Performance of the Full-scale Loop Hybrid Reactor Treating Coal Gasification Wastewater under Different Recirculation Modes

，Hongjun Han，F(xiàn)ang Fangand Wang Bing

（1．State Key Laboratory of Urban Water Resource and Environment，Harbin Institute of Technology，Harbin 150090，China；2．School of Municipal and Environmental Engineering，Shenyang Jianzhu University，Shenyang 110168，China）

Performance of the Full-scale Loop Hybrid Reactor Treating Coal Gasification Wastewater under Different Recirculation Modes

Qian Zhao1，Hongjun Han1?，F(xiàn)ang Fang1and Wang Bing2

（1．State Key Laboratory of Urban Water Resource and Environment，Harbin Institute of Technology，Harbin 150090，China；2．School of Municipal and Environmental Engineering，Shenyang Jianzhu University，Shenyang 110168，China）

This paper aims to investigate the simultaneous removal efficiencies of both COD and nitrogen in a single reactor treating coal gasification wastewater（CGW）.A novel loop hybrid reactor was developed and operated under different recirculation modes in order to achieve simultaneous removal of refractory compounds and total nitrogen（TN）in a full-scale CGW treatment plant.Mid-ditch recirculation was superior to other operational modes in terms of the NH3-N and TN removal，resulting in a TN removal efficiency of 52.3%. Although the system achieved equal COD removal rates under different recirculation modes，hydrophobic acid（HPO-A）fraction of effluent dissolved organic matter（DOMef）in mid-ditch recirculation mode accounted for 35.7%，compared to the proportions of 59.2%，45.3%and 39.4%for the other modes.The ultraviolet absorbance to dissolved organic carbon ratio test revealed that effluent under mid-ditch recirculation mode contained more non-aromatic hydrophilic components.Furthermore，appropriate recirculation and anoxic／oxic（A／O）partitions were also demonstrated to remove some refractory metabolites（phenols，alkanes，aniline，etc.），which reduced the chromaticity and improved the biodegradability.

coal gasification wastewater；refractory COD；total nitrogen；recirculation mode；DOM fractionation

1 Introduction

Nutrients removal is the most difficult task for a number of urban and industrial wastewater，such as landfill leachate［1-2］，pharmaceutical wastewater［3］，piggy wastewater［4］and black wastewater［5］. Biological processes，such as anaerobic-aerobic system，oxidation ditch or MBR are still the most economically viable techniqus.In recent years，attempts have been made to optimize the operating conditions in order to obtain simultaneous removal of organics and nitrogen with less reactors and hydraulic retention time（HRT）.Degradation of organic matters needs sufficient mixed liquor suspended solids（MLSS）and the nitrogen-removal potential depends on the reactor geometry，alternate／aerobic zones，aeration mode and intense internal circulation［6］.However，integrated removal of organic compounds and total nitrogen（TN）seems much more difficult for coal gasification wastewater（CGW）.CGW contains high concentration of hazardous and persistent compounds，many of which are typical nitrification inhibitors，such as benzene，toluene，ethylbenzene，xylene，phenols and polycyclic aromatic hydrocarbons（PAHs）［7-9］. Addition of powdered activated carbon（PAC）to an anaerobic or aerobic treatment system，i.e.powdered activated carbon technology（PACT），is known for its potential to enhance the removal efficiency of both refractory organic compounds and nitrification［10-12］. COD and TN in CGW were usually removed in different reactors or compartments［13］.Literatures about simultaneous removal of COD and TN are very limited to date.

In our previous lab-scale study，COD and total phenol（TPh）were remarkably removed via PACT［14］. But the NH3-N conversion rate was only 36.9%.In this study，oxidation-ditch operational modes were simulated using a full-scale PACT to treat CGW and the effectiveness and feasibility of the novel system were evaluated.The characterization of effluent dissolved organic matter（DOMef）was also performed by comparing DOM-fraction distributions and properties.

2 Materials and Methods

2.1 Process Description

A full-scale CGW zero liquid discharge（ZLD）demo-plant（capacity：360 m3／h）was built in China Coal Erdos Energy＆Chemical Co.，Ltd.PACT wasused after anaerobic process for further removal of COD and nitrogen.The quadrate PACT dimensions were：45 m long，30 m wide and 6 m high，with a total working volume of 8 000 m3.The corridors were 6 m in width，6 m in depth（as shown in Fig.1），with a total length of approx.210 m.The aeration system at the bottom in different area was controlled with different valves in order to obtain desirable anoxic／oxic zone as shown in Fig.2.

Fig.1 Process diagram of the hybrid reactor in phaseⅠ，Ⅱ，ⅢandⅣ

The aeration system could ensure well distribution of air while no air was supplied for the anoxic zone，resulting in a DO concentration less than 0.5 mg／L.10 stirrers were placed at 20 m-intervals along the ditch. PAC concentration of 1 g／L，HRT of 36 h and SRT of 30 d were used for PACT.At the end of the corridor was the inclined tube settling zone.The sludge at the bottom of the sedimentation zone was collected and transferred to the head of the ditch at a flow rate of 10 m3／h.DO and pH can be read from the fixed sensors.

2.2 Wastewater and Sludge

The main characteristics of theinfluent into the PACT tank are shown in Table 1.After the PACT tank was started up successfully in one month，MLSS concentration reached 3 037-4 472 mg／L and VSS／SS ratio was around 0.72.The sludge was grey-black with good settling properties.

Table 1 Main characteristics of the influent into the PACT tank

2.3 Operating Conditions and Testing Methods

After the start-up period，the system was operated stably for about one month with full-length aeration，namely phase I（day 1-33）.No mixed liquor recirculation was employed（Fig.1（a））.Methanol was fed as co-substrate with a dosage of 0.2 kg／m3at site 3［15］.During phase II（day 34-53），the mixed liquor at the end of the ditch was pumped back to the inletpoint（head-ditch recirculation mode，F(xiàn)ig.1（b）），with recirculation ratio of 200%.The same amount of methanol was added to the anaerobic／anoxic zone at site 1.During phase III（day 54-85），the mixed liquor was transferred to point A（mid-ditch recirculation mode，F(xiàn)ig.1（c））with other operating conditions unchanged.Point A was 66 m away from the inlet point.During phase IV（day 86-118），the mixed liquor at point 7 was recirculated to point B（called end-ditch recirculation mode，F(xiàn)ig.1（d））with the same recirculation ratio.Point B was 123 m away from the inlet point.During the 4 phases，duplicate samples were collected at 7 sampling sites daily，which were 0，42，66，105，123，156，213 m away from the inlet，respectively（as shown in Fig.1）.COD，TP，NH3-N，NO3-N and TN concentrations in each sampling site were analyzed once daily in accordance with standard methods［16］.DOC was measured by means of a carbon analyzer Shimadzu Co.，Japan.UV254was measured by a UV-vis spectrophotometer using a 1 cm quartz cell. During the stable operation in each phase，triplicate samples were collected from the effluent for DOC，DOM fraction and UV254（ultraviolet absorbance at 254 nm）analysis.The specific UV absorptivity（SUVA）index was defined as the ratio of the absorbance at 254 nm to the DOC content，representing the aromaticity and hydrophobic character of the DOMef.The filtrates were collected into precombusted glass amber bottles after filtrated through glass fiber filters（0.45 mm pore size）and stored in the dark at 4℃until analyzed.Gas chromatographymass spectrometry（GC-MS）was also performed［9］.

2.4 DOM Fractionation

TheDOMef fractionation method was derived from the procedures developed by Malcolm and Maccarthy［17］.The filtrates collected in Section 2.3 were fractionated into four fractions：hydrophobic-acid fraction（HPO-A），hydrophobic-neutral fraction（HPO-N），transphilic fraction（TPI），and hydrophilic fraction（HPI）by using XAD-4 and XAD-8 resins in series.Prior to adding to the columns，the resins were cleaned according to the procedures described by Leenheer［18］.

3 Results and Discussion

3.1 NH3-N Removal and Nitrogen Loss Pathway under Different Recirculation Modes

Fig.2 presents the profile of ammonia，nitrate and TN along the ditch under different recirculation modes. Ditch length means the distance between inlet point and the sampling point.During phase I，the NH3-N removal rate at sampling point 2 was only 3.3%.And no significant increase was observed after a shift in operational modes（phases II and III），indicating a severe suppression of the nitrifying bacteria activity by the high COD and TPh loadings at the head of the ditch.Ammonia oxidation rate increased with the increase of COD and TPh removal efficiencies along the ditch（as shown in Fig.3）.Eventually，58.2%of NH3-N was converted，resulting in nitrate concentration of 88.7 mg／L in the effluent.Without no aeration／anoxic alteration，it was reasonable to obtain little TN removal efficiency.Instead，an increase in TN concentration was observed at the some sampling points，which could be explained by the ammonia released from the cleavage of some nitrogen heterocyclic compounds（NHCs）［19-20］.

Fig.2 Profile of ammonia，nitrate and TN along the ditch in phaseⅠ，Ⅱ，ⅢandⅣ

Converting to head-ditch recirculation mode（phase II），the TN removal rate was not greatly improved，suggesting that pre-denitrification was not a feasible strategy for the TN removal enhancement.It could be attributed to the high concentration of nitrification inhibitors around the inlet point.Afterswitching to mid-ditch recirculation mode，the sharp increase in the NH3-N and TN removal rates were observed at sampling point 3 and 4.The increase in the nitrogen removal efficiency at point 3 was related with the dilution effect of internal circulation while the increase at point 4 could be attributed to denitrification occurring in the anoxic zone.During phase III，the nitrate concentration decreased in the effluent to 14.45 mg／L），compared to 88.65 mg／L in phase I. The methanol supplied at site 3 was quite beneficial for the biomass.They promoted denitrification as well as the degradation of refractory compounds as cosubstrate.Therefore，a fraction of COD was supposed to be consumed as electron donor for the denitrification. This was consistent with the COD variation along the ditch described in Section 3.2（as shown in Fig.3）. Within the first 66 m of the corridor，the refractory COD was remarkably removed，therefore reducing the inhibitory effect on the nitrifying bacteria denitrifying bacteria.52.3%of the TN was removed.Converting to phase IV，although the ammonia-to-nitrate conversion rate was high to 56.0%，the TN removal efficiency was only 36.5%，compared to 52.8%in phase III.It could be attributed to the little effective aerobic zones and inadequate retention time［21-22］.DO were 0.1，0.9，1.5，2.1，0.7，0.3 and 3.9 mg／L for sampling point 1-7，respectively，most of which were below the required level［6］.

Denitrification was the dominant mechanism of the TN removal in the hybrid process.The nitrogen removal amount via anoxic denitrification was approx.93.6，285.8 and 188.0 kg／d in phases II-IV，respectively，which strongly implied that A／O partitions and internal nitrate circulation were responsible for the enhancement of TN removal in the aerated-anoxic system.The midditch recirculation mode was superior to the headditch and end-ditch modes with respect to the TN removal from CGW in the full-scale loop hybrid reactor. Besides，compared with 3.5%（phase II），3.5%（phase III）and 2.4%（phase IV），little TN was removed via SND in phase I（data not shown），suggesting that appropriate recirculation mode was also beneficial for SND.

3.2 COD and TPh Removal Efficiency under Different Recirculation Modes

Without recirculation of mixed liquor，the PACT tank was substantially a type of loop reactor that generally exhibited plug-flow conditions.Converting to the recirculation modes，the reactor could be regarded as half-plug and half-mixing tank.Table 2 shows a contrast in the COD and phenol removal efficiencies under different operational modes.

During the 3 phases，the stable performance was marked by the constant COD and TPh concentration in the effluent.The averages of COD and TPh removal rates stayed at approx.70%and 85%even though the aeration area in the hybrid reactor were reduced by 20%in phases II-IV compared to phase I.Therefore，the biological reactions occurring in the anoxic zone might also contribute to the pollutants removal，resulting in a similar performance to that in phase I.

Table 2 COD and TPh removal efficiencies under during phaseⅠ-Ⅳ

Fig.3 shows the COD and TPh concentrations along the ditch.The amount of COD and TPh removed within the first half the ditch accounted for more than 70%of that was removed in the whole system. Therefore，high organic loading and relatively high removal efficiency at the first half of the ditch could be more likely to support active populations of the nitrifying bacteria and the denitrifying bacteria in the latter part of the reactor.Consequently，the configuration of the ditch-shaped reactor and A／O partitions favored the growth of the nitrifying bacteria and anaerobic bacteria.

Fig.3 Profile of COD and TPh along the ditch in phaseⅠ，Ⅱ，ⅢandⅣ

3.3 DOMef Fraction and UV254in the Effluent under Different Recirculation Modes

The DOM-fraction distribution pattern and the UV254are effective for evaluating the DOM characteristics.Despite of similar COD and DOC concentrations in the effluent，the corresponding fractional DOC might be varied，which can reflect different biological reactions occurring under different operational modes.The fractional DOC distributions in the effluents under different recirculation modes are shown in Fig.4（a）.

Fig.4 Fractional DOC and SUVA distribution of the effluent

HPO-A was the predominant aromatic fraction in the effluent（59.2%）during phase I，indicating the refractory characteristics and abundant presence of hydrophobic fractions.This observation was further confirmed by the much higher SUVA（approx.9.4 L／（m·mg））as shown in Fig.4（b）.In contrast，HPOA accounted for 45.3%，35.7%and 39.4%of DOMef under head-ditch，mid-ditch and end-ditch recirculation modes，respectively.HPO-A has low polarity and is the least degradable compounds［23］. Therefore，the refractory COD removal efficiency can be significantly enhanced by proper recirculation and A／O partitions.Transphilic fraction was another main ingredient of aromatic components（9.9%-26.3%）. The above-mentioned two fractions probably composed of PAHs，NHCs，aromatic protein related aromatic organics，etc.In general，HPO-A had the highest SUVA amongst the four fractions from all tested samples.As shown in Fig.4（b），the SUVA of the nonaromatic components（HPI fraction）under the four operational modes followed：mid-ditch recirculation（4.9 L／（m·mg））＞head-ditch recirculation（4.3 L／（m·mg））＞end-ditch recirculation（3.9 L／（m·mg））＞without recirculation（2.4 L／（m·mg））.It indicated that full-length aeration would produce less low-molecular non-aromatic components.

3.4 Analysis for the Organic Compositions of the Effluent

Organics compositions of the effluent inall phases are illustrated in Table 3.Volatile fat，the readily biodegradable intermediates from anaerobic hydrolysis and acidolysis，was not detected due to the immediate utilization by the biomass at the head of the ditch（data of the raw water is not shown）.Under recirculation modes，less phenols and recalcitrant substances were detected.2-methyl-phenol，3-ethyl-phenol，3，5-dimethyl-phenol，2，3，5，-trimethyl-phenol，1-ethyl-4-methoxy-benzene，aniline，2，6-bis（1，1-dimethylethyl）-naphthalene，concentrations，by contrast，had a remarkable reduction，indicating the advantages of introducing anoxic zones and internal circulation in the hybrid tank.One reason could be the denitrifying bacteria which utilized the relatively easily biodegradable organics or the intermediates from anoxic biological reactions as the electron donor to accomplish the TN removal.Another explanation was the facultative aerobes with higher activity，especially with the presence of methanol as co-substrate［15］.Phenols，cresols，alkanes and NHCs，were degraded or mineralized via different cleavage pathway under aerobic，anoxic and anaerobic conditions，resulting in different products［24-25］.In addition，it is noteworthy that aniline，which did not exist in the inlet，increased along ditch during phase I.In contrast，aniline obviously decreased in phases II-IV.Shi et al.have reported that aniline was the color-containing compound due to the existence of chromorphoric -CONH-，i.e.amino group in the molecule［26］.The increasing proportion of aniline along the ditch was in good agreement with the increase in chromaticity observed in phase I.Even though PAC present in the hybrid system has been found to partly remove chromophores from CGW，constant aeration probably led to the formation of aniline.Therefore，agreeing with fractionation，GC-MS analysis revealed that A／O partitions avoided the formation of some recalcitrant compounds，leading to a decrease in effluent chromaticity.

Table 3 Result of GC-MC analysis for effluent in phaseⅠ-IV

4 Conclusions

1）The novel loop hybrid reactor with mid-ditch recirculation and A／O partitions could achieve COD，TPh and TN removal efficiencies of 72.2%，84.9%and 52.8%，respectively.This paper provided reference for the simultaneous removal of refractory COD and nitrogen in CGW，coking wastewater or similar wastewater with high strength of organics and nitrogen，filling an international and domestic gap in this field.

2）The reactor was successfully applied in the fullscale plant.The corridor-shaped configuration of the loop reactor was beneficial for the simultaneous nutrients and nitrogen removal for CGW.

3）DOMef fractionation，SUVA and organic composition analysis revealed that the effluent under mid-ditch recirculation mode contained more nonaromatic hydrophilic components and less refractory metabolites（phenols，alkanes，aniline，etc.）.

4）It was suggested that some work needed to be done tooptimize the reactor with real-time control strategy in order to achieve higher nitrogen removal efficiency with lower external carbon resource.

［1］Aziz S Q，Aziz H A，Yusoff M S，et al.Removal of phenols and other pollutants from different landfill leachates using powdered activated carbon supplemented SBR technology.Environmental Monitor and Assessment，2012，184（10）：6147-6158.

［2］Peng Yongzhen，Zhang Shujun，Zeng Wei，et al.Organic removal by denitritation and methanogenesis and nitrogen removal by nitritation from landfill leachate.Water Research，2008，42（4／5）：883-892.

［3］Chen Zhaobo，Nie Shukai，Ren Nanqi，et al.Improving the efficiencies of simultaneous organic substance and nitrogen removal in a multi-stage loop membrane bioreactor-based PWWTP using an on-line knowledge-based expert system. Water Research，2011，45（16）：5266-5278.

［4］Ryu J H，Lee H L，Lee Y P，et al.Simultaneous carbon and nitrogen removal from piggery wastewater using loop configuration microbial fuel cell.Process Biochemistry. 2013，48（7）：1080-1085.

［5］De Graaff M S，Zeeman G，Temmink H，et al.Long term partial nitritation of anaerobically treated black water and the emission of nitrous oxide.Water Research，2010，44（7）：2171-2178.

［6］Guo Changzi，F(xiàn)u Wei，Chen Xuemei，et al.Nitrogenremoval performance and community structure of nitrifying bacteria under different aeration modes in an oxidation ditch.Water Research，2013，47（11）：3845-3853.

［7］Amor L，Eiroa M，Kennes C，et al.Phenol biodegradation and its effect on the nitrification process.Water Research，2005，39（13）：2915-2920.

［8］Li Huiqiang，Han Hongjun，Du Maoan，et al.Inhibition and recovery of nitrification in treating real coal gasification wastewater with moving bed biofilm reactor.Journal of Environmental Sciences，2011，23（4）：568-574.

［9］Wang Wei，Han Hongjun，Yuan Min，et al.Treatment ofcoal gasification wastewater by a two-continuous UASB system with step-feed for COD and phenols removal. Bioresource Technology，2011，102（9）：5454-5460.

［10］Cansado I，Mour?o P，F(xiàn)alc?o A I，et al.The influence of the activated carbon post-treatment on the phenolic compounds removal.Fuel Processing Technology，2012，103：64-70.

［11］Quan Xiangchun，Yang Zhifang，Shi Hanchang，et al.The effect of a secondary chlorophenol presence on the removal of 2，4-dichlorophenol（2，4-DCP）in an activated sludge system bioaugmented with 2，4-DCP degrading special culture.Process Biochemistry，2005，40（11）：3462-3467.

［12］Akta? ?，?e?en F.Adsorption，desorption and bioregeneration in the treatment of 2-chlorophenol with activated carbon. Journal of Hazardous Materials，2007，141（3）：769-777.

［13］Wang Zixing，Xu Xiaochen，Gong Zheng，et al.Removal of COD，phenols and ammonium from Lurgi coal gasification wastewater using A2O-MBR system.Journal of Hazardous Materials，2012，235／236：78-84.

［14］Zhao Qian，Han Hongjun，Xu Chunyan，et al.Effect of powdered activated carbon technology on short-cut nitrogen removal for coal gasification wastewater.Bioresource Technology，2013，142：179-185.

［15］Wang Wei，Han Hongjun，Yuan Min，et al.Enhanced anaerobic biodegradability of real coal gasification wastewater with methanol addition.Journal of Environmental Sciences，2010，22（12）：1868-1874.

［16］APHA.Standard Methods for Examination of Water and Wastewater.20th ed.Washington，DC：American Public Health Association，American Water Works Association，Water Environment Federation，1998.

［17］Malcolm R L，MacCarthy P.Quantitative evaluation of XAD-8 and XAD-4 resins used in tandem for removing organic solutes from water.Environment International，1992，18（6）：597-607.

［18］Leenheer J A.Comprehensive approach to preparative isolation and fractionation of dissolved organic carbon from natural waters and wastewaters.Environmental Science and Technology，1981，15（5）：578-587.

［19］Sun Qinghua，Bai Yaohui，Zhao Cui，et al.Aerobic biodegradation characteristics and metabolic products of quinoline by aPseudomonasstrain.Bioresource Technology，2009，100（21）：5030-5036.

［20］Bai Yaohui，Sun Qinghua，Zhao Cui，et al.Quinoline biodegradation and its nitrogen transformation pathway by aPseudomonassp.strain.Biodegradation，2010，21（3）：335-344.

［21］Peng Yongzhen，Hou Hongxun，Wang Shuying，et al. Nitrogen and phosphorus removal in pilot-scale anaerobicanoxic oxidation ditch system.Journal of Environmental Sciences，2008，20（4）：398-403.

［22］Hao X，Doddema H J，van Groenestijn J W.Conditions and mechanisms affecting simultaneous nitrification and denitrification in a Pasveer oxidation ditch.Bioresource Technology，1997，59（2／3）：207-215.

［23］Wei Liangliang，Wang Ke，Zhao Qingliang，et al. Fractional，biodegradable and spectral characteristics of extracted and fractionated sludge extracellular polymeric substances.Water Research，2012，46（14）：4387-4396.

［24］Harwood C S，Burchhardt G，Herrmann H，et al. Anaerobic metabolism of aromatic compounds via the benzoyl-CoA pathway.Fems Microbiology Reviews，1998，22（5）：439-458.

［25］Harwood C S，Gibson J.Anaerobic and aerobic metabolism of diverse aromatic compounds by the photosynthetic bacterium Rhodopseudomonas palustris.Applied Environmental Microbilogy，1988，54（3）：712-717.

［26］Shi Jianpeng，Wanyan hua，Hu Jiandong.Tentative study of aniline wastewater treatment by hybrid biofilm reactor. Water Treatment Technology，2006，32：57（in Chinese）.

X703.1

：

：1005-9113（2015）05-0031-07

10.11916／j.issn.1005-9113.2015.05.005

2015-01-19.

Sponsored by the State Key Laboratory of Urban Water Resource and Environment（Harbin Institute of Technology）（Grant No.2015DX02）.

?Corresponding author.E-mail：han13946003379＠163.com.