An Analysis of BTEX Occurrence in Stored Rainwater and Rainwater Runoff in Urban Environment

Abstract

Climate change and its effects, for instance drought, drive the search for alternative water sources. One of these sources is rainwater, especially the runoff from various roof surfaces in cities. In turn, its use in the city for the production of food as part of hydroponic and aquaponic systems requires knowledge of possible pollutants and their varied concentrations. In this article, the concentrations of benzene, toluene, ethylbenzene, and xylene (BTEX) in rainwater collected in cities from various surfaces and stored in various types of tanks (open and closed) are analysed. Tests were carried out on extracted specimens using gas chromatography with a flame ionisation detector (FID). BTEX compounds were determined using a chromatograph with a FID sensor and a capillary column. Organic substances were extracted from the water with dichloromethane. The possibility of occurrence of BTEX compounds in rainwater flowing down from various roof surfaces in the city was confirmed. The obtained results suggest future research directions for mitigating BTEX rainwater pollution in order to expand the scope of its subsequent use. Preliminary guidelines for its treatment based on the literature were proposed. The possibility of using urban rainwater for hydroponic and aquaponic systems was assessed in terms of BTEX content.

1. Introduction

Current conditions and predicted climate change necessitate the need to take radical actions to minimise the consumption of high-quality tap water supplied to city residents in order to prepare them for the occurrence of water shortages and drought. An example of such a scenario is Day Zero in Cape Town [1]. Attempts to develop food sovereignty among urban residents are also important. For these reasons, the possibility of replacing any volume of tap water with water from alternative sources is becoming very important. This also applies to water used in food production in hydroponic and aquaponic systems. Rainwater is one of the main alternative water sources in the city. One way to achieve these goals is to collect and store rainwater for use, which is known as rainwater harvesting (RWH). For this reason, it is extremely important to adapt the infrastructure in cities to collect and store rainwater wherever possible. The amount of rainwater that can be collected is extremely important, which is analysed in many articles, as exemplified by previous works [2,3].

However, its quality is also important, as it establishes the possibilities of its use and determines the necessary treatment methods. In this respect, factors that may affect this quality are crucial. Among the range of applications of the discussed alternative water source, the possibility of using the collected rainwater for food production in cities with the increasingly popular hydroponic and aquaponic systems should also be considered.

Rainwater in cities is currently stored in various conditions: from professional water collection and storage installations made of materials dedicated to this purpose, to do-it-yourself (DIY) structures, and even elements intended for other purposes (buckets, construction trays, large utility boxes, containers left after storing various materials, etc.). The elements intended for collecting rainwater are available as both closed and open versions. The location of rainwater tanks in relation to the potential source of pollution, such as a street with car traffic or gas station, also differs. Moreover, air quality and the surfaces from which rainwater is collected in the city have an impact on its quality.

For the above-mentioned reasons, rainwater from the city may contain various contaminants, limiting its further use or requiring specific treatment methods.

There are many articles and studies in the scientific literature devoted to the analysis of pollutants found in rainwater. There are studies on the quality of rainwater in the broader sense of the word [4,5] and works dealing only with a selected group of pollutants, e.g., pesticides [6,7] or heavy metals [8,9]. There are also individual works dealing with rainwater sourced from one type of surface or from several types of surfaces [9,10]. It should be noted, however, that the testing of rainwater from some surfaces is rare, which was indicated in the work [11]. Green roofs and photovoltaic panels can be mentioned here. Work analysing the quality of rainwater also applies to various locations. This includes cities, e.g., Paris (France) [12], Gdansk (Poland) [9], and Genoa (Italy) [13] as well as rural areas [14]. There are also studies on the quality of rainwater around the world, such as [6,15]. However, there is insufficient work on investigating the occurrence of benzene, toluene, ethylbenzene, and xylene (BTEX) in urban rainwater from various roof surfaces.

The method of storing rainwater on residents’ private plots (open tank vs. closed tank) is also rarely taken into account. Moreover, due to changing conditions in cities, including air quality and the frequency and volume of rainfall, as well as the increasing need to collect rainwater for further use, these studies should be intensified and expanded both in terms of the analysed pollutants and specific location. The results of these analyses and studies may also help to protect the quality of rainwater against additional contaminants and establish guidelines for its storage and its treatment. These activities will allow the use of rainwater in cities to be expanded to include other applications, such as the use of hydroponic or aquaponic systems for food production in cities.

It is for these reasons that the purpose of this article is (i) checking the presence of BTEX and performing an analysis of BTEX levels in rainwater from urban areas from various sources (various roof surfaces, including a green roof) and stored in various conditions (open and closed tanks), (ii) formulating suggested further research directions in terms of mitigating BTEX rainwater pollution in order to expand the scope of its subsequent use, and (iii) suggestions regarding the need for rainwater treatment, as indicated by the results of the conducted research.

2. BTEX: Characteristics and Occurrence

Benzene series pollutants (BTEX) consist of benzene, toluene, ethylbenzene and three xylenes (o-, m- and p-). They are main representatives of the toxic volatile compounds generated primarily by traffic activities in urban environments [16]. As contaminated rainwater may not be suitable for reuse due to the risk for human health risk, understanding the distribution of BTEX in the environment is becoming increasingly important [17]. It is also crucial to gather the knowledge concerning their characteristics.

2.1. Benzene

Benzene is a flammable, volatile liquid with a characteristic odour (sweet) and explosive properties [18,19,20,21,22]. At room temperature, it is a colourless [20,21,22] or light yellow liquid [20,21]. Benzene vapour is heavier than air and can fall to lower areas [22]. Importantly, due to the potential possibilities for water pollution, according to sources [20,21], benzene dissolves only slightly in water and floats on its surface. The solubility of benzene in water is 1790 mg/L at 25 °C [23].

Benzene in the environment comes from both natural sources (gases emitted from volcanoes or forest fires) and from human activity [20,21,23]. In urban air, small amounts of benzene come from tobacco smoke, gas stations, car exhaust fumes and industrial emissions [20,21]. It is important to note that the air around hazardous waste disposal sites and gas stations may contain higher concentrations of benzene than in other places [20,21].

This compound is commercially produced from coal sources and petroleum sources [22]. It is used as an extracting, degreasing and cleaning agent [19]. This compound is used in many industries as a solvent (printing inks, rubber, adhesives, varnishes) or a starting material for further synthesis as well as a component of flammable mixtures [18,19,22,24]. Benzene is used as an additive to gasoline due to increasing the octane number and reducing knocking [22]. Among the industries in which it is used, it is worth mentioning the chemical, pharmaceutical, rubber, plastics, paint and varnish industries [19,22,24].

Although the main route of benzene absorption by humans is through the respiratory tract [18,19,24,25], it is also absorbed through the digestive tract and skin [18,19,24]. In the case of oral benzene poisoning, this compound causes local irritation of the respiratory tract mucosa [24]. Food or beverages containing large amounts of benzene taken orally can, within minutes to hours, cause symptoms such as vomiting, stomach irritation, dizziness, sleepiness, convulsions, fast or irregular heartbeat, and even death [20,21]. Benzene is biotransformed in the body, and its main metabolite is phenol [19,24]. Benzene is a toxic substance. It causes acute poisoning, chronic poisoning, as well as long-term effects [18,19,24]. The toxic effects of benzene are first revealed in the central nervous system, while the highest concentrations of this compound are reached in adipose tissue and bone marrow [19]. Benzene and its metabolites damage the hematopoietic system (bone marrow) [19,24]. This compound is classified as an indisputable chemical carcinogen [19].

Under Directive 2022/431 [26] amending Directive 2004/37/EC [27], the occupational exposure limit values measured or calculated over an eight-hour reference period as a time-weighted average have been significantly reduced. From April 2026, it will be 0.2 ppm (0.66 × 10⁻³ mg/L), and currently, it is 0.5 ppm (1.65 × 10⁻³ mg/L) [26].

The study [28] indicated that an unknown high-affinity pathway is responsible for most of the metabolism of benzene at airborne concentrations below parts per million (<1 ppm, i.e., <3.26 × 10⁻³ mg/L at 1013.25 hPa and 20 °C). The subsequent paper [29] provided statistical evidence that benzene is metabolized to phenol and E,E muconic acid by two enzymes, not one, and that the putative high-affinity enzyme is mainly active below 1 ppm (<3.26 × 10⁻³ mg/L). This is important because essentially all humans are exposed to benzene, and the compound must be metabolized to produce toxic effects [29].

2.2. Toluene

Toluene is a volatile, clear, colourless liquid with a characteristic odour [19,30]. It is used as a solvent for paints and varnishes as well as a starting product in the synthesis of dyes and explosives [19,24]. Among the industries in which it is used, the chemical and pharmaceutical industries should be mentioned [19]. Lower volatility than benzene (twice) and lower toxicity determine the wide use of this compound in the industry [18,19,24]. It is obtained as a product of dry distillation of coal, from gas tar and by the pyrogenic decomposition of crude oil [19]. The sources of exposure to toluene vapours and skin contact are the industries in which it is used, losses during the distribution of motor fuels, and emissions from exhaust gases of car and aircraft engines [24].

This compound is absorbed through the respiratory tract through the skin (liquid toluene and its solutions) and also, although more slowly, through the digestive system [19,24]. Toluene, like benzene, is a toxic substance. It causes acute poisoning as well as chronic poisoning. The critical organ for toluene is the central nervous system [24]. Toluene is relatively quickly biotransformed to the low-toxic benzoic acid and then to the hippuric acid [18,19,25]. The solubility of toluene in water is 526 mg/L at 25 °C [31].

Australian drinking water guidelines [32] recommend that toluene concentrations should not exceed 0.8 mg/L due to the health criterion, while in the case of the aesthetic criterion—already a value above 0.025 mg/L of this compound may cause problems [32].

2.3. Ethylbenzene

Ethylbenzene is a flammable liquid that occurs in industrial conditions in crude oil, gasoline, coal tar and its fractions and products [18]. It is used in the chemical industry for the production of plastic masses, as a raw material for the synthesis of styrene, and in its pure form, it is used as a solvent for paints and varnishes [18,19,24]. The main source of exposure to ethylbenzene is the presence of ethylbenzene in gasoline and its use as a solvent [24]. The main route of absorption is the respiratory system, although it can also enter the body through direct contact with the skin [19,24]. It can cause acute and chronic poisoning [18,19]. Its toxic effect is similar to benzene but with a much lower intensity of effects [18].

2.4. Xylene

There are three isomers of xylene: o-xylene, m-xylene, and p-xylene, whereas mixed xylenes are a mixture of three isomers and a small amount of ethylbenzene [33,34]. The commercial product is a mixture of three isomers [24]. It is a flammable liquid that is colourless and has a characteristic odour [24].

Xylene is easily absorbed through the respiratory tract and the digestive tract but more slowly through intact skin [24]. This compound is a toxic substance. It has a narcotic effect on the central nervous system and strongly irritates the mucous membranes [24]. It causes acute poisoning and chronic poisoning [24].

2.5. The Occurrence of BTEX in Rainwater and Stormwater—Previous Studies

There are many investigations on the quality of rainwater collected directly and flowing down from various surfaces, which can be confirmed by studies [13,35,36,37]. For instance, in the study [13], such quality parameters as total suspended solids (TSSs), chemical oxygen demand (COD), selected heavy metals, ammonium ion (NH₄⁺) and pH were examined in road runoff and roof runoff. In the work [38], runoff from model roofs covered with various materials was examined in terms of the content of selected heavy metals, the assessment of electrical conductivity (EC) and pH. In turn, in the study [9], the content of metals in rainwater runoff was studied but from existing buildings. The work [35] is entirely devoted to the content of metals in rainwater flowing from roofs with different roofing materials. In the study [39], determining the quality of rainwater from a green roof, the content of biogenic compounds and selected metals was investigated. In another work [37], concerning the quality of runoff from green roofs, a similar scope of research was adopted, i.e., selected metals (although in a much wider scope), pH, EC, and total dissolved solids (TDSs). In the study [36], rainwater taken from the drainage of green roofs was tested for the content of biogenic compounds and selected metals. The pH, EC, TDS, and turbidity were also checked [36]. There are also studies on the content of polycyclic aromatic hydrocarbons in rainwater [40,41] or stormwater runoff [42].

However, the discussed studies did not deal with the content of BTEX compounds in the tested rainwater/stormwater. The number of articles on the occurrence of BTEX in rainwater, rainwater runoff from various surfaces in the city (above ground level, i.e., excluding roadways) and stormwater is still insufficient. This is all the more essential, especially since due to climate change, this water is increasingly considered an alternative water source for cities. In such a situation, each additional study in a new location is a novelty and brings new content to the existing state of knowledge. This is particularly important when it concerns rainwater runoff from various surfaces in the city (above ground level), rainwater from a green roof as well as samples corresponding to the actual conditions of water use by residents.

In the scientific literature, the subject matter of the occurrence of BTEX compounds in the environment appears quite frequently and is widely described. There are papers about the presence of BTEX in soil, groundwater and surface water, e.g., [43,44,45]. The occurrence of BTEX compounds in the air is the subject of much attention, as exemplified by studies on specific cities, e.g., Algiers City [46], Ahvaz [47] or Tehran [48], or specific types of facilities (e.g., gas stations [49]). It should be added, however, that although there are many papers regarding the presence of BTEX in the ambient environment, especially in the air, the same situation does not apply when it comes to scientific studies concerning their presence in rainwater. The number of papers regarding the occurrence of BTEX in rainwater and rainwater runoff from various surfaces (i.e., stormwater) in the city is insufficient, especially due to the fact that due to climate change, this water is increasingly considered as an alternative water source for cities.

Papers on the issue of the presence of BTEX in rainwater and its runoff from various surfaces appear in different contexts of this issue. A common context is the occurrence of build-up of volatile organics on urban roads due to traffic activity and the wash-off volatile organic compounds (VOCs) from these roads. Examples of studies on this topic include papers concerning Australia [50,51,52] and China [16,53]. A separate topic is the issue of the possibility of removing aromatic hydrocarbons from the air through rain. The work [54] should be mentioned here. The possibility of treatment stormwater contaminated with VOCs is also an important context. In this group of papers, the work [55] concerning the assessment of operation of coalescence separations installed on the way of flow of rainwater and meltwater to Slupia river, Poland, can be mentioned as an example.

Among the few studies on the investigation of BTEX concentration in rainwater or its runoff, the following papers should be mentioned: [56,57,58,59,60,61,62]. The studies can be basically divided into those concerning rainwater and stormwater.

In the study [59], rain samples from Wilmington, NC, USA were analysed for i.a. BTEX. The results obtained by authors are the evidence that gas-phase levels can be good predictors of VOC trends in rainwater. In the work [62], rainwater samples were taken in Yokohama, Japan, and tested for i.a. BTEX compounds. Rain droplets’ rainwater concentrations did not depend on the rainfall intensity, and the temporal variation of their concentrations was similar to that of gas-phase concentrations [62]. During the research of [56], rainwater samples were taken in the central part of Tehran, Iran and tested for i.a. BTEX. Mean concentrations for benzene, toluene, ethylbenzene and xylenes (all three isomers) were 25 µg/L, 37 µg/L, 16 µg/L and 21 µg/L, respectively.

In work [57], the content of VOCs (including BTEX) in stormwater samples collected from storm drains in Seoul, Korea, was investigated. According to the results of these studies, m-xylene was detected in almost all of the stormwater samples, and the presence of benzene, toluene, ethylbenzene and o-xylene was frequently recorded in the tested stormwater. The maximum recorded concentrations for benzene, toluene, ethylbenzene and xylene (the sum of m-, o- and p-) were 33 µg/L, 29,310 µg/L, 960 µg/L and 6740 µg/L, respectively.

In the study [60] of VOCs in stormwater, samples were collected from six different land use sites from a community of Beijing, China. Benzene, toluene, ethylbenzene, and m-, o-, and p-xylene were detected in all samples collected from locations such as highway junctions, city roads, gas stations, parks, campuses, and residential areas. These studies [60] indicated that land surfaces may be the main sources of VOCs in rainwater other than the ambient atmosphere. Statistical analysis results (ANOVA) indicated that both land use type and precipitation time intervals could significantly influence some stormwater VOCs variations, which include BTEX [60]. In turn, correlation analysis of results showed that BTEX compounds were transferred from chemical plants by the wind [60]. In the work [61], i.a., the content of BTEX compounds was analysed in stormwater samples collected from over 40 different sampling locations in North Carolina. M-, p-xylene and toluene were detected in more than half of all samples analysed, whereas benzene was only detected in less than 10% of all samples analysed. The maximum concentration measured for benzene was 0.15 µg/L, and for toluene, it was 32.84 µg/L [61]. The study conducted by [58] aimed to investigate the variability in the build-up of particle-bound volatile hydrocarbons on urban road surfaces and the resulting uncertainty inherent to the build-up process. It was indicated that VOCs exhibit significant variations in the load and composition in build-up during different seasons [58]. The obtained research results highlighted the potential of climate change influencing stormwater pollution [58].

2.6. BTEX in Urban Rainwater and Stormwater—Sources and Causes of Occurrence

In densely populated cities, there are anthropogenic sources of BTEX such as intense transportation, industrial and commercial activities [63]. In urban areas, the volatile compounds are believed to be mainly derived from vehicle emission and domestic coal combustion for heating in winter, far exceeding the limit of BTEX pollutants’ concentrations of indoor air quality [17].

The presence of volatile aromatic hydrocarbons in a wide range of environmental samples like water, air and soil increases the risk of human exposure through the ingestion, inhalation, or absorption through the skin [17]. According to the United States Environmental Protection Agency (US EPA), the maximum concentrations of benzene, toluene, ethylbenzene and xylenes permissible in drinking water were set at 0.005 mg/L, 1 mg/L, 0.7 mg/L and 10 mg/L, respectively [64]. Compared with TEX compounds, the distribution of benzene in water is relatively small. In the stormwater reuse scenarios by [16], the human health risk of BTEX pollutants has the following pathways: ingestion as drinking water, incidental ingestion during swimming, dermal contact exposure during swimming and dermal contact during shower and bath.

The rainwater quality was investigated, for instance, in Erbil Province (Iraq) from the point of view of various immediate uses such as irrigation, fish production, swimming and building [65]. The authors claim that for drinking and domestic purposes, only sedimentation and disinfection are needed. However, for the rainwater quality assessment in the city centre and outside Erbil city, there were mentioned parameters such as sulphate, turbidity, total solids, electric conductivity, pH, etc., but the effects of BTEX pollutants’ volatilization were not considered.

A hazard quotient and hazard index approach were used to estimate the risk provided in commercial, industrial and residential land areas of Shenzhen (China) [16]. The modelling approach developed in their study can be further improved by including more land types such as green land and parking lots. As concluded by the authors of [16], in order to enhance the accuracy of calculation outcomes and resulting maps, directly collecting stormwater samples in real rainfall events and testing BTEX pollutant concentrations for samples from each rainfall event could be considered.

The seasonal temperature and precipitation have an influence on the BTEX levels in natural waters. It was shown that rainwater could not effectively remove BTEX accumulated in the air, and only 0.1% of BTEX was removed by rainwater in Willington (NC, USA) over one year of monitoring [17]. Volatile BTEX species are accumulated in rainwater precipitation during the atmospheric phase under favourable meteorological conditions [66]. The authors of [66] mentioned the fact that the various formations of atmospheric water include rain, fog, dew and snow, which are complex mixtures characterized by different pH values and different content of salts, organic material and acids. More specifically, the physicochemical properties of rainwater samples depend on the sample origin and ambient temperature, rainfall intensity and air masses coming from different source regions and heights (wind speed and direction, pressure, humidity) and on the season [54]. The rainwater capacities for BTEX scavenging from ambient air were investigated by [54,66]. To better understand the possible mechanisms of BTEX partitioning between aqueous and gaseous phases during rainfall, the authors developed an analytical dynamic dilution system combined with a proton transfer reaction mass spectrometer [66]. It was assumed that the enhanced uptake of BTEX from the gaseous to aqueous phase is not primarily associated with the hydrogen bonding of volatile species gas bubbles with atmospheric water molecules. In the laboratory studies performed by [66], it was shown that mechanisms other than dissolution may also contribute to the linearity of the relationships described by Henry’s law. A clear positive correlation between the calculated enrichment factor and parameters characterizing interfacial adsorption (e.g., total van der Waals surface area) indicated that the adsorption phenomenon was a major mechanism having an impact on BTEX partitioning in ultra-pure water at a constant temperature of 25 °C [66]. The influence of different pH values and contents of salts, organic materials and acids on volatile compounds’ wet deposition for various formations of atmospheric water including rain, fog, dew and snow remains a matter of scientific debate. In further studies [54], the functional dependency of the BTEX enrichment factor on the gaseous concentrations, physicochemical properties of rainwater and meteorological parameters was examined by regression multivariate methods. A total of 53 sample pairs of air and rainwater were collected simultaneously during several rain events in Belgrade (Serbia) in 2015. The results revealed ones more that the registered amounts of BTEX in rainwater samples were higher than those predicted by Henry’s law, thus pointing to a possible adsorption mechanism.

According to studies [50,51], urban water quality can be significantly impaired by the build-up of pollutants such as VOCs on urban road surfaces due to vehicular traffic. In order to minimize the contamination of rainwater runoff from various surfaces, it is necessary to acquire the knowledge concerning the build-up processes of pollutants. Examples of investigations concerning the build-up processes of VOCs can be found in several papers [50,51]. Hong et al. [16] consider built-up BTEX loads on urban roads’ surfaces during the dry periods, which can be washed by stormwater runoff during rainfall periods. It was assumed that every rainfall event can wash off all BTEX pollutants. In terms of BTEX pollutant build-up on street surfaces in Shenzhen (China), benzene showed relatively lower build-up loads compared to toluene, which demonstrated the highest build-up loads [16]. Therefore, the authors of [16] decided to exclude benzene from the human health risk assessment in their study.

3. Materials and Methods

The main objective of this article is to analyse the practical conditions encountered when managing stormwater/stormwater runoff in cities. The research material consisted of rainwater/rainwater runoff samples collected from two Polish cities: Cracow and Wroclaw. These cities are among the 5 largest cities in Poland [67]. Rainwater samples were taken from open and closed tanks made of different materials (plastic, glass) and with different storage times. This was important, because it reflected the practical conditions occurring when using rainwater on plots used by residents. Open tanks were located in different locations in relation to the roads. Water in the tanks was collected from various surfaces:

-

Sheet metal (including sheet metal with bituminous elements used for repairs);

-

Terraces made of terracotta;

-

Roofs made of ceramic tiles;

-

Corrugated polyester boards reinforced with glass fibre;

-

Extensive green roofs (after passing through the green roof layers).

Figure 1 presents the roof made of corrugated polyester board reinforced with glass fibre—one of the surfaces from which running off rainwater was collected and sampled (a district located further from the city centre).

Figure 2 presents the surface of the green roof of Aquafarm USAGE in Wroclaw, Poland—the type of roof from the outflow of rainwater (after infiltration) was collected in a closed plastic reservoir (with drinking water approval) inside a container and tested for the occurrence of BTEX. Figure 3 shows the entire Aquafarm from the ground level. The Aquafarm was built as part of the USAGE project [68]. One of the goals of its construction was to confirm the suitability of rainwater/rainwater runoff for supplying the aquaponics system (assuming its treatment, among others, by the green roof).

In addition, rainwater was also collected directly, i.e., without contact with any surface (roof/terrace/drainage system) but in different areas:

-

In the city centre;

-

In the district located further from the city centre;

-

On the outskirts of the city but close to the approach path of airplanes to the airport.

In these cases, contact with the tank material was also analysed. Additionally, these samples allowed to estimate the impact of surfaces from which rainwater flows (or with which it comes into contact) on its quality, serving as an environmental background. It was preliminary analysis.

The methods of rainwater/rainwater runoff collection depend on the purpose of the research. For example, in the study [56], rainwater was collected in 500 mm diameter containers. In another study [59], water was collected in 4 dm³ glass beakers placed in automatic collectors. In turn, in the study [57], stormwater was collected in a storm drain, and a vial was submerged into the water.

Due to the application nature of the research, it was assumed that the sampling method should reflect the actual conditions of drawing of stored rainwater/rainwater runoff by citizens. Samples were collected in a manner adapted to the type of tank (open tank vs. closed tank vs. open container). Samples were collected, among other locations, from the following:

-

Via a tank tap (in the case of large closed tanks equipped with a tap);

-

By submerging a bottle (or a beaker in the case of small container/tank) under the water table (in the case of open tanks).

Samples were collected into dark glass bottles with a capacity of 1 dm³. Detailed characteristics of the location of each sampling point, along with the type (open/closed) and material of the tank/container, as well as the characteristics of the rainwater runoff surface (if applicable), are presented with the test results.

BTEX organic compounds were determined chromatographically using the SRI 8610C apparatus from SRI Instruments with a FID sensor and a capillary column GL-GC29477—Inter Cap Aquatic (75 m, 0.53 mm ID, 2.00 µm film).

Organic substances from water were extracted with pure dichloromethane (methyl chloride) in the following proportion: 50 mL water + 20 mL CH₂Cl₂. The extraction process consisted of shaking the mixture of water sample and methyl chloride five times for 1 min with the separation of fractions lasting 2 min between each shaking. BTEX extraction from rainwater was carried out from a sample with a volume of V_w = 50 mL using methyl chloride with a volume of V_ch.m. = 20 mL.

The mass of BTEX in the sample was determined based on an external standard. Figure 4a–d show the relationships between the peak area of BTEX in the chromatogram and the mass of BTEX in a 10 µL injection.

The partition coefficients K_H for BTEX were determined experimentally using the following relationships:

-

Mass balance for each substance of BTEX was calculated based on the general mass balance equations [69], and for this study, it is as follows:

$${\mathrm{C}}_0·{\mathrm{V}}_0={\mathrm{C}}_{\mathrm{W}}·{\mathrm{V}}_{\mathrm{W}}+{\mathrm{C}}_{\mathrm{m}.\mathrm{ch}..}·{\mathrm{V}}_{\mathrm{m}.\mathrm{ch}.}$$

(1)

where

C₀—initial concentration of the BTEX standard in the aqueous phase (μg/mL);

C_W—concentration of the BTEX standard in the aqueous phase at absorption equilibrium (μg/mL);

C_m.ch..—concentration of the BTEX standard in the methyl chloride at absorption equilibrium (μg/mL);

V₀—initial volume of water sample (mL);

V_W—volume of water sample after extraction (mL), V₀ = V_W;

V_m.ch.—volume of methyl chloride (mL)

-

Absorption equilibrium equation (Henry’s law [69]):

$${\mathrm{C}}_{\mathrm{m}.\mathrm{ch}.}={\mathrm{K}}_{\mathrm{H}}·{\mathrm{C}}_{\mathrm{w}}$$

(2)

where

K_H—Henry’s constant (ml water / ml methyl chloride).

From Equations (1) and (2), it follows that

$${\mathrm{C}}_0·{\mathrm{V}}_0={\displaystyle \frac{{\mathrm{C}}_{\mathrm{m}.\mathrm{ch}.}}{{\mathrm{K}}_{\mathrm{H}}}}·{\mathrm{V}}_{\mathrm{w}}+{\mathrm{C}}_{\mathrm{m}.\mathrm{ch}.}·{\mathrm{V}}_{\mathrm{m}.\mathrm{ch}.}$$

(3)

and from (3), K_H is equal:

$${\mathrm{K}}_{\mathrm{H}}={\displaystyle \frac{{\mathrm{C}}_{\mathrm{m}.\mathrm{ch}.}·{\mathrm{V}}_{\mathrm{w}}}{{\mathrm{C}}_0·{\mathrm{V}}_0-{\mathrm{C}}_{\mathrm{m}.\mathrm{ch}.}·{\mathrm{V}}_{\mathrm{m}.\mathrm{ch}.}}}\stackrel{{\mathrm{V}}_{\mathrm{w}}={\mathrm{V}}_0}{====}{\displaystyle \frac{{\mathrm{C}}_{\mathrm{m}.\mathrm{ch}.}}{{\mathrm{C}}_0-{\mathrm{C}}_{\mathrm{m}.\mathrm{ch}.}\cdot \frac{{\mathrm{V}}_{\mathrm{m}.\mathrm{ch}.}}{{\mathrm{V}}_{\mathrm{w}}}}}$$

(4)

The values of the K_H partition coefficients are listed in

Table 1. Henry’s constants for the water–methyl chloride system, temperature: 293 K.

Substance	C0 [μg/mL]	A [mV·s]	M [μg]	Cm.ch. [μg/mL]	CW (Equation (2)) [μg/mL]	KH (Equation (4))$\left[\frac{\mathbf{m}\mathbf{l}\mathbf{w}\mathbf{a}\mathbf{t}\mathbf{e}\mathbf{r}}{\mathbf{m}\mathbf{l}\mathbf{methyl}\mathbf{chloride}}\right]$
1	2	3	4	5	6	7
Benzene (78.1 g/mol)	0.17530	1.979	M = 0.001933 · A −1.5725·10−5 = 0.003810	0.3810	0.02295	16.6
Toluene (92.1 g/mol)	0.17246	2.120	M = 0.002213 · A −0.001021 = 0.003671	0.3671	0.02567	14.3
Ethylbenzene + m-xylene + p-xylene (106 g/mol) (mean value of KH)	0.51666	4.407	M = 0.003285 · A −0.004959 = 0.009518	0.9518	0.1360	7.0
o-xylene (106 g/mol)	0.17510	2.127	M = 0.002542 · A −0.001521 = 0.003886	0.3886	0.01963	19.8

M—mass of BTEX substance in 10 μL of methyl chloride’s injection; A—chromatographic peak area.

. BTEX aqueous solutions with a concentration of C₀ (Table 1, column 2) were extracted with methyl chloride (50 mL water + 20 mL CH₂Cl₂). A volume of 10 µL of the extract was introduced into the chromatographic column, and the peak area A was determined. Based on A, the mass M of BTEX was calculated (Table 1, column 4) using the relationships given in the graphs (Figure 4a–d). The concentration of BTEX in the extract (methyl chloride) C_m.ch.. was calculated based on the proportions (Table 1, column 5). Using Equation (4), the K_H coefficient was calculated (Table 1, column 7). In addition, the equilibrium concentration in water C_W (Equation (2)) was calculated at the extraction equilibrium state (Table 1, column 6).

Knowing the concentration of BTEX C_m.ch. in the methyl chloride fraction and K_H from Table 1, the concentration in the water sample C₀ can be calculated from the following formula (transforming Equation (4)):

$${\mathrm{C}}_0={\mathrm{C}}_{\mathrm{m}.\mathrm{ch}.}\left({\displaystyle \frac{1}{{\mathrm{K}}_{\mathrm{H}}}}+{\displaystyle \frac{{\mathrm{V}}_{\mathrm{m}.\mathrm{ch}.}}{{\mathrm{V}}_{\mathrm{w}}}}\right)$$

(5)

4. Results

The results calculated based on the methodology described above (in the Materials and Methods section) are presented in

Table 2. BTEX concentrations in stored rainwater/rainwater runoff samples collected from different measurement points.

Number of Sample	Characteristic of Water Sample Collection Point	Benzene	Toluene	Ethylbenzene + m-xylene + p-xylene	o-xylene
		μg/L
1	the centre of Cracow city, runoff from the roof, closed plastic tank	N.D.	N.D.	N.D.	N.D.
2	runoff from the building’s roof surface (roof made of ceramic tiles, terracotta), exposure to communication route, PVC gutter system, open tank	4.53	N.D.	15.45	N.D.
3	direct rainfall, open tank	6.00	N.D.	N.D.	N.D.
4	partly direct rainfall, partly runoff from the roof, runoff from the roof made of glass fibre-reinforced polyester, used for about 20 years, covered with a layer of pollution originated from Cracow’s air(especially smog), roof exposure to a nearby gas station, open tank	11.87	7.87	26.16	N.D.
5	the suburb of Cracow city, metal roof, airport nearby, closed plastic tank	N.D.	N.D.	55.05	N.D.
6	the centre of Cracow city, open black plastic container	32.55	N.D.	N.D.	N.D.
7	runoff from the building’s roof surface (roof made of ceramic tiles, terracotta), exposure to communication route, PVC gutter system, open tank	16.76	12.45	18.49	15.88
8	direct rainfall, open tank	63.43	24.51	N.D.	63.70
9	partly direct rainfall, partly runoff from the roof, runoff from the roof made of glass fibre-reinforced polyester, used for about 20 years, covered with a layer of pollution originated from Cracow’s air (especially smog), roof exposure to a nearby gas station, open tank	8.53	14.16	145.11	34.49
10	the suburb of Cracow city, airport nearby, open black plastic container	N.D.	N.D.	N.D.	N.D.
11	the suburb of Cracow city, airport nearby, open white plastic container	N.D.	N.D.	7.16	N.D.
12	the suburb of Cracow city, airport nearby, open glass container	N.D.	N.D.	N.D.	N.D.
13	rainwater collected by a PVC gutter system, infiltrated through the green roof, closed tank with drinking water approval	N.D.	N.D.	N.D.	N.D.
14	runoff from the roof made of sheet metal, gutters made of galvanized sheet metal, presence of communication routes near sampling point, open tank	N.D.	N.D.	24.46	N.D.

N.D.—not detected.

. The chromatographic column allows the complete separation of benzene, toluene and o-xylene. They correspond to separate peaks in the chromatogram. The separation of ethylbenzene, m-xylene, and p-xylene is practically impossible in this column. The peaks of these three substances overlap or are very close to each other. Therefore, their total content in the sample was determined.

The results were presented in two ways—in tabular form and graphically. The measurement results are listed in Table 2. If there was more than one obtained result, the average values were included. The location of water sampling points, the rainwater collection method (including materials with which stored rainwater/rainwater runoff came into contact) and the type of water collection tank/container are described in Table 2 in detail.

The summary of BTEX contaminant concentrations (when detected) in samples are listed in

Table 3. Summary of BTEX contaminant concentrations when present in samples.

	Mean μg/L	Median μg/L	Sample Standard Deviation μg/L	Maximum μg/L
benzene	20.5	11.9	21.2	63.4
toluene	14.7	13.3	7.0	24.5
ethylbenzene + m-xylene + p-xylene	41.7	24.5	48.0	145.1
o-xylene	38.0	34.5	24.1	63.7

. In addition, Figure 5 presents a box-and-whisker plots with the calculated 1st and 3rd quartile (taking into account the median) and outliers. Calculations were performed in MS Excel.

Additionally, to facilitate the analysis of the obtained results, they were presented in graphical form. Figure 6 presents combo charts of grouped columns, illustrating the concentrations of individual compounds from the BTEX group.

5. Analysis and Discussion of the Results

The highest concentrations of measured hydrocarbons exhibit ethylbenzene + m-xylene + p-xylene (measured as a sum of three substances). They were detected in seven samples. The highest concentration was found in the samples no. 4, 5, 9 and 14 (Table 2 or Figure 5)—all above 20 µg/L. Water sample no. 4 and no. 9, collected in an open tank, come from a roof made of corrugated polyester board reinforced with glass fibre, used for about 20 years (presented in Figure 1), covered with a layer of pollution originated from Cracow’s air (especially smog). The sample no. 5 concerns water flowing from a metal roof into a closed plastic tank. Dust and other organic substances accumulated on the roof surface (e.g., bird droppings, substances secreted by plants creeping on the roof surface) can adsorb volatile hydrocarbons, which are then washed away by rainwater. The likely source of BTEX compounds in sample no. 5 is pollution of the air generated by the presence of the airport, approach track for aircraft and nearby highway. The dominant westerly winds, characteristic of the sampling site, contribute to the inflow of air containing hydrocarbons from the airport, the nearby highway, and the gas station. BTEX compounds were also detected in sample no. 14, which was the runoff from a sheet metal roof into the open tank. In this case, the presence of transport routes may contribute to the increased content of these three detected hydrocarbons in rainwater and in water running off from a roof covered with dust that can adsorb hydrocarbons.

O-xylene was detected in the fewest samples, i.e., three. All three samples were collected from the same plot but from containers collecting rainwater or rainwater runoff from different surface types.

Toluene was detected in only four samples. These were samples taken from the same location. Toluene, after o-xylene, is the compound detected in the lowest number of samples. The source of toluene in samples no. 4, no. 7, no. 8 and no. 9 is probably the atmospheric air pollution, including smog, and other contaminants generated by external sources (transport routes, gas stations). The hydrocarbon may be adsorbed by dust or other substances covering the roof.

Benzene was detected in the largest number of samples taken, i.e., seven. The benzene content in no sample exceeded 20 µg/L. All samples in which benzene was detected came from open tanks. Samples no. 2, 3, 4, 7, 8 and 9 came from tanks located in the same area but differed in the type of runoff surface (direct precipitation/runoff from different building surfaces) and exposure to the gas station and traffic routes. The source of benzene in these samples could be the pollution of atmospheric air: for instance, urban smog. Moreover, the sampling area was located near the gas station and two transport routes. Similar results were obtained in research [70], which showed that gas station areas have a potential to export high loads of BTEX pollutants when rainfall events occur. The highest concentration of benzene was detected in samples no. 6 and no. 8 (Table 2). Its source is probably the pollution of atmospheric air, i.a., urban smog. Sample no. 6 was collected in the city centre from an open container, while sample no. 8 was taken from a tank collecting directly rainfall on a plot located away from the very centre of the city. It is possible that the materials of the tanks were also a source of hydrocarbons.

Most of the samples in which benzene was not detected came from locations on the outskirts of the city (samples no. 5, 10, 11, and 12; see Table 2). Sample no. 13 came from the city centre but from a location in a green area and sheltered by trees.

The obtained concentration values in the samples in which BTEX compounds were detected are higher than the average values recorded by [57,60,61] in stormwater, although the maximum concentration recorded by [61] for toluene (32.84 µg/L) was higher than in the present study (toluene 24.51 µg/L). In turn, all determined concentrations for toluene and almost all for benzene (except for sample no. 6 and 8) were lower than the average value for these compounds from the rainwater studies conducted by [56].

Comparing the recorded concentrations of toluene, o-xylene and the sum of ethylbenzene + m-xylene + p-xylene with the permissible values listed in [71] for drinking water (United States [64], Australia [72] and World Health Organization (WHO) [73]), it can be stated that they are significantly below the acceptable limits of BTEX in water. The aesthetic criterion [71,72] for the detected values of ethylbenzene may be exceeded (it is not possible to state unequivocally, because it was determined as a sum with m-xylene and p-xylene). The above-mentioned aesthetic guideline [71,72] for xylene is exceeded due to detected values for o-xylene for two samples. It is not possible to state unequivocally the number of samples in which this criterion may be exceeded, because other forms of xylene were determined as a sum with ethylbenzene. The only problem may be with the increased content of benzene. In comparison to the above-mentioned list [71], the benzene concentration was exceeded for four samples according to the WHO guidelines [71,73], for six samples according to the United States Environmental Protection Agency (US EPA) guidelines [64,71], and for all samples according to the Australian guidelines [71,72]. It should be noted, however, that these are guidelines for already purified water.

Due to a possible increased content of benzene, it is advisable to purify rainwater before using it, e.g., for hydroponics. BTEX aromatic hydrocarbons can be removed from water by adsorption on granular-activated carbon (GAC). There are a number of works on this issue [74,75]. According to the WHO guidelines, the required quality level for drinking water should be achieved using GAC or air stripping [76]. The removal of BTEX compounds from aqueous solutions has been successfully achieved using processes such as chemical oxidation, biological treatment, air stripping and adsorption [77,78]. Promising results in the removal of BTEX compounds from aqueous solutions were also obtained using montmorillonite modified with tetradecyl trimethyl ammonium bromide (TTAB—Clay) [77]. Benzene removal using activated carbon was 4.5 mg/g, and using TTAB—Clay, similar results were achieved, i.e., 3.98 mg/g [77,79]. At the detected concentrations, a GAC bed with appropriate parameters should be used, and in the future, it may be replaced by less expensive equivalents.

Particularly noteworthy is sample no. 13, which is rainwater that passed through the layers of the green roof of Aquafarm (Figure 2). No BTEX compounds were found in the sample. It should be added that Aquafarm was located in the centre of Wroclaw city and close to the road but in a place sheltered by tall trees and the high Water Tower building. Moreover, the green roof collected water only from direct rainfall. Rainwater from any surface was not redirected onto it. Furthermore, the reservoir storing water from the green roof was a tank approved for drinking water.

Changes in the BTEX content may probably be related to, among others, air quality, terrain sculpture, smog occurrence, the type and degree of contamination of surfaces from which rainwater runoff occurs, but also the material of the tank. The changes may be possibly also related to the frequency of precipitation, intervals between rainfall events, the period of collecting and time of storing rainwater in the tank, the type of tank, distribution rainwater use patterns and distances from specific infrastructure (for instance, gas stations and chemical plants).

Analysis conducted by [60] indicated that both land use type and precipitation time intervals could significantly influence some stormwater VOCs variations, including BTEX. The work [70] also indicates the role of climatic factors, for instance temperature, which could also have certain influences on the accumulation of BTEX in rainwater. In the work [60] based on correlation analysis, it was found that the BTEX compounds (benzene, toluene, ethylbenzene, xylenes) were transferred from chemical plants by the wind. In turn, rain droplets’ rainwater concentrations taken in Japan and tested for i.a. BTEX did not depend on the rainfall intensity, and the temporal variation of their concentrations was similar to that of gas-phase concentrations [62]. Furthermore, the research results obtained by [58] highlighted the potential of climate change to affect stormwater pollution [58].

First of all, however, it is necessary to verify the factors that rainwater users have a measurable influence on, i.e., the material of the tank, the type of tank, exposure to traffic routes and gas stations, as well as the type and condition of the runoff surface. Both the amount of rainfall and the time distances between rainfall events are subject to constant changes, and tracking them by residents and based on them making a dependence on the use (or not) of water seems extremely difficult to implement.

The analysis of the obtained results indicates that the research on BTEX compounds in stored rainwater/rainwater runoff should be continued.

6. Summary and Conclusions

The main objective of the conducted preliminary application studies was to confirm the possibility of the occurrence of BTEX compounds in stored rainwater/rainwater runoff from various surfaces (above ground level). An additional objective was to determine future directions of research in this field, especially in the area of factors that users can have a measurable influence on. The limitation of the conducted research program was its small scale. The assumption was that these studies are a preparation for new research directions. The investigations were planned in such a way that their results reflect the actual patterns/habits of rainwater use by citizens.

With appropriate wind directions, BTEX can be transported longer distances to places potentially free of volatile hydrocarbons. Moreover, roof coverings made of materials that do not contain volatile hydrocarbons may be also source of these substances because of the pollution on them. Dust settling on the roof surface, organic matter from the air or bird droppings, etc. can adsorb hydrocarbons. Precipitation will wash BTEX from these pollutants. The material from which tanks are made probably may also cause contamination of the rainwater collected in them. Due to the potential risk of contamination of rainwater flowing from roof surfaces and collected in tanks for further use, it is advisable to purify it.

Due to the low concentrations of hydrocarbons, their adsorption on granular-activated carbon (GAC) beds should be sufficient. This type of treatment system is recommended when stored rainwater/rainwater runoff with variable BTEX content will be intended for use in hydroponic or aquaponic systems.

To avoid the presence of BTEX compounds in rainwater, it is crucial to locate green roof/building surfaces intended for supplying hydroponics or aquaponics at a considerable distance from transport routes and gas stations and/or in sheltered places, e.g., by trees or high buildings. When collecting rainwater from a runoff surface, it is important to analyse the type of surface and, if possible, select a surface with the least impact on rainwater quality. These decisions are possible at the design stage. However, sometimes, there is a choice, for instance, of the roof surface side from which the water will be collected.

The most important conclusions and answers to the research questions are formulated below:

-

It has been shown that with the methodology used, it is possible to confirm the presence of BTEX in stored rainwater/rainwater runoff from the roof/terrace surfaces in the city in various locations (from the city centre to the outskirts);

-

In some locations in the city, it is possible that BTEX compounds may occur in stored rainwater/rainwater runoff;

-

The recorded TEX (without benzene) concentrations are relatively low compared to the values of acceptable limits for BTEX in drinking water;

-

The subject of future research on the occurrence of BTEX compounds in stored rainwater/rainwater runoff should be in particular the factors on which residents have a measurable impact;

-

If possible, it is advisable to use rainwater tanks with a drinking water approval certificate;

-

Due to the possibility of benzene presence, a GAC filter should be provided if the water is intended for, e.g., hydroponics or aquaponics.

The analytical research and literature studies carried out allowed us to propose the following future research directions in the field of investigations of the content of BTEX compounds in rainwater:

-

Studies of the possible impact of the tank material (or factors related to the installation of the tank);

-

Deepening the current research on the impact of the tank’s exposure (and runoff surface’s) to a gas station, airports, traffic routes and other facilities that may affect the content of BTEX;

-

Investigation of the role of components acting as a shield (the height of surrounding buildings/fencing walls/green belts should be taken into account);

-

Further research on the quality of water from the green roof;

-

Confirmation of obtaining BTEX concentration after treatment in the adsorption process on GAC below the normative values for drinking water.