. In this sense, 9500 Mt of TCM were estimated to be
emitted in 192 countries in 2007 (Penny et al., 2010). However, since the
entry into force of the Montreal Protocol, there has been a progressive
decrease in the environmental presence of TCM, with a decrease in its global
average concentration of 10 to 15 pptv decade
Given the above, in this paper, the potential interference of TCM in benzene measurements carried out by gas chromatography coupled to a photoionisation detector (GC-PID) is studied. A mechanism that explains the observed behaviour is also proposed.
An in-house-designed controlled atmosphere chamber was used to generate
dynamic test mixtures of benzene in the air with and without TCM (Fig. 1). This
chamber was used in previous works (Romero-Trigueros et al., 2016, 2017), and
only a brief description will be given here. Zero air was generated from
ambient air with a JUN-AIR compressor (Michigan, USA) provided with a drier,
which is capable of reducing the relative humidity of the air down to 5 %. This dry air flows through three consecutive scrubbers containing
silica gel with an indicator (orange gel) (Merck, Darmstadt, Germany) and K47
active charcoal (Chiemivall, Barcelona, Spain) to remove any traces of
remaining humidity and other gases present in the air. After purification, a
periodic check of organic pollutants in the zero air was carried out by gas
chromatography, ensuring they were below their limits of detection. Benzene
was incorporated to the zero air from a high-concentration gas mixture of
benzene in nitrogen. Two mixtures from Abelló Linde (Valencia, Spain)
were purchased at nominal concentrations of 1000 and 350
The final concentration of component
Schematic of the components of the controlled atmosphere chamber used to obtain gas mixtures of benzene in the air with and without potential interferent substances. (HI: humidity indicator; TI: temperature indicator; PI: pressure indicator).
Two identical type-approved BTEX Syntech Spectras GC955 chromatographs
equipped with PIDs (Groningen, Netherlands) were tested in this work. These
are widely used in European air pollution monitoring networks and were
identified as analysers I and II. The analytical process is
semi-continuous. While the GC-PID is analysing a sample, a new one is
sampled and sent to the pre-concentration system. The air sampling system
comprises a 35 mL capacity piston pump, and the suction operation is repeated
five times producing a total sample volume of 175 mL in each cycle. The
successive 35 mL samples of air flow to a pre-concentrator (consisting of a
column filled with Tenax), which retains the organic compounds and releases
the excess air. Once the five suction cycles are completed, the contaminants
retained in the pre-concentrator undergo thermal desorption and are carried
with Nitrogen 5.0 (99.999 % purity; Abelló Linde, Valencia, Spain)
towards the chromatographic column. The column was an AT-5 capillary column
(15 m length
A first set of experiments was carried out according to Standard EN
14662:2005-3 with analysers I and II for two different benzene nominal
concentrations (0.5 and 40
According to Standard EN 14662:2005-3, the parameters used to evaluate the
deviations caused by the interferents are the effect of organic compounds
(
Due to the different nature of the interference of the organic compounds that reach the PID, separate studies should be carried out for those that positively (increasing) and negatively (decreasing) affect the measurements of benzene. As explained later, TCM causes the concentration of benzene to decrease, whereas the rest of the compounds act positively; thus, independent tests studying only the influence of TCM were performed.
To study the effect of TCM on the GC-PID measurements of benzene, subsequent
tests were performed with Analyser I. This decision was supported by the
similar behaviour observed for both analysers when carrying out the tests
described in Sect. 2.2.1. In addition, a reproducibility test was carried
out in the lab. Both analysers simultaneously measured a gas
mixture containing 5
Analyser I was first calibrated with dynamic reference gas mixtures of
benzene (
The results obtained when carrying out the tests for evaluating the
interference of organic compounds according to Standard EN 14662:2005-3 are
shown in Table 2. Also, the results of a similar test with a nominal
concentration of benzene of 5
Results obtained when conducting the test to evaluate the
interference of organic compounds in benzene readings for analysers I and
II. Nominal concentrations tested: 0.5 and 40
A similar result (Locoge et al., 2010) was obtained with the same GC-PID and
a gas mixture of 5
Table 3 gathers the results of the tests performed according to Sect. 2.2.2.
It can be seen that the presence of TCM significantly decreases the readings
of benzene with respect to the reference gas mixture concentrations
(
Average readings of benzene concentrations obtained with Analyser I
when measuring benzene reference gas mixtures without TCM (
Calibration lines of Analyser I obtained by linear regression
(without TCM (
The experimental values of
Calibration lines for Analyser I with and without TCM at different concentration levels.
A generic representation of benzene readings as a function of the concentration of benzene and TCM in the reference gas mixture.
An analysis of the analytical method was done in order to understand the nature of this interference. For a substance to act as an interfering agent, it must have a retention time in the chromatographic column within the interval of identification of benzene, so that both species reach the detector within this interval. If this applies, the interference causes an increase or decrease in the detector signal. When the chromatograph has a PID, one of the following can occur. (i) If any organic compound other than benzene is ionised by the radiation of the detector lamp, the electric current increases, which leads to an increase in the readings of benzene. For this to happen, the ionisation potential of the interferent must be lower than that associated with the radiation of the lamp. (ii) The interferent causes a decrease in the benzene signal, which can be due to several reasons. One of them is that the radiation of the detector lamp is absorbed to a greater or lesser extent by the interferent, and the remaining energy is insufficient to completely ionise the benzene. This phenomenon is known as a “quenching effect” (Chou, 1999). This is the nature of the interference of humidity in benzene measurements (Romero-Trigueros et al., 2017). The second reason is that the interferent absorbs (blocks) part of the formed ions that participate in the quantification of benzene, leading to a decrease in the detected concentration. This mechanism is known as a quenching effect via electron capture (Senum, 1981). As discussed below, TCM acts in this way.
To explain the behaviour of benzene in the PID, we have proposed the model
shown in Fig. 4, which also serves as a basis to determine what happens in
the presence of TCM; Fig. 5. When the gasified benzene (
Schematic of the behaviour of benzene in the PID of the chromatograph in absence of TCM.
Schematic of the behaviour of benzene and TCM when they interact simultaneously in the PID detector.
When an air sample, containing benzene (
According to Fig. 4, the concentration of benzene read by the chromatograph
in the absence of TCM (
As can be deduced from Figs. 2 and 3,
As indicated previously, TCM was considered as a possible interfering contaminant to be evaluated according to Standard EN 14662:2005-3, but it was not included in the interferent list in its 2015 version. However, Sect. 8 of the current standard establishes that “some compounds, including carbon tetrachloride or butanol, may be present under site-specific conditions. In such cases, the responsibility for the proper determination of benzene falls on the network that operates the analyser by the appropriate choice of separation conditions (analytical column, temperature program of the column)”. However, technicians that operate air quality networks usually lack the knowledge and tools to choose the optimum conditions for the analysis. On the other hand, a correction of readings would require continuous measurements of TCM in the air and a knowledge of how TCM makes measurements deviate from their real value, which in turn, requires carrying out tests similar to those presented in this paper with dynamic dilution systems in controlled test atmospheres. This measure would not be easy to apply for economic and technical reasons, so the responsibility must not fall only on the network managers. It seems reasonable that the manufacturers of the equipment take actions for solving this problem – or, at least, for reducing the extent of the interference in their measurements – since they have the required technology and equipment. In any case, users of this type of equipment should be aware of the problem to try to minimise it. The discussion of this issue in the appropriate forum (e.g. the European Committee for Standardisation) also seems pivotal to reducing the uncertainty in benzene measurements by GC-PID in the presence of TCM concentrations.
The research described in this article has determined that TCM causes a significant interference in the measurement of benzene by GC-PID. This interference is negative; that is, readings of benzene are below their real ambient values, which may originate in a mismanagement of the air quality of a location with TCM present in its air relating to benzene.
The RE of the concentration of benzene measured as a
function of the concentration of TCM (
Interestingly enough, it is established in part 3 of the Standard EN 14662:2015 that the managers of the air quality monitoring network are responsible for determining the presence of TCM in the area where benzene is measured. If detected, they must act to eliminate the effect of the interferent. However, this approach would require continuous measurements of TCM in the air and a knowledge of how TCM makes measurements deviate from their real value, which in turn, requires carrying out tests similar to those presented in this paper with dynamic dilution systems in controlled test atmospheres. This may entail economic and technical issues so manufacturers of the chromatographs should try to solve this problem as they have greater technical and scientific capacity than network managers. In any case, all these issues should be discussed in the appropriate forum (e.g. the European Committee for Standardisation) in order to improve the uncertainty of benzene measurements and, thus, the management of air quality.
The data are available upon request to María Esther González (esthergd@um.es).
CRT carried out all the tests reported in this paper and the data process. MDM and EGF designed and developed the dynamic system used to generate the gas mixtures, designed the experimental methodology, and supervised the work. MEG prepared the paper and is the correspondence author.
The authors declare that they have no conflict of interest.
The authors would like to acknowledge the Consejería de Agua, Agricultura y Medio Ambiente of the Comunidad Autónoma de la Región de Murcia for its financial support and for the facilities to carry out this work. Cristina Romero-Trigueros acknowledges the financial support from a postdoctoral training and development fellowship (20363/PD/17) of Consejería de Empleo, Universidades y Empresa (CARM) by the Fundación Séneca – Agencia de Ciencia y Tecnología de la Región de Murcia. Edited by: Pierre Herckes Reviewed by: three anonymous referees