Emissions of PCDD / Fs and PCBs during the Cold Start-up of Municipal Solid Waste Incinerators

This study investigated the emissions of polychlorinated dibenzo-p-dioxins, dibenzofurans (PCDD/Fs) and polychlorinated biphenyls (PCBs) from 45-hr cold start-up operations (from the ambient temperature) of two municipal solid waste incinerators (MSWIs). Fifteen samples were collected during the whole process, while the sampling period was divided as follows: (1) the initial stage (25–250°C); (2) the second stage (250–850°C), and the final stage (850–1000°C). The peak of total (PCDD/F + PCB) (WHO2005-TEQ) concentration occurred during the second stage of the cold start-up. This could because the temperature in the combustion chamber reached 295–359°C, within the temperature window (250–450°C) of PCDD/F reformation by de novo synthesis. However, both PCDD/F and PCB (WHO2005-TEQ) concentrations were still 2.8–3.8 times higher than the regulated standards in Taiwan 45 hours after the cold start-up began, demonstrating significant memory effects. Higher PCB/(PCDD/F + PCB) TEQ fractions were observed during the initial stage of cold start-up. This is probably due to the higher volatilities of memorized PCB-126 and -169 than PCDD/F congeners, thus making them more easily released into the flue gas. On the other hand, the total (PCDD/F + PCB) emissions during 45-hr cold start-up were estimated at 46–68 mg WHO2005-TEQ.The annual (PCDD/F + PCB) emissions from normal conditions are generally estimated by the normal operating emissions factor without the cold start-up. The normal annual (PCDD/F + PCB) emission rates of the two MSWIs were 70.7–101 mg WHO2005-TEQ, including once cold start-up emission, and these increased to 142–480 mg WHO2005-TEQ, up by 65–458%, if one to eight times of cold start-up were included. Therefore, reducing unnecessary start-ups could be more important with regard to controlling the PCDD/F and PCB emissions than only attempting to increase the pollutant removal efficiency in the steady operations of MSWIs.


Polychlorinated
dibenzo-p-dioxins, dibenzofurans (PCDD/Fs or dioxins) and dioxin-like polychlorinated biphenyls (PCBs) are both included in the five main groups of POPs (persistent organic pollutants).PCDD/Fs, PCBs, and organochlorine pesticides are known collectively as organochlorine compounds, and also "endocrine disruptors", which are able to impair normal embryonic development and disrupt reproductive functions in adulthood (Müllerová and Kopecký, 2007).Both PCDD/Fs and PCBs are emitted from the related sources and enter the atmosphere in trace amounts, and can then be transported long distances before they are deposited (Lohmann and Jones, 1998;Lee et al., 2003;Lin et al., 2012a).This will result in widespread dispersal through the environment, and initiate the airdeposition-soil/water-food chain-wildlife/human exposure cycle (Lohmann and Jones, 1998;Hu et al., 2009;Chang, et al., 2013).
PCBs are industrially produced chemicals with a broad range of commercial applications, such as heat transfer fluids, organic diluents, and plasticizers.PCDD/Fs were first discovered in the flue gases and fly ash of municipal solid waste incinerators (MSWIs) in 1977 (Olie and Vermeulen, 1977), and are byproducts which form during the synthesis of the primary industrial halogenated aromatics, from other commercial processes, or during combustion (Lohman and Seigneur, 2001).Both PCDD/Fs and PCBs have become a serious issue in many countries because of their toxicological effects and consequent adverse health implications.
The emission sources of PCDD/Fs and dioxin-like PCBs are mainly related to human activities, with industrial combustion processes in particular being significant sources of PCBs and PCDD/Fs.One of the largest sources of PCDD/F emissions is poorly controlled incineration, with the previous studies noting that MSWIs are also an important source of dioxin-like compounds (Wang et al., 2010;Tuan et al., 2012).The combined regulation of both PCDD/Fs and PCBs is thus needed to control their emissions.However, the PCDD/F and PCB emissions in Taiwan are relatively low, because the MSWIs that operate in Taiwan are larger and have newer pollution control technologies than those in many other areas, and the related emissions limit (0.1 ng I-TEQ/Nm 3 ) is the most stringent in the world.
Because of this strict emissions standard for PCDD/F, many people may neglect the high emissions that occur during the cold start-up of MSWIs, a fact that has been highlighted in a number of recent studies.Indeed, the elevated PCDD/F concentrations during start-up can reach 96.9 ng I-TEQ/Nm 3 , and a high PCDD/F concentration (40 times higher than the Taiwanese emissions limit) can remain even 18 hours after the injection of activated carbon (Wang et al., 2007).The legislation that is in effect in most countries, including Taiwan, means that PCDD/F emissions from incinerators only need to be measured once or twice annually, and these are usually obtained under normal and good operating conditions.Since the emissions that occur during a cold start-up are much higher than those during normal operations, this could thus serious underestimated of the detected values.
The objective of this study is to evaluate the dibenzo-pdioxins, polychlorinated dibenzofurans (PCDD/F) and polychlorinated biphenyls (PCB) concentrations and their characteristics in the stack flue gases of municipal solid waste incinerators, as well as their continuous emission scenarios under cold-start conditions.Two incinerators (1 and 2) at a continuously operating MSWI in Taiwan were investigated.The whole sampling period continued until 45 hours after the cold start-up.Fifteen samples were collected (samples A-O) in each run of cold start-up.The results obtained in this study can provide useful information to help decisionmakers to develop better policies for the control of PCDD/Fs and PCBs.

Sample Collection
The emissions and characteristics of PCDD/Fs and PCBs in the stack flue gases of two incinerators at a continuously operating MSWI in Taiwan were investigated.The capacity of each incinerator is 1,350 tonnes per day.The operating conditions of the MSWI during two runs of cold start-up are shown in Tables A1 and A2.Since the MSWI was constructed relatively recently and equipped with good air pollution control devices, it perform well with regard to the control of dioxins.In addition, the flue gases of these two incinerators are regularly inspected to ensure that the dioxin emissions meet to the national regulated standard (0.1 ng I-TEQ/Nm 3 ).
Tables A1 and A2 show the sampling periods of two runs of cold start-up, and the sampling time is plotted against the combustion chamber temperature in Fig. 1.The sampling period was divided into three parts: the initial stage (combustion chamber temperature rose slowly form 25°C to 250°C), the second stage (the combustion chamber temperature rose quickly to 850°C), and the final stage (the combustion chamber temperature remained steady at over 1000°C).For MSWI 1, samples A-C were in the initial stage, standing for the very early combustion in whole startup process.Sample D-E were in the second stage, which was the important temperature range of de novo synthesis of PCDD/F formation (Chin et al., 2012;Lin et al., 2012b).In this range, the combustion chamber temperature was increased as fast as allowable to inhibit the PCDD/F reformation, therefore, the higher sampling frequency was utilized to record more detail information during this quickly changing procedure.Finally, the temperature of furnace reached a high enough level for steady operation, while the samples F-O were collected.For Incinerator 2, samples A-C were obtained in the initial stage, D-F in the second stage, and G-O in the final stage.The whole sampling period lasted for 45 hours after the cold start-up, and 15 samples were collected (samples A-O) in each run.

PCDD/F and PCB Sampling
The sampling procedures for the stack flue gases complied with the requirements of NIEA A807.75C.The particulate sampling train adopted in this study is comparable with that specified by US EPA Modified Method 5.The sampling time of each stack flue gas sample during cold start-up is listed in Fig. 1, with 15 samples being collected from each stack in each run.

Analyses of PCDD/Fs and PCBs
Analyses of the stack flue gases followed the US EPA Modified Method 23.All chemical analyses were executed by the Super Micro Mass Research and Technology Center at Cheng Shiu University, which is one of seven accredited laboratory in Taiwan for PCDD/F analyses.
Prior to analysis, each sample was spiked with a known amount of internal standard solution to ensure the recovery during the analyzing process.After being extracted for 24 h, the extract was concentrated, treated with concentrated sulfuric acid, and then subjected to a series of sample cleanup and fractionation procedures, including the use of a multilayer silica gel column, alumina column and activated carbon chromatography.The eluent was concentrated to approximately 1 mL and transferred to a vial.The concentrate was further concentrated to near dryness, using a stream of nitrogen.A high-resolution gas chromatograph/highresolution mass spectrometer (HRGC/HRMS) was used for PCDD/Fs analyses.The HRGC (Hewlett Packard 6970 Series gas, CA, USA) was equipped with a DB-5MS fused silica capillary column (L = 60 m, ID = 0.25 mm, film thickness = 0.25 µm) (J&W Scientific, CA, USA), and splitless injection was carried out.The oven temperature program was set as follows: begin at 150°C (held for 1 min), then increased at 30 °C/min to 220°C (held for 12 min), then increased at 1.5 °C/min to 240°C (held for 5 min), and finally increased at 1.5 °C/min to 310°C (held for 20 min).The HRMS (Micromass Autospec Ultima, Manchester, UK) was equipped with a positive electron impact (EI+) source.The analyzer mode of selected ion monitoring (SIM) was used with the resolving power set at 10,000.The electron energy and source temperature were specified at 35 eV and 250°C, respectively.The detailed instrumental analysis parameters of PCDD/Fs are given in our previous works (Lee et al., 2003;Wu et al., 2009;Mi et al., 2012).

PCDD/F and PCB Levels from the MSWIs during the Cold Start-up Operations
The PCDD/F and PCB concentrations in the flue gases of the Incinerator 1 during its cold start are shown as Table 1.The PCDD/F concentration was between 0.338-245 ng WHO 2005 -TEQ/Nm 3 .In the initial stage of cold start (samples A-C), the PCDD/F concentrations were between 1.20-1.73and averaged 1.48 ng WHO 2005 -TEQ/Nm 3 .In the second stage (samples D and E), the PCDD/F concentrations increased significantly to 77.7-245 and averaged 161 ng WHO 2005 -TEQ/Nm 3 .In the final stage (samples F-O), the PCDD/F concentrations remained at 0.338-18.5 and averaged 3.76 ng WHO 2005 -TEQ/Nm 3 .The largest increase in both PCDD/F mass and WHO 2005 -TEQ concentrations occurred when the combustion chamber temperature reached 295°C (sample D), which is within the temperature region (250°C-450°C) of PCDD/F reformation by de novo synthesis (Shaub and Tsang, 1983;Alcock and Jones, 1996;Huang and Buekens, 1996;Yu, et al., 2010;Lin, et al., 2011).The concentration peak also occurred at sample D.
On the other hand, the PCB concentrations were between 0.0295-50.5ng WHO 2005 -TEQ/Nm 3 (Table 1).In the initial stage of cold start-up (samples A-C), the PCB concentrations were between 0.292-0.570ng WHO 2005 -TEQ/Nm 3 and averaged 0.449 ng WHO 2005 -TEQ/Nm 3 .The PCB concentrations had a similar pattern to that of the PCDD/Fs in the second stage (samples D-E), when they rose significantly to 10.2-50.5 and averaged 30.4 ng WHO 2005 -TEQ/Nm 3 .The PCB concentrations then fell to 0.0295-1.37 and averaged 0.279 ng WHO 2005 -TEQ/Nm 3 in the final stage of cold start-up (samples F-O).As with the PCDD/Fs, the greatest increase in both PCB mass and WHO 2005 -TEQ concentration occurred when the temperature of the combustion chamber reached 295°C (sample D), when the concentration peak also occurred.
The results for Incinerator 2 are shown in Table 2.The overall PCDD/F concentrations were between 0.227-79.8ng WHO 2005 -TEQ/Nm 3 .In the initial stage of cold start-up (samples A-C), the PCDD/F levels were only between 0.350-5.60ng and averaged 2.84 ng WHO 2005 -TEQ/Nm 3 .In the second (samples D-F), the PCDD/F concentrations increased to 10.3-79.8ng and averaged 54.0 ng WHO 2005 -TEQ/Nm 3 , as the temperature inside the combustion chamber became suitable for de novo synthesis.In the final stage (samples G-O), the PCDD/F concentrations fell to 0.227-17.4ng WHO 2005 -TEQ/Nm 3 .The concentration peak occurred at sample E. The PCB concentrations in from Incinerator 2 during the cold start-up were between 0.0353-16.7 ng WHO 2005 -TEQ/Nm 3 .In the initial stage (samples A-C), the PCB concentrations were between 0.191-0.881,and averaged 0.470 ng WHO 2005 -TEQ/Nm 3 .In the second stage (samples D-F) the PCB concentrations increased to 0.783-16.

Congener Profiles of PCDD/Fs from the MSWIs during Cold Start-up Operations
Congener profiles of seventeen 2,3,7,8-subsituted PCDD/F mass concentrations in the stack flue gases during the first run of the cold start in Incinerator 1, and the second run of Table 1.PCDD/F and PCB concentrations in the flue gases of the Incinerator 1 during the cold start-up.and A2, respectively.The fractions (%) were calculated using the ratio of the concentration of each congener to that of total PCDD/Fs.Figs.A1 and A2 show that for each different run of the incinerator the PCDD/F congener profiles for mass were different in each period of cold start-up.In other words, the congener profiles for mass in the initial stage were different from those in the middle and final stages of the MSWI cold start-up operations.However, the congener profiles were very similar for both Incinerator 1 and Incinerator 2 in each of these stages.
A very high fraction of the OCDD mass in the stack flue gases of MSWI is thought to come from precursor reactions, such as pentachlorophenol (PCP) or other chlorinated hydrocarbons mixing with various waste materials, such as waste wood and contaminated soil, since it is known that PCDD/F congener profiles in pentachlorophenol (PCP) show high percentages of OCDD.It is assumed that the OCDD that flowed into the facility via MSW was decomposed during incineration to a greater extent than it was formed (Akai, 2001).

Congener Profiles of PCBs from MSWIs during the Cold Start-up Operations
The congener profiles of twelve congeners of PCB mass in the stack flue gases of the MSWI during the first run of the cold start-up in Incinerator 1, and the second run of the cold start in Incinerator 2 are shown in Figs.A5 and A6, respectively, with the data for both obtained in 2010.Each of the fractions (%) was calculated as the ratio of the concentration of each congener to that of total PCBs.Figs.A5  and A6 show that for each different run of the MSWIs, second runs (87.2%), while PCB-169 accounted for 13.9% and 12.5% mass in the first and second runs, respectively, indicating that these two congeners accounted for at least 99.7% of the total PCBs -WHO 2005 -TEQ.This can also be clearly observed apparently by comparing Figs. 4 and 5.It can be seen that although all 12 congeners were produced simultaneously, their lower toxicities in comparison to PCB-126 and PCB-169 resulted in the higher contribution fractions of the latter two with regard to total PCBs -WHO 2005 -TEQ.

Total PCDD/F + PCB Levels from the MSWIs during the Cold Start-up Operations
The   During the first run of the cold start-up, the waste started to be injected into Incinerator 1 when the temperature of the combustion chamber reached 422°C at sample E, in the final stage of the operations.The air pollution control devices (APCDs) were activated along with the waste input, including semi-dry lime sludge and activated carbon injection, from samples E, with the average speed of 1.89 m 3 /hr and 9 kg/hr, respectively.They worked for the purpose of absorbing PCDD/Fs and PCBs in the stack flue gases.Although the high amount of PCDD/F emissions in the initial and second stages may be due to the lack of activated carbon injection, the low combustion chamber temperature and incomplete combustion conditions under normal operating conditions could also be responsible for these.
The wastes were injected into Incinerator 2 during the second run of cold start-up, when the temperature in the combustion chamber was 359°C at sample E, during the second stage of the operations.Along with the waste injection, both semi-dry lime mud and activated carbon were also injected in during samples E-O, with the average speed of 0.913 m 3 /hr and 9 kg/hr, respectively, for the same purpose noted above.
The significant difference of the highest (PCDD/F + PCB) emissions between two start-up in Fig. 6 were observed.According to the aforementioned operation procedure, the APCDs were started at different combustion ambient temperatures (422 and 359°C), indicating the different period of time to adsorb the precursors (chlorine and carbon sources) before de novo synthesis occurred.The reduction of precursors were expected to decrease the maximum levels of PCDD/Fs and PCBs during the start-up procedure (Wang et al., 2012).Therefore, the post activation of APCDs could be an effective way to reduce the high PCDD/F and PCB Additionally, the input of the waste (sample E) could be one of the reasons for the high PCDD/F emissions, as the second highest (77.7 ng WHO 2005 -TEQ/Nm 3 ) and highest (79.8 ng WHO 2005 -TEQ/Nm 3 ) emissions occurred when the wastes started to be injected during the first and second runs of the cold start.
Although both the semi-dry lime mud and activated carbon were injected both when and after the temperature in the combustion chamber reached 850°C, the (PCDD/F + PCB) emission concentration (0.368 and 0.262 ng WHO 2005 -TEQ/Nm 3 in Incinerators 1 and 2, respectively) was still high at 45 hours after the cold start, and about 2.6-3.7 times higher than the national regulated standard in Taiwan, which demonstrates the significance influence of the memory effect in this context.
Furthermore, the high concentrations of such compounds will become residue onto the APCD inner surfaces, filter bags etc., and will be continuously released into the flue gas and increase the emission concentration.The above phenomenon is called the PCDD/F and PCB memory effects, which usually occurs after an unsteady operation, such as start-up operation.In the start-up process, the de novo synthesis dominates the high level of PCDD/Fs and PCBs and the memory effect extend continuous emission period.The early mentioned method to reduce the maximum emission could further shorten the memory effect by less PCDD/Fs and PCBs residual in the flue system.

Contribution Fractions of PCBs to the Total PCDD/F + PCB Concentrations
Fig. 7(a) shows the contribution fractions (%) of the PCB WHO 2005 -TEQ concentrations to the (PCDD/F + PCB) WHO 2005 -TEQ concentrations versus time in the stack flue gases of the MSWI during the first run of the cold start in Incinerator 1.The contribution fractions ranged between 5.7%-24.8%and averaged 11.4%.In the initial stage of cold start-up (samples A-C), the fractions were between 19.6%-24.8%and averaged 22.8%; in the second stage (samples D and E), the contribution fractions were between 11.6%-17.1% and averaged 14.4%; in the final stage (samples F-O), the contribution fractions were between 5.7%-8.7%and averaged 7.41%.These results indicate that a greater fraction of PCB TEQ was discharged during the initial stage of cold start-up operations.This is probably due to the fact that at the lower temperature of between 138 and 200°C in this initial stage the PCB congeners PCB-126 and PCB-169 were more volatile than PCDD/F congeners such as 2, 3,4,7,1,2,3,7,2,3,7,8,and this contributed more to the (PCDD/F + PCB) WHO 2005 -TEQ concentrations.Fig. 7(b) shows the contribution fractions (%) of the PCB WHO 2005 -TEQ concentrations to the (PCDD/F + PCB) WHO 2005 -TEQ concentrations over time in the stack flue gases of the MSWI during the second run of the cold start-up in Incinerator 2. The contribution fractions ranged between 6.9%-42.1% and averaged 16.6%.In the initial stage of cold start-up (samples A-C), the contribution fractions were between 6.9%-42.1%,and averaged 23.2%.In the second stage (samples D-F), the contribution fractions were between 7.1%-18.8%,and averaged 14.4%.In the final stage (samples G-O), the contribution fractions were between 10.4%-19.3%, and averaged 15.1%.As with Incinerator 1, this is probably because at the lower temperature of between 77 and 160°C, which occurred in the initial stage of cold start-up, the PCB congeners PCB-126 and PCB-169 were more volatile than PCDD/F congeners such as 2, 3,4,7,2,3,4,6,7,1,2,3,7,and 1,2,3,4,6,7,and thus contributed more to the (PCDD/F + PCB) WHO 2005 -TEQ concentrations.
According to the results for these two runs of cold startup, a peak occurred in the initial stage, and then decreased rapidly in the second stage, and finally a balance condition was achieved in the final stage.Since the fractions of the PCB WHO 2005 -TEQ concentrations were about 1/10-1/5 (11.4%-16.6%), the significance of PCBs to the stack flue gases of MSWIs cannot be neglected.

Significance of Cold-Start Times on the Annual Total (PCDD/F+ PCB) Emissions
Incinerators are usually shut-down and started up at least once a year for maintenance, with this happening more often for older MSWIs.Therefore, we simulate sceneries with different numbers of cold start-ups and investigate the contribution fractions (%) related to this with regard to the annual total (PCDD/F + PCB) (WHO 2005 -TEQ) emissions (Table 3).
The (PCDD/F + PCB) (WHO 2005 -TEQ) emissions during cold start-up (45hr) were about twice as much as that of a whole year's normal operations (8547 hr), without considering the PCDD/F emissions contributed by the long lasting memory effect, because in this study the sampling period ended two days after cold start-up.

Increase in Annual total (PCDD/F + PCB) (WHO 2005 -TEQ) Emissions with Increasing Numbers of Cold Startups
The amount of annual (PCDD/F + PCB) (WHO 2005 -TEQ) emissions from normal operating conditions with one time of the cold start-up was 100.9 (mg WHO 2005 -TEQ) in Incinerator 1 and 70.7 (mg WHO 2005 -TEQ) in Incinerator 2 in 2010.However, if the number of cold start-ups was increased to two, four, six and eight times yearly, the amount of total (PCDD/F + PCB) (WHO 2005 -TEQ) emissions will be raised by approximately 65%, 196%, 327%, and 458%, respectively (Table 4).In other words, the average annual (PCDD/F + PCB) (WHO 2005 -TEQ) emissions will be 142, 255, 367, and 480 mg WHO 2005 -TEQ, respectively.Therefore, it is very important to try and prevent unnecessary cold start-ups in order to control PCDD/F and PCB emissions in MSWIs.respectively) 45 hours after the cold start-up began, at about 2.6-3.7 times higher than the national regulated standards in Taiwan.These results demonstrate the significance of PCDD/F and PCB memory effects in MSWI.A higher toxicity fraction of PCB-TEQ was emitted during the initial stage of cold start-up than in the second one.This is probably because at the lower temperature in the initial stage the PCB congeners PCB-126 and PCB-169 were more volatile than the PCDD/F congeners, such as 2,3,4,7,8-PeCDF and 2,3,4,6,7,8-HxCDF, and thus contributed more to the (PCDD/F + PCB) WHO 2005 -TEQ concentrations.
For general MSWI operating conditions, the amount of annual (PCDD/F + PCB) (WHO 2005 -TEQ) emissions might include one cold start-up emission in addition to normal emissions, which were estimated as 100.9 (mg WHO 2005 -TEQ) in Incinerator 1 and 70.7 (mg WHO 2005 -TEQ) in Incinerator 2 in 2010.However, if the number of cold startups was increased to between two and eight times annually, the total (PCDD/F + PCB) (WHO 2005 -TEQ) emissions will increase by approximately 65%-458%.Therefore, avoiding or reducing the number of cold start-up operations might be a more efficient and effective method to decrease the real amount of PCDD/F and PCB emissions than only controlling the pollutant levels during the steady operation of MSWIs.However, the start-up process is originally not a well control combustion condition.Two to four samples a year of normal emission are hard to represent and calculate an accurate annul emission of MSWI.Therefore, a continuous and long-term sampling (adsorption, optical method, or insitu chromatography) is suggested to collect more real and complete emission information for MSWIs in the future studies.

Fig. 1 .
Fig. 1.The sampling time corresponding to the combustion chamber temperature during the second run of the cold startup, (a) Incinerator 1 and (b) Incinerator 2.
7 and averaged 11.3 ng WHO 2005 -TEQ/Nm 3 .In the final stage (samples G-O), the PCBs were broken down or removed, and thus the concentrations fell to 0.0353-2.50 and averaged 0.445 ng WHO 2005 -TEQ/Nm 3 .Somewhat different to the PCDD/F concentrations in Incinerator 2, while the largest increases in both PCB mass and WHO 2005 -TEQ concentration occurred when the temperature in the combustion chamber reached 359°C (sample E), the highest PCB WHO 2005 -TEQ concentrations occurred later, at sample F.
variations in the total (PCDD/F + PCB) (WHO 2005 -TEQ) concentration in the stack flue gases with time during the first and second runs of cold start-up are shown in Fig. 6.Fig. 6(a) shows that the peak of the total (PCDD/F + PCB) (WHO 2005 -TEQ) concentration (sample D) was at 10 hours after combustion started, during the second stage of the cold start-up operations.At this time the temperature in the combustion chamber reached 295°C, within the temperature window (250°C-450°C) of PCDD/F de novo synthesis.

Fig. 4 .
Fig. 4. Congener profiles of total PCB (WHO 2005 -TEQ) concentration in the stack flue gases of the Incinerator 1 during its cold start-up in 2010, (a) mass and (b) toxicity.

Fig. 5 .
Fig. 5. Congener profiles of total PCB (WHO 2005 -TEQ) concentration in the stack flue gases of the Incinerator 2 during its cold start-up in 2010, (a) mass and (b) toxicity.

Fig. 6 .
Fig. 6.Total (PCDD/F + PCB) concentration (ng WHO 2005 -TEQ/Nm 3 ) in the stack flue gases of the Incinerator during the first run of the cold start-up in 2010, (a) Incinerator 1 and (b) Incinerator 2.

Fig. 7 .CONCLUSIONS
Fig. 7. Percentage of PCB (WHO2005-TEQ) concentration in the total (PCDD/F+PCB) concentration in the stack flue gases during the first run of the cold start, (a) Incinerator 1 and (b) Incinerator 2.

Figure
Figure A1 Congener profiles of PCDD/F mass concentration in the stack flue gases of the MSWI 1 during its cold start in 2010

Table 2 .
PCDD/F and PCB concentrations in the stack flue gases of the Incinerator 2 during the cold start-up.
TableA1The operating conditions of the MSWI 1 during its cold start-up TableA2The operating conditions of the MSWI 2 during its cold start-up