Fuel 203 (2017) 749–756

Contents lists available at ScienceDirect

Fuel

journal homepage: www.elsevier .com/locate / fuel

Full length article

Behavior of mercury emitted from the combustion of coal and driedsewage sludge: The effect of unburned carbon, Cl, Cu and Fe

http://dx.doi.org/10.1016/j.fuel.2017.04.1040016-2361/� 2017 Elsevier Ltd. All rights reserved.

⇑ Corresponding author.E-mail address: slee@chungbuk.ac.kr (S.-S. Lee).

Sang-Sup Lee a,⇑, Jennifer Wilcox b

aDepartment of Environmental Engineering, Chungbuk National University, Cheongju 28644, Republic of KoreabDepartment of Chemical & Biological Engineering, Colorado School of Mines, Golden, CO 80401, USA

a r t i c l e i n f o

Article history:Received 8 January 2017Received in revised form 19 March 2017Accepted 24 April 2017Available online 15 May 2017

Keywords:CoalDried sewage sludgeMercuryCombustion

a b s t r a c t

As the behavior of mercury generated from the combustion of mercury-containing fuels is highly depen-dent upon mercury speciation, many studies have investigated mercury speciation following combustionof various fuels. Unburned carbon and the content of chlorine (Cl), copper (Cu), and iron (Fe) compoundsin ash are reported to affect the behavior of mercury. This study was conducted to understand the impor-tant factors impacting behavior of mercury from single coal combustion and co-combustion of dried sew-age sludge with coal. Factors reported in previous studies were selected as potential factors affectingmercury oxidation and retention. Six coal and six dried sewage sludge samples were used to conduct sin-gle coal combustion and co-combustion tests. Each fuel sample was combusted in a lab-scale drop tubecombustion system. Concentrations of elemental and oxidized mercury were determined using theOntario Hydro Method. Ash samples collected on the bottom of the combustion system were analyzedto determine unburned carbon and mercury retention percentages. Unburned carbon and Cl were foundto be important in both oxidation and retention of mercury for single coal combustion. Cl, Cu, and Fe con-tent increased the oxidation of elemental mercury generated from co-combustion of dried sewage sludgewith coal.

� 2017 Elsevier Ltd. All rights reserved.

1. Introduction

Mercury is generated in flue gas following combustion of fuelscontaining mercury, such as fossil fuels and waste fuels. Threeforms of mercury exist in the combustion flue gas: elemental,oxidized, and particulate. All of the gaseous mercury exists aselemental mercury under the high temperature in the furnacebased on thermochemical equilibrium. The elemental mercurycan be oxidized by subsequent cooling and homogeneous andheterogeneous reactions [1,2]. The amount of mercury controlledby conventional air pollution control devices is highly dependentupon mercury speciation [3,4]. Particulate mercury is easily cap-tured by particulate matter control devices such as electrostaticprecipitators (ESPs) and a fabric filter (FF) [3,5]. A significantamount of oxidized mercury can be captured by alkali and alkalineearth metal compounds used in a flue gas desulfurization (FGD)system [6]. However, elemental mercury captured by conventionalair pollution control devices is negligible. Previous studies havereported that the formation mechanism of each mercury species

is complex and affected by many factors. Gibb et al. [7] combustedfour coal samples and investigated the fate of mercury generatedby the combustion. They found that mercury retention in ash isclosely related to the unburned carbon content of the ash. Asunburned carbon content increased, mercury retention increasedand gaseous mercury emission decreased. Wang et al. alsoreported that unburned carbon content is important for both oxi-dation and adsorption of mercury [8]. Results obtained from theanalysis of fly ash collected in ESPs in pulverized bituminouscoal-fired boilers also showed a consistent trend of increasing mer-cury retention with increasing unburned carbon content of ash [9].However, Sable et al. found that mercury retention decreased withincreasing unburned carbon content, implying that other factorsalso affect mercury capture on ash [10,11]. Chlorine content of fuelis reported to be closely related to oxidation of elemental mercurybecause the primary mercury oxidation product in flue gas isbelieved to be HgCl2 [9]. A study on the speciation of mercury infly ashes using the temperature programmed decompositiontechnique reported that chlorides played an important role inmercury oxidation and retention [12,13]. In addition, ferric oxide(Fe2O3) and cupric oxide (CuO) in ash are reported to increasethe oxidation of elemental mercury [14,15].

750 S.-S. Lee, J. Wilcox / Fuel 203 (2017) 749–756

Recently, energy conversion by combustion of dried sewagesludge has been suggested as a method to dispose of sewage sludgesince the Ocean Dumping Act now regulates its disposal in theocean. As dried sewage sludge has a heat value comparable to thatof low rank coal, the co-combustion of dried sewage sludge withcoal is one of the most promising methods for sewage sludge dis-posal. Sewage sludge is very different from coal in terms of mineralmatter content, mercury content, and other properties. Due tothese different properties, co-combustion of dried sewage sludgewith coal may have different mercury emission behavior comparedto the single combustion of coal. Studies have been conducted tounderstand mercury emission behavior from co-combustion ofsecondary fuel with coal. Tests were conducted on co-combustion of secondary fuels such as chicken manure, live resi-due, and wood with coal using a 50 kW entrained-flow pulverizedfuel combustor [10]. Secondary fuels were added at 10% and 20%(on an energy basis) to the coal. Mercury oxidation increased withan increase in chicken manure. Co-combustion of chicken manureresulted in a higher fraction of oxidized mercury in the flue gasthan that obtained from co-combustion of olive residue or woodwith coal. This was ascribed to the higher chlorine content ofchicken manure, indicating the importance of chlorine in mercuryoxidation.

Previous studies have reported that several factors such asunburned carbon, and metal and chlorine contents of fuel affectthe behavior of mercury emitted from the combustion of solidfuels. However, only some of these factors were investigated ineach study. To the best of our knowledge, this is the first studyon the importance of each factor depending on the type of fuelbeing used. In addition, very few studies have been conducted toelucidate the behavior of mercury from co-combustion of driedsewage sludge with coal. In this study, concentrations of elementaland oxidized mercury were determined in the flue gas from singlecombustion of coal and co-combustion of dried sewage sludge withcoal. The unburned carbon percentage, and the chlorine (Cl), cop-per (Cu), and iron (Fe) content of the fuel were selected as potentialfactors affecting the behavior of mercury. The effects of these fac-tors on mercury oxidation and retention were investigated to gaina better understanding of the behavior of mercury from both singlecoal combustion and co-combustion of dried sewage sludge withcoal.

2. Materials and method

2.1. Materials

Six coal and six dried sewage sludge samples were tested in thisstudy. Two different ranks of coal samples with different composi-tions were used. Dried sewage sludge samples were obtained fromdifferent wastewater treatment plants. The samples were sieved toobtain the particles between 53 and 105 lm. Ultimate and proxi-mate analyses of coal and dried sewage sludge samples were con-ducted, and the results are summarized in Table 1. The mercurycontent of each sample was determined using a solid mercury ana-lyzer (DMA-80, Milestone, Inc., Shelton, CT, U.S.A.). Copper (Cu)and iron (Fe) content was determined using inductively coupledplasma optical emission spectroscopy (ICP-OES) techniques inorder to investigate their effects on the oxidation of elementalmercury. Chlorine (Cl) content was determined using the oxygenbomb combustion-ion chromatography (EN 14582) method touse it as a potential factor in both oxidation and retention of mer-cury. Carbon (C) and mercury (Hg) content of the samples wereused to determine post-combustion unburned carbon and mercuryretention percentages, respectively. Proximate analysis of the sam-ples was conducted using a thermogravimetric analyzer (Q600, TA

Instruments Inc. U.S.A.). The higher heating values of the sampleswere determined using the Dulong’s formula:

Hh ¼ 8100Cþ 34250ðH� O8Þ þ 2250 S kcal=kg fuel ð1Þ

where C, H, O and S are the content of carbon, hydrogen, oxygen andsulfur in weight mass fraction. As shown in Table 1, the metal con-tent of dried sewage sludge samples is much higher than that ofcoal samples. Due to the very low ash content of coal sample B,the unburned carbon and mercury retention percentages were notdetermined for the ash from combustion of coal sample B. The mer-cury content of the dried sewage sludge samples is 10–40 timeshigher than that of the coal samples. As the sewage sludge sampleswere pre-dried using a specially designed heat-flow rotary kilndryer, they have similar moisture content to the coal samples. Forco-combustion, the coal sample A was selected because it has anaverage level of Hg, Cl, Cu and Fe content among the coal samplesused in this study. Each dried sewage sludge sample was pre-mixed with the coal sample A at a mass ratio of 1:3 or 1:5. The coalsample F was also used in co-combustion tests for comparisonpurposes.

2.2. Experimental method

Experiments were conducted using a lab-scale combustion andgas analysis system as shown in Fig. 1. The system consists of threeparts: (1) fuel feeding, (2) combustion, and (3) gas analysis. Thefeeder (Rovo feeder, Fine Techniques Co., Korea) is capable of pre-cision feeding with high feed accuracy and constancy (repeatability0.5%). Fuel was combusted in a drop tube furnace. The drop tubereactor was made of quartz, with an inside diameter of 4.4 cm.The drop tube reactor was covered by an electrical furnace witha height of 1 m. Three thermocouples were located outside thedrop tube reactor along with the height of the furnace to controlthe temperature. The temperature outside the drop tube reactorwas maintained at 950 �C i.e., above the auto-ignition temperatureof the fuel. The combustion flue gas was analyzed using a gas ana-lyzer (Vario Plus, MRU, Germany) and the Ontario Hydro Method.The gas analyzer determined the concentrations of carbon monox-ide (CO) and carbon dioxide (CO2) in order to check the stabilityand extent of combustion. Mercury speciation was conductedusing the Ontario Hydro Method. A heat tape with a temperatureof 110 �C covered the reactor and tubing between the combustorand the impinger train to prevent water vapor from condensingon the inside of the reactor and tubing.

Each single and mixed fuel was combusted at an air/fuel ratio of1.2. The injection rate of each fuel was determined in order toobtain the air/fuel ratio of 1.2 at an air flow rate of 4 L/min. Fuelwas injected and burned at a constant rate in the drop tube furnacefor 11 min. The combustion test conditions are summarized inTable 2. Some of the flue gas was sampled for gas and mercuryanalysis. One stream of the sampled gas passed through the gasanalyzer to measure concentrations of CO and CO2 every 2 s. Theother stream of sampled gas passed through an impinger train con-taining a 1 N potassium chloride (KCl) solution and a 4 wt% potas-sium permanganate (KMnO4)/10 wt% sulfuric acid (H2SO4)solution. The concentration of oxidized mercury in the flue gaswas determined from the amount of mercury absorbed in the KClsolution. The concentration of elemental mercury was determinedfrom the amount of mercury absorbed in the KMnO4 solution.Based on the concentrations of oxidized ([Hg2+]out) and elemental([Hg0]out) mercury, the mercury oxidation percentage was deter-mined by the following equation:

Mercury oxidationð%Þ ¼ ½Hg2þ�out½Hg2þ�out þ ½Hg0�out

� 100 ð2Þ

Table 1Ultimate and proximate analysis results of coal and dried sewage sludge samples.

Sample Rank Ultimate analysis (dry basis) Proximate analysis

C(%)

H(%)

N(%)

S(%)

Hg(ppm)

Cl(ppm)

Cu(ppm)

Fe(%)

Ca(%)

Moisture(%)

Volatilematter (%)

Fixedcarbon (%)

Ash(%)

Higher heatingvalue (kcal/kg)

Coal A Bituminous 74.4 4.4 1.4 0.48 0.048(±0.01) 160 14 0.16 0.93 5.8 30.3 55.7 8.2 7180Coal B Subbituminous 68.7 5.1 1.2 0.17 0.014(±0.002) 170 10 0.12 0.48 10.7 39.5 45.2 4.6 6860Coal C Bituminous 73.2 3.8 1.7 0.62 0.073(±0.005) 150 15 0.11 0.91 2.9 25.0 58.3 13.8 7010Coal D Bituminous 71.1 4.2 1.9 0.44 0.025(±0.004) 330 10 0.14 0.06 1.0 30.4 51.3 17.3 6990Coal E Subbituminous 70.0 4.6 1.6 0.77 0.028(±0.002) 30 80 0.28 0.14 17.6 35.1 42.2 5.1 7040Coal F Bituminous 74.7 4.3 2.0 0.62 0.032(±0.001) 245 17 0.22 0.66 6.5 26.3 54.9 12.3 7200Sludge1 – 22.3 4.1 4.1 1.07 0.79(±0.04) 470 2901 3.0 1.77 6.7 46.1 5.5 41.7 2660Sludge2 – 28.5 4.9 4.8 1.08 1.92(±0.28) 130 565 1.12 2.53 5.3 48.6 8.3 37.8 3420Sludge3 – 28.9 4.8 4.7 0.99 2.54(±0.57) 170 490 0.93 2.49 6.3 47.1 8.1 38.5 3460Sludge4 – 21.8 3.9 3.9 1.09 1.22(±0.16) 300 1375 1.77 2.00 8.2 43.4 5.8 42.6 2620Sludge5 – 35.2 5.4 5.6 0.55 0.67(±0.05) 100 182 0.78 0.83 3.2 46.4 5.7 44.7 4270Sludge6 – 34.4 5.5 5.7 0.92 1.97(±0.10) 180 448 0.98 2.31 9.4 53.0 10.2 27.4 4060

Fig. 1. Schematic diagram of the combustion and mercury measurement system.

Table 2Summary of combustion test conditions.

Item Conditions

Fuel Coal A–F, Dried sewage sludge 1–6Coal A: sludge (5:1), Coal A: sludge(3:1)Coal F: sludge (3:1)

Temperature outside thecombustor

950 �C

Inside diameter of the combustor 4.4 cmLength of the combustor (furnace) 1 mAir flow rate 4 L/min at 20 �CAir/fuel ratio 1.2

S.-S. Lee, J. Wilcox / Fuel 203 (2017) 749–756 751

Ash leaving the furnace was collected on the bottom of the fur-nace and analyzed for the concentrations of unburned carbon andmercury. The unburned carbon concentration in the ash was deter-mined by Korean standard loss on ignition (KS L 5405) method. Themercury concentration in the ash was determined using the solid

mercury analyzer. Based on the unburned carbon ([C]ash) and mer-cury ([Hg]ash) concentrations in the ash, the unburned carbon andmercury retention percentages were determined by the followingequations:

unburned carbonð%Þ ¼ ½C�ash½C�sample

� ½ash�sample � 100 ð3Þ

Mercury retentionð%Þ ¼ ½Hg�ash½Hg�sample

� ½ash�sample � 100 ð4Þ

where [C]sample, [ash]sample and [Hg]sample are the carbon, ash andmercury concentrations, respectively, in the sample. The unburnedcarbon percentage was estimated as loss on ignition because it hasbeen reported to affect the behavior of mercury [15–17].

3. Results and discussion

3.1. Combustion stability test

In addition to mercury speciation, combustion flue gas was ana-lyzed for concentrations of CO and CO2 to check the stability andextent of combustion. Fig. 2 shows the concentrations of CO andCO2 in the flue gas generated from the combustion of coal A. Asshown, the gas concentrations were consistent after approximately1 min from the start of combustion. The average CO2 concentrationcalculated after 1 min of combustion was more than 90% of thetheoretical CO2 concentration. The average CO concentration wasas low as approximately 50 ppm. This suggests that combustionis stable and complete. The combustion flue gas was sampled formercury speciation during 10 min after the first minute from thestart of combustion, in order to obtain samples during stable andcomplete combustion. Stable and complete combustion was con-firmed for all single combustion and co-combustion tests con-ducted in this study. Each combustion test under the sameconditions was conducted four times, but only the results obtainedunder stable and complete combustion were used. Their averageand standard error values were also determined to investigatethe oxidation and retention of mercury with respect to the contentof Cl, Cu and Fe.

3.2. Single combustion of coal

Single coal combustion was conducted for six coal samplesshown in Table 1. The concentrations of elemental and oxidizedmercury in the flue gas generated from single coal combustionwere determined, and mercury oxidation percentages were deter-mined using Eq. (2). The results are presented in Figs. 3 and 4. As

Fig. 2. Concentrations of CO and CO2 in flue gas from combustion of coal A.

Fig. 3. Concentrations of elemental and oxidized mercury in flue gas from singlecoal combustion.

Fig. 4. Mercury oxidation percentage for single coal combustion.

Fig. 5. Mercury oxidation percentage with unburned carbon percentage for singlecoal combustion.

752 S.-S. Lee, J. Wilcox / Fuel 203 (2017) 749–756

shown in Fig. 3, different mercury emissions were found depend-ing on the mercury content of the coal. Total mercury concentra-tions in the flue gas were found to be less than 3 lg/m3 for allcoal combustion tests. The mercury oxidation percentage inFig. 4 shows a different trend of results from the total mercury con-centration in Fig. 3. This suggests that mercury emission concen-tration is not related to the mercury oxidation percentage.

In order to identify the important factors affecting the oxidationof elemental mercury emitted from the single combustion of coal,the unburned carbon percentage was determined from the carbonconcentration in ash obtained from each combustion test using Eq.(3). Fig. 5 presents the mercury oxidation percentage with theunburned carbon percentage. As shown, the mercury oxidationpercentage is related to the unburned carbon and increases withan increase in the unburned carbon percentage. The mercury oxi-dation percentages of more than 30% located above the trend lineare obtained from combustion of coal D and F which contain a highlevel of chlorine among the coal samples. In Fig. 6, the mercury oxi-dation percentage is presented with respect to the Cl, Cu, and Fecontent of the coal samples. Chlorine content increases the oxida-tion of elemental mercury in flue gas as reported in several previ-ous studies. However, Cu and Fe were found to have little effect onthe oxidation of gaseous elemental mercury from single coalcombustion.

The mercury retention percentage was also determined fromthe mercury concentration in ash obtained from each combustiontest using Eq. (4). In Figs. 7 and 8 it is presented with respect tothe unburned carbon percentage and Cl content, respectively,which were suggested as important factors in mercury oxidation.Similar to the mercury oxidation percentage, the mercury reten-tion percentage increased in proportion to increases in theunburned carbon percentage and Cl content. Although Cl contentshowed closer relationship with mercury retention percentagethan with mercury oxidation percentage, this may be ascribed that

Fig. 6. Mercury oxidation percentage with Cl, Cu, and Fe content for single coal combustion.

Fig. 7. Mercury retention percentage with unburned carbon percentage for thesingle coal combustion.

Fig. 8. Mercury retention percentage with Cl content for single coal combustion.

Fig. 9. Concentrations of elemental and oxidized mercury in flue gas from co-combustion of dried sewage sludge with coal A.

S.-S. Lee, J. Wilcox / Fuel 203 (2017) 749–756 753

the mercury retention percentage for the coal sample B was notdetermined due to its very low ash content. These results indicatethat the unburned carbon percentage and Cl content are importantin both oxidation and retention of mercury for single coal combus-tion. This also suggests that mercury oxidation is closely related tomercury retention for single coal combustion.

3.3. Co-combustion of dried sewage sludge with coal

Coal sample A was pre-mixed with each of the dried sewagesludge samples at a mass ratio of 3:1 or 5:1 and combusted. Forthis co-combustion, the outlet elemental and oxidized mercuryconcentrations, and the mercury oxidation, unburned carbon, and

mercury retention percentages were determined as described forsingle coal combustion in the previous section. Fig. 9 shows theconcentrations of elemental and oxidized mercury emitted fromthe co-combustion of dried sewage sludge with coal A comparedto those from single combustion of coal A. As shown, gaseous mer-cury emissions significantly increased following co-combustionwith the dried sludge samples. This is ascribed to the higher mer-cury content of the dried sewage sludge samples compared to thatof the coal samples. In addition, higher mercury oxidation percent-ages were found in the co-combustion flue gases, compared tothose in the combustion flue gas of coal A, as shown in Fig. 10. Sim-ilarly to single coal combustion, total mercury concentration wasnot related to the mercury oxidation percentage for theco-combustion.

Mercury oxidation percentages determined from eachco-combustion test are presented with respect to the unburnedcarbon percentage in Fig. 11. In contrast to single coal combustion,unburned carbon was not related to the oxidation of gaseouselemental mercury from the co-combustion of dried sewage sludgewith coal. Fig. 12 shows that Cu and Fe content, which were notfactors in mercury oxidation for single coal combustion, affect

Fig. 10. Mercury oxidation percentage for co-combustion of dried sewage sludgewith coal A.

Fig. 11. Mercury oxidation percentage with unburned carbon percentage for co-combustion of dried sewage sludge with coal A.

Fig. 13. Cu and Fe content with respect to Cl content of samples used for co-combustion of dried sewage sludge with coal A.

754 S.-S. Lee, J. Wilcox / Fuel 203 (2017) 749–756

oxidation of elemental mercury from co-combustion. This may beascribed to much higher Cu and Fe content of dried sewage sludgesamples compared to coal samples. While unburned carbon and Cuand Fe content played differing roles in mercury oxidation between

Fig. 12. Mercury oxidation percentage with Cl, Cu, and Fe cont

single coal combustion and co-combustion, Cl content played animportant role in mercury oxidation for both single coal combus-tion and co-combustion, as shown in Fig. 12. To determine whethera potential confounding factor is associated, Cu and Fe content arepresented in Fig. 13 with respect to Cl content. Cu and Fe contentare closely related to Cl content, indicating that these are potentialconfounding factors. However, mercury oxidation percentagesincreased with co-combustion of dried sewage sludge comparedto those obtained from single combustion of coal A, as shown inFig. 10. In addition, co-combustion with dried sewage sludgesamples 2 and 5, which have a lower Cl content than coal A,demonstrated higher mercury oxidation percentages than singlecombustion of coal A. This indicates that factors other than Clcontent also affect oxidation of elemental mercury fromco-combustion of dried sewage sludge. Due to more dominanteffects of Cu and Fe on mercury oxidation, unburned carbon mayhave little effect on the oxidation of elemental mercury from theco-combustion.

Mercury retention percentages are presented with respect tothe unburned carbon percentage and Cl content in Figs. 14 and15, respectively. Fig. 14 shows that unburned carbon hardly affectsmercury retention in co-combustion (similarly to mercury oxida-tion). Although Cl content increased mercury oxidation, it wasnot related to retention of mercury from co-combustion, as shownin Fig. 15. This may be ascribed to lower mercury retention per-centages of 4.5–21.9% for co-combustion compared to 6.6–78.4%for single coal combustion. During co-combustion of dried sewagesludge with coal under our combustion test conditions, the oxida-tion of elemental mercury by Cl, Cu, and Fe compounds in ash may

ent for co-combustion of dried sewage sludge with coal A.

Fig. 14. Mercury retention percentage with unburned carbon percentage for co-combustion of dried sewage sludge with coal A.

Fig. 15. Mercury retention percentage with Cl content for co-combustion of driedsewage sludge with coal A.

Fig. 16. Mercury oxidation percentage with Cl content for both single coalcombustion and co-combustion of dried sewage sludge with coal A (Red symbolfor single coal combustion and blue symbol for co-combustion). (For interpretationof the references to colour in this figure legend, the reader is referred to the webversion of this article.)

Fig. 17. Mercury oxidation percentage with Cl content for co-combustion of driedsewage sludge with coal F compared to that with coal A.

S.-S. Lee, J. Wilcox / Fuel 203 (2017) 749–756 755

be more dominant than retention of mercury on ash. However, infull-scale application, a significant amount of oxidized mercurygenerated from co-combustion may be captured onto fly ash inparticulate matter control devices such as ESPs and an FF.

Among the factors affecting mercury oxidation and retentioninvestigated in this study, Cl content was the only factor to affectthe oxidation of elemental mercury from both single coal combus-

tion and co-combustion. Fig. 16 presents the mercury oxidationpercentage obtained from both combustion types with respect tofuel Cl content. Although Cl content was also shown to be some-what related to oxidation of elemental mercury from both combus-tion types, co-combustion test results show a larger slope ofincrease in mercury oxidation with Cl content than those for thesingle coal combustion test. This indicates the increase in mercuryoxidation is affected by other factors such as high Cu and Fe con-tents in ash from co-combustion of dried sewage sludge with coal.Additional co-combustion tests were conducted using coal F. Theirmercury oxidation results are presented with respect to fuel Clcontent in Fig. 17 in addition to the results from co-combustionof coal A. As shown in the figure, co-combustion test results of coalF show a similar trend of mercury oxidation results to those of coalA. This indicates that the behavior of mercury from co-combustionfound in this study could be generally observed in the co-combustion of coal with dried sewage sludge. The effects of coaltype and combustion condition on the behavior of mercury fromco-combustion with dried sewage sludge will be investigated infuture research.

4. Conclusion

Using a lab-scale drop tube combustion system, single combus-tion tests for six kinds of coal and co-combustion tests for six driedsewage sludge samples with coal were conducted. Concentrationsof elemental and oxidized mercury emitted from each single coalcombustion and co-combustion test were determined. To gain abetter understanding of mercury emission behavior, mercury oxi-dation and retention percentages were determined after each com-bustion test. Unburned carbon percentage and Cl, Cu, and Fecontent were selected as potential factors affecting mercury oxida-tion and retention. All factors investigated in this study were foundto be potential factors affecting mercury oxidation during combus-tion. However, important factors in mercury oxidation differdepending on fuel properties. While unburned carbon was impor-tant in mercury oxidation for single combustion of coal that con-tains a low level of metals, Cu and Fe content were important formercury oxidation following co-combustion of coal and dried sew-age sludge that contains large amounts of Cu and Fe. Cl contentwas found to be important for both single coal combustion andco-combustion. Co-combustion of dried sewage sludge with coaldemonstrated higher gaseous mercury emissions and mercury oxi-dation percentages than single coal combustion. The higher mer-cury oxidation percentage may be ascribed to mercury oxidationby large amounts of Cu and Fe in ash, in addition to that by Cl.Unburned carbon and Cl content were also found to be important

756 S.-S. Lee, J. Wilcox / Fuel 203 (2017) 749–756

factors in the retention of mercury from single coal combustion.Although the factors selected in this study were not found to berelated to mercury retention for co-combustion, a significantamount of oxidized mercury generated from co-combustion maybe captured onto fly ash in full-scale particulate matter controldevices. Therefore, the content of Cl, Cu, and Fe, which affect mer-cury oxidation, may also be related to retention of mercury for co-combustion of dried sewage sludge with coal in full-scale opera-tions. It is known that the oxidation of mercury can be facilitatedthrough wall effects and by many surfaces [18–20], and the impactof the large surface-to-volume ratio of the small lab system [21,22]versus large commercial power plants will be investigated infuture research. In addition, phosphorous is known to be an impor-tant constituent in sewage sludge [23], and the impact of phospho-rous upon mercury speciation can be determined in future work.

Acknowledgements

This work was supported by the Korea Institute of Energy Tech-nology Evaluation and Planning(KETEP) and the Ministry of Trade,Industry & Energy(MOTIE) of the Republic of Korea (No.20153010102030) and by Basic Science Research Program throughthe National Research Foundation of Korea (NRF) funded by theMinistry of Education (2015R1D1A1A01060942). The authorsthank Sin-Wook Kang, Hae-In Kim, Tanveer Ahmad and Wan KyuLim for their comment and experimental assistance.

Appendix A. Supplementary data

Supplementary data associated with this article can be found, inthe online version, at http://dx.doi.org/10.1016/j.fuel.2017.04.104.

References

[1] Kilgroe JD, Sedman CB, Srivastava RL, Ryan JV, Lee CW, Thorneloe SA. Control ofmercury emissions from coal-fired electric utility boilers: interimreport. National Risk Management Research Laboratory, United StatesEnvironmental Protection Agency; 2001.

[2] Lee SS, Sasmaz E, Padak B, Wilcox J. Homogeneous and heterogeneous mercuryreaction chemistry in coal combustion flue gases. Pittsburgh Coal Conference;2009.

[3] Zhou J, Wu S, Pan Y, Su Y, Yang L, Zhao J, et al. Mercury in municipal solidswaste incineration (MSWI) fly ash in China: Chemical speciation and riskassessment. Fuel 2015;158:619–24.

[4] Varejao EVV, Bellato CR, Fontes MPF. Mercury fractionation in streamsediments from the Quadrilatero Ferrifero gold mining region, Minas GeraisState, Brazil. Environ Monit Assess 2009;157:125–35.

[5] Park KS, Seo YC, Lee SJ, Lee JH. Emission and speciation of mercury fromvarious combustion sources. Powder Technol 2008;180:151–6.

[6] Tang S, Wang L, Feng X, Feng Z, Li R, Fan H, et al. Actual mercury speciation andmercury discharges from coal-fired power plants in Inner Mongolia, NorthernChina. Fuel 2016;180:194–204.

[7] Gibb WH, Clarke F, Mehta AK. Fate of coal mercury during combustion. FuelProcess Technol 2000;65:365–77.

[8] Wang F, Wang S, Meng Y, Zhang L, Wu Q, Hao J. Mechanisms and roles of flyash compositions on the adsorption and oxidation of mercury in flue gas fromcoal combustion. Fuel 2016;163:232–9.

[9] Kolker A, Senior CL, Quick JC. Mercury in coal and impact of coal quality onmercury emissions from combustion systems. Appl Geochem2006;21:1821–36.

[10] Sable SP, Jong WD, Meij R, Spliethoff H. Effect of secondary fuels andcombustor temperature on mercury speciation in pulverized fuel co-combustion: Part 1. Energy Fuels 2007;21:1883–90.

[11] Sable SP, Jong WD, Spliethoff H. Combined homo and heterogeneous model formercury speciation in pulverized fuel combustion flue gases. Energy Fuels2008;22:321–30.

[12] Lopez-Anton MA, Diaz-Somoano M, Martinez-Tarazona MR. Mercury retentionby fly ashes from coal combustion: influence of the unburned carbon content.Energy Fuels 2007;46:927–31.

[13] Lopez-Anton MA, Perry R, Abad-Valle P, Diaz-Somoano M, Rosa M, Martinez-Tarazona MR, et al. Speciation of mercury in fly ashes by temperatureprogrammed decomposition. Fuel Process Technol 2011;92:707–11.

[14] Ghorishi SB, Lee CW, Jozewicz WS, Kilgroe JD. Effects of fly ash transition metalcontent and flue gas HCl/SO2 ratio on mercury speciation in waste combustion.Environ Eng Sci 2005;22:221–31.

[15] Bhardwaj R, Chen X, Vidic RD. Impact of fly ash composition on mercuryspeciation in simulated flue gas. J Air Waste Manage Assoc 2009;59:1331–8.

[16] Abad-Valle P, Lopez-Anton MA, Diaz-Somoano M, Martinez-Tarazona MR. Therole of unburned carbon concentrates from fly ashes in the oxidation andretention of mercury. Chem Eng J 2011;174:86–92.

[17] Wilcox J, Sasmaz E, Kirchofer A, Lee SS. Heterogeneous mercury reactionchemistry on activated carbon. J Air & Waste Manage Assoc 2011;61:418–26.

[18] Presto AA, Granite EJ. Critical review: survey of catalysts for oxidation ofmercury in flue gas. Environ Sci Technol 2006;40:5601–9.

[19] Presto AA, Granite EJ, Karash A, Hargis RA, O’Dowd WJ, Pennline HW. A kineticapproach to the catalytic oxidation of mercury in flue gas. Energy Fuels2006;20:1941–5.

[20] Presto AA, Granite EJ. Noble metal catalysts for mercury oxidation in utilityflue gas. Granite Platinum Metals Rev 2008;52:144–54.

[21] Pennline HW, O’Dowd WJ, Freeman MC, Granite EJ, Karash A, Lacher C. Amethod to remove mercury from flue gas: the thief process. Fuel ProcessTechnol 2006;87:1071–84.

[22] Granite EJ, Freeman MC, O’Dowd WJ, Hargis RA, Pennline HW. The thiefprocess for removal of mercury from flue gas. J Environ Manage2007;84:628–34.

[23] Granite EJ. A pilot plant design for the integrated gasification process [MSthesis]. New York, New York: The Cooper Union; 1989.