diffusive samplers and high-frequency analyzer: A critical comparison

Applied Geochemistry 72 (2016) 51e58

Contents lists available at ScienceDirect

Applied Geochemistry journal homepage: www.elsevier.com/locate/apgeochem

Hydrogen sulfide measurements in air by passive/diffusive samplers and high-frequency analyzer: A critical comparison S. Venturi a, b, *, J. Cabassi a, b, F. Tassi a, b, F. Capecchiacci a, b, O. Vaselli a, b, S. Bellomo c, S. Calabrese d, W. D’Alessandro c a

Dipartimento di Scienze della Terra, University of Florence, Via G. La Pira 4, 50121 Florence, Italy CNR e Istituto di Geoscienze e Georisorse, Via G. La Pira 4, 50121 Florence, Italy Istituto Nazionale di Geofisica e Vulcanologia, Sezione di Palermo, Via Ugo La Malfa 153, 90123 Palermo, Italy d Dipartimento di Scienze della Terra e del Mare, University of Palermo, Via Archirafi 36, 90123 Palermo, Italy b c

a r t i c l e i n f o

a b s t r a c t

Article history: Received 10 February 2016 Received in revised form 23 May 2016 Accepted 6 July 2016 Available online 7 July 2016

In this study, hydrogen sulfide (H2S) measurements in air carried out using (a) passive/diffusive samplers (Radiello® traps) and (b) a high-frequency (60 s) real-time analyzer (Thermo® 450i) were compared in order to evaluate advantages and limitations of the two techniques. Four different sites in urban environments (Florence, Italy) and two volcanic areas characterized by intense degassing of H2S-rich fluids (Campi Flegrei and Vulcano Island, Italy) were selected for such measurements. The concentrations of H2S generally varied over 5 orders of magnitude (from 10
1e103 mg/m3), the H2S values measured with the Radiello® traps (H2SR) being significantly higher than the average values measured by the Thermo® 450i during the trap exposure (H2STa), especially when H2S was <30 mg/m3. To test the reproducibility of the Radiello® traps, 8 passive/diffusive samplers were contemporaneously deployed within an 0.2 m2 area in an H2S-contaminated site at Mt. Amiata (Tuscany, Italy), revealing that the precision of the H2SR values was ±49%. This large uncertainty, whose cause was not recognizable, is to be added to that related to the environmental conditions (wind speed and direction, humidity, temperature), which are known to strongly affect passive measurements. The Thermo® 450i analyzer measurements highlighted the occurrence of short-term temporal variations of the H2S concentrations, with peak values (up to 5732 mg/ m3) potentially harmful to the human health. The Radiello® traps were not able to detect such temporal variability due to their large exposure time. The disagreement between the H2SR and H2STa values poses severe concerns for the selection of an appropriate methodological approach aimed to provide an accurate measurement of this highly toxic air pollutant in compliance with the WHO air quality guidelines. Although passive samplers may offer the opportunity to carry out low-cost preliminary surveys, the use of the high-frequency H2S analyzer is preferred when an accurate assessment of air quality is required. In fact, the latter provides precise real-time measurements for a reliable estimation of the effective exposure to hazardous H2S concentrations, giving insights into the mechanisms regulating the dispersion of this air pollutant in relation to the meteorological parameters. © 2016 Elsevier Ltd. All rights reserved.

Keywords: Active analyzers Passive/diffusive samplers Gaseous contaminants Air quality monitoring Hydrogen sulfide

1. Introduction Air pollution is the top environmental risk factor for premature death in Europe (EEA, 2014). The development of efficient methodologies to: (i) monitor the behavior and fate of air pollutants, (ii)

* Corresponding author. Department of Earth Sciences, Via G. La Pira 4, 50121, Florence, Italy. E-mail address: stefania.venturi@unifi.it (S. Venturi). http://dx.doi.org/10.1016/j.apgeochem.2016.07.001 0883-2927/© 2016 Elsevier Ltd. All rights reserved.

identify the contaminant sources and (iii) evaluate the hazard for human health and environment, is a primary issue in environmental monitoring. The most broadly applied technique for monitoring air pollutants, including H2S, NO2, O3, Hg, benzene, toluene, ethylbenzene and xylene (BTEX), involves the use of passive/diffusive samplers (e.g. Shooter et al., 1995; Flores et al., 1996; nova  et al., 2006; Campos Bernard et al., 1999; Tang et al., 2002; Kla et al., 2010; Nash and Leith, 2010; Zabiegala et al., 2010; Byanju et al., 2012; Colomer et al., 2012; D’Alessandro et al., 2013;

52

S. Venturi et al. / Applied Geochemistry 72 (2016) 51e58

Pavilonis et al., 2013; Al-Awadhi, 2014; Huang et al., 2014). These devices consist of a tube packed with a specific absorbent able to trap the pollutant diffusing through the tube according to Fick’s First Law (i.e. gas molecules move from high to low concentration zones). The low cost and no power supply requirement of the passive traps allow their use at multiple locations, in order to easily obtain a general view about the spatial distribution of the air pollutant (Hangartner, 2000). However, this method implies a relatively long exposure time (up to several days), preventing the detection of short-term temporal variations (faster than the exposure time) of the pollutant concentrations (Krupa and Legge, 2000). In addition, the efficiency of the traps is strongly affected by several environmental factors, e.g. temperature, humidity and wind speed (Brown, 2000; Krupa and Legge, 2000; Delgado-Saborit and EsteveCano, 2006), whose effects are difficult to quantify. Analytical instrumentations equipped with high-sensitivity detectors have successfully been used to provide active (i.e. collecting air samples by a pump set at a specific flow rate) measurements of various air pollutants at high temporal frequency, such as CO2, H2S, SO2, NOx and O3 (e.g. Carapezza et al., 1984; Badalamenti et al., 2001; Kourtidis et al., 2004; Peralta et al., 2013; Olafsdottir et al., 2014). Nevertheless, this approach is generally expensive and requires frequent instrument calibration and an in situ power source available. Selecting the most appropriate technique for measuring air pollutant concentration requires consideration of the trade-offs between advantages and limitations of passive and active methods which have to be carefully evaluated. Accordingly, the present study focused on the comparison between two different analytical approaches: a) Radiello® traps and b) a Thermo® 450i analyzer, to measure hydrogen sulfide (H2S) concentrations in air. Indoor and outdoor H2S measurements in a wide concentration range were performed in areas characterized by both natural and anthropogenic H2S sources. The main aim was to evaluate the reliability and the efficiency of the two techniques for investigating and monitoring this air pollutant in compliance with the ambient air quality standards dictated by the World Health Organization (WHO, 2000, 2003). 2. Origin of hydrogen sulfide (H2S) and its impact on human health Hydrogen sulfide is an air pollutant commonly emitted by both natural sources and anthropogenic activity. The latter includes sewage treatment plants, concentrated animal feeding operations, oil and gas refineries, geothermal power plants, paper mills and vehicular traffic (e.g. Kourtidis et al., 2004, 2008; Colomer et al., 2012). Natural H2S sources are thermal and cold sulfur-rich springs, wetlands, manure, coal pits, and gas and oil reservoirs (e.g. Bates et al., 1992; WHO, 2003). In particular, huge amounts of H2S are emitted from volcanic and hydrothermal systems (e.g. ndez et al., 2015), where this gas D’Alessandro et al., 2013; Herna may represent a significant hazard for human health (Sigurdsson et al., 2015). For example, in 1971, 1976 and 1997, 13 people were likely asphyxiated by H2S emitted from the Kusatsu-Shirane and Adatara volcanoes (Honshu, Japan) (Cook and Weinstein, 2011; Williams-Jones and Reimer, 2015). Long-term exposures to low H2S concentrations can have severe health consequences on respiratory, cardiovascular and nervous systems (e.g. Legator et al., 2001; Bates et al., 2002, 2013). At 15 mg/ m3, H2S provokes eye irritation, while at slightly higher concentrations it induces damage to the upper respiratory tract and causes loss of smell, whereas coma occurs after a single breath when the H2S values are 1400 mg/m3 (McManus, 1999; WHO, 2000, 2003). The maximum allowable concentration of H2S for ambient air

reported by the air quality guideline (WHO, 2000) is 150 mg/m3 (average value in 24 h), although it should not exceed 7 mg/m3 (average value in 30 min) to avoid significant odor annoyance due to its typical rotten-egg smell (detectable at 0.7e42 mg/m3, depending on individual sensitivity; Schiffman and Williams, 2005).

3. Methodology 3.1. Measurement sites Measurements of H2S concentrations in air were carried out in three different Italian localities (Fig. 1a), i.e. 1) Florence (Tuscany, central Italy; Fig. 1b), an urban context affected by anthropogenic H2S sources, 2) Solfatara crater (Fig. 1c), a tuff cone in Campi Flegrei showing an intense fumarolic activity and strong diffuse soil gas emission (e.g. Caliro et al., 2007; Tassi et al., 2013) and 3) Vulcano Island in the Aeolian Archipelago (Fig. 1d), where S-rich fumaroles occur at the base of the recently active volcanic cone (La Fossa crater) and on the crater rim (e.g. Capaccioni et al., 2001; Granieri et al., 2006). At each site, continuous H2S measurements using the Thermo® 450i analyzer were carried out during the exposure of the Radiello® traps placed at 1.5 m above the ground level and <15 cm away from the analyzer. The exposure time of the Radiello® traps (ranging from 52 to 371 min) was selected on the basis of the real-time measured H2S concentrations. Air temperature, as well as wind speed and direction were measured using a portable Davis Vantage Vue weather station. A reproducibility test for H2S concentrations in air was performed at Mt. Amiata (central Italy; Fig. 1a) by deploying 8 Radiello® traps close to each other within an 0.2 m2 area (Fig. 1e). The selected site was located close to the Acqua Passante well that was drilled for mining exploration in 1938. Presently, it discharges about 1,800, 10 and 0.82 ton/year of CO2, CH4 and H2S, respectively, from a chimney at about 6 m above the ground level (Nisi et al., 2014) (Fig. 1e). The reproducibility test data of the Radiello® traps were compared with those obtained by the simultaneous acquisition of H2S measurements at the same site by using the Thermo® 450i analyzer.

3.2. Radiello® traps The Radiello® traps consist of (i) an adsorbent cartridge, (ii) a diffusive body, and (iii) a supporting plate. The polyethylene adsorbent cartridge is placed inside the diffusive body, the latter consisting of a micro-porous high-density polyethylene cylinder connected to the polycarbonate supporting plate. The radial design of the diffusive body optimizes the efficiency of the trap, since it allows the pollutant uptake from all the directions (D’Alessandro et al., 2013; Pavilonis et al., 2013). The adsorbent cartridge is impregnated with zinc acetate [Zn(CH3COO)2] that reacts with H2S producing zinc sulfide (ZnS), according to the following reaction:

ZnðCH3 COOÞ2 þ H2 S/ZnS þ 2C2 H4 O2

(1)

Zinc sulfide is then extracted from the cartridge with MilliQ water (10 mL) and analyzed by visible spectrometry after adding 0.5 mL of ferric chloride-amine solution (Fondazione Maugeri, 2007). The H2S concentrations in air (H2SR, in mg/m3) are computed according to the following equation (Fondazione Maugeri, 2007):

S. Venturi et al. / Applied Geochemistry 72 (2016) 51e58

53

Fig. 1. Site locations (Fig. 1a) of the H2S surveys carried out by Thermo® 450i analyzer and Radiello® traps: city of Florence (Fig. 1b), Solfatara crater (Fig. 1c), and Vulcano Island (Fig. 1d). The location of the site at Mt. Amiata where the reproducibility test of the Radiello® trap method was carried out is also shown (Fig. 1e).

54

S. Venturi et al. / Applied Geochemistry 72 (2016) 51e58


.  34:082  H2 SR ðppbÞ H2 SR mg m3 ¼ 24:45   34:082 m   1000 ¼ 24:45 QK  t

(2)

where 34.082/24.45 is the conversion factor between ppb and mg/ m3 for H2S (Thorsteinsson et al., 2013), m is the S2
mass (in mg), t is the exposure time (in min) and QK is the sampling rate of the Radiello adsorbent cartridge, whose dependence on temperature is given by (Fondazione Maugeri, 2007):

 QK ¼ 0:096

TðKÞ 298

3:8 (3)

where T(K) is the ambient temperature. Eq. (3) is considered valid at wind speed ranging from 0.1 to 10 m/s (Fondazione Maugeri, 2007). The detection limit of this method, for 1 h exposure at 25  C, is 25 mg/m3 (corresponding to a mass, m, of 0.105 mg), whereas the uncertainty (2s) is ±8.7% (Fondazione Maugeri, 2007). However, recent studies (Koutrakis et al., 1993; Gair and Penkett, 1995; Brown, 2000; Krupa and Legge, 2000; Pennequin-Cardinal et al., 2005) pinpointed that weather conditions (e.g. humidity, temperature, wind speed and direction) can affect the efficiency of the Radiello® traps, although a reliable evaluation of their effective influence on passive H2S measurements is not available. 3.3. Thermo® 450i analyzer The Thermo® 450i analyzer uses a pulsed fluorescence technology to determine the concentrations of the main gaseous sulfur species (H2S and SO2) in air (Thermo Scientific, 2012). The instrument measures SO2 concentrations, whereas H2S is analyzed after its oxidation to SO2. The air is pumped into the analyzer and directed to either (i) an H2S-SO2 converter or (ii) the fluorescence chamber, i.e. bypassing the converter. The operative air flux, regulated by a pump, is 1 l/min. In the fluorescence chamber, pulsating UV light excites the SO2 molecules. As the excited SO2 molecules decay to lower energy states, they emit a UV light that is proportional to the SO2 concentration. A photomultiplier tube (PMT) detects the UV light emission of the decaying SO2 molecules. When the air sample passes through the converter, H2S is oxidized to SO2 (efficiency >80%) and the instrument measures the concentrations of Combined Sulfur (CS), i.e. the sum of SO2 and H2S. When the air flux bypasses the converter, the analyzer only determines SO2. Hence, the difference between the two signals allows H2S content to be calculated (Thermo Scientific, 2012). Although the duration of a single measurement is 1 s, a satisfactory precision (±1%; Thermo Scientific, 2012) is obtained in 1 min, by averaging the results of 60 repeated measurements (linearity ±1%; Thermo Scientific, 2012). The detection limits under field conditions ranged from 0.1 to 0.7 mg/m3 (Percy et al., 2012). The accuracy of the instrument was checked prior the measuring campaigns using gases with certified concentrations. During the field surveys, the instrumentation was power-supplied by a high-performance portable gel battery. 4. Results The average, minimum and maximum H2S concentrations (H2STa, H2STmin and H2STmax, respectively) measured at each site using the Thermo® 450i analyzer (H2ST) during the exposure of the Radiello® traps, as well as the H2SR concentrations, the Radiello® exposure time (min), the air temperature ( C), and the wind speed (m/s) and direction are reported in Table 1. Sites #1, #2, and #4 (Fig. 1b), representing, respectively, a solid

waste disposal site, an indoor environment and a busy street in Florence, were characterized by H2STa concentrations lower than the odor annoyance threshold (i.e. 7 mg/m3), with H2STmax values up to 3.19, 3.85 and 3.34 mg/m3, respectively (Table 1). The H2SR values at these sites were lower than the detection limits for the respective exposure times (i.e. 10.8, 4.2 and 5.2 mg/m3, respectively). On the contrary, the H2SR concentration at #3, a sewer pipe (Fig. 1b) was 233 mg/m3, whilst H2STa value was 191 mg/m3 (H2STmax up to 870 mg/ m3). At Solfatara crater, the H2STa values at #5 to #13 ranged from 233 to 2825 mg/m3, i.e. significantly higher than the WHO (2000) limit for ambient air, with a H2STmax value up to 5732 mg/m3 measured at #11, located at <20 m from the main fumarolic emissions of the crater, i.e. Bocca Grande and Bocca Nuova (BG and BN, respectively; Fig. 1c). Relatively high H2STmax values (>1000 mg/ m3) were also measured close to Fangaia (#10), a bubbling mudpool located in the central part of the crater, and at Stufe (literally, “Stoves”) (#5, #6, #12), in the northern sector of inner crater flank (Fig. 1c). The latter consist of two excavated caves emitting hot vapors that in the past were used as sudatoria (sweat rooms). The H2STa values at #14 and #15, few tens of meters W of the main fumarolic area (Fig. 1c), were one order of magnitude lower than those measured in the other Solfatara sites (65.5 and 45.2 mg/m3, respectively), likely due to the prevailing wind direction (SW), i.e. they were upwind with respect to the fumaroles (Fig. 1c). The H2SR values measured in this volcanic area ranged from 45.2 (#15) to 3422 (#11) mg/m3. Relatively high H2SR concentrations (>800 mg/ m3) were mostly measured close to the hydrothermal fluid emissions (#5, #10, #12). Most sites (#16, #18, #21, #22, #23 and #24) at Vulcano Island showed relatively low (4.54 mg/m3) H2STa values, although the concentrations of H2STmax at #16, #21 and #22 were higher than the odor annoyance threshold (7 mg/m3). At the fumarolic area at the base of the active volcanic cone, sites #17, 19 and 20 (Fig. 1d) detected relatively high H2STa concentrations (from 31 to 464 mg/ m3). At #20, close (<30 m) to the so-called hippopotamus pound pool used by tourists for thermal therapy (Fig. 1d), the H2Smin value was 168 mg/m3 and all the measured values were higher than the limit concentrations reported by WHO (2000). The highest H2SR value (1174 mg/m3) was measured at the hippopotamus pound pool, whilst the other H2SR values were ranging from 16.9 mg/m3 (#23) to 306 mg/m3 (#17). 5. Discussion 5.1. H2STa vs. H2SR An exhaustive evaluation of the efficiency and reliability of analytical methods aimed to accurately measure H2S in air over a wide range of concentrations is essential to investigate the effective impact of this air pollutant on the environment and its effects on human health. The two approaches applied for the present study are conceptually different. The Thermo® 450i analyzer provides real-time, high frequency (60 s) H2S measurements (H2ST), whereas the Radiello® samplers trap H2S (H2SR) over a relatively large time period (from hours to days), thus providing cumulative concentrations of this air pollutant. Notwithstanding, the two dataset can reliably be compared by considering the average values of the H2S measurements carried out by the Thermo® 450i analyzer (H2STa) during the exposure time of the Radiello® traps. Both the H2STa and H2SR values are consistent with the guideline values for airborne pollutants that are commonly expressed as concentrations averaged over specific time intervals. The H2STa vs. H2SR and H2STa vs. H2SR/H2STa binary diagrams (Fig. 2a,b) show that the data of the two methods were in a good agreement at relatively high

S. Venturi et al. / Applied Geochemistry 72 (2016) 51e58

55

Table 1 The average, minimum and maximum H2S concentrations (H2STa, H2STmin and H2STmax, respectively, in mg/m3) measured using the Thermo® 450i analyzer, H2S values from the Radiello® traps (H2SR in mg/m3), exposure time (t, in min) of the passive traps, air temperature (T, in  C), and the wind speed (average value in m/s) and direction are reported (b.d.l. ¼ below detection limit). Site

Description

Date

t min

T C

H2SR mg/m3

H2STa mg/m3

H2STmin mg/m3

H2STmax mg/m3

Wind speed m/s

Wind direction

#1 #2 #3 #4 #5 #6 #7 #8 #9 #10 #11 #12 #13 #14 #15 #16 #17 #18 #19 #20 #21 #22 #23 #24

landfill indoor environment sewer pipe urban center Solfatara crater Solfatara crater Solfatara crater Solfatara crater Solfatara crater Solfatara crater Solfatara crater Solfatara crater Solfatara crater Solfatara crater Solfatara crater Vulcano island Vulcano island Vulcano island Vulcano island Vulcano island Vulcano island Vulcano island Vulcano island Vulcano island

April 2015 January 2015 January 2015 January 2015 February 2015 February 2015 February 2015 February 2015 February 2015 July 2014 July 2014 July 2014 July 2014 July 2014 July 2014 September 2014 September 2014 September 2014 September 2014 September 2014 September 2014 September 2014 September 2014 September 2014

151 371 93 345 52 87 100 52 206 62 60 60 60 70 120 140 123 142 126 117 120 94 181 184

20 23.5 16 13 15 15 13 17 19 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27

b.d.l. b.d.l. 233 b.d.l. 1053 506 452 798 196 847 3422 858 297 65.5 45.2 41.2 306 33.5 25.6 1174 18.4 61.4 16.9 38.7

0.98 0.76 191 0.42 944 397 328 599 223 914 2825 769 484 28.4 14.8 4.07 144 2.15 31.0 464 2.54 4.54 1.43 1.27

b.d.l. b.d.l. 10.3 b.d.l. 476 78.3 134 311 126 117 1103 518 181 b.d.l. b.d.l. 0.23 70.0 b.d.l. 2.19 168 b.d.l. b.d.l. b.d.l. b.d.l.

3.19 3.85 870 3.34 3252 1142 618 832 421 1932 5732 1138 985 141 141 7.81 228 6.52 111 789 7.96 27.8 5.34 3.49

e e e e e e e e e 2.92 4.46 4.31 8.15 7.2 1.21 0.95 0.56 0.14 0.42 b.d.l. 1.11 2.25 0.57 2.30

e e e e e e e e e NW WNW WNW SW SSW SW SW SW NW NNW e SSW NW NE W

Fig. 2. (a) H2SR vs. H2STa and (b) H2SR/H2STa vs. H2STa binary diagrams: urban areas (green square), Solfatara crater (blue circle) and Vulcano Island (yellow triangle). The H2S concentrations are in mg/m3. The H2S concentration ranges for the values of detectable odor (0.7 mg/m3; Schiffman and Williams, 2005), odor annoyance (7 mg/m3) and WHO (2000) ambient air threshold (150 mg/m3) are marked with different colors (from light to dark pink). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

concentrations (>150 mg/m3), whereas at <30 mg/m3 (#14 and #15) the H2SR/H2STa ratios were significantly >1, up to 30.5 (#24). The inconsistency between the results provided by the two

methodological approaches could prevent a reliable evaluation of the air quality. For instance, the H2SR concentrations at Vulcano Island were higher than the odor annoyance threshold, whereas the corresponding H2STa values, with the exception of those measured at the #19, #17 and #20 sites, were below this limit (Table 1). The strong difference between the two datasets at relatively low H2S concentrations may be due to either an overestimation of the Radiello® traps or an underestimation of the Thermo® 450i analyzer. Both these solutions cannot be discarded a priori, since the true H2S values, i.e. those that should be measured using a third independent high-accuracy method, are not available. The accuracy of the Thermo® 450i analyzer, mostly related to the efficiency of the H2S-SO2 converter and detector, is periodically tested in laboratory using a standard gas mixture (SOL manufacturer), whose H2S concentration is 1 mmol/mol. The pump of the instrument regulates the inlet flux of air, thus wind speed and direction, which may strongly vary during outdoor measurements, have a negligible effect on this method. The Thermo® 450i measurements are also independent on both gas composition and air temperature (operating temperature: 15e35  C). The accuracy of the H2S measurements carried out by Radiello® traps are known to be influenced by several factors. Sulfur-bearing gas compounds, such as carbon disulfide (CS2) and mercaptans, are expected to be absorbed by passive traps. Although the abundances of these gases are commonly considered negligible, they may affect the H2S measurements in areas characterized by a specific contaminant source. For example, the bubbling gas exhalations at the Levante beach (Vulcano Island) showed CS2 concentrations ranging from 2.8 to 8.9 nmol/mol (Tassi et al., 2012), whereas those measured in the Solfatara fumarolic gases were from 156 to 368 nmol/mol (Tassi et al., 2015). However, the corresponding H2S contents were remarkably higher (0.42e2.02 and 5.3e15 mmol/ mol, respectively; Tassi et al., 2012, 2015), therefore the influence of CS2 on the Radiello® H2S measurements (interference rate: 0.026 mg/m3 H2S per mg/m3 of CS2; http://www.sigmaaldrich.com/ analytical-chromatography/air-monitoring/radiello/faq.html) was likely negligible. Lab experiments, carried out by Mason et al. (2011) and

56

S. Venturi et al. / Applied Geochemistry 72 (2016) 51e58

Pavilonis et al. (2013), demonstrated that the accuracy of the Radiello® measurements, which is strongly dependent on the correct estimation of the sampling rate (QK), increases at increasing H2S concentrations and exposure durations. The H2S concentrations measured in traps exposed for 24 h at 52 mg/m3 were found to be underestimated of up to 50% with respect to the true value. Differently, in this study the H2SR concentrations were higher than the H2STa values, with the only exceptions of those measured at #9, #10, #13 and #19 (Table 1). Although the true H2S concentrations were not known, the apparent contradiction between lab tests (Mason et al., 2011; Pavilonis et al., 2013) and outdoor measurements (this study) may be related to the influence of the air temperature on the Radiello® traps that, according to eq. (2), controls the QK value. The computation of the H2SR data was carried out considering the average value of air temperature measured during the trap exposure, i.e. the temporal variations of this parameter, able to cause changes on the sampling rate, were not taken into account. Other environmental parameters, such as air humidity and wind speed and direction, are not included in eq. (2), although they are expected to significantly affect the QK value (e.g. Gair and Penkett, 1995; Pennequin-Cardinal et al., 2005). For instance, when the wind speed is too low the gas molecules cannot effectively diffuse through the trap surface, whereas high wind speed may cause a shortening of the effective length of the diffusion path (Brown, 2000). Notwithstanding these studies, the effects of the environmental conditions on the QK value of Radiello® traps are particularly complex to be reliably predicted and evaluated. To provide insights into the uncertainty on the H2SR data, a reproducibility test was carried out by deploying 8 passive traps close to the Acqua Passante chimney (Fig. 1e) that discharges high amounts of H2S-rich gases (Nisi et al., 2014). The H2SR concentrations measured for this test ranged from 62.2 (R5) to 130 (R3) mg/m3 (Table 2). As shown by geochemical surveys (Table 1), these values were significantly higher than those of H2STa measured during the exposure of the Radiello® traps (41 mg/m3; Table 2), although they fall within the range of the H2Smin and H2Smax values (<0.1 and 395 mg/m3, respectively; Table 2). The most striking result of this test is that the H2SR uncertainty (2s) was ±49%, much higher than that reported by the supplier (±8.7%; Fondazione Maugeri, 2007). Considering that the 8 traps were contemporaneously exposed at the same site, such a low reproducibility was not caused by the influence of the environmental conditions, which were the same for all the traps. Thus, it should be admitted that other unknown cause(s) were significantly affecting the reproducibility and precision of the passive traps. It is worth noting that the measured uncertainty was still not sufficient to explain the differences observed

between the H2SR and H2STa concentrations for H2S < 30 mg/m3 (Fig. 2a,b), confirming that the influence of these unknown factors tends to strongly affect the precision of the Radiello® trap at low air concentrations. 5.2. H2ST and H2SR data vs. air quality Measurements of air contaminants carried out at highfrequency allow short-term variations to be recognized that cannot be recorded by the passive traps, the latter providing only cumulated concentration data over the exposure time. Although they can still be adequately applied for evaluating the compliance to the 24 h WHO threshold in conditions of constantly high H2S concentrations, passive measurements are ineffective in evaluating the impact of harmful gases when hazardous values are only occasionally exceeded. At #3, for example, the H2STa and H2SR concentrations were 191 and 233 mg/m3, respectively, i.e. slightly higher than the threshold value for ambient air (WHO, 2000), whereas the temporal pattern of the H2ST concentrations (Fig. 3a) highlights the occurrence of three peaks with values representing critical conditions. A similar situation occurred at #15, where the relatively low H2STa and H2SR values (14.8 and 45.2 mg/m3) masked the sharp H2ST increases occurred after 50 min of measurement (H2STmax ¼ 141 mg/m3; Fig. 3b). At #19, the H2ST concentrations showed a decreasing trend after the occurrence of several H2ST peaks (Fig. 3c) that were not recognizable on the basis of the H2STa and H2SR concentrations (31 and 25.6 mg/m3, respectively; Table 1). The peak values and temporal trends of the H2S concentrations shown by high-frequency H2ST data reported in Fig. 3aec were likely related to episodic variations of the weather conditions. This information, necessary for a reliable assessment of the air quality in the investigated areas, was completely masked by the Radiello® trap measurements. To further support this observation, the H2ST concentrations at #12 (Solfatara crater, Fig. 1c) are coupled with the wind direction (Fig. 4), the latter being continuously recorded by the meteorological station. At the beginning of the survey, for about 12 min, the wind was blowing from SE, i.e. from the Stufe emitting site (Fig. 1c), and the H2ST concentrations were up to 1138 mg/m3, a value that may cause damages to the human respiratory system. Then, the wind direction progressively shifted to SW, W and NW. Since no significant amount of hydrothermal gas discharges occurred along the wind path, the H2ST concentrations sharply decreased down to <700 mg/m3. Both the H2STa and H2SR concentrations (769 and 858 mg/m3, respectively) did not provide any indication of these abrupt variations in terms of H2S concentrations and wind. 6. Conclusions

Table 2 H2SR concentrations (mg/m3) measured in the 8 traps, contemporaneously exposed close to the Acqua Passante chimney (Mt. Amiata). The H2STa, H2STmin and H2STmax concentrations (mg/m3) measured during the exposure of the Radiello® traps are also reported (b.d.l. ¼ below detection limit). Radiello® trap ID

t min

H2SR mg/m3

R1 R2 R3 R4 R5 R6 R7 R8

120 120 121 120 119 120 119 119

101 128 130 78.6 62.2 111 89.5 84.4

T ( C) H2STa (mg/m3) H2STmin (mg/m3) H2STmax (mg/m3)

13 41 b.d.l. 395

The present study has demonstrated that H2S measurements in air from areas affected by natural and anthropogenic sources are strongly depending on the applied methodological techniques. A significant decoupling in terms of H2S concentrations was recorded between passive/diffusive samplers and Thermo® 450i analyzer, particularly when H2S was <30 mg/m3. These differences pose severe concerns for a proper selection of the method to be used for H2S measurements in compliance with the air quality guideline. The H2SR concentrations in the study areas were, in most cases, overestimated with respect to those provided by the Thermo® 450i analyzer (H2STa). This is in contrast with the results of laboratory experiments performed to test the accuracy of the Radiello® traps, suggesting that a correct estimation of their sampling rate QK is rather difficult, especially for outdoor measurements, which are characterized by unpredictable variations of the meteorological parameters.

S. Venturi et al. / Applied Geochemistry 72 (2016) 51e58

57

Fig. 4. H2ST (mg/m3) vs. time (min) diagram for the measurements carried out at #12 (Solfatara crater). Symbols and colors as in Fig. 3. The dark pink background (see Fig. 2) indicates that all the H2ST values were above the WHO (2000) ambient air threshold (150 mg/m3). The 5-min averaged wind directions recorded at #12 are also reported. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

short-term temporal variations of the air pollutant concentrations. Hence, this analytical approach is not able to detect peak values that may have a significant environmental impact and represent a risk for the human health. On the whole, real-time and high-frequency H2S measurements acquired using the Thermo® 450i analyzer are to be preferred for outdoor surveys since they are able to provide a more realistic description of the relationship between the H2S concentrations and the environmental parameters, offering the opportunity to gain insights into the behavior of this pollutant in air under different chemical-physical conditions. Passive/diffusive samplers remain a valid alternative to evidence the presence of air contamination problems on large areas when budget limitations or logistical problems do not allow the use of active real-time instrumentations. Acknowledgements This work was financially supported by the V2-INGV-DPC Project, the Municipality of Abbadia San Salvatore (Siena, Itay) and the laboratories of Fluid Geochemistry and Stable Isotopes of the Department of Earth Sciences of the University of Florence. We thank the reviewer, T. Lopez, for her valuable comments and thoughtful review to the manuscript. References

Fig. 3. H2ST (mg/m3) vs. time (min) diagram for the measurements carried out at (a) #3, (b) #15 and (c) #19 sites. The green line refers to the H2SR concentration measured at each site. Colored areas as in Fig. 2a,b. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

In addition, the reproducibility test at Mt. Amiata demonstrated that the efficiency of the Radiello® traps was also depending on unknown factor(s) not related to the weather conditions. Besides their large uncertainty, the Radiello® measurements cannot record

Al-Awadhi, J.M., 2014. Measurement of air pollution in Kuwait city using passive samplers. Atmos. Clim. Sci. 4, 253e271. http://dx.doi.org/10.4236/ acs.2014.42028. Badalamenti, B., Liotta, M., Valenza, M., 2001. An automatic system for continuous monitoring of CO2, H2S, and SO2 and meteorological parameters in the atmosphere of volcanic areas. Geochem. Trans. 5 http://dx.doi.org/10.1039/ b104622p. Bates, T.S., Lamb, B.K., Guenther, A., Dignon, J., Stoiber, R.E., 1992. Sulfur emissions to the atmosphere from natural sources. J. Atmos. Chem. 14, 315e337. Bates, M.N., Garret, N., Shoemack, P., 2002. Investigation of health effects of hydrogen sulfide from geothermal source. Arch. Environ. Health 57 (5), 405e411. Bates, M.N., Garrett, N., Crane, J., Balmes, J.R., 2013. Associations of ambient hydrogen sulfide exposure with self-reported asthma and asthma symptoms. Environ. Res. 122, 81e87. Bernard, N.L., Gerber, M.J., Astre, C.M., Saintot, M.J., 1999. Ozone measurement with passive samplers: validation and use for ozone pollution assessment in montpellier, France. Environ. Sci. Technol. 33, 217e222. http://dx.doi.org/10.1021/ es971140k. Brown, R.H., 2000. Monitoring the ambient environment with diffusive samplers: theory and practical considerations. J. Environ. Monit. 2, 1e9. Byanju, R.M., Gewali, M.B., Manandhar, K., 2012. Passive sampling of ambient nitrogen dioxide using local tubes. J. Environ. Prot. 3, 177e186. http://dx.doi.org/

58

S. Venturi et al. / Applied Geochemistry 72 (2016) 51e58

10.4236/jep.2012.32022. Caliro, S., Chiodini, G., Moretti, S., Avino, R., Granieri, D., Russo, M., Fiebig, J., 2007. The origin of the fumaroles of La Solfatara (Campi Flegrei, south Italy). Geochim. Cosmochim. Acta 71, 3040e3055. http://dx.doi.org/10.1016/j.gca.2007.04.007. Campos, V.P., Cruz, L.P.S., Godoi, R.H.M., Godoi, A.F.L., Tavares, T.M., 2010. Development and validation of passive samplers for atmospheric monitoring of SO2, NO2, O3 and H2S in tropical areas. Microchem. J. 96 (1), 132e138. http:// dx.doi.org/10.1039/b104622p. Capaccioni, B., Tassi, F., Vaselli, O., 2001. Organic and inorganic geochemistry of low temperature gas discharges at the Baia di Levante beach, Vulcano Island, Italy. J. Volcanol. Geotherm. Res. 108, 173e185. Carapezza, M., Gurrieri, S., Nuccio, P.M., Valenza, M., 1984. CO2 and H2S concentrations in the atmosphere at the Solfatara of Pozzuoli. Bull. Volcanol. 47 (2), 287e293. , H.E., Iglesias, E.M., 2012. Estimation of hydrogen sulfide Colomer, F.L., Morato emission rates at several wastewater treatment plants through experimental concentration measurements and dispersion modeling. J. Air & Waste Manag. Assoc. 62 (7), 758e766. http://dx.doi.org/10.1080/10962247.2012.674008. Cook, A., Weinstein, P., 2011. Volcanic and Geothermal Processes: Health Effects. Reference Module in Earth Systems and Environmental Sciences - Encyclopedia of Environmental Health, pp. 664e671. Delgado-Saborit, J.M., Esteve-Cano, V., 2006. Field study of diffusion collection rate coefficients of a NO2 passive sampler in a Mediterranean coastal area. Environ. Monit. Assess. 120, 327e345. http://dx.doi.org/10.1007/s10661-005-9065-9. D’Alessandro, W., Aiuppa, A., Bellomo, S., Brusca, L., Calabrese, S., Kyriakopoulos, K., Liotta, M., Longo, M., 2013. Sulphur-gas concentrations in volcanic and geothermal areas in Italy and Greece: characterising potential human exposures and risks. J. Geochem. Explor. 131, 1e13. http://dx.doi.org/10.1016/ j.gexplo.2012.08.015. EEA, 2014. Air Quality in Europe: 2014 Report. European Environment Agency, Copenhagen. Flores, M.I., Ray, J.D., Joseph, D.B., 1996. Passive Sampling Devices: a Cost-effective Method of Obtaining Air Quality Data in Protected Areas. National Park Service Strategic Plan, Final Draft, NPS D-1151, September 1996. Fondazione Maugeri, 2007. Instruction Manual for Radiello Sampler. http://www. radiello.com/english/index_en.html. Gair, A.J., Penkett, S.A., 1995. The effects of wind speed and turbulence on the performance of diffusion tube samplers. Atmos. Environ. 29 (18), 2529e2533. Granieri, D., Carapezza, M.L., Chiodini, G., Avino, R., Caliro, S., Ranaldi, M., Ricci, T., Tarchini, L., 2006. Correlated increase in CO2 fumarolic content and diffuse emission from La Fossa crater (Vulcano, Italy): evidence of volcanic unrest or increasing gas release from a stationary deep magma body? Geophys. Res. Lett. 33, L13316. http://dx.doi.org/10.1029/2006GL026460. Hangartner, M., 2000. Limits and Advantages of Diffusion Sampling in Ambient Air Pollution Monitoring. 7th International Conference on Atmospheric Science and Applications to Air Quality, 31 October e 2 November 2000, Taipeh, Taiwan. ndez, P.A., Melia n, G., Giammanco, S., Sortino, F., Barrancos, J., Pe rez, N.M., Herna  n, E., Lo pez, M., Donovan, A., Mori, T., Notsu, K., 2015. Contribution of CO2 Padro and H2S emitted to the atmosphere by plume and diffuse degassing from volcanoes: the Etna volcano case study. Surv. Geophys. 36, 327e349. http:// dx.doi.org/10.1007/s10712-015-9321-7. Huang, J., Lyman, S.N., Hartman, J.S., Gustin, M.S., 2014. A review of passive sampling systems for ambient air mercury measurements. Environ. Sci. Process. Impacts 16, 374e392. http://dx.doi.org/10.1039/C3EM00501A. nov , L., Urbanova , P., Holoubek, I., 2006. Passive air Kla a, J., Kohoutek, J., Hamplova sampler as a tool for long-term air pollution monitoring: Part 1. Performance assessment for seasonal and spatial variations. Environ. Pollut. 144, 393e405. http://dx.doi.org/10.1016/j.envpol.2005.12.048. Kourtidis, K., Kelesis, A., Maggana, M., Petrakakis, M., 2004. Substantial traffic emissions contribution to the global H2S budget. Geophys. Res. Lett. 31, L18107. http://dx.doi.org/10.1029/2004GL020713. Kourtidis, K., Kelesis, A., Petrakakis, M., 2008. Hydrogen sulfide (H2S) in urban ambient air. Atmos. Environ. 42, 7476e7482. http://dx.doi.org/10.1016/ j.atmosenv.2008.05.066. Koutrakis, P., Wolfson, J.M., Bunyaviroch, A., Froelich, S.E., Hirano, K., Mulik, J.D., 1993. Measurement of ambient ozone using a nitrite-coated filter. Anal. Chem. 65, 209e214. Krupa, S.V., Legge, A.H., 2000. Passive sampling of ambient, gaseous air pollutants: an assessment from an ecological perspective. Environ. Pollut. 107, 31e45. Legator, M.S., Singleton, C.R., Morris, D.L., Philips, D.L., 2001. Health effects from chronic low-level exposure to hydrogen sulfide. Arch. Environ. Health Int. J. 56 (2), 123e131. Mason, J.B., Fujita, E.M., Campbell, D.E., Zielinska, B., 2011. Evaluation of passive samplers for assessment of community exposure to toxic air contaminants and related pollutants. Environ. Sci. Technol. 45 (6), 2243e2249. http://dx.doi.org/ 10.1021/es102500v.

McManus, N., 1999. Safety and Health in Confined Spaces. Lewis Publisher/CRC Press, Boca Raton, FL. Nash, D.G., Leith, D., 2010. Use of passive diffusion tubes to monitor air pollutants. J. Air & Waste Manag. Assoc. 60, 204e209. http://dx.doi.org/10.3155/10473289.60.2.204. Nisi, B., Vaselli, O., Tassi, F., de Elio, J., Ortega, M., Caballero, J., Rappuoli, D., Mazadiego, L.F., 2014. Origin of the gases released from the Acqua Passante and Ermeta wells (Mt. Amiata, central Italy) and possible environmental implications for their closure. Ann. Geophys. 57 (4), S0438. http://dx.doi.org/10.4401/ ag-6584. Olafsdottir, S., Gardarsson, S.M., Andradottir, H.O., 2014. Spatial distribution of hydrogen sulfide from two geothermal power plants in complex terrain. Atmos. Environ. 82, 60e70. http://dx.doi.org/10.1016/j.atmosenv.2013.10.013. Pavilonis, B.T., O’Shaughnessy, P.T., Altmaier, R., Metwali, N., Thorne, P.S., 2013. Passive monitors to measure hydrogen sulfide near concentrated animal feeding operations. Environ. Sci. Process. Impacts 15 (6), 1271e1278. http:// dx.doi.org/10.1039/c3em00082f. Pennequin-Cardinal, A., Plaisance, H., Locoge, N., Ramalho, O., Kirchner, S., Galloo, J.C., 2005. Performances of the Radiello® diffusive sampler for BTEX measurements: influence of environmental conditions and determination of modelled sampling rates. Atmos. Environ. 39, 2535e2544. http://dx.doi.org/ 10.1016/j.atmosenv.2004.12.035. n, M., Salcido, A., Celada-Murillo, A.T., NavarroPeralta, O., Castro, T., Duro rquez, C., García, J., de la Rosa, J., Torres, R., VillegasGonz alez, R., Ma n-Sierra, S., Imaz, M., Martínez-Arroyo, A., Saavedra, I., de la Martínez, R., Carreo Luz Espinosa, M., Torres-Jaramillo, A., 2013. H2S emissions from Cerro Prieto geothermal power plant, Mexico, and air pollutants measurements in the area. Geothermics 46, 55e65. Percy, K.E., Hansen, M.C., Dann, T., 2012. 4. Air quality in the Athabasca oil sands region 2011. In: Percy, K.E. (Ed.), Alberta Oil Sands: Energy, Industry and the Environment, Developments in Environmental Science, vol. 11. Elsevier, Oxford, pp. 47e92. Schiffman, S.S., Williams, C.M., 2005. Science of odor as a potential health issue. J. Environ. Qual. 34 (1), 129e138. Shooter, D., Watts, S.F., Hayes, J., 1995. A passive sampler for hydrogen sulfide. Environ. Monit. Assess. 38, 11e23. Sigurdsson, H., Houghton, B., McNutt, S., Rymer, H., Stix, J., 2015. The Encyclopedia of Volcanoes, second ed. Academic Press. Tang, H., Sandeluk, J., Lin, L., Lown, J.W., 2002. A new all-season passive sampling system for monitoring H2S in air. Sci. World J. 2, 155e168. http://dx.doi.org/ 10.1100/tsw.2002.87. Tassi, F., Capecchiacci, F., Cabassi, J., Calabrese, S., Vaselli, O., Rouwet, D., Pecoraino, G., Chiodini, G., 2012. Geogenic and atmospheric sources for volatile organic compounds in fumarolic emissions from Mt. Etna and Vulcano Island (Sicily, Italy). J. Geophys. Res. 117, D17305. http://dx.doi.org/10.1029/ 2012JD017642. Tassi, F., Nisi, B., Cardellini, C., Capecchiacci, F., Donnini, M., Vaselli, O., Avino, R., Chiodini, G., 2013. Diffuse soil emission of hydrothermal gases (CO2, CH4 and C6H6) at the Solfatara di Pozzuoli crater (Campi Flegrei, southern Italy). Appl. Geochem. 35, 142e153. http://dx.doi.org/10.1016/j.apgeochem.2013.03.020. Tassi, F., Venturi, S., Cabassi, J., Capecchiacci, F., Nisi, B., Vaselli, O., 2015. Volatile organic compounds (VOCs) in soil gases from Solfatara crater (Campi Flegrei, southern Italy): geogenic source(s) vs. biogeochemical processes. Appl. Geochem. 56, 37e49. http://dx.doi.org/10.1016/j.apgeochem.2015.02.005. Thermo Fisher Scientific Inc, 2012. Thermo Scientific Model 450i Hydrogen Sulfide & Sulfur Dioxide Analyzer Manual. Link. https://www.thermoscientific.com/ content/dam/tfs/ATG/EPD/EPD%20Documents/Product%20Manuals%20&% 20Specifications/Air%20Quality%20Instruments%20and%20Systems/Ambient% 20Gas/D19731~.pdf.  hannsson, T., 2013. StatisThorsteinsson, T., Hackenbruch, J., Sveinbjӧrnsson, E., Jo tical assessment and modeling of the effects of weather conditions on H2S plume dispersal from Icelandic geothermal power plants. Geothermics 45, 31e40. http://dx.doi.org/10.1016/j.geothermics.2012.10.003. WHO, 2000. Air quality guidelines for Europe, second ed. WHO Regional Publications. European Series, No. 91. WHO, 2003. Hydrogen Sulfide: Human Health Aspects. Concise International Chemical Assessment Document 53. World Health Organization, Geneva, Switzerland. Williams-Jones, G., Reimer, H., 2015. Hazards of Volcanic Gases. The Encyclopedia of Volcanoes, second ed. Elsevier Inc, pp. 985e992. Zabiegala, B., Urbanowicz, M., Szymanska, K., Namiesnik, J., 2010. Application of passive sampling technique for monitoring of BTEX concentration in urban air: field comparison of different types of passive samplers. J. Chromatogr. Sci. 48, 167e175. http://www.sigmaaldrich.com/analytical-chromatography/air-monitoring/radiello/ faq.html.