Introduction

Looking to technical literature, a number of articles, books and monographs deal with organic pollutants and their characterization in environmental matrices. Thousands of compounds exist that affect all sectors of our world and impact on the living organisms and man. Clean air contains ca. 2 ppm of volatile hydrocarbons, many of which of natural origin (methane, isoprene and terpenes), while petroleum and fuel gas residues as well as pesticides, additives, plasticizers, fragrances and pharmaceuticals pollute inhabited and industrial areas (Hester and Harrison 1995). Both airborne particulates and emissions consist of organic matter in percentages ranging from 20 to 80 % (Seinfeld and Pankow 2003); it is a complex mixture of non-polar hydrocarbons (i.e. alkyl, aryl and polynuclear compounds) and polar substances, namely oxygenated (organic acids, phenols, alcohols and carbonyls) and halogenated (dioxin-like, perfluorinated) products, compounds of sulphur, nitrogen and phosphor, and metal organics (Turpin et al. 2000; Fraser et al. 2002; Alves 2008). Organics affect also waters and soils, characterizing both the surface and depths of Earth (Lake et al. 1979; Mai et al. 2002; Nadal et al. 2004).

Two complementary approaches, which can be called “deductive” and “inductive”, are followed to chemically characterize the environment. The former one consists of studying a few compounds, previously selected. This is the case of polynuclear aromatic hydrocarbons (PAHs) associated to aerosols or of fluorotelomer alcohols dispersed in surface waters. When that occurs, the chemical and physical properties of matrices and compounds are known, which favours the setup of reliable procedures for determining the analyte concentrations. The second approach is based on the chemical screening of a matrix extract, performed through general use methods that enrich the analytes and, conversely, cut off most potential analytical interferences. Through the first way, the target substances are determined with high accuracy, precision and sensitivity in well-characterized matrices. This approach looks like a tool subdued to several purposes, e.g. to evaluate the contamination degree of the environment, implement the environmental legislation, verify the effectiveness of detoxification processes and assess the health risk. The second approach is rather uncommon and a bit uncertain, but improves the knowledge of our world, providing background information for dedicated studies of new toxicants, new key compounds and new chemical kinetics.

Occurrence of organic pollutants in the environment and analytical approach

Focusing on the atmosphere, most persistent organic pollutants (POPs) and emerging organic contaminants (EOCs) partition between gas and particulate phases; moreover, organics tend to accumulate on the fine and ultra-fine particles, although exceptions and peculiarities exist. Therefore, to assess the true impact of substances on organisms and materials both vapours and size segregated particulate fractions must be collected and analysed separately. Nonetheless, investigations are usually focussed on only one substrate (e.g. PM10). Except for acidic and basic substances, semi-volatile organics have no group-specific sorbents (e.g. XAD-2). Thus, they are enriched as a whole from air, leaving the task of removing interferences to sample handling and instrumental analysis. Airborne particulates are collected onto inert membranes (Teflon or quartz) by operating at high or medium volume conditions (ca. 70 or 1.0–2.3 m3/h, respectively). The analytes are recovered from substrates by operating the soxhlet, microwave, ultrasonic bath or accelerated solvent extraction using polar solvents or mixtures. Thermodesorption/cryofocusing has found some applications, as the step following the vapour recovery from SPME cartridges exposed to atmosphere, or direct transfer of an aerosol aliquot into a micro-chamber acting as injector (Bates et al. 2008). With regard to compound identification and quantification, key points are neutrality, polarity, hydrophobicity and chemical stability. These properties influence the choice of the separation devices (columns, traps, cartridges and bio-accumulators), of gas, liquid or ion chromatography techniques adopted for analysis, and of the detection systems (flame ionization, electron capture, mass spectrometry (MS), fluorimetry and electrochemistry). In particular, various MS techniques have been adopted for this purpose: quadrupole and ion trap (EI), tandem (MS–MS), time-of-flight, high-resolution and triple quadrupole mass spectrometry. Each of them presents advantages and limitations depending on the nature of the analytes. For instance, flame ionization (FID) coupled to gas chromatography (GC) was the first tool to detect PAHs; thanks to low cost, ease use and wide linearity, it was the choice technique of the reference method for benzo[a]pyrene in urban air (IME 1994). Nevertheless, FID was soon replaced by mass spectrometry detection (MSD) owing to poor selectivity and impossibility of using perdeuterated homologues as reference or surrogate standards. Ion trap MS does not offer better performances than quadrupole MS, unless the tandem mode is applied to dirty matrices. As soon as GC–HRMS or high-performance liquid chromatography with fluorescence detection (HPLC-FD) has been adopted to perform instrumental analysis, the cleanup step included in the methods has been cut. Long-exposure diffusive samplers have been patented, suitable to collect semi-volatile organics from air. Nevertheless, their use is restricted to vapours of persistent substances like PAHs and PCBs (Söderström et al. 2005; Klánová et al. 2006; Tuduri et al. 2012). Filter/cartridge sets formally collect all PAHs and POPs from air, although only sampling trains encompassing denuders and filters seem capable of discriminating vapours from particulate contaminants (Gundel et al. 1995; Tsapakis and Stephanou 2003). Polyurethane foam (PUF), charcoal, polyvinylbenzene resins (XAD), reversed-phase surface modified membranes and silicone gum are adopted as cartridge sorbents or coated on high-efficiency denuders (Table 1).

Table 1 Block diagram of techniques used for measuring organic contaminants including PAHs, phthalates and psychotropic substances
Full size table

Neutral highly polar chemicals are important components of the solvent-extractable organic fraction of particulates, looking both to mass percentage and long-term toxicity (Fabiani et al. 2008). Nevertheless, they gained the concern of scientists later than non-polar n-alkanes and low-polar PAHs due to the hard identification and quantification. In fact, the highly polar fraction looked as a very complex mixture of chemicals often occurring at low trace levels, prone to decomposition and unable to provide specific signals suitable for identification through gas chromatography coupled with common detectors. Derivatization could be a useful tool, but not for all analytes. On the other hand, HPLC was insufficient until various hiphenated MS techniques were developed. The availability of high-resolution detectors improved the characterization of environmental matrices by providing the identification of unexpected substances like flame retardants, insecticides and drugs, and reducing the limit of detection of compounds by orders of magnitude.

Excellent reviews have been published on EOC analysis in waters (Richardson 2009) and in the atmosphere (Xie and Ebinghaus 2008), while several papers analyse specific classes of compounds, e.g. organic phosphates, flame retardants with special focus on polybrominated diphenyl ethers, and fragrances. Even much richer is literature concerning polychlorodibenzodioxins/furans, polychlorobiphenyls and organochloride pesticides.

The focus of this short review encompasses three groups of pollutants, i.e. PAHs, phthalate esters and drugs, which enjoyed different concerns along the last 30 years. In fact, PAHs belong by far to most known chemicals and are recognized as toxicants. Phthalates are principally used to modulate the polyvinyl chloride strength and flexibility. Apart from a number of industrial processes, phthalate esters occur in personal care, medical and consumer products (Petersen and Breindahl 2000; Rahman and Brazel 2004). Phthalate esters are ubiquitous in the environment; on the other hand, they are endocrine disruptors, mutagenic and probable promoters of obesity (Shaffer et al. 1945; International Agency for Research on Cancer 1982; USDHHS 1993; Bornehag et al. 2004; Martino-Andrade and Chahoud 2010; La Merrill and Birnbaum 2011; Silva et al. 2004; European Commission 2001, 2005). Their investigations in the environment are relatively recent and overall restricted to indoor air (Fromme et al. 2004; Wormuth et al. 2006). Drugs (both illicit and legal) are acquiring only today the concern that they probably merit (Cecinato and Balducci 2007; Balducci et al. 2009; Viana et al. 2010, 2011). Among psychotropic substances, nicotine has a peculiar importance with regard to pollution of indoor environments, owing to the recognized impact of tobacco smoke on human health (FDA 1996; WHO 2008, 2009, 2011). By contrast, very few studies have been focussed till now on the presence of nicotine, caffeine and the corresponding by-products in open air, and the environmental data available appear very scarce, compared to the worldwide diffusion of these substances.

Most PAH investigations focus their objectives on the high-boiling fraction since most of them are recognized as probable or possible carcinogens. Since measurements are made for health preservation purposes, the focus is on breathable or fine particulates, although some attention is paid also to settled dust. HPLC-FD and GC–MS are the techniques of choice for this purpose (Winberry et al. 1988). The use of low-bleed and inert GC phases allows to extend the study to high molecular weight congeners, e.g. dibenzo/naphthopyrenes and fluoranthenes (Menichini and Merlo 2012). Both positive and negative artefacts are known of the PAH measurements (e.g. the adsorption of PAH on soot/filter surfaces or conversely their volatilization, and the PAH decomposition induced by oxidants); thus, dedicated procedures and tools are adopted to minimize them (e.g. ozone scrubbers).

A number of PAH derivatives have been found more toxic than parent compounds due to presence of methyl-, nitro- or halogen substituents in the molecule (Konig et al. 1983; Allen et al. 1997; Durant et al. 1996; Hannigan et al. 1998).

Nitro-PAHs are potent mutagens, not requiring enzymatic activation to display their toxicity. As important components of old-engine diesel vehicles, they occur in the air as a consequence of either direct emission or photochemical reactions. In Italy, nitro-PAH measurements were made since 1994 (Liberti et al. 1984; Fuoco et al. 1999; Cecinato et al. 1998). Accounting for small percentages of PAHs, nitro-PAHs required sensitive and selective detectors to be identified and quantified. Thus, GC-NICI-MSD (Di Filippo et al. 2009), GC-MSMS or HPLC-FD after reduction and derivatization has been adopted.

As for oxygenated PAHs, four main categories have been investigated, namely phenols/diols (e.g. 2-naphthol, 9-hydroxyphenanthrene and 1-hydroxypyrene), ketones or quinones (9H-fluorene-9-one, anthracene-9,10-dione, 7H-benz[de]anthracene-7-one and benz[a]anthracene-7,12-dione), carboxyalhehydes (napthalene-1-carboxaldehyde, pyrene-1-carboxaldehyde) and oxygenated heterocyclics (e.g. xanthene and xanthone, naphthofuranes). Dedicated studies on oxygenated PAHs made between 1983 and 1998 (Ramdahl 1983; Cretney et al. 1985; Kamens et al. 1989; Strandell et al. 1994; Oda et al. 1998) and, more recently, by Rosario Sienra (2006), Park et al. (2006), Albinet et al. (2007), and Tsakapis and Stephanou (2007), were based overall on characterization through GC–MSD, however, HPLC-MSMS was proposed by Delhomme et al. (2008). A review of Walgraeve et al. (2010) discuss most results in this topic.

As for airborne phthalates, they were first enriched from air at high aspiration flow either onto quartz/glass fibre filters (dust) or polyurethane foam (vapours). Solvent extraction was operated in soxhlet by using benzene/methanol 4:1, hexane/diethyl ether 9:1 or dichloromethane as solvent (Thuren and Larsson 1990; Xie et al. 2006); afterward, the dried residue was processed through column chromatography (silica). The phthalates were eluted with the neutral polar fraction and characterized through capillary gas chromatography coupled with mass spectrometric detection, EI-SIM mode. Deuterated homologues acted as internal reference compounds. Recently, airborne phthalates were measured in Nanjing, China (Wang et al. 2008) after a dedicated procedure was optimized including alkyl phenols (Xie et al. 2006). The positive-ion chemical ionization was applied to phthalates and organophosphates in indoor dust. Special attention was paid to avoid the volatilization from polyurethane foam cartridges and filters, by reducing the sampling period to 12 h (Bergh et al. 2010).

The analytical procedures optimized for monitoring psychotropic compounds were substantially derived from those applied for drugs and their metabolites or by-products in aqueous media, with the exception of sampling (Zaromb et al. 1993). At this regard, cocaine, cannabinoids and heroin exist predominantly as particles, whilst nicotine, caffeine and amphetamines are expected to partition between gaseous and particulate phases, although the true respective percentages remain unknown (Häger and Niessner 1997; Lai et al. 2008). Psychotropic substances were investigated in the frame of an extensive characterization of organic aerosols including PAHs, n-alkanes and oxy-PAHs (Cecinato et al. 2009; Ladji et al. 2009). Cocaine was detected at concentrations as low as 0.005 ng/m3 (limit of detection, LoD, equal to 0.001 ng/m3), while the three cannabinoids identified in aerosols (i.e. cannabinol, cannabidiol and Δ9-tetrahydrocannabinol) were determined only when their sum exceeded 0.01 ng/m3 (LoD ~ 0.003 ng/m3 for each of them). The procedure comprised the sample collection at 55 m3/day, solvent extraction in soxhlet, fractionation on alumina column and GC-MSD (EI, SIM) analysis. Better analytical performances were obtained by applying HPLC-MSD and/or derivatization to cannabinoid analysis (Postigo et al. 2008, 2009; Balducci et al. 2009). HPLC-MSD allowed also to verify the occurrence of traces of amphetamines and heroin in the air of Spain at a few picogram levels.

Environmental EOC measurements

PAHs were detected in aerosols since 1950 (Waller 1952; De Maio and Corn 1996; Gordon and Bryan 1973; Gordon 1976). PAH measurements have been carried out everywhere, including polar and remote regions (Fellin et al. 1996; Ciccioli et al. 1994, 1996; Halsall et al. 1994; Khalili et al. 1995; Alves et al. 2001; Yassaa et al. 2001; Chang et al. 2004). In Italy, the first spotty measurements of airborne PAHs were made in the late 1960s (Zoccolillo et al. 1972). In the 1990s, investigations were made in many cities experiencing different degrees and types of pollution (Menichini 1992; Ferrovie dello Stato 1990) (see Table 2). Chemical characterizations were carried out by means of capillary GC-FID; reversed-phase HPLC-UV/FD was used only to reprocess congeners not resolved by GC phases.

Table 2 Atmospheric concentrations (nanograms per cubic metre) of carcinogenic PAHs, detected in Italian cities during winter 1990 (Ferrovie dello Stato 1990)
Full size table

A big impulse to investigations on PAHs was provided by environmental legislation (IME 1994; GURI 1999). Indeed, starting from 1994, the Authorities of Italian Regions must monitor particulate PAHs in the principal 15 Italian cities and implement policies aimed to reduce or maintain the yearly average concentration of benzo[a]pyrene (BaP) below 1.0 ng/m3. Two more recent decrees have modified this concentration limit into air quality target and extended the measurements to seven carcinogenic PAHs in urban and rural localities, in order to assess the average exposition of population to this category of pollutants (GURI 2007, 2010). Nowadays, the 1.0 ng/m3 target is usually reached in Italy, although important exceptions exist for areas hosting heavy industry concentrations like steel plants, refineries and harbours (e.g. in Taranto and Trieste). The PAH concentrations are declining since 20 years; nonetheless, BaP often exceeds 1.0 ng/m3 in winter, despite the annual average is lower (see Table 3).

Table 3 Atmospheric concentrations (nanograms per cubic metre) of carcinogenic PAHs in Rome and Milan, Italy
Full size table

First restricted to PAH measurements, the investigations have been recently addressed to five major goals, i.e.: (1) to determine airborne PAHs in remote or developing regions (Bin Abas et al. 2004; Fanou et al. 2006; Pozo et al. 2009; Ladji et al. 2009; Sofowote et al. 2010; Callen et al. 2011; Amador-Muñoz et al. 2011); (2) to improve the knowledge of the PAH toxicity apart from cancer promotion, assess the “relative toxic potency” of each congener compared to BaP and dioxin and evaluate the true exposition of humans to PAHs (Rosario Sienra 2006; Villalobos-Pietrini et al. 2006; Srogi 2007; Martorell et al. 2010; Wei et al. 2011); (3) to assess the relative contributions of emission sources to the airborne PAH budget (Li and Kamens 1993; Gogou et al. 1996; Schauer et al. 1996; Cass 1998; Sharma et al. 2007); (4) to implement environmental legislations (EPA 1999; EC 2001; Ravindra et al. 2008); and (5) to know the partition of PAHs between gas and particulates, and their distribution along the aerosol size fractions. Worth mentioning are also the development of new procedures and instrumentations (Klanová et al. 2006; Liu et al. 2007; Li et al. 2010; Costa Menezes et al. 2011), and investigations in indoor environments, e.g. in work places and offices, houses, churches, schools and shops, both in the presence and absence of tobacco smokers (Li and Ro 2000; Lodovici et al. 2004; Ding et al. 2007; Lu et al. 2008; Mannino and Orecchio 2008; Delgado-Saborit et al. 2011; Titcombe et Simcik 2011).

According to measurements made before 2000, in Italy the semi-volatile nitronaphtalenes were the most abundant nitro-PAHs, while toxic nitrofluoranthenes and nitropyrenes ranged from 10 pg/m3 to ca. 1 ng/m3 (Ciccioli et al. 1996; Cecinato 2003). By contrast, nowadays single NPAH concentrations exceeding 0.1 ng/m3 are unusual (Di Filippo et al. 2010). As for oxygenated PAHs, the concentrations in the atmosphere range from a few picograms up to tens of nanograms per cubic metre (Park et al. 2006; Albinet et al. 2007; Walgraeve et al. 2010). Unfortunately, no measurements have been made till now in Italy.

The occurrence of phthalate esters in the air at concentrations exceeding 300 ng/m3 was reported by Thomas in 1973 and Cautreels et al. in 1977 (Table 4). In the same years, the use of gas chromatography coupled with electron capture detection was proposed as an alternative to flame ionization since ECD appeared selective and high responding vs. both halogenated and poly-oxygenated molecules (Giam et al. 1975), whilst MS instruments were much more expensive. Phthalates were identified in dusts of Antarctica and Himalaya. Both vapour and aerosol fraction were investigated over the North Sea (Xie et al. 2005) and extensively measured in Chinese cities, where total airborne concentrations ranged from ≈60 to ≈2,200 ng/m3 (Wang et al. 2006). Unlike other contaminants, the highest phthalate values were observed in summer, and this finding was associated to volatilization from substrates. Despite the ubiquitous occurrence of phthalates in the atmosphere (Staples et al. 1994), the concern about these substances is overall restricted to indoor environments (Rudel et al. 2003; Fromme et al. 2004; Geiss et al. 2009), where concentrations are much higher than in open air (Table 4). In general, phthalates looked just as interfering compounds of true pollutants. Teil et al. (2006) compared the phthalate burdens of the air and rain water, and more recently Weschler et al. (2008) included indoor settled dust. The phthalate metabolites in humans have been studied by Koch et al. (2003, 2007) and Calafat et al. (2004).

Table 4 Concentrations of phthalate esters (nanograms per cubic metre) observed in outdoor and indoor air
Full size table

Nowadays, both tobacco smoking and illicit substances reach the size of global problems involving health, security, economy, ethics and policy (European Monitoring Centre for Drugs and Drug Addiction 2009; Italian Police 2009; Central Intelligence Agency 2008; United Nations Office on Drug and Crime 2009). Nevertheless, scarce attention is paid to the presence and aftermaths of psychotropic substances in the environment. Till now, nicotine has been widely studied only indoors, whilst measurements of other drugs have been made solely outdoors (Michael et al. 1996; Schorp et al. 2002; Nebot et al. 2005; Sureda et al. 2009). Illicit substances have been listed within the emerging pollutants (Richardson 2009) since they have been ascertained to affect the surface and waste waters (Castiglioni et al. 2006; Boleda et al. 2007; Postigo et al. 2008, 2009; van Nuijs et al. 2009; Mari et al. 2009). Quantitative relationships have been observed between the contents of cocaine, Δ9-tetrahydrocannabinol and heroin metabolites in waste waters, and the drug prevalence estimated at local and national scales (Bones et al. 2007; Zuccato et al. 2008; Domènech et al. 2009). By contrast, the measurements in open atmospheres have been started later since the atmospheric contamination induced by drugs is supposedly negligible. For instance, the studies focussed on tobacco smoking consider tar and gaseous toxicants in spite of nicotine (IARC 1982), and caffeine is practically disregarded. Moreover, only in 2007 and 2009 was the ambient air contamination by cocaine, cannabinoids and heroin clarified (Cecinato et al. 2007; Postigo et al. 2008). Nonetheless, legal and illicit psychotropic substances affect the whole world. Nicotine and caffeine can reach 100 and 30 ng/m3, respectively, while illicit substances sometimes exceed 1 ng/m3 and are often much more abundant than known toxicants, e.g. BaP and organic halogenides.

Till now, only two extensive studies have been carried out about this topic, namely the PSALM and AriaDrugs projects (Cecinato et al. 2010; Department for Anti-Drug Policies of the Italian Presidency of Ministers’ Council DPA 2010). As parts of PSALM, launched by CNR-IIA and based upon the voluntary participation of Regional Agencies for the Preservation of the Environment, two in-field campaigns were conducted in January and June 2009 in 38 Italian localities (see Table 5). AriaDrugs, funded by DPA, consists of continuous measurements carried out over 1 year in eight Italian big cities. Numerous outcomes and information were drawn from the former project, and even more are expected from AriaDrugs, whose end is scheduled for 2012. In particular, the variables potentially influencing the drug burdens in the air (e.g. meteorology, levels of oxidants and soot) will be evaluated, and the chance of deriving chemical indicators for the illicit drug abuse will be explored. Nevertheless, the health impact on populations will be elucidated only if long and extensive programmes will be implemented in regions of drug production, trafficking and consumption (European Monitoring Centre for Drugs and Drug Addiction 2009).

Table 5 Average contents (nanograms per cubic metre) of cocaine, cannabinoids and benzo(a)pyrene in the air of Italian cities in 2009 (Cecinato et al. 2011)
Full size table

Conclusions

On the basis of the renewed general concern, PAHs merit to be regarded again as “emerging contaminants”, going beyond the sheer monitoring in the atmosphere. It is quite surprising, for instance, that the influence of oxidants on atmospheric PAHs, although known, remains insufficiently studied, although tests conducted on artificial matrices or in artificial conditions have displayed quite conflicting results. Special attention would be paid to assess the real exposition to PAHs of the entire population or selected categories, taking into account reliable emission factors from human activities and natural events, as well as point and diffusive sources. In this regard, widely shared methods suitable to evaluate the various forms of toxicity induced by PAHs on humans are necessary to uniform legislation and epidemiological approach across the world (Collins et al. 1998).

In our opinion, phthalate esters will gain more environmental concern in the next future. Due to variety of uses they find in consumer and industrial products, favoured by their ability to modulate chemical and physical properties according to needs and by low production costs, it looks very hard their replacement with equivalent materials. Nevertheless, the degree of contamination reached in the environment, the knowledge of their toxicity and the persistence in the environment will lead community to develop phthalate-free technologies.

As for psychotropic substances, several questions are still waiting for answers, namely:(1) what are the predominant sources of environmental pollution at local and regional scales; (2) how persistent are psychotropic substances in the atmosphere, and how much the aerosols are representative of total burdens affecting the environment; nicotine in particular seems worth of more extensive studies in open air; (3) how realistic is the attempt of deriving drug prevalence indicators from the environmental monitoring; and (4) what are the detrimental effects induced by drugs and their by-products on human health, in particular on non-consumer people.

While niche-level researches appear reasonably suitable to tackle most emerging items, only large-scale programmes co-ordinated at international level can succeed in improving the general level of knowledge and, finally, the quality of our life.