Monitoring of Dioxins/Furans (PCDD/Fs) in Smoke of Charcoal Grilled Meat-Restaurants in Amman-Jordan

PCDDs /Fs were investigated in outdoor air of charcoal grilled meat-restaurants in Jordan. Restaurants were given the codes (S, Z, G, B, and J). Levels of PCDDs/Fs depend on the quantity of used charcoal, amount of grilled meat, and sampling season. Results in (ng TEQ/m 3 ) found in the order: S restaurant (12.4) > J (3.4) > B (1.8) > G (1.6) > Z (0.2). Total emissions in (g TEQ / year) were in the order: S restaurant (1.295) > J (0.132) > B (0.092) > G (0.025) > Z (0.003). This study is the rst of its kind in MENA region. Polychlorinated dibenzo-p-dioxins (PCDDs) and polychlorinated dibenzofurans (PCDFs) concentrations were studied in outdoor air of charcoal grilled meat-restaurants in Amman/Jordan. The studied restaurants were given the codes (S, Z, G, B, and J). The PCDD/Fs compounds were extracted from the lter using the green extraction method called QuEChERS and detected in extracts using GC/MS. The concentrations of PCDD/Fs in the chimneys were found to depend on many factors that were measured such as type and quantity of grilling fuels, amount quantities of grilled food after marinating and sampling season. The results showed that PCDD/Fs concentrations in toxic equivalents (TEQs) in (ng TEQ/m 3 ) were found in the following order: S restaurant (12.4) > J (3.4) > B (1.8) > G (1.6) > Z (0.2) where some samples contained 2, 3, 7, 8-TCDD in considerable amounts. The total emissions of PCDD/Fs in (g TEQ / year) for all restaurants samples were in the following order: S restaurant (1.295) > J (0.132) > B (0.092) > G (0.025) > Z (0.003). This study is featured as the rst of its kind in the Middle East and North Africa that reveals the dangers of using charcoal for grilling because it causes contamination of food and air. It also calls for advising people to minimize the consumption of grilled food since food will contain a portion of the carcinogenic B (a) P and 2,3,7,8-TCDD and also advising for using lters for restaurant chimneys.


Introduction
Air pollution causes nowadays the atmosphere to be not clean as in the old time before the industrial revolution due to the increase of the concentrations of harmful chemicals in the atmosphere. Air pollution is caused from the entrance of one or more of contaminants that deteriorate the imbalance of air quality.
These contaminants may be found as solid particles, liquid droplets, or mixtures of gases which affects negatively on human, animals, plants and all living organisms by man-made and natural sources (Arthur 1977;Hewitt and Jackson 2007). There are many sources of human activities such as power plants, re neries, factories and vehicles emissions (Aschner and Costa 2015;Manisalidis et al. 2019). In addition, the chemical and fertilizer industries, agricultural emissions, residential heating and cooking ,domestic cleaning activities, wars, nuclear tests are also other human-made sources (Hewitt and Jackson 2007 ;Pénard-Morand and Annesi-Maesano 2004;Manisalidis et al. 2019). On the other hand, natural sources such as volcanic erosions, forests re, dust storms and oceans (Vallero 2007;Vesilind et al. 2013).Cooking activities produces many harmful pollutants that affect negatively the atmosphere and human health (Buonanno et al. 2009). Commercial restaurants kitchens are a main contribution source of outdoor air pollution Chen et al. 2007). In addition, food emissions have negative effects on the workers' health in restaurants (Svendsen et al. 2003) and on people who live near them .Cooking operations of food at high temperature cause emission of pollutants including volatile organic compounds (VOCs) and ne particulate organic matters , aliphatic hydrocarbons, Polycyclic aromatic hydrocarbons, dioxins/furans, heterocyclic aromatic amines (HCAs), BTEX (benzene, toluene, ethylbenzene, and xylenes), carbonyl compounds like (aldehydes, ketones ,carboxylic acids), and others (Kabir and Kim 2011;Lin et al. 2010 ;Schauer et al. 1999).When charcoal is used in cooking processes at high temperatures, these harmful chemicals are also emitted from the charcoal itself and the food cooked, so it can be a major contributor to the deterioration of the outdoor air quality of the surrounding environment (Pandey et al. 2009;Kabir et al. 2010). PCDD/Fs are abundant persistent organic pollutants that have been classi ed in the Stockholm Convention (UNEP 2001). PCDD/Fs are planar tricyclic aromatic compounds with 1 to 8 chlorine atoms. They are stable nonpolar compounds, have low solubility in water, low vapour pressure, have lipophilic character and they are bio-accumulative in the environment through the food chain (Fiedler et al. 1990). The 2, 3, 7, 8-substituted congeners of PCDDs are considered as the most toxic (carcinogenic) anthropogenic compounds to the environment and humans (Fiedler 1996;Fiedler et al. 1990). Dioxins can be emitted from many anthropogenic sources into the atmosphere. They can be spread in other components of the environment such as soil, sediments, seawater, air, vegetation, and animals (Mosca et al. 2010).They are produced as unwanted byproducts during the incomplete combustion of organic compounds from the anthropogenic sources (industrial and thermal processes) (Alcock and Jones 1996;Fries 1995). The most common source for PCDD/Fs exposure is food sources especially seafood and animal products during food processes (cooking) (Fries 1995;Hutzinger and Fiedler 1993).
This study will examine the emissions effects of charcoal grill restaurants on the outdoor air quality and to determine the levels of PCDD/Fs in stacks that depend on many factors and conditions that will be monitored, measured and investigated. Therefore, the main objective of this work is to monitor PCDD/Fs in the smoke of Amman's charcoal grilled meat restaurants in the chimneys. This study has the distinction of being the rst of its kind to be conducted in Jordan, the Middle East and North Africa (MENA). It is also one of the very few studies in the world that has examined concentrations of PCDD/ Fs emissions in the outdoor environment from restaurant stacks that use coal as grill fuel.
Other instruments used were analytical balance, drying oven, vortex mixer, centrifuge, anemometer, pump and Reacti-Therm with air compressor.

Sampling and sample preparation
Smoke samples from ve charcoal grilled meat-restaurants in Amman city were collected for PCDD/Fs analysis. The restaurants were given the symbols [S, Z, G, B and J]. The glass bre lters (37mm diameter, A/E type, PALL, Mexico) were used for collecting smoke samples. They were placed in Gelman cassette (37 mm 2 pieces, Supelco). Eighteen samples (chimneys sample and eld blank for each) were collected by using pump system for smoke suction from the restaurants' chimney which contains the emitted PCDD/Fs. It was faced the chimney at a distance of 1 meter and the sampling period takes six to eight hours. Sampling was reiterated for each restaurant and location according to many conditions and factors that were measured, monitored and documented. These conditions were : weather temperature, stack emissions ow rate, temperature of emissions, wind speed, humidity, cross sectional area of the chimney, height of the chimney (height of the building), sampling seasons (summer and winter) , type of charcoal used and type of the meat. Beside main samples many simultaneous blank samples (reference) were collected from each restaurant by using the same pump and lter type but at a distance of 50 m from the chimney at the same parameters.

Sample extraction and clean up
The extraction and clean-up steps for PCDDs/Fs are summarized as follows: The lter containing the collected sample was added to 50 ml empty polypropylene centrifuge tube. After that, 10 ml acetonitrile, 5 ml deionized water, and 70 µL of the internal standard (1, 2, 3, 4-tetrachloro [ 13 C12] dibenzo-p-dioxin, 200 ppb) were added to the centrifuge tube followed by strong shaking. Then QuEChERS extraction pack with salts mixture was added followed by vigorous shaking using vortex for about 5 minutes, and then was centrifuged for 10 minutes at 3500 rpm. 6 mL of the supernatant of the extraction step were transferred to a QuEChERS dispersive SPE clean up tube followed by vigorous shaking for about 5 minutes and centrifugation for 10 minutes at 8000rpm. Thereafter, 5 ml were evaporated to dryness at 40 C and the residues were reconstituted in 350 µL solvent-mix of (1:1 of dichloromethane/ hexane) and transferred to a 350 µL insert-GC vials.

PCDD/Fs quantitative identi cation
The GC/MS quantitative analysis of the dioxins/furans was done in two ways namely: scan mode and Selective ion monitoring mode (SIM). First, the mass spectrometer was applied in the full scan mode by scanning the mass over a wide range of masses (mass to charged ratio m/z). Each PCDD/Fs was distinguished by its retention time and mass fragments after scanning of masses established. SIM mode is then used and allows the mass spectrometer to detect speci c compounds of PCDD/Fs to get higher sensitivity with higher accuracy and precision instead of stepping a wide range of masses. Each compound was represented by a peak of a de nite retention time (Stashenko and Martinez 2014). Standards of 17 dioxins/furans congeners with their mix standards were injected in triplicate onto the GC-MS column in the scan mode to identify the retention time of each PCDD/F. After that PCDD/Fs were sorted into groups in the SIM mode where each compound was designated by its retention time and two major m/z fragments as shown in Table 1

Linear range
Dilution series of different concentrations were prepared and injected onto the GC-MS column to construct the calibration curves. Six concentration levels in (µg/L) of PCCD/Fs were used for the calibration curve to study the linearity of the 17 PCDD/Fs congeners. From the calibration curves, the linear equations and the coe cients of determination (R 2 ) were calculated. The coe cients of determination were found between 0.9902 and 0.9983 indicating a good linearity in the studied concentration range.

Detection limit and limit of quantitation
Limit of detection (LOD) of the instrument was obtained for PCDDs/Fs by diluting the mix standard solution of PCDD/Fs until the ratio of signal to noise (S/N) is equal to 3, while the limit of quantitation (LOQ) was obtained when the (S/N) ratio is equal to 10. The results show that for all studied PCDD/Fs, the LODs were between 0.0043and 0.0550 μg/L while the LOQs were between 0.0143 and 0.1832 μg/L.
Extraction recoveries for PCDD/Fs samples PCDD/Fs recovery experiments were done by using three blank samples. The blank lter for each sample was split in half. The rst half was extracted and cleaned-up by QuEChERS procedure. However, the second half was spiked with the three concentrations levels of PCDD/Fs standard solutions (low, medium, and high) in ppb and with the internal standard [1, 2, 3, 4 -Tetrachloro [ 13 C12] dibenzo-p-dioxin] ; to give a nal concentration in each sample of 1 ppm in 350 µL). Once the second half was prepared for each blank sample, it was extracted and cleaned up by the above mentioned procedure. The recovery results were ranged between 82 and 99% for the low concentration; 83-100% for the medium concentration and 82-101% for the high concentration. All recoveries were found to be within the acceptable range for trace analysis (84-102%) (Gonzalez and Herrador 2007).

Instrument precision for PCDD/Fs
The instrument precision was determined by the injection of three standard solutions in triplicate. The results show that the instrument precision was very good due to the results of the coe cients of variation (CV).The results of all CV were less than the accepted limit value for trace analysis (15%) (Gonzalez and Herrador, 2007).

Analysis of PCDD/Fs samples
Eighteen smoke samples from the chimneys of ve charcoal grilled meat-restaurants and also eighteen.
Eighteen samples eld blank samples for each were collected. The samples were undergone all previous steps of preparation, extraction, and clean up. The extracts were injected in triplicate onto the GC/MS column. Every sample was given a symbol that indicates the location of restaurant. For every studied sample the surrounding conditions were gathered. The most prominent of these were: weather conditions such as (wind speed, humidity, and the average temperature), information about the stack itself as (stack outlet velocity, stack outlet temperature), and also sampling hours, sampling seasons, and the volume of air in m 3 . All PCDD/FS concentrations were calculated after subtracting the corresponding eld blanks in µg/m 3 . Table 2 summarizes the pro les of the studied ve restaurants. Table S1 indicates PCDD/Fs results of the samples of the ve restaurants in µg/m 3 and the converted concentrations to the ng TEQ/m 3 smoke sample.

Results And Discussion
The smoke of ve charcoal grilled meat-restaurants in Amman city were chosen to do this study. The restaurants were given the symbols [S, Z, G, B and J]. This study focused basically on S restaurant that many experiments ( Table S2 and Figure 4. Charcoal grilled meat-restaurants generate PCDD/Fs compounds. The concentrations of these pollutants and compounds strongly dependent on many factors and conditions mentioned previously. From these factors that were determined and studied, the results were discussed. The variations in results between samples of the ve studied restaurants are caused by many possible factors. One of these studied factors is sampling time, (morning or evening). As we mentioned before we focused on S restaurant. For this restaurant the samples S2, S4, S6, and S7 those taken in the evening were the highest concentrations except for S4. The reason is simple, because the food consumption is more in the evening than in the morning, and thus the increase in the amount of charcoal used. Eventually the increase of chimneys emissions means more production of PCDD/Fs pollutants, and this considered as a major source of urban outdoor air pollution. In addition, S restaurant was the highest restaurants in food consumption, therefore amount of charcoal used was higher than the other studied restaurants, which means that restaurant's food and charcoal consumption is another factor that explains the difference between restaurant emissions. Also, one of the important factors that play the main role in the difference between the studied restaurants was the type and quantity of oil used. Finally, the meat type such as local or Romanian is also an important factor due to the amount of fat that the meat contained.
Comparison was made between the results of this study and previous study (Lin et al. 2011). Lin studied the levels of seventeen (PCDD/Fs) in the stacks of three fast food restaurants in Taiwan.

Conclusions
The results indicate that there are obvious variations of the PCDD/Fs concentrations from emitted emissions when compared with the previous studies. These variations in PCDD/Fs concentrations related to many conditions as quantities of food, food oil consumption rates, grilling fuels, marinating style, cooking times, Stack outlet velocity, Sampling season, and stack diameters. This study provides essential information that using charcoal as fuel for grilling is a major source for emissions of PCDD/Fs. The risk also revolves that the food containing considerable amounts of a ''possibly carcinogenic'' 2,3,7,8-TCDD compound.

Figure 2
Concentrations of PCDD/Fs compounds (µg/m3) for S restaurant chimney smoke at different sampling days.

Figure 4
Total emissions of PCDD/Fs to air for the ve restaurants chimneys' smoke in g TEQ/year.