TREATM ENT OF REFINERY OILY SLUDG E U SIN G ULTRASOUND, BIO-SURFACTANT, AND ADVANCED O X IDA TIO N PROCESSES by Ju Zhang B. S., Agricultural U niversity o f Hebei, 2005 M. S., Being Normal U niversity, 2008 THESIS SUBMITTED IN PARTIAL FULFILLM ENT OF THE REQUIREM ENTS FOR TH E D EG REE OF M ASTER OF SCIENCE IN NATURAL RESOURCES AND ENVIRONM ENTAL STUDIES (ENVIRONM ENTAL SCIENCE) UNIVERSITY OF NORTHERN BRITISH COLUM BIA September 2012 © Ju Zhang, 2012 1+1 Library and Archives Canada Bibliotheque et Archives Canada Published Heritage Branch Direction du Patrimoine de I'edition 395 Wellington Street Ottawa ON K1A0N4 Canada 395, rue Wellington Ottawa ON K1A 0N4 Canada Your file Votre reference ISBN: 978-0-494-94115-7 Our file Notre reference ISBN: 978-0-494-94115-7 NOTICE: AVIS: The author has granted a non­ exclusive license allowing Library and Archives Canada to reproduce, publish, archive, preserve, conserve, communicate to the public by telecommunication or on the Internet, loan, distrbute and sell theses worldwide, for commercial or non­ commercial purposes, in microform, paper, electronic and/or any other formats. L'auteur a accorde une licence non exclusive permettant a la Bibliotheque et Archives Canada de reproduire, publier, archiver, sauvegarder, conserver, transmettre au public par telecommunication ou par I'lnternet, preter, distribuer et vendre des theses partout dans le monde, a des fins commerciales ou autres, sur support microforme, papier, electronique et/ou autres formats. The author retains copyright ownership and moral rights in this thesis. Neither the thesis nor substantial extracts from it may be printed or otherwise reproduced without the author's permission. L'auteur conserve la propriete du droit d'auteur et des droits moraux qui protege cette these. Ni la these ni des extraits substantiels de celle-ci ne doivent etre imprimes ou autrement reproduits sans son autorisation. In compliance with the Canadian Privacy Act some supporting forms may have been removed from this thesis. Conform em ent a la loi canadienne sur la protection de la vie privee, quelques formulaires secondaires ont ete enleves de cette these. W hile these forms may be included in the document page count, their removal does not represent any loss of content from the thesis. Bien que ces formulaires aient inclus dans la pagination, il n'y aura aucun contenu manquant. Canada ABSTRACT Oil refinery sludge can be generated throughout the oil production process. It consists o f a large amount o f petroleum hydrocarbons (PHCs) and other hazardous materials w hich should be disposed o f appropriately. In order to find effective m ethods to treat the oily sludge, three different approaches w ere investigated in this research, including biorem ediation, oil recovery, and advanced oxidation processes (AOPs). In term s o f the biorem ediation approach, the oily sludge was mixed with soil, and a screened bacterium strain was then introduced w ith the supplem ent o f nutrients and the addition o f bio-surfactant. The reduction rate o f total petroleum hydrocarbons (TPH) in oily sludge spiked soil was up to 50.8% after 40 days o f biodegradation. W ith regard to recovering oil from the oily sludge, three processes w ere investigated, including ultrasonic treatment alone, freeze/thaw treatment alone, and the combined ultrasonic and freeze/thaw treatment. The experim ental results revealed that the com bined process could achieve satisfactory perform ance o f oil recovery. In term s o f the advanced oxidation processes (AOPs), the ultrasonic treatm ent alone, the Fenton process alone, and the combination o f ultrasound and Fenton’s reagents, were exam ined for their abilities to reduce petroleum hydrocarbons (PHCs) content in oily sludge. The Taguchi experimental results indicated that the com bination o f ultrasound and Fenton reagents achieved the best effect, with the highest TPH reduction rate o f 88.1% being observed in the experimental conditions. TABLE OF CONTENTS A BSTRA C T.............................................................................................................................................. ii TABLE OF C O N T E N T S.......................................................................................................................iii LIST OF TABLES................................................................................................................................... vi LIST OF F IG U R E S ............................................................................................................................. viii LIST OF A BBR E V IA TIO N S...............................................................................................................xi ACK NO W LEDG EM ENT................................................................................................................... xii Chapter 1 General Introduction...........................................................................................................1 Chapter 2 Rem ediation o f refinery oily sludge using isolated strain and biosurfactant.. 6 A b stract.....................................................................................................................................................7 1. Introduction........................................................................................................................................ 8 2. M aterials and M ethods...................................................................................................................10 2.1. Oily sludge and soil sam p le................................................................................................. 10 2.2. Chem icals................................................................................................................................. 11 2.3. Bacteria isolation and identification.................................................................................. 11 2.4. Bioremediation experim ent.................................................................................................. 14 2.5. Analysis o f total petroleum hydrocarbon (T P H )........................................................... 15 3. Results and discu ssio n ...................................................................................................................17 3.1. TPH biodegradation...............................................................................................................17 3.2. Analysis o f optimal biorem ediation con d itio n s...............................................................19 4. C onclusion.......................................................................................................................................25 Chapter 3 Oil recovery from refinery oily sludge via ultrasound and freeze/th aw 26 A bstract.................................................................................................................................................. 27 1. Introduction......................................................................................................................................28 2. Materials and m ethods.................................................................................................................. 31 2.1. Oily slu d g e...............................................................................................................................31 2.2. Chemicals and B io-surfactant............................................................................................. 32 2.3. Experiments on oil recovery using three different p ro cesses...................................... 32 2.4. Experiments on factors affecting the com bined treatm ent process............................. 35 2.5. Sample extraction and analysis............................................................................................ 36 2.5.1. TPH concentration in the recovered oil la y e r............................................................ 36 2.5.2. TPH concentration in separated aqueous p hase.........................................................38 2.5.3. PHCs an aly sis................................................................................................................... 39 3. Results and d iscu ssio n .................................................................................................................. 41 3.1. Com parison o f m ethods....................................................................................................... 41 3.2. Effects o f different factors on the com bined treatment process................................... 46 3.2.1. Effects o f ultrasonic pow er and treatm ent du ratio n..................................................46 3.2.2. Effects o f sludge/water ra tio .......................................................................................... 48 3.2.3. Effects o f bio-surfactant (rham nolipids)..................................................................... 49 3.2.4. Effects o f salt addition (sodium chloride)................................................................... 51 3.3. PHC fraction analysis for recovered o il..............................................................................52 4. C o nclusion.......................................................................................................................................54 C h a p te r 4 T rea tm e n t o f oily sludge th ro u g h ad v an ced o x id atio n p ro c e sse s ...................... 57 A b stract.................................................................................................................................................. 58 1. Introduction......................................................................................................................................59 2. Materials and m eth o d s.................................................................................................................. 6 6 2.1. Oily sludge and chem icals................................................................................................... 6 6 2.2. Ultrasonic apparatus.............................................................................................................. 67 2.3. Experim ent on oxidation o f PHCs in oily sludge by three different p ro cesses 67 2.4. Experim ent on factors affecting the com bined process (U S/F enton)..........................70 2.5. Sample extraction after advanced oxidation processes..................................................72 2.6. PHCs an aly sis.........................................................................................................................75 3. Results and discussions................................................................................................................. 76 3.1. Comparison o f m ethods.........................................................................................................76 3.1.1. TPH reduction using ultrasonic irradiation alo n e ................................................... 76 3.1.2 TPH reduction through Fenton’s reaction process.................................................... 79 3.1.3 TPH reduction through the combination o f ultrasound and Fenton’s reaction process (US/Fenton).........................................................................................................84 3.1.4 Petroleum hydrocarbons distribution in samples and fractional analysis after oxidation processes.......................................................................................................... 8 8 3.2. Degradation o f petroleum hydrocarbons after different treatm ents through US/Fenton p ro cess........................................................................................................... 90 iv 3.3. Impact o f factors on TPH degradation through the com bined (US/Fenton) process94 3.2.1. The effect o f sludge content........................................................................................... 97 3.2.2. The effect o f ultrasonic treatm ent tim e ......................................................................100 3.2.3. The effect o f the ratio o f H 2 O 2 to Fe2+.......................................................................101 3.4. Impact o f factors on degradation o f PH C s in different fractions through the com bined (US/Fenton) process.........................................................................................102 4. C onclusion .......................................................................................................................I l l Chapter 5 General C onclusion...................... 113 Limitations and future research....................................................................................................... 115 R EFER ENC E........................................................................................................................................ 117 v LIST OF TABLES Table 2.1 Summary o f clean soil p ro p erties............................................................................... 10 Table 2.2 Summary o f oily sludge properties.............................................................................10 Table 2.3 Taguchi experimental d esig n ..................................................................................15 Table 2.4 ANOVA for biorem ediation o f T P H .........................................................................20 Table 2.5 ANOVA for F3 fraction degradation.........................................................................22 Table 2.6 ANOVA for F4 fraction degradation.........................................................................22 Table 3.1 Characteristics o f the oily sludge................................................................................32 Table 3.2 Influencing factors and their corresponding lev els.................................................36 Table 3.3 Summary o f oil recovery results using different m ethods.....................................42 Table 3.4 PHCs fraction distribution for samples before and after US+F/T treatm ent... 53 Table 4.1 Experiment factors and their three le v e ls .................................................................71 Table 4.2 L27 array orthogonal experimental d e sig n ........................................................... 72 Table 4.3 TPH reduction in oily sludge through US treatm ent alone................................... 77 Table 4.4 TPH reduction in samples under Fenton process alo n e.........................................80 Table 4.5 TPH content in aqueous and solid phases o f samples* before and after advanced oxidation processes...............................................................................................89 Table 4.6 Remaining TPH mass in samples after different US/Fenton treatm en ts 92 Table 4.7 Degradation o f petroleum hydrocarbons in sam ples after different U S/Fenton treatments and the S/N ratio re su lts.................................................................................... 93 Table 4.8 Analysis o f Variance for TPH degradation after different treatm ents through US/Fenton p ro cess..................................................................................................................97 Table 4.9 Analysis o f variance for Fraction 2 (F2) degradation after different U S/Fenton treatm ents................................................................................................................................ 103 Table 4.10 Analysis o f variance for Fraction 3 (F3) degradation after different treatm ents through US/Fenton treatm ents..........................................................................................104 Table 4.11 Analysis o f variance for Fraction 4 (F4) degradation after different treatm ents through US/Fenton treatm ents............................................................................................104 LIST OF FIGURES Figure 2.1 Isolation procedure for PHCs degrading m icroorganism s.................................. 12 Figure 2.2 Soil contaminated with PHCs in a land-farm ing s ite ........................................... 13 Figure 2.3 Enrichment o f b a c te ria ................................................................................................ 13 Figure 2.4 Isolated strain: Luteibacter s p ....................................................................................14 Figure 2.5 A flask w ith soil.............................................................................................................16 Figure 2.6 Setup o f the biorem ediation experim ents................................................................ 16 Figure 2.7 TPH reduction rate for different trea tm e n ts........................................................... 18 Figure 2.8 TPH concentration remained in soil after b iorem ediation .................................. 19 Figure 2.9 Effect o f independent factors on TPH reduction...................................................21 Figure 2.10 Effect o f independent factors on F3 fraction reduction..................................... 21 Figure 2.11 Effect o f independent factors on F4 fraction reduction..................................... 23 Figure 2.12 TPH fraction distribution before and after bioremediation: “Be” - before bioremediation and “A f ’- after biorem ediation................................................................24 Figure 3.1 Oily sludge sam p le.......................................................................................................31 Figure 3.2 Ultrasonic treatment sy stem .......................................................................................33 Figure 3.3 Ultrasonic treatment for oil re c o v e ry ...................................................................... 34 Figure 3.4 Evaporation o f extraction so lu tio n ........................................................................... 37 Figure 3.5 Round flask w ith residual after evaporation...........................................................38 Figure 3.6 Aqueous sample after oil reco v ery........................................................................... 39 Figure 3.7 Samples for GC an aly sis............................................................................................ 40 Figure 3.8 Oil recoveries for different treatm ent methods (error bar represents standard dev iatio n )..................................................................................................................................42 Figure 3.9 TPH concentrations in separated oil layer (a) and w ater (b) (error bar represents standard deviation).............................................................................................. 43 Figure 3.10 Oil recovery versus ultrasonic pow er for the combined process (error bar represents standard deviation) (experim ental condition: ultrasonic treatm ent duration o f 10 min, sludge/water ratio o f 1:4, without the addition o f rham nolipids and N aC l)...................................................................................................................................47 Figure 3.11 Oil recovery versus ultrasonic treatm ent duration for the com bined process (error bar represents standard deviation) (experimental condition: ultrasonic treatm ent pow er o f 6 6 W, sludge/water ratio o f 1:4, without the addition o f rhamnolipids and N aC l)..........................................................................................................48 Figure 3.12 Oil recovery versus sludge/water ratio for the combined process (error bar represents standard deviation) (experim ental condition: ultrasonic pow er o f 6 6 W , ultrasonic treatment duration o f 1 0 min, w ithout the addition o f rham nolipids and N aC l).......................................................................................................................................... 49 Figure 3.13 Oil recovery versus bio-surfactant concentration for the com bined process (error bar represents standard deviation) (experimental condition: ultrasonic pow er o f 6 6 W, ultrasonic treatm ent duration o f 10 m in, sludge/water ratio o f 1:4, w ithout the addition o f N aC l)...............................................................................................................50 Figure 3.14 Oil recovery versus N aC l concentration for the combined process (error bar represents standard deviation) (experim ental condition: ultrasonic pow er o f 6 6 W, ultrasonic treatment duration o f 10 min, sludge/water ratio o f 1:4, w ithout the addition o f rham nolipids)....................................................................................................... 52 Figure 3.15 GC profiles o f samples from oil recovery treatm ent (A represents oil sample after US + F/T treatment and B represents original oily sludge sam ple).................... 53 Figure 4.1 Oxidation o f oily sludge using ultrasound.............................................................. 67 Figure 4.2 Fenton’s reaction process for degradation o f oily sludge.................................... 6 8 Figure 4.3 Samples after US/Fenton treatm ent..........................................................................69 Figure 4.4 Liquid-liquid extraction...............................................................................................73 Figure 4.5 M echanical shaking for the extraction o f PHCs from solids...............................74 Figure 4.6 Extraction solution cleanup using silica gel co lu m n ............................................ 75 Figure 4.7 Remaining TPH mass in samples after different ultrasonic treatm ent durations (error bar stands for standard deviation; the value at the time o f 0 stands for the initial TPH m a ss).................................................................................................................... 77 Figure 4.8 Remaining TPH mass in samples after Fenton process alone (error bar stands for standard deviation; the value at the dosage o f 0 stands for the initial TPH mass) 80 Figure 4.9 TPH mass remained in samples after three different methods (US alone for 5 minutes; Fenton alone with 15 m l........................................................................................ 85 Figure 4.10 TPH and fraction reduction via three different processes (US alone for 5 minutes; Fenton alone w ith 15 ml H 2 O 2 ; US/Fenton w ith 5 m inutes o f ultrasound and 15 ml o f H 2 0 2) (error bars stand for standard deviation)........................................85 Figure 4.11 Fractional distribution o f petroleum hydrocarbons in sam ples before and after three different processes (US alone for 5 minutes; Fenton alone w ith 15 ml H 2 O 2 ; US/Fenton w ith 5 m inutes o f ultrasound and 15 ml o f H 2 O 2 )........................... 89 Figure 4.12 M ain effect plot o f factors on the TPH degradation through U S/Fenton process........................................................................................................................................95 Figure 4.13 Interaction o f factors on TPH degradation through US/Fenton process: (a) interaction o f sludge content and ratio o f H 2 O 2 to Fe 2 +;(b) interaction o f sludge content and US time; (c) interaction o f ratio o f H 2 O 2 to Fe2+ and US tim e............... 96 Figure 4.14 M ain effect o f factors for F2 degradation through US/Fenton p ro cess 104 Figure 4.15 Interaction o f factors on F2 degradation through US/Fenton process: (a) interaction o f sludge content and ratio o f H 2 O 2 to Fe2+; (b) interaction o f sludge content and US time; (c) interaction o f ratio o f H 2 O 2 to Fe2+ and US tim e 105 Figure 4.16 M ain effect o f factors for F3 degradation through US/Fenton p ro cess 106 Figure 4.17 Interaction o f factors on F3 degradation through US/Fenton process: (a) interaction o f sludge content and ratio o f H 2 O 2 to Fe2+; (b) interaction o f sludge content and US time; (c) interaction o f ratio o f H 2 O 2 to Fe2+ and US tim e 107 Figure 4.18 M ain effect o f factors for F4 degradation through US/Fenton p ro cess 108 Figure 4.19 Interaction o f factors on F4 degradation through US/Fenton process: (a) interaction o f sludge content and ratio o f H 2 O 2 to Fe2+; (b) interaction o f sludge content and US time; (c) interaction o f ratio o f H 2 O 2 to Fe2+and US tim e 109 x LIST OF ABBREVIATIONS ANOVA Analysis o f variance API American Petroleum Institute ASTM American Society for Testing and Materials CCME Canadian Council o f M inisters o f Environment EPA Environmental Protection Agency DCM Dichlorom ethane GC Gas chrom atograph GC-FID Gas chrom atograph-Flam e ionization detector PAHs Polycyclic arom atic hydrocarbons PCBs Polychlorinated biphenyls PHCs Petroleum hydrocarbons TPH Total petroleum hydrocarbons US Ultrasonic irradiation ACKNOWLEDGEMENT This thesis reflects the final part o f my M.Sc. study at the University o f N orthern British Columbia. First o f all, I would like to express my sincere and deep gratitude to m y Co-supervisors, Professor Jianbing Li and Professor Ronald W. Thring. Their w ide knowledge, logical w ay o f thinking, and insightful instructions have been instrum ental for my research and study. W ithout their patience, guidance, and support, I could never have been able to accomplish this research. I also would like to thank Dr. M ichael Rutherford and Dr. Liang Chen for serving as m y supervisory com mittee members. I greatly appreciate their insightful com m ents, guidance and discussions. I would like to express m y appreciation to Dr. Quanji W u at the UNBC Central Equipment Laboratory for analyzing all o f m y experim ental samples. His patient w ork and instruction on improving the accuracy o f chem ical analysis are greatly appreciated. Furthermore, I wish to thank Dr. U m esh Pashotam for his patient and helpful instruction as the supervisor o f m y teaching assistant position at UNBC. I wish to thank Dr. Xinghui Xia from Beijing N orm al U niversity as m y supervisor during m y study in China. H er insightful instruction has been invaluable for m y research. W ithout her kind support and help, I could never have been able to get the opportunity to study abroad. During m y thesis work, I have collaborated w ith m any colleagues in the UNBC soil and groundwater remediation research laboratory. I would like to extend m y w arm thanks to Xinyuan Song, Xiaoqin Yan, Lailin Chen, Zhifeng Liu, Gopal Saha, and Dongqing M ao for their kind sharing their knowledge, skills, and thoughts. I owe m y sincere thanks to m y loved family, m y parents, grandparents, and m y younger brother for their supports and understanding. W ithout their supports and encouragem ent, I could not have come to UNBC to pursue the M .Sc. degree. Special thanks go to m y mother, for her love and support throughout m y life. Chapter 1 General Introduction The petroleum industries can generate a great deal o f oily sludge wastes during crude oil transportation, storage, and further refining processes (A driana and Nei, 2002; M arin et al, 2006; Ram aswam y et al, 2007). Generally, the oily sludge is a complex water-in-oil (W /O) emulsion, typically consisting o f 30-50 wt% o f oil, 30-50 wt% o f water, and 10-12 wt% o f solids (Reynolds and Heuer, 1993; Ram aswam y et al, 2007). It contains a significant am ount o f petroleum hydrocarbons (PHCs), and m any o f the PHCs are o f high m olecular-weight alkanes, (Overcash and Pal, 1979). Due to the com plex com position and the high content o f PHCs, oily sludge has been known as a hazardous waste. Im proper disposals could lead to environmental pollution and pose serious risk to the environm ent and the hum an health (Elektorowicz et al, 2005). Various treatm ent methods, such as pyrolysis, incineration, land-filling, and land-farm ing (Shie et al, 2000; Che et al, 2008), have been explored to deal with the oily sludge waste. However, the treatment o f oily sludge using these conventional m ethods is relatively time-consuming, ineffective, and expensive (Shie et al, 2000; Buyukkamaci and Kucukselek, 2007). It is thus o f importance to find im proved methods. Although the biorem ediation technology has been w idely used for the reduction o f organic pollutants in soil and water, its effective application to the oily sludge treatm ent still needs more investigation. The biodegradation o f PHCs in oily sludge could be im proved by using special bacterial strains with high biodegradation capacity and by enhancing the m icroenvironm ent for bacterial activity (Banat et al, 2000; Cameotra and Singh, 2008). Effective bacterial strains could be screened and isolated from soils contaminated w ith PHCs for a long period. T he isolated strains can then be reintroduced into the contam inated soil medium. In term s o f im proving the bacterial activity, adding sufficient nutrients into soils could be the foremost strategy to 2 stimulate bacterial growth (Cam eotra and Sigh, 2008). A nother im provement m ethod is to add bio-surfactants to soil so that the solubility o f hydrophobic organic pollutants can be enhanced and thus more contaminant m olecules present in soil pore w ater can be approachable for bacteria living in soil. Consequently, the enhanced bioavailability can lead to the improved biorem ediation performance. Another m ethod to im prove the perform ance o f conventional treatm ent processes is through oil recovery before disposal (Elektorow icz, et al, 2006; Zubaidy and A bouelnasr 2010). The oily sludge waste could be regarded as a valuable resource, and the oil recovery can significantly reduce the concentration o f PHCs and the volume o f sludge for conventional treatment. In fact, the treatment o f sludge containing over 10% o f oil could result in economic benefit from oil recovery (Ramaswamy, et al, 2007). M any studies have reported the recycling o f oil from different oil wastes using various approaches, including solvent extraction, the use o f ultrasound and microwave, and the m eans o f freeze/thaw. However, a variety o f problem s are still associated with these methods, such as high cost and low recovery efficiency. The advanced oxidation processes (AOPs) applied in wastewater treatm ent (Adewuyi, 2005; Pang et al, 2011) m ight also be plausible approaches for the treatment o f oily sludge. They are chem ical oxidation methods w ith the capability o f treating m any types o f organic and inorganic contaminants (Adewuyi, 2005). In general, the prominent principle o f A O Ps is the generation o f free hydroxyl radicals (*OH) w hich are extremely reactive species. These free radicals serve as the powerful oxidants for the destruction o f contam inant m olecules (Gogate et al, 2004). The use o f ultrasonic and ultraviolet irradiation as w ell as the application o f oxidation agents (e.g., hydrogen peroxide, Fenton/Fenton-like reactions, ozone) 3 have been investigated as viable AOPs in many studies (G ogate and Pandit, 2003; Sivasankar and Moholkar, 2009). However, the oxidation efficiency is usually limited for single AOP. In order to im prove the efficiency o f AOPs, m any researchers have tried to develop hybrid AOPs by com bining oxidation agents with ultrasonic or ultraviolet treatment. The combination o f m ethods was proved to yield synergistic effect for the destruction o f hydrophobic organic m olecules (Chand et al, 2009). However, few studies have been reported to treat the oily sludge wastes using hybrid AOPs. The objective o f this study was to find effective m ethods for the treatment o f oily sludge. The performances o f three different bioremediation, oil recovery, and enhanced m ethods w ere examined, advanced oxidation. In terms o f the including enhanced bioremediation method, a specific bacterial strain was isolated from soils contam inated w ith PHCs and was then applied to the soil rem ediation process, while the nutrients and bio-surfactants were added to the soil to im prove bacterial activity. The Taguchi design m ethod was applied to arrange different experiments on investigating the effects o f various factors. In term s o f oil recovery from oily sludge, the efficient o f combing ultrasound and freeze/thaw was investigated on recovering oil from oily sludge. The impacts o f several factors on the oil recovery treatm ent perform ance through the combined m ethod were evaluated. These factors include ultrasonic power, ultrasonic treatment duration, addition o f bio-surfactant, addition o f salt, and the ratio o f sludge to water. In terms o f the investigation o f advanced oxidation processes, the PHCs in oily sludge w ere oxidized via ultrasound, Fenton process, and the combination o f ultrasound and Fenton process. A variety o f factors were evaluated on the oxidation efficiency, and they include ultrasonic power, treatm ent duration, initial sludge concentration, and the ratio o f hydrogen peroxide and ion Fe (II). 4 Their individual and com bined effects on the reduction o f PHCs in oily sludge were examined. The experimental results o f these three different approaches w ould provide a sound basis for developing environm entally friendly and econom ically com petitive m ethods for oily sludge treatment. 5 Chapter 2 Remediation of refinery oily sludge using isolated strain and biosurfactant 1 1 This work was published as J. Zhang, J. B. Li, L. L. Chen, R. Thring, Remediation o f refinery oily sludge using isolated strain and biosurfactant, Proceedings o f 2011 International Symposium on Water Resources and Environmental Protection (ISWREP2011), Xi’an, China, May 20-22, 2011, pp. 1649-1654, IEEE Press. 6 Abstract A series o f laboratory experiments were conducted to investigate the biorem ediation o f oil refinery sludge using an isolated bacterial strain and a bio-surfactant. The strain called uncultured Luteibacter sp. isolated from soils from a previous sludge landfarm ing site w as used for bioaugmentation. A commercial biosurfactant o f rhamnolipid (JBR 425) was used for improving the bioavailability o f petroleum hydrocarbons (PHCs) to the bacteria strain. The impacts o f initial total petroleum hydrocarbon (TPH) concentration, nutrient supply, and biosurfactant concentration on the remediation process w ere examined. The degradation efficiencies o f n-alkenes between C10-C16, C16-C34 and C34-C50 were also compared. The results can provide sound scientific basis for developing cost-effective rem ediation m ethods to treat oil refinery sludge. K ey words: Oily sludge, bioremediation, biosurfactant, total petroleum hydrocarbons 7 1. Introduction The petroleum industry can generate a large am ount o f oily sludge during production, transportation, and refining processes (Kuriakose and M anjooran, 2001; M arin et al, 2006). Due to the high content o f petroleum hydrocarbons (PH Cs) and complicated com position, oily sludge has been known as a hazardous and recalcitrant w aste product and m ight cause severe environmental pollution if directly disposed into the environment (O vercash and Pal, 1979; Bhattacharyya and Shekdar, 2003). During the past years, various biological treatm ent approaches have been investigated to convert the toxic recalcitrant compounds into non-toxic forms by microbial consortium (Overcash and Pal, 1979). Biological treatm ent techniques involve two categories, including biostim ulation and bioaugmentation. The bioaugm entation approach introduces m icroorganisms identified/screened from natural environm ent and enriched in laboratory condition into contam inated soil (C am eotra et al, 2008). M any studies have isolated different strains capable o f degrading PHCs from oil contam inated sites. For example, Ekp et al (2008) obtained microbes from the crude oil sludge environm ent and found that the isolated P. aeruginosa could degrade 97.2% o f the PH C s added into the medium. Cameotra et al (2008) tested the capability o f one consortium o f tw o strains (Pseudomonas aeruginosa and rhodococcus) to biodegrade hydrocarbons in sludge and they reported that 90% o f the contaminants were depleted in liquid culture after 6 weeks. O n the other hand, biostimulation focuses on the addition o f specific nutrients or other additives as stimuli to contaminated medium in order to enhance the effectiveness o f degrading pollutants by microorganisms. For example, W alworth et al (2007) focused on optim izing nutrient amendments in a sub-antarctic soil and they found that the rate o f biodegradation w as maximized when the nitrogen (in the form o f N H 4 C1) content was 125 mg/kg. 8 M achin-Ramirez et al (2008) used a fertilizer solution to investigate the effect o f nutrient addition on slurry-phase biodegradation o f weathered oily sludge waste and 24% o f PHCs biodegradation w as attributed to the addition o f fertilizer in the slurry. O ther experim ents reported a wide range o f C:N and C:P ratio from 9-200 and 60-800 to enhance PHCs biodegradation (Huesemann, 1994). In general, bioaugm entation and biostim ulation are used to improve the microenvironment o f biodegradation. However, the efficiency of biodegradation is also usually restricted by the low bioavailability of hydrophobic organic compounds in soil. In order to enhance the solubility and desorption o f PH C s entrapped in soil, chemical and biological surfactants are generally introduced for stim ulation (Calvo et al, 2009). Among them, bio-surfactants produced by microorganisms have been receiving increasing interests in biorem ediation studies due to their low toxicity and wide environmental com patibility (Ron and Rosenberg, 2001). Several bio-surfactants have been reported in the area o f biorem ediation o f organic pollution in water and soil, including rhamnolipids, sophorolipids, and surfactin (Banat et al, 2000). Although m any experiments have been previously carried out to investigate the biorem ediation o f gasoline and diesel contaminated soil, only a few works focused on oily sludge where the long carbon-chain petroleum hydrocarbons dominate (Rahm an et al, 2003;G hazali et al, 2004), especially for n-alkanes from nC20-nC50. In this study, one isolated strain from a PHCs land-farm ing site was applied to examine the biodegradation process o f oily sludge. Taguchi experim ental design (Castorena-Cortes et al, 2009) was applied to investigate the effect o f initial petroleum hydrocarbon concentration, nutrient addition, and biosurfactant (JBR 425) addition on the biodegradation o f long-chain hydrocarbons and to determ ine the optim al condition for biorem ediation o f oily sludge. 9 2. Materials and Methods 2.1. Oily sludge and soil sample The oily sludge collected from an oil refinery plant in w estern Canada w as stored in a capped stainless-steel bucket at 4°C. Uncontam inated soil was collected from a sam pling site w ithin a forest located near the road (close to Oskpika and Tyner Blvd) in Prince George, BC, Canada. The sampling depth was about 60 cm to 80 cm. The soil was collected w ith stainless steel shovel and was air dried at room tem perature for 3-4 days. The soil with low concentration o f petroleum hydrocarbons w as passed through a # 2 0 sieve to rem ove coarse particles and was then stored in fridge at 4°C. Properties o f uncontam inated soil and oily sludge w ere listed in Tables 2.1 and 2.2. Table 2.1 Summary of clean soil properties Properties Percentage (% by mass) Elements Concentration (mg/kg) 1.81 0.04 0.097 0.18 0.37 Ca Mg Fe Zn Cu 5900 6200 28507 161.6 12.9 Total carbon Inorganic carbon Total nitrogen Total phosphorus Total potassium Soil texture Percentage o f sand Percentage o f silt Percentage o f clay 10.9% 53.5% 35.6% Table 2.2 Summary of oily sludge properties Concentration (m g/g) Elements Concentration TPH 40.99 Total carbon 86.10% (by mass) F2 fraction 2.39 Al 27.50 mg/kg F3 fraction 13.55 Fe 448 mg/kg F4 fraction 67.09 Ni 11.6 mg/kg Percentage o f F4 (by mass) 61.10% Cr 0.6 mg/kg Percentage o f F3 (by mass) 33.05% Cd 0.043 mg/kg Properties 10 2.2. Chem icals Decane (nCIO, > 99% purity, Sigm a-Aldrich), hexadecane (nC16; > 99% purity, Sigma-Aldrich), tetratriacontane (nC34; > 99% Sigma-Aldrich), Pentacontane(nC50, > 99% purity, Sigma-Aldrich) were used as standard substances for GC analysis; dichlorom ethane (DM C) (> 99% purity, HPLC grade, Sigm a-Aldrich) and cyclohexane (> 99% purity, HPLC grade, Sigma-Aldrich) were used to extract soil samples. N H 4 N O 3 and KH 2 PO 4 purchased from Sigma were used as nutrients for biorem ediation experiments. The rham nolipid (JBR425) bio-surfactant was purchased from Jeneil Biosurfactant Company (Saukville, W I, USA). It is a m ixture o f two rhamnolipids C 2 6 H 4 8 0 9 (RLL) and C 3 2 H 5 8 0 i 3 (RRLL). 2.3. Bacteria isolation and identification The mineral salt medium (M SM ) and trace elem ent solution were used for bacterial isolation (Noordman and Janssen, 2002). M SM stock solution was prepared as following: 26.5 g /L Na 2 H P 0 4 .12H 2 0 , 7 g/L KH 2 P 0 4, 1 g/L M g S 0 4.7 H20 and 2.5 g/L (NH 4 ) 2 S 0 4. The trace element solution was prepared as following: 2.65 g/L C aCl2, 1 g/L FeS 0 4 .7 H 2 0 , 0.05 g/L Z n S 0 4 .7H 2 0 , 0.05 g/L H3BO3, 0.05 g/L CoC12 .6H 2 0 , 0.02 g/L M n S 0 4 .5H 2 0 , 0.015 g/L Na 2 M o 0 4 .2 H 2 0 , and 0.01 g/L N iCl 2 .6H 2 0 . The pH value o f M SM was adjusted to 7.3 ± 0 . 1 with 1M NaOH or 1M HC1 according to the background pH o f samples. The M SM was sterilized by autoclaving for 30 min at 121 °C. Stock solutions o f oily sludge collected from an oil refinery in w estern Canada were prepared in cyclohexane, and were used for providing different carbon sources for microbial growth. The procedure o f bacteria isolation was dem onstrated in Fig 2.1 based on previous w ork (Chen et al, 2011). To enrich the PHCs degrading bacterium , 5 g o f contam inated soil 11 samples from one land-farming site (Fig 2.2) were added into a 250-ml conical flask w ith 50 ml o f sterilized M SM and 5 mL o f oily sludge solution after volatilization o f cyclohexane in the fume hood. The flasks were shaken in a mechanical shaker (150 rpm ) at room temperature (20 ± 2°C) for two w eeks as the first enrichm ent step. A fter that, the second enrichment w ere conducted by transferring 5 mL o f the culture after 1st enrichm ent process to another flask containing 50 mL o f sterilized M SM w ith the same amount o f stock solution as the first enrichment (Fig 2.3). The PHCs degrading microbes in the 1st and 2nd enriched culture were determined using M SM agar plates sprayed with oily sludge as the sole carbon and energy source. After incubation on the M SM agar plates for 3-5 days at 20 ± 2°C, any visible colony grow th was recorded as positive growth. Individual colony w as identified by its color and m orphology and further purified by streaking on nutrient agar plate. O ne isolated bacterium was used in this study, and it was identified as the uncultured Luteibacter sp., which was used for biorem ediation experim ent (Fig 2.4). 1st Enrichment Contaminated soil sampling Carbon source Oily sludge 2nd Enrichment Repeated streaking for purification Streaking o f enriched culture Figure 2.1 Isolation procedure for PHCs degrading microorganisms 12 Fig u re 2.2 Soil contaminated with PHCs in a land-farming site Figure 2.3 Enrichment o f bacteria 13 F ig ure 2.4 Isolated strain: Luteibacler sp 2.4. Biorem ediation experim ent The uncontaminated soil was autoclaved at 121 °C for 30 minutes and was then dried in an oven for 12 hours at 105°C. A certain am ount o f refinery oily sludge w as placed into a j a r with 10-20 ml o f cyclohexane. The dried sterile soil was added into the ja r w ith continuous stirring to reach a given TPH content (mg/kg). The spiked soil sample in the ja r was then placed in the fume hood for 2-3 days in order to com pletely vaporize cyclohexane from the sample. Taguchi L-9 experimental design w as applied to investigate the optim al condition o f bioremediation. The initial sludge and nutrient content as w ell as the rham nolipids (RL) concentration were studied with three levels (Table 2.3). The 250-ml flasks (Fig 2.5) w ere set up w ith the following treatments: (1) 25 g o f spiked soil with different TPH contents; (2) inocula o f Luteibacter sp. (OD 0.8), and (3) different nutrient solution to adjust C:N:P ratio. The flasks were sealed using rubber stoppers with two holes and aerated through a peristaltic 14 pum p for 40 days (Fig 2.7). Total petroleum hydrocarbon (TPH) concentration and PHC fractions (i.e. F2 fraction representing C10-C16, F3 fraction representing C16-C34, and F4 fraction representing C34-C50) w ere determ ined at the beginning and the end o f an experiment. The experiments were conducted in triplicated. ANOVA was applied to analyze the experimental results. 2.5. Analysis o f total petroleum hydrocarbon (TPH) 2 g o f soil sample was added in a 40-m l vial w ith 10 ml o f dichlorom ethane (DCM ) and then the vial was placed on an orbital shaker for mechanical extraction at 150 rpm for 30 minutes. The extraction was conducted three times and about 30 ml o f extraction solution was collected into a vial. Table 2.3 Taguchi experimental design Test N o. TPH content (mg/kg) Nutrients (C:N:P) Rhamnolipid (mg/kg) 1 2297 100:10:1 40 2 2297 100:20:5 400 3 2297 100:50:10 1000 4 4674 100:10:1 400 5 4674 100:20:5 1000 6 4674 100:50:10 40 7 9690 100:10:1 1000 8 9690 100:20:5 40 9 9690 100:50:10 400 Then the extraction solution w as transferred into a glass colum n for cleanup. The colum n was packed with silica gel and anhydrous sodium sulphate and was rinsed w ith 2 0 ml o f solvent (1:1 cyclohexane/DCM ) before use. After the extraction solution passed through, another 20 ml o f solvent (1:1 cyclohexane/DCM ) was poured to elute the column. A round 15 F ig u re 2.5 A fla sk w ith soil Figure 2.6 Setup o f the bioremediation experiments flask was placed under the colum n to collect the extraction after cleanup and then the extraction solution was evaporated by a rotary evaporator to reduce the volum e to 1 ml. After evaporation, the residue in round flask was transferred into 2 -ml sam ple vial with cyclohexane for GC analysis.Gas C hrom atograph (Varian CP-3800) w ith flam e ionization detector (GC-FID) was used for TPH analysis. The param eters for GC analysis were: 30-m metal colum n (Restek MXT-1) with 0.53 mm ID; injector temperature at 320 °C; detector temperature at 350°C; nitrogen gas as carrier gas w ith a constant flow at 7.5 ml/m in; oven temperature starting at 40°C for 4 min, ram ping to 140 °C at 10 °C /min, reaching to 340°C at 20 °C /min, and then being held at 340 °C for 11 minutes. External standard m ethod w as used to calculate TPH concentration in sample solution. The decane (CIO), hexcane ( C l 6 ), tetratriacontane (C34), and pentacontane (C50) w ere used as standard com pounds to determine the TPH fractions. Fraction 1(F1) was defined as hydrocarbons before CIO, fraction 2 (F2) was defined as hydrocarbons from CIO to C 16, fraction 3 (F3) was the group o f hydrocarbons from C16 to C34, and fraction 4 (F4) was the group o f hydrocarbons from C34 to C50. 3. Results and discussion 3.1. TPH biodegradation The TPH reduction rate and residual concentration o f TPH in soil after 40-day biorem ediation w ere shown in Fig 2.7 and Fig 2.8, respectively. The residual concentration o f F2 fraction was less than 150 mg/kg for all o f the treatment samples, and was not shown in Fig. 2.8. The highest (i.e. about 50.8%) and low est (i.e. about 5.8%) TPH reduction rate was 17 observed with treatments L 8 and L I, respectively. In terms o f TPH fractions, however, treatment LI was associated with much higher reduction rate o f F3 (47.9%) and relatively no reduction o f F4. For treatments L7, L 8 and L9 w hich had the highest TPH contents (i.e. highest initial TPH concentration about 9690 mg/kg), it was observed from Fig 2.8 that the residual TPH concentration in soil sam ples after biorem ediation was less than 6100 m g/kg w ith F4 fraction being less than 4600 mg/kg. A s for treatments L7 and L 8 , the reduction rates for F2, F3, and F4 fractions were all above 30%, and the TPH reduction rate were 40.1% and 50.8%, respectively. As a result, the experim ental results indicated that the isolated bacterium (uncultured Luteibacter sp.) was an effective bacterial strain for the biodegradation o f long-chain petroleum hydrocarbons, and the com bination o f bioaugm entation and biostim ulation was illustrated as an effective m ethod for the remediation o f oily sludge contamination. In fact, the strain o f Luteibacter sp. was investigated for degrading PAHs w ith other soil bacterial strains (Cea et al, 2010). 55% F2 F3 F4 TPH 45% :Stt g 35% c 25% * ;I 5% L2 L3 L4 L5 L6 L7 L8 L9 xt T reatm en ts F igure 2.7 TPH reduction rate for different treatments 18 8000 7000 '55 22223 F3 F4 |p 6 0 0 0 cz k^ I 5000 tph c= § o o 4000 Jj 3000 3 -§ 2000 >■, 1000 0 L1 L2 I L3 L4 L5 T re a tm e n ts L6 L7 L8 L9 Figure 2.8 TPH concentration remained in soil after bioremediation 3.2. Analysis o f optimal bioremediation conditions Signal to noise ratio (S/N ratio) analysis w as applied to determine the perform ance o f three factors on the biodegradation o f TPH in oily sludge. Fig. 2.9 presents the results, and it indicates that all the three factors showed significant im pacts on TPH reduction. The optim al biorem ediation conditions were: initial TPH content at 9690 mg/kg (i.e. level 3), nutrient addition at the ratio o f 100:50:10 (i.e. level 3) for C:N:P, and rhamnolipid addition at 400 m g/kg (i.e. level 2). Table 2.4 lists the ANOVA results, and it indicates that under the 0.05 significance level, the initial TPH content was the m ost significant factor to affect the bioremediation o f PHCs in oily sludge, followed by rham nolipid addition concentration (P < 0.05). 19 Table 2.4 ANOVA for bioremediation o f TPH F a cto r S q u a re M S* df F valu e P v a lu e Initial TPH Nutrient addition 0.0885 0.0443 2 0.0002 0.0165 0.0082 2 45.7925 4.2553 0.0707 Rhamnolipid addition 0.0260 0.0130 2 6.7250 0.0294 (MS: means o f squares) Fig. 2.10 presents the S/N ratio analysis results in term s o f F3 reduction after bioremediation, and it indicates that the optimal condition for F3 degradation was: initial TPH content at 2297 mg/kg (i.e. level 1), nutrient addition at the ratio o f 100:25:5 (i.e. level 2), and rhamnolipid content at 400 m g/kg (i.e. level 2). Table 2.5 lists the ANOVA results for F3 reduction, and it indicates that under the significance level o f 0.05, there was no significant difference among all the three factors. In term s o f F4 fraction reduction, Fig. 2.11 and Table 2.6 present the S/N ratio analysis and ANOVA results, respectively, and the optimal condition for F4 degradation was: initial TPH content at 9690 m g/kg (i.e. level 3), nutrient addition at the ratio o f 100:50:10 (i.e. level 3), and rhamnolipid content at 1000 mg/kg (i.e. level 3). However, the S/N ratio at 400 mg/kg o f rham nolipid was close to that at 1000 mg/kg, indicating that further increase o f rham nolipid concentration above 400 m g/kg did not effectively contribute to the increase o f F4 reduction. ANOVA results in Table 2.6 indicate that the initial TPH content was the m ost significant factor (P<0.05), followed by rhamnolipid concentration and nutrient addition, in term s o f the degradation o f F4 fraction in oily sludge. 20 -32 -2 8 o go -24 - 20, 1 3 2 Initial TPH c o n te n t 2 3 Nutrient addition 1 1 2 R ham addition 3 Figure 2.9 Effect o f independent factors on TPH reduction S/N ratio -3 2 -28 -24 - 20 , 3 1 2 Initial TPH con ten t 2 3 Nutrient addition 1 2 3 Rham addition 1 Figure 2.10 Effect o f independent factors on F3 fraction reduction 21 Table 2.5 ANOVA for F3 fraction degradation F a cto r S q u a re M S* df F va lu e Initial TPH 0.5236 0.2618 2 1.1795 0.370 Nutrient 0.4736 0.2368 2 1.0667 0.407 0.4792 0.2396 2 1.0794 0.398 addition Rham addition P v a lu e (MS: means o f squares) Table 2.6 ANOVA for F4 fraction degradation F a c to r S q u a re M S* df F v a lu e P v a lu e Initial TPH 0.1512 0.0756 2 77.6373 0.0001 Nutrient addition 0.0251 0.0125 2 12.8806 0.0067 Rham addition 0.0301 0.0151 2 15.4707 0.0043 (MS: means o f squares) The initial TPH concentration in soil was found as a significant factor influencing the biorem ediation efficiency by Luteibacter sp. The degradation o f F3 fraction at low initial TPH content (i.e. level 1) in soil was better than those at higher initial TPH contents (i.e. levels 2 and 3) as shown in Fig. 2.10. However, the best biodegradation o f F4 fraction was achieved for the treatments with the highest initial TPH content as shown in Fig. 2.11. It was also observed from Fig. 2.7 that the reduction o f F3 fraction was higher than that o f F4 among all the treatments at low initial TPH content (i.e. treatm ents L1-L6), but w as low er for treatments at high initial TPH content (i.e. treatments L7-L9). Fig. 2.12 presents the PHCs fraction distribution before and after biorem ediation, and it was observed that the proportion o f F3 in TPH was reduced from 16.0% to 8 .8 %, indicating that the isolated bacterial strain (Luteibacter sp.) preferred to use hydrocarbons in the F3 group as its carbon and energy sources for grow th rather than those heavier com ponents such as F4 (C34-C50). Also, the petroleum hydrocarbon compounds in the F4 group are m ore 22 hydrophobic and m ore strongly attached to soil particles, and hardly released into the w ater phase. In terms o f PHCs fraction distribution at higher initial TPH content, the F4 fraction in TPH was decreased while the F3 fraction w as increased. For example, the F4 fraction accounted for 80.2% o f TPH before biorem ediation, but it was decreased to 75.7% for treatment L 8 . O n the contrary, F3 fraction in TPH w as increased from 18.1% to 22.2%. Previous studies reported that long-chain alkane could be monooxidized or cooxidized during microbial growth (Kester and Foster, 1963). As a result, some intermediate hydrocarbons were generated during the degradation o f com pounds in the F4 group by bacterial strain, and these intermediate products m ight contribute to the increase o f F3 fraction in TPH. r 40 -35 -30 o CO -25 CO -20 -15, -10 3 2 1 Initial TPH co n ten t 1 2 3 Nutrient addition 1 2 3 Rham addition Figure 2.1J Effect o f independent factors on F4 fraction reduction Previous studies (Franzetti and Caredda, 2009) reported that the biodegradation performance o f bacteria could be enhanced when the ratio o f carbon, nitrogen and phosphorus element (C:N:P) was appropriately adjusted, but could be inhibited w hen 23 excessive am ount o f nutrients was added. In this study, such negative effect o f nutrient addition was not observed on the degradation o f TPH and F4 fraction, but it was observed on F3 fraction reduction. The F4 group w as a dom inant long-chain hydrocarbon fraction in oily sludge used in this study, and therefore its degradation by uncultured Luteibacter sp. m ight need a relatively higher concentration o f nutrient addition to the soil m edium as com pared w ith that for shorter carbon-chain compound. 90% 80% o J B e-F 2 Af-F2 B e-F 3 ixxxa Af-F3 B e-F 4 HHillflHI A f-F4 70% 60% CO 50% TD c 2 40% 99% pure), Nonadecane (nC19; >99% pure), Tetratracon (nC34; >99% pure), and pentacontane (C50) were purchased from Sigma and used as standard compounds for PHCs analysis. Rham nolipid (JBR-425) purchased from Jeneil Bio-surfactant Co. (LLC, USA) was used as the bio-surfactant. It is a 25% aqueous solution m ixed w ith two rhamnolipids: C 2 6 H 4 8 O 9 and C 3 2 H 5 8 O 1 3 . FeS 0 4 -7 H 2 0 . The dichloromethane (DCM) (>99% , HPLC) and cyclohexane (>99%, HPLC) w ere used as solvents for sample extraction. The toluene (>99%, HPLC) w as used as solvent to dilute extraction samples for GC analysis. Silica gel (purchased from Sigma) activated at 105 °C for 12 h was used to clean up extraction solution, and anhydrous sodium sulfate dried at 400 °C for 1 2 h was used to absorb water in the extraction solution. 2.3. Experim ents on oil recovery using three different processes 32 The experiments were conducted to com pare the efficiencies o f oil recovery from refinery sludge by using three different treatm ent processes, including ultrasonic treatm ent alone (US), freeze/thaw alone (F/T), as well as the com bined ultrasonic and freeze/that treatment (US + F/T). In terms o f ultrasonic treatm ent alone, 10 g o f oily sludge w as placed inside a 120 ml beaker which contained 40 ml o f deionized water, and w as then treated by placing the 0.5 in. diameter titanium sonic probe into the center o f the sample. The sonic probe was operated by a 20 kHz M isonix Sonicator 3000 generator. The treatm ent w as conducted at a working pow er o f 66 W for 10 min. Figs. 3.2 and 3.3 illustrate the ultrasonic treatment system. Converter Ultrasonic Probe o o o o o o o o o o o o Beaker Water Oil sludge Ultrasonic G enerator Figure 3.2 Ultrasonic treatment system After ultrasonic treatment, the treated sample was observed to have a reduced viscosity than the original sludge and was transferred into a 50-ml centrifuge tube. The sample w as then centrifuged for 15 min at 5000 rpm. The oil and aqueous phases after centrifugation were then separated using a separation funnel. The mass o f oil layer separated was then measured and considered as the oil recovery. The total petroleum hydrocarbon (TPH) 33 concentrations in the separated oil layer and aqueous phase were also analyzed. Figure 3.3 Ultrasonic treatment for oil recovery In terms o f freeze/thaw treatm ent alone, 10 g o f oily sludge w as put into a 50-m l centrifuge tube with 40 ml o f DI w ater and was frozen within a freezer under - 2 0 °C for 12 h. The frozen sample w as then thawed at an am bient tem perature o f 24 °C. A fter thawing, the sample in the tube was centrifuged for 15 m in at 5000 rpm. The oil and aqueous phases after centrifugation were separated using a separation funnel, and the mass o f oil layer separated from the sample as well as the TPH concentrations in oil layer and aqueous phase w ere then measured. In term s o f the combined ultrasonic and freeze/thaw treatment process, 10 g o f oily sludge was put into a 120 ml beaker w ith 40 ml o f deionized water, and was then treated by 34 ultrasound at 66 W for 10 min. After ultrasonic treatment, the sample w as transferred to a 50-ml centrifuge tube and was centrifuged for 15 min at 5000 rpm. After centrifugation, the sample in the tube w ent through the freeze/thaw process, and the rem aining treatm ent procedures were the same as those for the freeze/thaw treatm ent alone. 2.4. Experim ents on factors affecting the com bined treatm ent process The individual impacts o f five different factors on the oil recovery rate from the combined ultrasonic and freeze/thaw treatm ent process were further exam ined to better understand this process and provide useful inform ation for its effective operation. Five factors w ere selected, including ultrasonic power, ultrasonic treatment duration, sludge/water mixing ratio, bio-surfactant concentration, and salt (NaCl) concentration. Table 3.2 summarizes the corresponding levels o f the experim ental factors, while the interaction effects o f different factors were not investigated in this study. In terms o f ultrasonic power, experiments were conducted at four levels from 21 to 66 W w ith an ultrasonic treatm ent duration o f 10 min (i.e. the m axim um treatm ent duration level in this study) and a sludge/water ratio o f 1:4 (i.e. the m edium m ixing ratio level selected in this study), w ithout the addition o f bio-surfactant and salt. In term s o f ultrasonic treatment duration, four levels (0.5, 1, 5 and 10 min) were examined at the ultrasonic power o f 66 W (i.e. the m axim um pow er level selected in this study) and sludge/water ratio o f 1:4 w ithout the addition o f bio-surfactant and salt. In terms o f other factors, four different sludge/water m ixing ratios from 1:8 to 1:1 as well as five different levels o f bio-surfactant concentration (from 0 to 700 mg/L) and salt concentration (from 0 to 5.0% by m ass) were investigated at the ultrasonic power o f 66 W and 10 min o f treatment time. For the exam ination o f sludge/w ater ratio, no rhamnolipids and salt were added to the sludge slurry system; for the exam ination o f bio-surfactant concentration, a sludge/water ratio o f 1:4 was maintained w ithout the addition o f salt, and for the exam ination o f salt concentration, a sludge/water ratio o f 1:4 w as applied without the addition o f rhamnolipids. The experim ental procedures for each factor’s exam ination were the same as described before. Table 3.2 Influencing factors and their corresponding levels Influencing factors Level descriptions Ultrasonic power (W) 2 1 ,3 3 ,4 8 , 66 Ultrasonic duration (min) 0.5, 1, 5, 10 Sludge/water ratio 1:8, 1:4, 1:2, 1:1 Rhamnolipids concentration (mg/L) 0, 40, 100, 400, 700 NaCl concentration (% by mass) 0, 0.3, 1.0, 3.0, 5.0 2.5. Sample extraction and analysis 2.5.1. TPH concentration in the recovered oil layer 1 g o f the recovered oil layer sample was dissolved with 20 ml o f solvent (cyclohexane) in a 40-ml vial and then the vial was placed on an orbital shaker for mechanical extraction at 150 rpm for 1 h. After shaking, the extraction solution was transferred into a glass colum n for cleanup to remove moisture, particulate, and unw anted polar organic com pounds (CCM E, 2001). The column was packed with silica gel and anhydrous sodium sulfate and rinsed with 20 ml o f solvent (1:1 cyclohexane/DCM ) before use. After the extraction solution passed through, another 20 ml o f solvent (1:1 cyclohexane/DCM ) was poured to elute the column. A round flask with mass o f Mo (mg) was put under the column to collect the extraction solution after cleanup, and then the extraction solution in the flask was evaporated using a rotary 36 evaporator (Fig 3.4 ) to remove the solvent contained in the extraction solution. Figure 3.4 Evaporation o f extraction solution After evaporation, the round flask (Fig. 3.5) with residue w as put in the fume hood for 30 min at room temperature and only petroleum hydrocarbons (PHCs) w ere left in the flask (CCM E, 2001). The mass o f the flask containing PHCs was then m easured as M T (mg). Consequently, the concentration o f TPH in the recovered oil can be calculated by Eq. (1). Similarly, TPH concentration in the original sludge can also be obtained. q _ M T~ M 0 M (1) where M is the mass o f oil layer sample or original sludge sample used for extraction analysis (g), Mo is the mass o f a round flask and C is the TPH concentration in oil layer or original sludge (mg/g). 37 Figure 3.5 Round flask with residual after evaporation The oil recovery was defined as the ratio o f the mass o f PHCs in the recovered oil to the mass o f PHCs in the original sludge sample, and can be obtained using Eq. (2): C R = — x M ^-------oJl-layer_ y j 0()o/o (2 ) ^sludge * ^sludge where R is oil recovery (%), C01i-iayer and C s]udge are TPH concentrations (mg/g) in the recovered oil layer and original sludge, respectively, Moii-iayer (g) is the total mass o f recovered oil layer from separation, and M siudge (g) is the mass o f oily sludge used for each experimental treatment. 2.5.2. TPH concentration in separated aqueous phase The separated aqueous phase (Fig 3.6) from the sample after oil recovery treatm ent consists o f petroleum hydrocarbons and could be considered as wastewater, and the TPH 38 concentration in such aqueous phase should also be analyzed. This was com pleted using about 40 ml o f aqueous phase w ater through liquid-liquid extraction w ith 15 ml o f cyclohexane for three times (SW -846 EPA, 1993). ■H R F ig u re 3.6 Aqueous sample after oil recovery A bout 45 ml o f extraction solution was collected and then cleaned up through a glass column packed w ith silica gel and anhydrous sodium sulfate as described above. The remaining procedures were the same w ith that for measuring TPH in oil layer. A s a result, TPH concentration in the aqueous phase C water (mg/L) w as obtained using Eq. (3): r w a t e=r Mt~M q (3) w here V is the volum e o f aqueous phase used for sample extraction (L). 2.5.3. PHCs analysis After evaporation using a rotary evaporator, petroleum hydrocarbons in the round flask w ere transferred into a 15-ml sample vial by using cyclohexane, and 2 ml o f solution in the 39 vial was then sent for the analysis o f PHCs using a Varian CP-3800 Gas C hrom atograph with flame ionization (GC-FID) (Fig 3.7). F ig u re 3.7 Samples for GC analysis External standard m ethod was used for identification. T he decane (CIO), hexadecane ( C l6), tetratriacontane (C34), and pentacontane (C50) were used as external standards to determ ine the concentration o f PHCs and TPHs fractions (CCM E, 2001), w here fractions F I, F2, F3 and F4 were defined as the group o f hydrocarbons from C6 to CIO, CIO to C l 6, C16 to C34, and C34 to C50, respectively. The GC analysis conditions were: ZB-capillary colum n (Phenomenex Torrance, CA) with 30 m x 0.25 m m ED (inner diameter) and 0.25-pm film thickness; inject volume o f 1 pL; injector and detector (FID) tem peratures at 320 °C; carrier gas (helium) at a constant flow rate o f 1.5 m L/min during analysis. The splitless injection m ode was performed on the 1079 PTV injector and after 0.7 m in the split m ode was activated at split ratio o f 10:1. The capillary colum n temperature program was initially held at 50°C for 1 min, then ramped at 15.0 °C/m in to 110 °C and further increased at 10.0 °C/m in to 300°C and then held for 11 min. The total running time for a sample w as 45 minute. 40 3. Results and discussion 3.1. Com parison o f methods The experimental results o f using different oil recovery m ethods are sum m arized in Table 3.3, and it was indicated that F/T m ethod alone worked m ore effectively in terms o f the oil recovery rate (with an oil recovery o f 65.7%) than the tw o other m ethod (Fig. 3.8). It is recognized that the water/oil em ulsion is stabilized by the existence o f an em ulsifying film consisting o f surfactant m olecules which could prevent w ater droplets from contacting each other (Chen and He, 2003). During the freezing o f w ater droplets, some surfactant m olecules that form the water-in-oil em ulsion would be expelled from the ice lattices at the oil—w ater interface and diffuse into the oil phase, and during thaw ing process m ore surfactant m olecules could be diffused aw ay from the interface, leading to the lack o f surfactant molecules on the emulsifying film. Thus the w ater droplets could coalesce and form larger water droplets which facilitate the water/oil separation, while some surfactants m ay form micelles inside the w ater droplets w ith trace amount o f oil (Chen and He, 2003; Lin and He, 2007). It was found from the experiments that ultrasonic treatm ent alone was associated w ith the lowest oil recovery (i.e. 58.9%), while the com bination o f ultrasound and freeze/thaw achieved an oil recovery o f 64.2% which was close to that for F/T alone. This m ay indicate that ultrasound alone could not effectively break the emulsifying film o f surfactant molecules. However, ultrasound alone achieved the highest TPH concentration in the recovered oil layer (i.e. highest purity) (Fig. 3.9a). The TPH concentration in the recovered oil was 625 mg/g for F/T, 933 mg/g for US, and 851 mg/g for US + F/T, while the TPH concentration in 41 the original oily sludge was 610 mg/g. As a result, the recovered oil from ultrasonic treatm ent alone contained 93.3% o f TPH. In fact, for a multiphase system when solid phase exists, petroleum hydrocarbon molecules are either strongly adsorbed onto the surface o f solid particles or trapped inside the sediment matrix, and this would prevent the separation o f oil from the multiphase system. Previous studies (Feng and Aldrich, 2000; Breitbach et al, 2003; Hamdaoui et al, 2005) proved that ultrasound could effectively promote the desorption o f T able 3.3 Summary o f oil recovery results using different methods. M ass o f recovered TPH concentration in TPH recoveiy TPH concentration in oil layer (g) recovered o il layer (m g/g) rate (%) w astew ater (m g/L ) F /T m ethod 6.41 625 65.7% <25 U S m ethod 3 .8 4 933 58.9% 1550 U S + F/T m ethod 4.6 0 851 64.2% 200 M ethod 610 TPH in original sludge (m g/g) 100 80 o V a* j- 60 40 o 20 F& T US T r e a tm e n t m e th o d F& T+US F igure 3.8 Oil recoveries for different treatment methods (error bar represents standard deviation) 42 1000 (a) B KX 800 s _o "■C 600 03 v -rr& T T y .■y s y y / 4 -* c w fi ou S3 H 400 '> % < ///, mm -fr / ' / " / ' /y / / ' / / / yy y y y / y ■.• S S , / • ✓✓ ' , / v yy v / / > s ' y - yy yyy / ' SS y //> / ,• z ,> V ^ ; z• / ' / 0 Oily sludge (b) ' / / ■V : ft. 200 '/ . F&T US F&T+US M ethods 2000 1600 c ® 3 '*■* tr 1200 £2Q gM M c su o c o 35 800 u £ 33 400 H ft. m 0 F&T M eth od s u s . F&T+US Figure 3.9 TPH concentrations in separated oil layer (a) and water (b) (error bar represents standard deviation). organic compounds. Due to ultrasonic irradiation in liquid media, m icro-bubbles could form and then collapse or implode when they reach some critical size (Hamdaoui et al, 2007), leading to localized high pressure and tem perature shockwave. The collapse or im plosion occurring in the vicinity o f particle surface would release the adsorbed or trapped organic compound molecules from the solid particles or sedim ent matrix into the solution. M oreover, the collapse can bring high-speed liquid micro-jets w ith strong shear force w hich can then break the aggregates o f solid particles and result in the detachm ent o f oil and w ater from the 43 solid particles (Feng and Aldrich, 2000; Mason, 2007). Consequently, ultrasound could considerably enhance the separation o f oil from solid phase and m ore adsorbed or trapped oil could enter the oil layer after oil/water separation. T he experimental results in this study indicated that freeze/thaw alone was not effective for desorption. Although it obtained the highest oil recovery, the concentration o f TPH in the recovered oil was the lowest (i.e. 62.5%). Through application o f ultrasound, the TPH concentration in the recovered oil for combined US + F/T process increased significantly from 62.5% to 85.1% w hich was close to that for ultrasonic treatment alone (i.e. 93.3%). The TPH concentrations in aqueous phase after w ater/oil separation for different treatment methods are presented in Fig. 3.9 b. It was found that for ultrasonic treatm ent alone, the separated aqueous phase contained very high concentration o f TPH (1550 mg/L). Such high concentration m ay result in high cost for further treatm ent o f the w astew ater after oil recovery. In contrast, freeze/thaw treatm ent alone w as associated w ith the low est TPH concentration in wastewater (i.e. less than 25 mg/L). T he application o f ultrasound could promote the desorption o f adsorbed or trapped PHCs and then effectively release them to the solution, while freeze/thaw was not effective in prom oting the desorption. In fact, for the samples treated by freeze/thaw alone, it was observed in the experiments that the solid particles were hardly separated and settled in the bottom o f the tube. Thus the recovered oil layer also contained high contents o f solids, and the TPH concentration in the recovered oil (i.e. 62.5%) was close to that in the original sludge (i.e. 61%). The observed sludge settleability after freeze/thaw treatment in this study w as not in agreement w ith previous study conducted by Jean et al. (1999) who used freeze/thaw to treat the oily sludge samples taken from the dissolved air flotation (DAF) unit o f a w astew ater treatment plant. The crude 44 oil tank bottom sludge used in this study contained more solid content (i.e. 15%) than the DAF sludge (i.e. 7.8%), and this em ulsion had an oil/solid ratio o f about 4.0 (Table 3.2). The oil recovery from sludge requires not only the separation o f oil from w ater but also the detachment o f oil m olecules from solids. It was suggested that the F/T m ethod was effective to drive hydrocarbon m olecules away from w ater to form free oil layer (Chen and He, 2003). However, as described before, the F/T m ethod could not provide a strong driving force to remove the adsorbed oil to the aqueous phase from solid particles within the crude oil tank bottom sludge. For samples treated by ultrasound or com bined ultrasound and freeze/thaw, it w as observed in the experiments that the solid particles were significantly separated from aqueous phases and settled in the bottom o f the tube, and this indicates that ultrasound could enhance the settling o f solid particles. As a result, the recovered oil for ultrasound alone and com bined ultrasound and freeze/thaw contained m uch less solids but w ith m uch higher TPH (i.e. 93.3% and 85.1%) as compared with that for freeze/thaw alone. Due to ultrasonic irradiation, some desorbed PHCs from solids also entered the aqueous phase, but w ith the effect o f freeze/thaw, some hydrocarbon molecules could be expelled from the expansion o f w ater droplets turning into ice and then enter the oil phase. As a result, it was observed that the TPH concentration in the separated aqueous phase for com bined ultrasound and freeze/thaw w as m uch low er (i.e. 200 mg/L) than that for ultrasound alone (i.e. 1550 mg/L) but higher than that for freeze/thaw alone (i.e. <25 mg/L). This was in agreement w ith previous studies that F/T m ethod was effective for separating oil from the aqueous phase (Chen and He, 2003; Jean et al, 1999). As described above, the com bined process brought higher oil recovery than that for the ultrasonic treatment alone, and much higher TPH concentration in the recovered oil layer 45 than that for the F/T method. Consequently, the com bined ultrasonic and freeze/thaw treatment was identified in this study as more effective than the other two m ethods and w as further exam ined for oil recovery from the refinery crude oil tank bottom sludge. 3.2. Effects o f different factors on the com bined treatm ent process 3.2.1. Effects of ultrasonic power and treatm ent duration In terms o f the com bined ultrasonic and freeze/thaw treatm ent process, Figs. 3.10 and 3.11 present the oil recovery results under the impacts o f ultrasonic pow er and treatm ent duration. It was observed from Fig. 3.10 that the oil recovery o f the com bined process could be improved by increasing ultrasonic pow er at low level. For example, the recovery rate w as increased from 57.7% at ultrasonic pow er o f 21 W to 63.6% at 33 W. However, further increase in ultrasonic power at level above 33 W w as not associated w ith significant enhancement o f oil recovery. The oil recovery was only increased to 64.1% at ultrasonic pow er o f 66 W. Similar results o f the limitation o f ultrasonic pow er have also been reported in many studies in other areas (Feng and Aldrich, 2000; M ason and Lorimer, 2002). It has been reported that the phenom ena o f ultrasonic cavitation could play a significant role in enhancing the desorption o f adsorbed molecules, and the effect o f cavitation depends on the size o f bubbles while m ore energy could be stored w ithin the bigger bubbles (Breitbach et al, 2003). In this study, the low frequency o f ultrasound (20 kH z) was applied. In order to produce shockwave and high speed microjets, im plosion radius o f bubbles is around 170 pm at 20 kHz (Schueller and Yang, 2001). However, bubbles w ith radius at a few m icrons usually become unstable and collapse at such ultrasonic frequency, leading to the rare existence o f larger bubbles. Thus the inhibition on the size o f cavitation m icrobubbles could 46 be attributed to no further significant increase o f oil recovery rate even though the ultrasonic pow er was increased from 33 W to 66 W. 100 90 80 fc* 70 > 60 « 50 3 40 30 o 20 10 10 20 30 40 50 60 70 Ultrasonic power (W) Figure 3.10 Oil recovery versus ultrasonic power for the combined process (error bar represents standard deviation) (experimental condition: ultrasonic treatment duration o f 10 min, sludge/water ratio o f 1:4, without the addition o f rhamnolipids and NaCl). In term s o f the im pact o f ultrasonic treatm ent duration, it can be found from Fig. 3.11 that the oil recovery was increased to 64.2% w ithin 1 m in o f ultrasonic treatm ent followed by freeze/thaw. N o further significant increase o f oil recovery w as observed when ultrasonic irradiation was increased to 10 min. The oil recovery w as 64.1% when the treatm ent duration was 10 min w hich was close to that at 1 min o f treatment. This m ay be explained by the fact that when adsorbed organic molecules w ere desorbed from solid particles into aqueous phase even through long duration o f ultrasonic treatm ent re-adsorption might happen in the system during freeze/thaw process and would thus com prom ise the effect o f ultrasonic desorption (Feng and Aldrich, 2000). 47 100 90 80 70 60 50 o 40 30 20 10 0 2 4 6 8 10 12 U ltrasonic treatm ent tim e (minute) Figure 3.11 Oil recovery versus ultrasonic treatment duration for the combined process (error bar represents standard deviation) (experimental condition: ultrasonic treatment power o f 66 W, sludge/water ratio of 1:4, without the addition o f rhamnolipids and NaCl). 3.2.2. Effects o f sludge/water ratio Fig 3.12 presents the oil recoveries at different sludge/w ater ratios for the com bined treatment process. It can be observed that oil recovery w as increased from 41.9% to 80.0% as the slurry content was increased from sludge/w ater ratio o f 1:8 to 1:2 and then slightly dropped to 72.2% at sludge/water ratio o f 1:1. The increase o f sludge content in the slurry system could result in more oil recovery. However, further increase o f sludge content when the sludge/water ratio was above 1:2 could result in increased viscosity o f the slurry w hich then could impede the formation and collapse o f cavitation bubbles. As a result, the effect o f sonication was weakened and oil recovery rate was decreased when the sludge content w as too high in the slurry. Similar phenom ena was also reported by Feng and A ldrich (2000) who used ultrasonic irradiation to remove diesel from solid in slurry state, and they indicated that the increase o f solid concentration above 50% significantly inhibited cavitation process in 48 oily sand-w ater system. In terms o f the TPH concentration in the recovered oil, it was observed to be 658, 846, 851, and 659 m g/g for the sludge/w ater ratios o f 1:1, 1:2, 1:4 and 1:8, respectively. The TPH concentration in the recovered oil for sludge/water ratio o f 1:2 was close to that for sludge/water ratio o f 1:4. Consequently, b y considering oil recovery and the quality o f the recovered oil, the effective sludge/water ratio was 1:2 when using the com bined treatment process. 100 90 80 70 60 50 o 40 30 20 10 0.00 0 .2 5 0 .5 0 0 .7 5 1.00 1 .2 5 S lu d g e/w a ter r a tio Figure 3.12 Oil recovery versus sludge/water ratio for the combined process (error bar represents standard deviation) (experimental condition: ultrasonic power o f 66 W, ultrasonic treatment duration o f 10 min, w ithout the addition o f rhamnolipids and NaCl). 3.2.3. Effects of bio-surfactant (rham nolipids) The effect o f rhamnolipid addition to the oily sludge slurry system on the com bined treatment process is shown in Fig. 3.13. It can be found that the oil recovery was increased from 64.1% to 73.2% as the concentration o f rhamnolipids in w ater was increased from 0 to 100 mg/L. However, the oil recovery dropped to 61.8% and 62.5% at rhamnolipids 49 concentration o f 400 and 1000 mg/L, respectively. 100 90 80 70 V > © w 9J S. 60 50 40 30 20 10 0 0 100 200 300 400 500 600 700 800 Biosurfactant concentration (mg/L) Figure 3.13 Oil recovery versus bio-surfactant concentration for the combined process (error bar represents standard deviation) (experimental condition: ultrasonic power o f 66 W, ultrasonic treatment duration of 10 min, sludge/water ratio o f 1:4, without the addition o f NaCl). Rhamnolipids are a class o f glycolipid bio-surfactants usually produced by specific bacterial strains (Mulligan, 2005), and the critical m icelle concentration (CM C) has been reported between 10 and 230 mg/L (N itschke et al, 2005). The addition o f rham nolipids into the oily sludge slurry samples could affect the adsorption energy o f petroleum hydrocarbons with solid particles and lower the energy required for desorption, and thus the organic compounds could be easily removed at the same ultrasonic pow er application. Furthermore, the presence o f surfactant could lower the surface tension and affect the formation and collapse o f ultrasonic cavitation bubbles. Therefore, an apparent increase o f oil recovery was observed with addition o f rhamnolipids up to 100 mg/L. However, w hen the adsorbed hydrocarbon molecules was continuously removed from solid particles by ultrasonic irradiation, the free sites on solid particle surface could allow for the adsorption o f 50 bio-surfactants if there were a large amount o f rham nolipids in the oil/solids/w ater multiphase system. The ultrasonic cavitation would also exert an effect on the adsorption and desorption o f rhamnolipids, and thus not all o f the ultrasonic energy introduced into the system was used to remove oil from solid particles. As a result, the oil recovery rate dropped when the rhamnolipid concentration was above 100 mg/L. 3.2.4. Effects of salt addition (sodium chloride) Fig. 3.14 presents the effect o f sodium chloride (NaCl) concentration on the com bined treatment process. It was found that the oil recovery was increased from 64.1% to 74.2% as the addition o f NaCl was increased from 0 to 1%. As the salt concentration was further increasing, the oil recovery started decreasing and dropped to 59.0% w hen the N aCl concentration was 5%. Thus the low salinity showed a positive impact while the high salinity illustrated a negative im pact on oil recovery from oily sludge. Low salinity brine injection has been studied for oil recovery and it has been reported that the presence o f ions would affect the adsorption o f oil onto clay or mineral layers (Cissokho et al, 2009; Lager et al, 2006). Abram ov et al. (2009) indicated that the addition o f N a+ helped to break the bond between oil and sand soils by increasing the negative charges on the soil surfaces. In addition, the presence o f NaCl could also enhance cavitation bubble implosion and thus enhance the sonic pow er intensity (Suri et al, 2010). As a result, the addition o f salt could introduce a positive effect on oil recovery using the com bined treatment process. However, excessive am ount o f NaCl (i.e. above 1%) in the sludge slurry system could reduce the concentration o f PHCs in the aqueous phase and lead to a negative impact on the desorption o f PHCs from solid particles by ultrasound (Dukkanci and Gunduz, 51 2006). Therefore, the effective salinity was found to be about 1% for oil recovery from oily sludge. 100 90 80 £ 70 s- 60 a» ► o 50 w v u 40 30 20 10 0 0 1 2 3 4 5 6 Salt (NaCl) concentration (%w/w) F igure 3.14 Oil recovery versus NaCl concentration for the combined process (error bar represents standard deviation) (experimental condition: ultrasonic pow er o f 66 W, ultrasonic treatment duration o f 10 min, sludge/water ratio o f 1:4, without the addition o f rhamnolipids). 3.3. PHC fraction analysis for recovered oil Fig. 3.15 presents the GC profiles o f samples from the combined ultrasonic and freeze/thaw treatment, and it can be found that there w as no significant difference betw een PHC fraction distribution in the original sludge sam ple and the recovered oil sample, indicating that the effect o f ultrasonic destruction o f petroleum hydrocarbons was not significant (i.e. no significant shift o f peaks towards the left o f GC profile), and the m ain mechanism o f enhanced oil recovery through the com bined process was ultrasonic enhance desorption. 52 Figure 3.15 GC profiles o f samples from oil recovery treatment (A represents oil sample after US + F/T treatment and B represents original oily sludge sample). Table 3.4 PHCs fraction distribution for samples before and after US+F/T treatment Samples PHCs fraction distribution (%) F2 F3 F4 Original sludge 23.0% 63.9% 13.1% Oil recovered at 66 W for 10 minutes with no salinity and surfactant 21.8% 64.0% 13.2% Oil recovered at 66 W for 1 minute with no salinity and surfactant 21.8% 64.9% 13.3% Oil recovered at 66 W for 10 minutes with 3% NaCl 21.9% 64.7% 13.4% Oil recovered at 66 W for 10 minutes with 40 mg/L biosurfactant 21.6% 64.9% 13.5% Table 3.4 lists the PHC fraction distributions in the original oily sludge sample and several recovered oil samples under different treatm ent conditions. It can be found that the difference among the proportions o f PHC fractions in the recovered oil under different ultrasonic treatment conditions was very small, and the average F2, F3, and F4 fractions in the recovered oil were 21.77%, 64.64%, and 13.37%, respectively. The F2 fraction in the recovered oil was slightly lower (i.e. about 1.2%) than that in th e original sludge, while the F3 and F4 fractions in the recovered oil were slightly greater (i.e. about 0.8% and 0.2%) than that in the original sludge. This may indicate that the application o f ultrasound could destruct some light oil compounds such as F2, leading to slight increase o f the proportion o f F3 and F4 fractions in the recovered oil. However, the effect o f ultrasonic destruction was not very significant. It has been reported that the destruction o f organic com pounds such as long-chain hydrocarbons was resulted from the production o f hydroxyl radicals (Feng and A ldrich, 2000). This usually occurs when the ultrasound is in the mid frequency from 200 to 400 kH z (Breitbach et al, 2003; Petrier and Francony, 1997). In this study, the condition for the production o f hydroxyl radicals was not reached at low frequency o f 20 kHz, thus there w as no significant destruction to change the PHC proportions in the recovered oil. However, the energy released from ultrasonic cavitation under this frequency w as enough to overcom e the affinity o f hydrocarbon molecules with solid particles and to increase oil desorption. 4. Conclusion Oil recovery from refinery oily sludge was investigated in this study using three different approaches, including ultrasonic treatm ent alone, freeze/thaw alone, and com bined ultrasonic and freeze/thaw treatment. By com prehensively considering oil recovery efficiency, as w ell as TPH concentrations in the recovered oil and in the separated wastewater, the com bination o f ultrasound and freeze/thaw was identified as an effective method w ith satisfactory performance. Under the experimental conditions, it achieved an oil recovery rate o f 64.2%, and TPH concentrations o f 85.1% and 200 mg/L in the recovered oil and wastewater, respectively. The experimental results revealed that ultrasound could prom ote the separation o f oil from solids while freeze/thaw could prom ote the separation o f w ater and oil in the m ultiphase 54 system. Several different factors were further examined to investigate their individual im pacts on the performance o f the com bined treatment process, and it w as observed that under the experimental conditions, the oil recovery was im proved with ultrasonic pow er at low pow er level, but further increase in ultrasonic power at level above 33 W was not associated w ith significant enhancement o f oil recovery. The oil recovery was increased to its peak w ithin 1 m in o f ultrasonic treatment, and thereafter no further significant increase o f oil recovery w as observed. The results o f examining the individual impacts o f other factors on the com bined treatment process indicated an effective sludge/w ater ratio o f 1:2, a rham nolipids concentration o f 100 mg/L, and a NaCl concentration o f 1%, respectively. A n oil recovery rate o f up to 80.0% was observed under the experimental conditions o f ultrasonic pow er o f 66 W, ultrasonic treatm ent duration o f 10 min, sludge/water ratio o f 1:2, and no addition o f bio-surfactant and salt. The analysis o f PHC fraction distributions in the recovered oil samples indicated that the effect o f ultrasonic destruction o f organic com pounds was insignificant, and the m ajor mechanism o f oil recovery enhancement was through enhanced ultrasonic desorption o f PHCs from solid particles under the low frequency application o f 20 kHz. In this study, the individual impacts o f different factors on the com bined treatm ent process w ere examined through a series o f laboratory experiments. However, the interaction effects among these factors, the identification o f m ajor influencing factors, and the optimal combination o f these factors w ere not examined. This could be investigated through factorial experimental design m ethod in future studies. In addition, the recovered oil from the crude oil tank bottom sludge in this study was not a pure m ixture o f petroleum hydrocarbons, and m ay need further treatment. The value o f the recovered oil as fuel (such as asphaltene content, ash content, salt content, and heat o f 55 combustion) and the detailed cost/benefit o f the proposed treatm ent process were not analyzed, and should be exam ined in future studies. In summary, the combined ultrasonic and freeze/thaw process could represent an environm entally friendly and econom ically competitive alternative for the effective treatm ent o f oily sludge waste from the petroleum industry, and is w orth o f further investigations. 56 Chapter 4 Treatment of oily sludge through advanced oxidation processes 57 Abstract The effective disposal o f oily sludge generated from petroleum industry is a w orldw ide concern. Due to its com plex in com position, oily sludge is relatively hard to clean up using traditional approaches. In this study, three different advanced oxidation processes w ere investigated for their capabilities to reduce petroleum hydrocarbons (PHCs) in oily sludge, including ultrasonic treatm ent alone, Fenton process alone, and the com bination o f ultrasound and Fenton process (US/Fenton). The results revealed that the com bined process could achieve the best reduction in total petroleum hydrocarbons (TPH). The impacts o f four different factors on the com bined process were further exam ined, including the initial sludge content, the molar ratio o f hydrogen peroxide to Fe2+,(H 20 2/F e2+) the ultrasonic power, and the ultrasonic treatm ent duration. The highest TPH reduction rate o f 88.1% was observed by US/Fenton process with a sludge content o f 20 g/L, a H 20 2/Fe2+ molar ratio o f 4:1, ultrasonic treatment time for 5 minutes, and ultrasonic pow er o f 60 W. Key w ords: Oily sludge, total petroleum hydrocarbons, advanced oxidation processes, Ultrasound, Fenton, 58 1. Introduction Oily sludge is one o f the m ajor wastes generated from petroleum industry. It is form ed when petroleum w aste settles down to the bottom o f the crude oil storage tank, or when the oily material is collected from oil-water separator, or even when heavy petroleum hydrocarbons (PHCs) accumulate after petroleum crudes are distilled in the refinery system (Bhattacharyya and Shekdar 2003; Kaushik et al, 2012). The oily sludge is a com plex consisting o f water, inorganic solid particles, and various petroleum hydrocarbons. It typically contains about 30-50% o f oil, 30-50% o f water, and 10-12% o f solids by m ass (Reynolds et al, 1993; Ram aswam y et al, 2007). The PHCs in oily sludge include a large amount o f long-chain alkanes and alkenes as well as polycyclic aromatic hydrocarbons, coke, asphaltenes, and resins (Overcash and Pal 1979; Kaushik et al, 2012). The utility o f these heavy hydrocarbons is very limited and uncontrolled dispose o f them w ithout any treatm ent could pose danger to the environment. Therefore, many technologies have been developed to treat oily sludge before its disposal. The advanced oxidation processes (AOP) have received considerable attention in the past decades as potential effective m ethods for the destruction and degradation o f recalcitrant materials (Gogate and Pandit, 2004). During the AOPs treatm ents, a sufficient am ount o f hydroxyl radicals can be generated through ultrasonic irradiation, ultraviolet radiation, photocatalysis, ozone, and/or Fenton/Fenton’s like reactions (Torres et al., 2007a, b). The hydroxyl radicals are strong and non-selective oxidants which can oxidize various recalcitrant compounds due to the high oxidation potential (E°=2.8) o f such radicals (Al-Kdasi et al, 2004). The final products o f oxidation reactions include carbon dioxide, short-chain organic acids, and inorganic ions, w hich are usually less toxic and favourable to 59 biodegradation (Adewuyi, 2005). The reactions o f hydroxyl radicals w ith pollutants are essentially determined by the generation o f radicals and the transfer o f radicals to target compounds during various advanced oxidation processes (M ahamuni and Adewuyi, 2010). Generally, the reactions between hydroxyl radicals and pollutants are very fast, w ith the value o f first order kinetic constant being observed in the range o f 108-109 M 'S '1even when extremely low radical concentration exists (i.e., about 10‘12 to 10"'° M) (Esplugas et al, 2002). Ultrasonic irradiation (US) and Fenton reagents are A OPs applied in a variety o f fields. Ultrasound is regarded as a “clean” technology to remove contaminants w ithout the generation o f secondary pollutants (Pang et al, 2011),and it iseasy to m anage the use o f ultrasonic generators. U ltrasonic treatment can generate the process by OH radicals due to the acoustic cavitation, w hich involves the form ation and subsequent expansion o f m icrobubbles under the periodic pressure variations (M ohajerani et al, 2010). When acoustic waves are induced, the micro-bubbles are formed and rem ain stable oscillation until they reach to a critical size and then violently collapse. With the sudden im plosion o f m icro-bubbles, the temperature can rise up to 5000 K (Suslick et al, 1986) inside the residual bubbles. Thus chemical bonds between hydrogen and oxygen (H-O) in w ater molecules are broken inside the bubbles, causing the generation o f hydroxyl radical (-O H), hydrogen radical (H-), hydroperoxyl radical (H 0 2-) and hydrogen peroxide. The serial chain reactions under ultrasonic irradiation can be listed as following (Adewuyi 2005): H 20 ^ H- + OH (1) HO- + -OH -*• H zO + O- (2) HO- + H20 -» H 2O z+ O- (3) H- + -OH -> H 20 (4) 60 H- (5) + H- —►H2 o2 (6) HO- + OH -» H 2+ 0 2 (7) O +o- HO- (aq)+ • OH(aq) H 20 2(aq) (8) H- + 0 2 —►HO z • (9) H 0 2 + H- (10) H 20 2 ho2+ho2--> h2o2+o2 (11) 0 2— 2 0 (12) o2+ o -* o3 (13) The hot spot theory is usually used to explain the destruction and degradation o f organic compounds in liquid. Two m ajor reactions can take place due to acoustic cavitation, including: (a) pyrolytic or com bustive reaction resulting from high tem perature in the hot spot area, and (b) free radical attack between cavitation bubble and bulk liquid phase (M ohajerani et al, 2010). M any studies have reported the effect o f ultrasonic irradiation on the removal o f organic compounds. For example, Torres et al. (2007a) reported that the bisphenol-A concentration was under detection limit after treatm ent with ultrasonic cavitation (300 kHz and 80 W) for 90 minutes, and the oxidation intermediates w ere formed during ultrasonic treatment. Liang et al. (2007b) investigated the favourable ultrasonic frequency at the input power o f 50 W and they found that the degradation o f 4-chlorophenol was faster at higher frequency (i.e. 200 kHz) than at lower frequency (e.g., 28 kHz, 50 kHz). However, when used alone for the oxidation o f chemicals, the sono-oxidation rates are relatively low (Liang et al, 2007a; Virkutyte et al, 2010). 61 A nother advanced oxidation m ethod is the use o f Fenton ‘s reagents for degradation o f compounds. As com pared to other oxidants, Fenton’s reagents (hydrogen peroxide and iron) are inexpensive and environm entally friendly (Pignatello et al, 2006). Because o f its simplicity and strong oxidation power, the Fenton’s process has been frequently used as an advanced process for the oxidation o f various organic contaminants. D uring the chain reactions associated w ith Fenton’s reagents (Eq. 14-18) (Pignatello et al, 2006), hydrogen peroxide serves as an oxidizing agent. A metal salt, usually ferrous (Fe2+) salt or metal oxide, generally works as a catalyst for the chain reactions. The ferric ion (Fe3+) can also be used as a catalyst for the generation o f hydroxyl radicals, and the related reactions are usually called “Fenton-like” reactions. The generation o f hydroxyl radicals during Fenton reactions can be described below: Fe2+ + H 20 2 - -> Fe3+ + *OH + OH' (14) Fe3+ + H 20 2 - -> Fe-O OH2+ + H+ (15) Fe-OOH2+ —H►Fe2+ + H O z- (16) h o 2 + h o 2- --► h 2o 2 (17) Fe3++ H 0 2- ~ -> Fe2+ + 0 2 + OH' (18) During Fenton’s reactions, a large am ount o f hydroxyl radicals are generated simultaneously if the hydrogen peroxide approaches to F e2+ ion (Eq. 14). The sam e am ount o f Fe3+ is also produced along with the formation o f free radicals, which later precipitates in the form o f am orphous ferric oxyhydroxides (Fe-OOH), causing the increase o f pH from strongly acidic to neutral, and generating an undesirable sludge. It has been reported that the favorable pH range to maximize oxidation in w ater is approxim ately 3.0-4.0 (Lu et al. 2010a). 62 For example, Pham et al (2011) reported a DEHP decrease o f 31% in the slurry by Fenton pre-treatment with a pH o f 4.3. The application o f Fenton process has been w idely reported in many studies w hich focused on the treatm ent o f pollutants either in w ater or in soil. For instance, Sedlak and Andren (1991) investigated the ability o f Fenton’s reagent to degrade the chlorinated aromatic hydrocarbons. M asom boon et al (2010) examined the degradation o f 2.6-dimethylaniline using Fenton process, and almost completely rem oval of 2.6-dimethylaniline could be obtained in 4 hours o f electro-Fenton treatment w ith a pH value o f 2, ferrous ion 1 mM , and hydrogen peroxide concentration o f 20 mM. They also proposed a general pathway for the degradation o f 2,6-dim enylaniline initiated by OH (M asom boon et al., 2010). V illa et al (2008) evaluated the effect o f Fenton process on the degradation o f DDT and diesel in soil, and they observed that 80% o f diesel and 75% o f DDT present in contaminated soil w ere removed after 64 hours o f reaction. A long with the degradation o f pollutants, 80% o f organic m atter naturally present in soil w as also degraded during the Fenton process (Villa et al., 2008). The oxidation efficiency o f AOPs in treating chemical w aste can be affected by tw o aspects, including the production rate o f free radicals and the extent o f the produced radicals contacting with contaminant m olecules (Gogate and Pandit 2004). The application o f individual AOPs m ight not be able to achieve high degradation efficiency and good economics. However, there is a sim ilarity between the pollutant oxidation m echanism s among various AOPs, and the individual AOPs might be complementary to one other. A s a result, it is expected that the com bination o f AOPs can achieve better results as com pared to single AOP. In fact, the com bination o f ultrasonic irradiation and the use o f oxidants (hydrogen peroxide, ozone, and Fenton/Fenton-like agents), has become popular recently 63 since the hybrid m ethod has been revealed m ore effective in degrading some recalcitrant compounds. For example, N eppolian et al (2004) reported that the degradation rate o f /?ara-chlorobenzoic acid (p-CBA) was 7.3 xlO '3 m in '1 under the com bination process o f ultrasound and Fenton-like reactions as com pared to the value o f 4.5 x 10'3 m in '1 under the process o f only using 20 kH z ultrasound. Sun et al (2007) observed that the optim um concentration o f Fe2+ for the decolorization o f acid black 1 (AB1) solution was 0.025 m M under the com bination o f ultrasound and Fenton process when the concentration o f H 2 O 2 was 8.0 m M and the pH value was around 3.0. Virkutyte et al (2010) investigated the effect o f sono-Fenton-like process on the degradation o f naphthalene w hen the m ineral iron in soil was used as the catalysts, and they observed that the optim um degradation efficiency o f naphthalene reached 97% w hen 600 m g/L o f H 2 O 2 w as added into the soil w ith initial naphthalene concentration o f 200 m g/kg while the ultrasonic pow er was at 200 W and 400 W. The chain reactions in such Sono-Fenton system can be described by Eqs. 19-26 (G ogate and Pandit, 2004; M ohajerani et al, 2010). In those com bined processes, hydroxyl radicals are generated by the decom position o f hydrogen peroxide, during w hich Fe2+ ions are converted to F3+. Meanwhile, the application o f ultrasonic energy could isolate Fe2+ from Fe-O OH 2+ by the reaction o f Eq. 26 (in w hich US stand for the application o f ultrasonic irradiation), and the isolated Fe2+ in its turn could react w ith hydrogen peroxide to generate hydroxyl radicals (Eq.19). As a result, the iron catalysts could be regenerated during the process com bining ultrasonic irradiation w ith Fenton reagents. Fe2+ + H 20 2 — ►Fe3++ O H + OH (19) Fe3+ + H 20 2 — > Fe-O OH2+ + H+ (20) Fe3+ + HOz* — > Fe2+ + 0 2 + H+ (22) Fe2+ +-OH — > Fe3+ + OH (23) F2+ + HOz- + H+ — > Fe3+ + H 20 2 (24) Fe3++ H 0 2- — ►Fe2+ + 0 2 + OH' (25) US Fe-OOH2+ — > Fe2+ + H O 2- (26) Given the advantages for the degradation o f organic contaminants, A OPs have been widely studied to treat various wastes, especially w hen the pollutants are not am enable for biodegradation. O ily sludge is complex mixture o f various petroleum hydrocarbons, m any o f which are hard to be destructed by the m eans o f conventional remediation technologies. Therefore, it is reasonable to introduce A OPs into oily sludge treatment. However, the previous studies have m ainly focused on the degradation o f individual target contaminants, and only a few studies have extended the application o f AOPs to treat contam inated systems containing a m ixture o f hazardous organic compounds. The objective o f this study is then to evaluate the degradation efficiency o f the advanced oxidation processes for oily sludge treatment, and the ultrasonic irradiation (US), the Fenton reagents, and the combined US and Fenton process (US/Fenton) were examined. A variety o f factors, including the initial sludge content, the m olar ratio o f hydrogen peroxide to Fe2+, ultrasonic power, and ultrasonic treatm ent duration, w ere investigated for their effects on oxidation o f petroleum hydrocarbons (PHCs) through the combined m ethod (US/Fenton). The results would provide a sound basis for developing efficient and econom ically com petitive methods for oily sludge treatment. 65 2. Materials and methods 2.1. Oily sludge a n d chem icals The oily sludge was obtained from the bottom o f a crude oil tank in an oil refinery plant in western Canada (Prince George). The sludge was kept at 4°C in a capped stainless-steel bucket. The sludge was very sticky and its viscosity was not available to measure. The summary o f its characteristics is listed in Table 3.1. The concentration o f petroleum hydrocarbon (PHCs) in sludge was analyzed based on the sam ple extraction process w hich will be described in Section 2.5 for solid sample extraction. T he content o f metal elem ents was measured with Inductively Coupled Plasma (ICP) in accordance with the m ethod given in ASTM D5185 (2009), the w ater content was m easured using the m ethod A STM D1744 (1992), and the solid content was calculated with the content o f w ater and PH Cs in sludge. Hexadecane (nC16; >99% pure), nonadecane (nC19; >99% pure), tetratracon (nC34; >99% pure), and pentacontane (nC50) were purchased from Sigm a and they w ere mixed in toluene as standard compounds for PHCs analysis and fractional analysis. The dichloromethane (DCM ) (>99%, HPLC) was used for sam ple extraction. C yclohexane (>99%, HPLC) was used with DCM to rinse columns packing w ith silica gel. Toluene (>99% , HPLC) was used as solvent o f samples w ith PHCs for GC analysis. Silica gel (purchased from Sigma) was activated at 105°C for 12 h and then used to clean up extraction solution, and anhydrous sodium sulfate dried at 400 °C for 12 w as used to dewater extraction samples. F eS 0 4 7H20 (purchased from Sigma) and H 2 O 2 (30% w/w) solution w ere used as Fenton’s reagents deployed either w ith ultrasound or without ultrasound. 66 2.2. Ultrasonic apparatus The ultrasonic apparatus used in this experim ent was a M isonix Sonicator 3000 generator with a titanium sonic probe and the diam eter o f the tip horn was 0.5 inch (Fig. 4.1). The frequency o f ultrasound originated from this generator was constant at 20 kHz, but the pow er could be adjusted from 0 up to about 75 W. W hen ultrasonic irradiation w as applied, the sonic probe was inserted into the sample and the tip hom w as always kept under liquid during the treatment process. F ig u re 4.1 Oxidation o f oily sludge using ultrasound 2.3. Experiment on oxidation o f PHCs in oily sludge by three different processes Laboratory experiments were carried out to exam ine the effect o f three different 67 advanced oxidation processes on oily sludge treatment, including ultrasonic irradiation alone, the Fenton’s reaction alone, and the combination o f ultrasound and Fenton’s reaction. In terms o f the ultrasonic (US) process alone, 1 g o f oily sludge was put into a 100 ml beaker with 25 ml deionized water. The ultrasonic probe was then placed into the sludge/water system for ultrasonic oxidation (Fig 4.1). The ultrasonic pow er w as fixed at 60 W and the treatm ent duration time was set up as 1, 3, 5, and 8 minutes, respectively. After ultrasonic irradiation, the mixture in the beaker was transferred into a 50 ml tube for centrifugation to separate the solid from the liquid for further analysis o f petroleum hydrocarbons w ithin liquid and solid phases. if F igure 4.2 Fenton's reaction process for degradation o f oily sludge In terms o f the Fenton’s reaction (Fenton) alone, 1 g o f sludge was put into a 100 ml beaker, and then 0.63 g o f FeS 0 4 ' 7 H 2 0 was added into the system (Fig 4.2). The volum e o f 68 hydrogen peroxide added to the beaker was set at 5, 10, 15, and 20 ml, respectively. The total aqueous volum e in the beaker was consistent w ith that for ultrasonic treatment, and therefore the volum e o f DI w ater added varied according to the dosage o f hydrogen peroxide solution consumed in samples during the Fenton process, and they w ere 20, 15, 10, and 5 ml, receptively. Due to the violent reaction, hydrogen peroxide w as gradually added into the system until the appropriate volum e by the use o f 1 ml pipette. The mixture in the beaker w as manually stirred during the Fenton process. F ig u re 4.3 Samples after US/Fenton treatment In terms o f the combination o f ultrasound and Fenton process (US/Fenton), the reaction was also conducted in a 100 ml beaker. 1 g o f oil sludge was added into the beaker w ith 10 ml o f deionized w ater and 0.63 g o f F e S 0 4-7H20. The ultrasonic probe w as then placed 69 under the liquid for ultrasonic treatment w ith the pow er o f 60 W. 15 ml H 2 O 2 solution was added into the system at the rate about 3 m l per m inute during the US/Fenton treatm ent for 5 minute. As shown in Fig 4.3, the “iron sludge” can be formed due to the reaction o f Fenton’s reagents, either through the Fenton process alone or through the com bined process o f ultrasound and Fenton reaction. Thus, the petroleum hydrocarbons (PHCs) rem ained after treatment would be distributed am ong both the aqueous phase and solid phases (e.g., “iron sludge” and solid particles in oily sludge). Two distinctive extraction procedures were conducted to extract PHCs from both the liquid and solid phases. Since the mass o f “iron sludge” varies from sample to sample, it is difficult to determ ine the concentration o f PHCs in the solid phase. Accordingly, the reduction in the total mass o f PHCs in a sample after treatment was used to analyze the efficiency o f the treatm ent processes based on Eqs. (1) and (2 ). Mass o f PHCs in one sample (mg) = m as s o f PHCs in liquid phase (mg) + m as s o f PHCs in solid phase ( m g ) (1) Reduction Rate Mass o f PHCs be fo r e t r e a m te n t (mg) — Mass o f PHCs a f t e r t r e a t m e n t ( m g ) Mass o f PHCs b e f o r e t r e a t m e n t ( m g ) x 100% (2) 2.4. Experim ent on factors affecting the com bined process (US/Fenton) The impacts o f different factors on the combined process o f ultrasound and Fenton’s reaction were also investigated. These factors include the initial oily sludge content (sludge content), the m olar ratio o f H 2 O 2 to Fe2+ (H20 2/Fe2+), the ultrasonic power, and the ultrasonic 70 treatm ent duration. Each factor was exam ined at 3 levels. The details o f experim ental factors and their levels were listed on Table 4.1. The Taguchi orthogonal experim ental design m ethod was used to arrange the experiments, and the L27 array was arranged using the software M initab 16 as shown in Table 4.2. This experimental design m ethod can allow for the exam ination o f the main effect o f each factor and the interaction effect betw een the sludge content and the molar ratio o f H 20 2/Fe2 +, between the sludge content and US time, and between the m olar ratio o f H20 2/Fe2 + and US time. Table 4.1 Experiment factors and their three levels Levels Factors 1 2 3 (A) Initial oil sludge content ( sludge content) (g/L) 20 40 60 (B) Molar ratio o f H20 2 to Fe2+ (H20 2 /Fe2+) 4:1 10:1 50:1 (C) Ultrasonic power (US power) (W) 20 40 60 (D) Ultrasonic treatment time (US time) (min) 1 3 5 There w ere 27 experimental runs according to the design, and each run was replicated for two times. The volume o f H20 2 used in the experiment was set as a constant (i.e. 15 ml) for all o f the 27 experimental treatments. The experimental operating procedure for each run w as similar to that as described in section 2.3, including (a) adding different am ount o f oily sludge (e.g. 0.5 g, 1.0 g, 1.5 g) into a 100 ml beaker w ith 10 ml o f de-ionized water, and producing a total liquid volume o f 25 ml after adding 15 m l o f H20 2 solution (the initial sludge content listed in Table 4.1 was determ ined by dividing the mass o f oily sludge by the volum e o f 25 ml), (b) different amount o f F e S 0 4-7H20 was used to adjust the m olar ratio o f H20 2/Fe2+ to 4:1, 10:1, and 50:1, respectively, and (c) after starting the ultrasonic treatment, 71 hydrogen peroxide was added into the beaker at the rate o f 15 ml, 5 ml, and 3 ml per m inute for the 1-minute, 3-minute, and 5-minute treatm ent, respectively. L27 array orthogonal experimental design Table 4.2 E xp erim en tal test # S lu d g e con ten t H 20 2 /F e2+ U S tim e U S pow er (g /L ) (m o le /m o le ) (m inute) (W ) L 1 20 4:1 1 20 L 2 20 4:1 3 40 L 3 20 4:1 5 60 L 4 20 10:1 1 40 L 5 20 10:1 3 60 L 6 20 10:1 5 20 L7 20 50:1 1 60 L 8 20 50:1 3 20 L 9 20 50:1 5 40 L 10 40 4:1 1 40 L 11 40 4:1 3 60 L 12 40 4:1 5 20 L 13 40 10:1 1 60 L 14 40 10:1 3 20 L 15 40 10:1 5 40 L 16 40 50:1 1 20 L 17 40 50:1 3 40 L 18 40 50:1 5 60 L 19 60 4:1 1 60 L 20 60 4:1 3 20 L 21 60 4:1 5 40 L 22 60 10:1 1 20 L 23 60 10:1 3 40 L 24 60 10:1 5 60 40 L 25 60 50:1 1 L 26 60 50:1 3 60 L 27 60 50:1 5 20 2.5. Sam ple extraction after advanced oxidation processes The sample in the beaker after oxidation treatm ent was transferred into a 50 ml tube for centrifugation for 30 min at 5000 rpm. After centrifugation, the supernatant (about 25 ml) 72 was transferred into a separating funnel for liquid extraction, and the solids residue left at the bottom o f the centrifugation tube was used to extract PHCs in the solids. In terms o f the liquid-liquid extraction, 15 ml o f DCM w as added into the separating funnel containing about 25 ml o f supernatant (Fig 4.4), and the funnel w as m anually shaken for several m inutes (SW -846 EPA, 1993). A fter equilibrium for 30 minutes, the upper layer in the funnel was collected into a glass tube. A nother 15 ml o f DCM was then added to the separating funnel for another liquid-liquid extraction, and such extraction w as conducted for three times. All o f the extraction solution was collected into the glass tube for further cleanup. F ig u re 4.4 Liquid-liquid extraction In terms o f the solids extraction, 10 ml o f DCM was added into the 50 m l centrifugation tube containing solids at the bottom. The tube w as placed on a mechanical shaker for 30 minutes o f shaking at 120 rpm (Fig. 4.5), and was then sent for centrifugation at 5000 rpm 73 for 15 minutes. A fter centrifugation, the liquid was transferred into a glass tube. A nother 10 ml o f DCM was added into the centrifugation tube for a second extraction, and such extraction was conducted for three times. All o f the extraction solution w as collected into the glass tube for further cleanup. To analyze the initial PHCs concentration in the oily sludge before any treatment, 25ml DI w ater was added into a tube with lg oily sludge and the tube was centrifuged for 30minutes at 5000rpm. And then sim ilar liquid extraction and solids extraction procedures were implemented. Figure 4.5 Mechanical shaking for the extraction o f PHCs from solids The extraction solution was sent to a silica gel colum n (Fig. 4.6) for cleanup to rem ove moisture, particulate, and unwanted polar organic compounds (CCM E, 2000). The colum n was packed with silica gel and anhydrous sodium sulfate and rinsed w ith 20 ml o f solvent 74 (1:1 cyclohexane/DCM ) before use. After the extraction solution passed through, another 20-30 ml o f solvent (1:1 cyclohexane/DCM ) was poured to elute the column. A round flask was used to collect extraction at the rear o f the column. The cleaned extraction was then reduced to about 1 ml by using a rotary evaporator, and toluene was used as solvent to transfer the concentrated extraction into a 15-ml vial bringing to a final volum e o f 12 ml. 2 ml o f this extraction solution was then transferred into a 2-ml vial for GC analysis. Figure 4.6 Extraction solution cleanup using silica gel column 2.6. PHCs analysis 2 ml o f the final extraction solution was used for PHCs analysis and TPH fraction analysis using a Varian CP-3800 Gas Chrom atograph w ith flam e ionization (GC-FID ) (Fig 75 3.9 in Chapter 3). The external standard m ethod was used for the analysis. D ecane (CIO), hexadecane ( C l6), tetratriacontane (C34), and pentacontane (C50) were used as external standards to determ ine the concentration o f PHCs and PHC fractions (CCM E, 2001), w here Fraction 1 (F I), Fraction 2 (F2), Fraction 3 (F3) and Fraction 4 (F4) is defined as the group o f petroleum hydrocarbons from C6 to CIO, CIO to C l 6, C16 to C34, and C34 to C50, respectively. The GC analysis conditions were set up as: ZB-capillary colum n (Phenom enex Torrance, CA) with 30 m x 0.25 mm ID (inner diam eter) and 0.25-pm film thickness; inject volume o f 1 pL; injector and detector (FID) tem peratures at 320 °C; carrier gas (helium ) at a constant flow rate o f 1.5 m L/min during analysis. The split-less injection m ode was performed on the 1079 PTV injector and after 0.7 min the split mode w as activated at split ratio o f 10:1. The capillary column temperature program was initially held at 50°C for 1 min, then ramped at 15.0 °C/m in to 110 °C and further increased at 10.0 °C/m in to 300 °C and then was held for 11 min. The total running time for a sample w as 45 minutes. 3. Results and discussions 3.1. Comparison o f methods 3.1.1. TPH reduction using ultrasonic irradiation alone Table 4.3 and Fig. 4.7 present the results o f TPH reduction in oily sludge using ultrasonic irradiation alone. The initial TPH mass in the system w ith 1 g o f oily sludge was 431.5±11.4 mg (i.e. the total mass o f PHCs measured from both liquid and solid phases). A fter ultrasonic treatment alone, the TPH mass remained in the system ranged from 334.0±8.6 mg to 76 363.8±22.6 mg, with the highest TPH reduction rate being 22.6% after US treatm ent for 5 minutes (Table 4.3). The TPH reduction rate was slightly increased to 22.6% from less than 20% when the ultrasonic treatment time w as prolonged from 1 minute to 5 minutes. However, longer treatment than 5 minutes in this study did not improve the ultrasonic treatm ent performance, and the TPH reduction rate was slightly decreased to 16.3% after US treatm ent for 8 minutes. Table 4.3 TPH reduction in oily sludge through US treatment alone Ultrasonic treatment duration (min) TPH reduction rate 5 CL| H 300 ® 200 6 s 1 18.0% 3 15.7% 5 8 22.6% 16.3% too 0 1 3 5 US treatm ent tim e (minute) 8 F igure 4.7 Remaining TPH mass in samples after different ultrasonic treatment durations (error bar stands for standard deviation; the value at the time o f 0 stands for the initial TPH mass) The performance o f ultrasonic irradiation on degradation o f organic hydrocarbons in liquid systems can be affected by the cavitation phenom ena (Liang et al, 2007 a). The thermal cracking due to the collapse o f cavitation bubbles and the free radicle reactions 77 initiated by hydroxyl radicals at the interfacial region and in bulk liquid are the two m ajor mechanisms for the degradation o f organic com pounds via ultrasonic irradiation. The therm al cracking o f petroleum hydrocarbons under ultrasonic irradiation has been reported by m any researchers based on their studies in upgrading heavy gas oil and vacuum residue (Gopinath et al, 2006; Kaushik et al, 2012). The cavitation bubbles filled w ith vapor and/or gas form ed via ultrasound can violently collapse to form “hot spots” in the corresponding region, w here the temperature and pressure can reach up to 5000 K and several hundred atm ospheres (Suslick et al, 1986). The cumulative energy generated by these bubbles is extrem ely high, and thus the thermal scission o f carbon-carbon (C-C) bonds could occur in heavy petroleum hydrocarbon molecules. And also some organic com pounds w ith higher vapor pressure could directly decompose inside the cavitation bubbles due to thermal pyrolysis (Liang et al, 2007 a; Pang et al, 2011). M eanwhile, the hydroxyl radicals (OH ) are form ed inside the cavitation bubbles by w ater pyrolysis when these bubbles collapse intensely. These radicals can then be transferred to the bubble interface and bulk liquid and react w ith hydrophobic organic com pounds present in bulk liquid (Liang et al 2007a). Therefore, the free-radical reactions caused by the generation o f hydroxyl radicals also contribute to the reduction o f petroleum hydrocarbons under ultrasound treatm ent alone. The petroleum hydrocarbons in oily sludge usually contain a large num ber o f carbon atoms, and they have low vapor pressure and low solubility in water. M ost o f such heavy petroleum hydrocarbons are hardly able to permeate into the cavitation bubbles. Thus, the free radical reactions can occur near the interface o f cavitation bubbles and more often in the bulk liquid w here the heavy hydrocarbon com pounds are largely present. Moreover, it has been reported that the presence o f solid particles in liquid can affect 78 the collapse o f cavitation bubbles and generate high-speed microjets o f liquid through the cavity (Suslik 1990). In this study, solid particles present in oily sludge produced a heterogeneous system for ultrasound treatment. The liquid microjects resulting from ultrasonic irradiation in such system could enhance the transport o f hydroxyl radicals into the interface region and the bulk liquid (Liang et al, 2007 b). This w ould facilitate the contact o f hydroxyl radicals with the petroleum hydrocarbons. In addition, the ultrasonic irradiation could promote the desorption o f petroleum hydrocarbons from the solid phase and their subsequent dispersion into the liquid phase (Gaikwad et al, 2008). T he dispersion o f petroleum hydrocarbons in the bulk liquid could also enhance the contact o f petroleum hydrocarbon m olecules with the free hydroxyl radicals, and thus facilitate the oxidation reactions. 3.1.2 TPH reduction through Fenton’s reaction process Table 4.4 and Fig. 4.8 present the TPH reduction results under the Fenton process alone. After Fenton’s oxidation process alone, the rem aining TPH mass in the sam ple systems (i.e. the total PHCs mass measured from both liquid phase and solid phase) ranged from 372.7 + 11.9 mg to 379.6±36.4 mg, as compared to the initial TPH mass o f 431.5±11.4 mg. It can be found that the relatively higher TPH reduction (i.e. 13.8%) was achieved w ith the H 2 O 2 dosage o f 15 ml and 20 ml, equivalent to 600 m l/L and 800 m l/L o f 30% (w/w) hydrogen peroxide solution in the system. The am ount o f H 2 O 2 added into the system was m uch m ore excessive than the amount needed for the degradation o f all petroleum hydrocarbons. For example, if 5 ml o f H 2 O 2 was added, the number o f m oles o f hydrogen peroxide was about 50 mmol. The initial TPH mass in 1 g o f oily sludge was about 0.43g (±0.01g), and thus the 79 moles o f carbon atoms in hydrocarbons were about 29 mmol (i.e. assum ing that this oily sludge is m ade o f saturated hydrocarbons and that the mass o f carbon elem ent counts for about 80% o f the total mass o f PHCs), which was less than the moles o f H 2 O 2 present in system. However, further increase in the H 2 O 2 dosage to 10 ml, 15 ml, and even 20 ml, did not result in any further improvement in oxidation o f petroleum hydrocarbons (Fig.4.8). Table 4.4 TPH reduction in samples under Fenton process alone H20 2 dosage (ml) 5 10 15 20 TPH reduction rate 13.2% 12.0% 13.8% 13.8% 400 60 E U 3 00 a Pm ® Vi Vi 200 93 S 100 0 5 10 15 20 dosage of hydrogen peroxide (ml) Figure 4.8 R e m a in in g TPH mass in samples after Fenton process alone (error bar stands for standard deviation; the value at the dosage o f 0 stands for the initial TPH mass) Fenton’s reactions have been reported for the oxidization o f organic contaminants in w ater and soil, and the oxidation mechanism w as dem onstrated in the sequence o f reactions (Eqs. 14 to 18). The free hydroxyl radicals (-O H) have been proved to be essential for the degradation o f organic pollutants in the course o f Fenton’s process. These free radicals are 80 strong electrophilic and nonselective oxidizing species, with an oxidation potential (E°) o f 2.8. The pathways o f petroleum hydrocarbon destruction can be described below (Buxton et al., 1988; Von Sonntag and Schuchmann, 1997): •OH + R-H —> R* + H20 (27) •OH + C=C — > H O -C -O (28) — * F a t h e r reactio n s •OH+ OH / \ (29) H w here R represents carbon chain, and R- stands for the intermediate product o f carbon-centered radicals. The *OH radicals react w ith organic compounds by abstracting H from the C-H, N-H, or O-H bonds (Eq.27), adding to C=C bonds (Eq.28), or adding to the aromatic rings (Eq.29). In this study, the oily sludge contains a large am ount o f petroleum hydrocarbons, including straight-chain alkanes, branched alkanes, alkenes, and aromatic compounds. These hydrocarbons can react w ith *OH radicals in accordance w ith reactions (Eqs. 27- 29) during the Fenton’s process. R•+ 0 2 R -H + + HO- (30) R- + 0 2 -*• R -O O - ->-► R-O- (31) r .+ (32) OH -> ROH 2H20 2 0 2 + 2H20 (33) Sequentially, the intermediate products (R ) can either react with 0 2to generate peroxyl radicals (R-OO ) when air is present in the liquid or react with -OH to form alcohols (ROH) as the oxidation products (Adewuyi, 2005). The reactions o f carbon-centered radicals are described in Eqs.30-32. In addition, the net reaction o f Fenton’s process is the conversion o f H 2 O 2 into O 2 and w ater w ith iron as catalyst (Eq.33) (Pignatello et al, 2006). The decom position o f H 2 O 2 could also provide extra O 2 for the oxidation o f hydrocarbons. In the term ination step, the interm ediate radicals o f R-, R -O O , and R.-0* could couple or disproportion with each other, leading to the decom position o f hydrocarbons and the generation o f various by-products (Von Sonntag and Schuchm ann, 1997; Stark, et al 2011). M eanwhile, alcohols formed in reaction (Eq. 32) could further react w ith OH to form ketones, esters, and finally lead to carboxylic acids (Benner, et al, 2000). M any studies in the remediation o f petroleum contam ination confirm ed the occurrence o f various oxidation products after Fenton or Fenton-like treatment. For instance, Lu et al (2010b) investigated the use o f Fenton-like oxidation to treat soil contam inated w ith petroleum hydrocarbons. They reported that the concentration o f hydrocarbons was decreased in soil. However, residues in soil after Fenton treatment contained higher proportion o f m ore condensed compounds CnH 2 n+ z 0 2 with the z value o f much less than 0, indicating the presence o f fused-ring structure, as compared with com pounds initially present in contam inated soil (Lu et al., 2010b). As discussed above, although there are advantages o f the free-radical oxidation o f petroleum hydrocarbons in the course o f Fenton process alone, the TPH reduction in oily sludge through Fenton’s oxidation alone in this study was relatively low. This m ay be explained by the inadequate contact o f hydroxyl radicals with petroleum hydrocarbons. A large amount o f hydroxyl radicals were form ed in the aqueous phase under Fenton’s reaction process, while the majority o f petroleum hydrocarbons in o ily sludge w ere present in non-aqueous form or attached into solid particles. In such case, it was difficult for these 82 radicals to contact substantial petroleum hydrocarbon m olecules for achieving high degree o f PHC oxidation. A lternative m ethods such as ultrasonic irradiation m ay be required to improve the desportion and dispersion o f oil into the aqueous phase. M oreover, not all the hydroxyl radicals take part in in the oxidation o f PHCs, and som e amount o f them could be consumed by reacting w ith H 2 O 2 m olecules and Fe2+ ions, in w hich case H 20 2 and Fe2+ serve as radical scavengers (Eqs. 34 and 35). HO- + H 20 2 -* H O z- + H20 (33) HO- + Fe2+ -+ Fe3+ + O H - (34) In summary, the results o f TPH oxidation by Fenton’s reaction process indicated that the petroleum hydrocarbons in oily sludge could be oxidized by the free radicals to some extent, and the Fenton oxidation was considered as a relatively low effective approach for oily sludge treatm ent with regard to its TPH reduction rate.However, the Fenton’ reaction process is still a very attractive option to treat m any organic com pounds, especially for the resistant compounds w ith complex structures. W hen com paring w ith other oxidation processes, substantial hydroxyl radical are generated through Fenton’s reaction process, serving as strong and nonselective oxidants. The Fenton’s reagents (e.g., hydrogen peroxide and iron) are relatively inexpensive; the process is easily operated and does not require any energy supply. Therefore, it is reasonable to com bine conventional Fenton’s reaction process with other technologies to improve the radical-leading oxidation and/or enhance the cycle o f Fe2+ for further generation o f free radicals in the process. 83 3.1.3 TPH reduction through the com bination o f ultrasound and F enton’s reaction process (US/Fenton) Laboratory experiments were carried out to investigate the effect o f the com bination o f ultrasonic irradiation and Fenton’s reaction (US/Fenton) on the TPH reduction in oily sludge. 5 m inutes o f ultrasonic treatment tim e and the hydrogen peroxide dosage o f 15 ml w ere selected for the US/Fenton process based on the TPH reduction results o f each m ethod (i. e. US alone and Fenton alone). In Fig 4.9 is presented the remaining TPH m ass in oily sludge samples treated by three oxidation processes (i. e. US alone, Fenton alone, and U S/Fenton), and TPH reduction rate for samples is listed in Fig. 4.10. TPH m ass (measured from both the liquid and the solid phases) in the sample after US/Fenton treatment was 245.4±23.8 mg, w ith 43.1% TPH reduction (Fig 4.10). This reduction o f 43.1% in TPH by U S/Fenton is greater than the sum o f reduction rate by US alone for 5 m inutes and Fenton’s reaction alone with H 2 O 2 dosage o f 15 ml. This indicates the emergence o f som e synergistic effect w hen combining US w ith Fenton for the degradation o f petroleum hydrocarbons in oily sludge. Moreover, the reduction in different PHC fractions w as shown in Fig. 4.10. The reduction in Fraction 2 was less than 30% by US alone and by Fenton alone (i. e. 23.7% and 29.6%, respectively); the reduction o f 20.6% in Fraction 3 (F3) was observed for ultrasonic treatment alone and only 8.9% for Fenton’s reaction treatm ent alone; Fraction 4 (F4) was reduced by 36.9% for ultrasonic treatm ent alone and 12.7% for Fenton’s reaction treatm ent alone. However, during US/Fenton process the highest reduction o f 56.7% was observed for Fraction 2, followed by 46.5% for F4 and 39.1% for F3. 84 400 M £ 300 u a tO 200 yi s 100 Initial US/Fenton Fenton US M ethods Figure 4.9 TPH mass remained in samples after three different methods (US alone for 5 minutes; Fenton alone with 15 ml H20 2; US/Fenton with 5 minutes o f ultrasound and 15 ml o f H20 2) (Error bar stands for standard deviation) 70 □ US £ -I-* A 50 a 40 pa o « (j 3 T3 V oa □ Fenton H U S/Fenton 60 30 1 20 10 i 0 F2 F3 1 F4 1 PHCs Petroleum hydrocarbon Fractions Figure 4.10 TPH and fraction reduction via three different processes (US alone for 5 minutes; Fenton alone with 15 ml H:0 : ; US/Fenton with 5 minutes o f ultrasound and 15 ml o f H20 2) (error bars stand for standard deviation) As mentioned above, the use o f Fenton’ reagents alone cannot achieve high degree o f petroleum hydrocarbon oxidation due to the inefficient contact o f hydroxyl radicals in the 85 aqueous phase with hydrocarbon molecules in the oily sludge. However, the ultrasonic irradiation would significantly improve the desorption and dispersion o f hydrocarbons into the aqueous phase. As a result, com bining US and Fenton’s reaction process would provide w ith a large am ount o f hydroxyl radicals in situ and at the sam e time enhance the contact o f substantial radicals with hydrocarbons in the oily sludge, thereby increasing the oxidation efficiency. Another apparent benefit o f the com bined process w as to improve the recycling o f iron catalyst. As discussed before, ultrasonic irradiation could isolate Fe2+ from the com plex Fe-OOH formed through serial reactions catalyzed by Fe2+, and the isolated Fe2+ can in turn react with hydrogen peroxide to generate more free radicals for oxidization o f petroleum hydrocarbons. A t the same time, the application o f ultrasound during Fenton’s reaction process m ight help reduce the formation o f “ iron sludge” . In addition, the heterogeneous system m ight affect the TPH reduction under the com bined process. The system under the US/Fenton process w as heterogeneous due to the presence o f solid particles in oily sludge. The liquid-solid interface w ould provide extra active area for the reaction o f hydroxyl radicals with the absorbed organic compounds. A large amount o f hydrocarbons were attached on the solid surface and the radicals were largely generated in the bulk liquid. However, the m icrojets resulting from ultrasonic irradiation in such heterogeneous system could deliver the radicals from bulk liquid to the interface region where considerable hydrocarbons are present, thus enhancing the mass transfer in the system and improving the contact o f radicals w ith the target hydrocarbons. Moreover, as shown in Table 3.1, the oily sludge contains some metals such as Fe and Cu. These metals might be present at their oxidant state in solid particles. The solid particles containing metals could be regarded as extra catalyst for the generation o f hydroxyl radicals 86 if the hydrogen peroxide reached onto the solid surface. Furthermore, the cavitation o f ultrasound m ight increases the local tem perature and pressure and make it m ore convenient for the oxidation o f hydrocarbons under the com bined process. W hen the ultrasonic irradiation was applied in the system containing petroleum hydrocarbons, the energy released due to collapse o f cavitation bubbles can cause the cleavage o f carbon-carbon bonds in hydrocarbon m olecules, producing alkyl radicals. M eanwhile, reactions related to Fenton’s process can provide with a large amount o f hydroxyl radicals, w hich attack the hydrogen in hydrocarbon molecules to form alkyl radicals. Therefore, during the com bined process, the generation o f alkyl radicals would be improved by the thermal cleavage due to ultrasound and by the attacking o f hydrogen radicals associated with Fenton’s reactions. Consequently, m ore alkyl radicals are formed under combined process than that under either individual process, and these unstable intermediate radicals would either react w ith each other or w ith hydroxyl radicals to form m ore stable oxidation products. Thus the oxidation o f hydrocarbons would benefit from the combination o f ultrasound and Fenton process. In terms o f the degradation o f petroleum hydrocarbon fractions, the decom position o f long-chain hydrocarbons (i.e. Fraction 4) were more significant than the other two groups under US treatment alone (Fig 4.10). This indicates that the release o f ultrasonic energy can provide m ore convenient conditions for the cleavage o f heavy hydrocarbons. M eanwhile, as for the Fenton process alone (Fig 4.10), the effect o f F enton’s reactions on the degradation o f Fraction 2 (F2) was relatively m ore significant com pared w ith degradation o f F3 and F4, indicating that the lighter hydrocarbons are more easily decom posed under the attack o f free radicals. Combining US and Fenton process, therefore, appears to im prove the TPH 87 reduction in samples by enhancing the degradation o f petroleum hydrocarbon in all fractions (i.e. F2, F3, and F4). However, as for each o f the three treatments, the reduction rate for Fraction 3 is less than that for Fraction 4. This m ight be attributed to the accum ulation into Fraction 3 o f intermediate products resulting from the decom position o f heavy PHCs in Fraction 4. Some by-products from the degradation o f heavy hydrocarbons m ight not be easily for further decom position or even com plete mineralization. In summary, with the com bination o f ultrasonic irradiation with Fenton’s reaction, the degradation o f hydrocarbons in oily sludge was im proved in com parison to US alone and Fenton process alone and thus the combined US/Fenton process was proved to be a reliable and m ore effective m ethod to treat oily sludge. 3.1.4 Petroleum hydrocarbons distribution in samples and fractional analysis after oxidation processes The distribution o f petroleum hydrocarbons in the liquid and the solid phases after treatments was also investigated and Table 4.5 lists the results. D ue to the high viscosity o f oily sludge, most o f the petroleum hydrocarbons are present in the form o f non-aqueous phase or adsorbed onto the solid particles. As a result, the TPH mass in aqueous phase was very low before any treatm ent (i.e. 31.4 mg out o f 431.5 mg TPH in the initial samples with lg oily sludge and 25 ml water). However, the TPH mass in the aqueous phase was increased (i.e. 65.3 mg) after US treatment alone for 5 minutes, which indicated the em ulsification effect o f ultrasonic irradiation for the dispersion o f hydrophilic compounds. Furthermore, pH was decreased to 1.9 after US/Fenton process and some iron sludge was form ed (Fig 4.3). The petroleum hydrocarbons remained in the aqueous phase m ight need further treatm ent 88 such as bioremediation. The oxidation products, such as carboxyl com pounds and other oxidants, might be more am enable for biodegradation by m icroorganisms. However, the value o f pH for samples in this study was too low (about 2 for US/Fenton). The strong acidic condition is not in favor o f m icrobial activity for biodegradation. Therefore, pH should be adjusted if biorem ediation technologies would be subsequently applied for further contamination elimination. Table 4.5 TPH content in aqueous and solid phases o f samples* before and after advanced oxidation processes Fenton alone (15 ml o f H20 2) Oxidation process U S alone US/Fenton (5 minutes o f (15 ml H20 2 + 5minutes ultrasound) o f ultrasound) Initial TPH in aqueous phase (mg) 49.7 65.3 48.3 31.4 TPH in solid phase (mg) 322.0 268.7 197.2 400.1 TPH (aqueous + solid) 371.7 334.0 245.5 431.5 pH 2.3 5.0 1.9 * l g oily sludge in each sample was treated with different processes and the volume o f liquid was assumed 25 ml for all samples. □ F2, U S/Fenton, 16.71% □ F3, U S/Fenton, 74.28% 0 F2, Fenton, 18.40% □ F3, Fenton, 0 F2, US, 21.96% □ F3, US, 71.42% B F4, U S/Fenton, 7.89% □ F4, Fenton, 8.43% B F4, US, 6.62% IgggggliIflllliilsS 0 F2, Orignal, 21.72% 0 F3, Orignal, 69.46% □ F4, Orignal, 8.82% Figure 4.11 Fractional distribution o f petroleum hydrocarbons in samples before and after three different processes (US alone for 5 minutes; Fenton alone with 15 ml H20 2; US/Fenton with 5 minutes o f ultrasound and 15 ml o f H 20 2) 89 Fig 4.11 presents the distribution o f petroleum hydrocarbons in different fractions in samples before and after three oxidation processes: US, Fenton, and US/Fenton. The samples with the highest TPH reduction rate after each process w ere chosen for comparison. It can be found that there was difference in the distribution o f petroleum hydrocarbons in three fractions after oxidation treatments. For example, after the US/Fenton treatment, the percentage o f Fraction 2 was decreased to 17.0% from 21.7% and to 8.3% from 8.8% for Fraction 4, while the percentage o f Fraction 3 increased to 74.7% from 69.5%. Sim ilar patterns were observed in the samples after Fenton’s treatment. M eanwhile, for samples treated with US alone, the percentage o f F2 was 22.0% com pared with the value 21.7% in the original oil sludge samples, the percentage o f F3 increased to 71.4% from 69.5% , while the percentage o f F4 decreased to 6.6%. The observation o f increase in the percentage o f F3 in all the samples implied the accumulation o f by-products due to the oxidation/decom position o f F4 hydrocarbons. However, the structures o f petroleum hydrocarbons after oxidation treatments would be different from those in the oily sludge. The carboxyl acid m ight be formed as the main by-products. The alkanes and branched alkanes would also be oxidized into ketones and alcohols with less carbon numbers. Further study should be conducted to investigate the com position o f oxidation products in detail. 3.2. Degradation o f petroleum hydrocarbons after different treatments through US/Fenton process The TPH degradation results o f individual oxidation processes (i. e. US alone, Fenton alone, and US/Fenton) indicated the synergistic effect on the oxidation o f oily sludge by combining ultrasonic irradiation with Fenton’s reaction. It is thus o f importance to investigate 90 the effect o f different factors on the combined process (US/Fenton). The orthogonal experiments were further carried out in this study to exam ine the impact o f four factors on the degradation o f petroleum hydrocarbons through the com bined US/Fenton treatment. Table 4.6 lists the mass o f TPH and the m ass o f petroleum hydrocarbons (PHCs) in different fractions (F2, F3, and F4) in samples before and after US/Fenton treatm ent for each orthogonal experimental run, and Table 4.7 lists the degradation rate o f TPH and PH Cs in different fractions for each run. The S/N ratio was also listed in Table 4.7. It can be found from Table 4.6 that the TPH mass in the samples (m easured from both liquid phase and solid phase) was decreased to the range o f 22.6±3.5 m g/L to 91.6 ±3.8 mg from the initial TPH mass o f 170.2±17.2 mg (w ith 0.5 g o f oily sludge initially added in the sam ple and sludge content 20g/L) for treatments LI to L9. The TPH degradation rate was in the range o f 51.9% to 88.1% (Table 4.7). W hen 1 g o f oily sludge was added to the system (sludge content 40 g/L) for treatments L10 to L I 8, the initial TPH mass was increased to 444.6 ± 20.7 mg. The TPH mass remained in the samples dropped to the range o f 71.7 ± 13.9 m g to 256.6 ± 32.2 mg, and the TPH degradation rate ranged from 42.3% to 83.9%. The largest initial TPH mass (i.e. 638.5±22.6 mg) occurred for treatments LI 9 to L27 when 1.5 g of oily sludge was added in samples (sludge content 60 g/L). The rem aining TPH mass in samples o f these experimental runs ranged from 119.5±2.5 mg to 408.5±24.4 mg/L, and the corresponding TPH degradation rate ranged from 36.0% to 81.3%. Among all 27 runs, the best TPH reduction rate 88.1%was observed for treatment L3 w hen sludge content w as set at 20 g/L, the ratio o f H 2 O 2 to Fe2+ at 4:1, US time for 5 m inutes and US pow er set at 60W. 91 Table 4.6 Remaining TPH mass in samples after different US/Fenton treatments Mass o f PHCs in different fractions after different US/Fenton treatments (mg) Initial mass o f TPH and PHCs in different fractions (mg) Ll TD1I TPH SD for .r m . TPH F2 SD for F2 F3 SD for F3 F4 SD for F4 TPH SD for TPH F2 170.2 17.2 38.4 4.2 128.4 16.3 3.4 0.6 SD for F2 F3 SD for F3 F4 SD for F4 90.4 9.3 16.5 2.2 70.4 7.5 3.5 0.4 L2 23.2 1.1 2.6 0.2 20.0 1.2 0.6 0.2 L3 22.6 3.5 2.4 0.1 19.8 3.5 0.4 0.1 L4 47.9 6.7 7.6 0.9 38.1 5.7 2.2 0.1 L5 37.8 5.3 4.9 1.1 30.3 4.7 2.6 0.5 L6 52.4 5.1 8.4 2.0 42.5 3.5 1.5 0.1 L7 91.6 3.8 13.2 0.5 75.6 2.7 2.8 0.6 L8 65.9 12.67 11.1 3.4 53.0 9.4 1.8 0.2 L9 68.1 0.4 10.8 0.2 55.6 0.3 1.7 0.1 192.5 3.6 44.1 1.7 144.7 2.7 3.7 0.8 L10 444.6 20.7 100.1 4.2 337.9 16.3 6.59 2.3 LI 1 71.7 13.9 12.2 2.9 57.7 10.8 1.8 0.4 L12 238.9 35.8 39.4 7.9 195.6 27.2 3.9 0.7 0.3 L13 215.0 10.1 35.5 1.8 176.8 90.9 2.7 L14 248.4 15.2 44.6 0.2 200.5 91.3 3.3 0.1 LIS 139.3 24.9 21.9 5.5 114.3 18.5 3.1 0.9 1.0 204.3 33.3 4.1 0.1 L16 256.6 32.2 48.2 L17 200.7 30.2 36.3 3.2 160.8 33.0 3.6 0.4 L18 161.1 8.3 29.5 4.7 128.3 12.9 3.3 0.0 LI9 638.5 22.6 161.2 7.2 382.9 20.7 93.9 5.2 281.7 15.4 7.3 0.1 L20 L21 463.1 27.9 14.2 224.3 119.5 31.3 2.5 41.1 179.0 91.3 26.5 0.2 4.2 23.6 4.9 1.2 4.6 0.2 1.1 L22 398.0 16.9 91.3 5.2 298.2 12.6 8.5 0.9 L23 339.6 20.6 56.8 0.5 272.6 18.6 10.2 2.4 L24 212.4 5.7 40.8 1.7 165.8 3.8 5.8 0.2 L25 348.4 6.2 80.4 3.1 260.1 2.3 7.9 0.9 L26 348.2 9.9 78.1 1.2 261.6 8.9 8.5 0.1 L27 408.5 24.4 89.9 3.7 308.7 18.8 9.9 1.8 1.8 92 Table 4.7 Degradation o f petroleum hydrocarbons in samples after different US/Fenton treatments and the S/N ratio results Degradation rate (%) Treatment conditions Ll L2 L3 L4 Sludge content (g/L) 20 20 20 20 Ratio o f H20 2 to Fe2+ 4 4 4 10 US time(min) 1 3 5 1 US power (W) 20 40 60 40 S/N ratio TPH 52.6 87.8 88.1 74.8 F2 52.9 92.5 93.1 78.1 F3 53.6 86.6 86.9 74.9 F4 11.8 84.5 90.1 45.8 TPH 21.44 38.54 39.09 33.22 F2 34.47 39.32 39.38 37.85 F3 34.58 38.75 38.78 37.49 F4 21.44 38.54 39.09 33.22 31.17 L5 20 10 3 60 80.2 86.0 80.0 36.2 31.17 38.69 38.06 L6 20 10 5 20 72.5 75.8 71.8 70.9 37.01 37.59 37.12 37.01 L7 20 50 1 60 51.9 62.2 50.1 30.7 29.74 35.88 34.00 29.74 L8 20 50 3 20 65.5 68.1 65.0 56.1 34.98 36.66 36.26 34.98 L9 20 50 5 40 64.2 69.2 63.3 58.0 35.27 36.80 36.03 35.27 L10 40 4 1 40 56.7 56.0 57.2 43.5 32.77 34.96 35.15 32.77 L ll 40 4 3 60 83.9 89.1 84.6 71.4 37.07 39.00 38.55 37.07 Ll 2 40 4 5 20 46.4 61.0 42.1 40.9 32.23 35.71 32.49 32.23 L13 40 10 1 60 51.6 64.6 47.7 58.3 35.31 36.20 33.57 35.31 L14 40 10 3 20 44.1 55.5 40.7 50.3 34.03 34.89 32.19 34.03 36.42 34.58 L15 40 10 5 40 68.7 78.1 66.2 53.6 34.58 37.85 L16 40 50 1 20 42.3 51.8 39.5 38.1 31.62 34.29 31.93 31.62 L17 40 50 3 40 54.9 63.8 52.4 45.5 33.16 36.10 34.39 33.16 Ll 8 40 50 5 60 63.8 70.6 62.0 50.4 34.05 36.98 35.85 34.05 40.2 22.3 26.97 32.99 32.08 26.97 Ll 9 60 4 1 60 40.9 44.6 L20 60 4 3 20 64.9 74.5 61.3 70.7 36.99 37.44 35.75 36.99 L21 60 4 5 40 81.3 85.4 80.3 67.4 36.57 38.63 38.09 36.57 32.73 31.03 32.02 L22 60 10 1 20 37.7 43.3 35.6 39.9 32.02 L23 60 10 3 40 46.8 64.7 41.1 28.0 28.94 36.22 32.28 28.94 L24 60 10 5 60 66.7 74.7 64.2 59.8 35.53 37.47 36.15 35.53 L25 60 50 1 40 45.4 50.2 43.8 43.9 32.85 34.01 32.83 32.85 L26 60 50 3 60 45.5 51.6 43.5 39.6 31.95 34.25 32.77 31.95 L27 60 50 5 20 36.0 44.2 33.3 30.5 29.69 32.91 30.45 29.69 93 3.3. Im pact o f factors on TPH degradation through the combined (U S/Fenton) process The main effect o f individual factors on TPH degradation w as presented in Fig 4.12. The m ain effect plot (i.e. mean S/N ratios) in Fig 4.12 shows the contribution to the change o f response (i.e. S/N ratio) due to the change in one o f the influence factors from one level to another. It can be found that the S/N ratio decreased with the increase in sludge content and with the increase in the ratio o f H 2 O 2 to Fe2+, w hile S/N ratio w as increased w hen ultrasonic irradiation time was extended. However, the pattern o f S/N ratio change w ith ultrasonic power is different from that with the other three factors. The S/N ratio increased along w ith ultrasonic pow er increasing from level 1 to level 2 (i.e. from 20 W to 40 W), but it then slightly decreased w hen the pow er was increased to 60 W (i.e. level 3) from 40 W (i.e. level 2). The greater change in S/N ratio occurred w hen sludge content and US tim e increasing from level 1 to level 3, indicating that the sludge content and US time w ere the significant factors on petroleum hydrocarbon degradation using the US/Fenton process. And it is shown in Fig4.12 that the optimal condition for TPH reduction would be: sludge content at 20 g/L, the ratio o f H 2 O 2 to Fe2+ at 4:1, the ultrasonic irradiation tim e for 5 m inutes, and the ultrasonic pow er at 40 W, respectively. An interaction plot represents the interaction betw een two different factors w ith multiple levels. In an interaction plot, the levels o f one param eter are set on the x-axis and a separate line stands for the m ean S/N ratio o f each level for the other parameter. It can be very clear to illustrate the interaction o f two factors in the interaction plot: the further apart the separate lines, the stronger interaction between the two factors. However, if the separate lines are parallel to each other, no m atter how far they are apart from each other, there is no interaction 94 between the tw o factors. In this study, three individual interaction plots w ere obtained and presented in Fig 4.13a, b, and c. Since there are three levels for each factor, three curves were displayed in each plot, and each o f which represents the mean S/N ratio at one level for a factor. The degree o f interaction between the factors depends on the departure o f a curve from the trend o f the previous curve in the order o f successive increasing or decreasing o f the category factor (Torres et al, 2009; Wang et al., 2009). It was observed in Fig 4.13 that the higher degree o f interaction appeared betw een the ratio o f H 20 2 to Fe2+ and the US time. The minimum interaction occurred between the sludge content and the US time. Furtherm ore, the interaction plot o f the ratio o f H20 2 to Fe2+ and the US tim e showed that the greater TPH degradation was obtained when the ratio w as at 4:1 and the US treatm ent tim e for 3 minutes. 38 sludge content Ratio o f H 2 0 2 to Fe2+ ■*— US tim e US power 37 M ean 36 35 34 33 4:1 Sludge content (g/L) 10:1 50:1 Ratio o f H 2 0 2 to Fe2+ US tim e(m inute) US power(W ) F ig u re 4.12 Main effect plot o f factors on the TPH degradation through US/Fenton process 95 39 ♦ ratio 4:1 — ratio 10:1 —*— ratio 50:1 _© 37 M ean '•C C5 Wl 35 33 31 20g/L 40g/L Sludge content (g/L) 60 g/L —•— US tim e 1 m in —■— US tim e 3 m in —*— US tim e 5 m in Mean 39 37 u Xfl 35 33 31 20g/L 40g/L Sluge content (g/L) ■*— US tim e 1 min US tim e 3 min ■a— US tim e 5 min Mean 39 o 60g/L 37 4:1 10:1 The ratio o f H20 2 to Fe 2+ 50:1 F igure 4.13 Interaction o f factors on TPH degradation through US/Fenton process: (a) interaction o f sludge content and ratio o f H20 2 to Fe2*;(b) interaction o f sludge content and US time; (c) interaction o f ratio o f H20 2 to Fe2+ and US time 96 Statistical analysis o f the experimental data is necessary to further understand the oxidation treatm ent process in addition to the S/N ratio analysis. In this study, the analysis o f variance o f m eans (ANOVA) was carried out to verify the im pact o f the design factors and the interactions o f factors on the TPH degradation in oily sludge. The ANOVA was implemented by using MINITAB 16, and the results w ere show n in Table 4.8. There are several parameters generated during the ANOVA process, including the degree o f freedom (DF), the sequential sums o f squares (Seq SS), the adjusted sum o f squares (Adj SS), and the adjusted means squares (Adj MS). F-test was perform ed w ith 95% confidence interval. It was verified that sludge content and US time had significant im pact on TPH degradation in oily sludge as shown in Table 4.8 (i.e. with P value less than 0.05). The other two factors didn’t show significant impact on the US/Fenton treatm ent perform ance. The ANOVA results also illustrated that there was no significant interaction betw een factors to affect the TPH degradation, w ith the P values o f the interaction effects all greater than 0.1. Table 4.8 Analysis o f Variance for TPH degradation after different treatments through US/Fenton process Sources DF Seq SS Adj SS Sludge content 2 39.646 Ratio o fH 20 2toFe2+ 2 18.395 US time 2 24.982 US power 2 22.422 Sludge content*Ratio o f H20 2 to Fe2+ (a) 4 5.597 Sludge content*US time (a) 4 Ratio o f H20 , to Fe2+ *US tim e 2Fe3+ + ROH + H20 (36) 3.4. Im pact o f factors on degradation o f PH Cs in different fractions through the combined (US/Fenton) process In addition to the analysis o f TPH degradation results, it is also interesting to investigate the performance o f the combined process (i.e. US/Fenton) on the degradation o f petroleum hydrocarbons (PHCs) in different fractions. The residual mass and the degradation rate results o f three PHC fractions (i. e. Fraction 2, Fraction 3, and Fraction 4) w ere shown in Table 4.6 and Table 4.7, respectively. Tables 4.9, 4.10, and 4.11 list the ANOVA results for the degradation o f Fraction 2, Fraction 3, and Fraction 4, respectively. 102 Figs. 4.14 and 4.15 present the m ain effect plot and the interaction plots for the degradation o f Fraction 2 (F2 degradation), respectively. As shown in Fig 4.14, the pattern o f each factor for F2 degradation was similar to that for the TPH degradation (Fig. 4.12). As in Fig 4.14, F2 degradation was decreased w ith the increase in sludge content and the ration o f H 2 O 2 to Fe2+, and the better performance o f F2 degradation w as obtained w hen the sludge content was 20 g/L and the ratio o f H 2 O 2 to Fe2+w as 4:1. M eanwhile, the S/N ratio w as increased with the increase o f US time from 1 m inute to 3 minutes, but no significant enhancement was observed when the US time was extended to 5 minutes. Sim ilar pattern occurred for the effect o f US power. The enhancement o f S/N ratio was observed w hen the US pow er grew from 20 W to 40 W, but no further im provem ent w as found w hen the US pow er was increased to 60 W. W ith respect to the interaction o f factors for F2 degradation (Fig 4.15), it can be found that significant interaction occurred betw een the ratio o f H 2 O 2 to Fe2+ and the US time. The ANOVA result revealed the relative significance o f such interaction, w ith P value o f this interaction w as 0.058 (Table 4.9). Table 4.9 Analysis o f variance for Fraction 2 (F2) degradation after different US/Fenton treatments Sources DF Seq SS Adj SS Adj MS F P Sludge content 2 22.255 22.255 11.1275 13.49 0.006 Ratio o f H2O2t 0 Fe2+ 2 12.496 12.496 6.2481 7.58 0.023 US time 2 28.367 28.367 14.1834 17.20 0.003 US power 2 15.848 15.848 7.9239 9.61 0.013 Sludge content*Ratio o f H20 2 to Fe:' 4 2.147 2.147 0.5368 0.65 0.647 4 13.919 13.919 3.4798 4.22 0.058 6 4.948 4.948 0.8246 26 103.434 Total (a): * denotes the interaction between two factors 103 Table 4.10 Analysis o f variance for Fraction 3 (F3) degradation after different treatments through US/Fenton treatments Sources DF Seq SS Adj SS Adj MS F P Sludge content 2 51.230 51.230 25.6148 7.88 0.021 Ratio o f H20 2 to Fe2+ 2 21.616 21.616 10.8082 3.32 0.107 U S time 2 23.055 23.055 11.5275 3.55 0.096 U S power 2 26.365 26.365 13.1825 4.06 0.077 Sludge content*Ratio o f H20 2 to Fe2+ (a) 4 7.094 7.094 1.7735 0.55 0.710 Sludge content*US time 2 /Fe2+ ratio was set at 4:1, the US pow er w as set at 60 W, and the ultrasonic treatm ent time was set at 5 minutes. The sludge content and ultrasonic treatm ent tim e were confirm ed to have significant impact on the degradation o f petroleum hydrocarbons. There was no substantial interaction between factors w ith respect to total petroleum hydrocarbon (TPH) degradation. The TPH reduction rate w as decreased with the increase in sludge content, w hile TPH degradation increased as ultrasound treatm ent time increased. There was no significant interaction between factors w ith respect to TPH degradation. However, significant interaction w as observed between the ultrasonic treatm ent tim e and the H 2 C>2 /Fe2+ ratio w hen considering F2 degradation and F4 degradation. In summary, the advanced oxidation process by the combination o f ultrasonic irradiation with Fenton’s reagents w as proved to be an effective treatment approach to reduce petroleum hydrocarbons in oily sludge, and the degradation efficiency o f such m ethod could be im proved under optim al experimental conditions. Oxidation products o f petroleum hydrocarbons after advanced oxidation treatment m ight need further investigation. 112 Chapter 5 General Conclusion 113 Oily sludge is one o f the major wastes generated in the petroleum industry. Traditional methods to dispose o f oily sludge are relatively tim e-consum ing and inefficient. It is thus o f importance to explore novel approaches to treat oily sludge before direct disposal o f it into the environment. In this thesis research, three distinctive approaches w ere investigated for their capabilities to treat oily sludge. Chapter 2 is focused on the enhanced biorem ediation technology to im prove the degradation o f petroleum hydrocarbons when oily sludge was spiked into soil. O ne bacterium strain Luteibacter sp. was isolated from petroleum hydrocarbons contaminated soil and was then reintroduced into the soil to biodegrade the hydrocarbons, along w ith the addition o f nutrients and bio-surfactants (rhamnolipids). The impacts o f initial TPH content in soil/sludge mixture, nutrient addition, and the bio-surfactant w ere examined by the use o f Taguchi experimental design. The results indicated that the optimal biorem ediation condition for TPH reduction was: initial TPH content at 9690 mg/kg, nutrient addition at the ratio o f 100:50:10 for C:N:P, and the addition o f rham nolipid at 400 mg/kg. The degradation results o f the heavy petroleum hydrocarbons suggested that the isolated bacterium was effective for degrading long-chain PHCs. Chapter 3 investigates the oil recovery approach to treat oily sludge. Three methods, including ultrasonic irradiation alone, freeze/thaw treatm ent alone, and the treatm ent o f combining ultrasound with freeze/thaw, w ere applied to recover oil from the oily sludge. The results indicated that the combined method was considered as an effective approach to recover oil from sludge with regard to oil recovery efficiency and TPH concentrations in the recovered oil and in the separated wastewater. U nder the experimental conditions, oil recovery rate o f 64.2% was achieved by using the com bined treatment process, w ith TPH 114 concentrations o f 85.1% and 200 mg/L in the recovered oil and wastewater, respectively. W hen considering the impact o f different factors, the increase in oil recovery was observed as the ultrasonic power was set at a lower level o f 33 W, a sludge/water ratio o f 1:2, a rhamnolipids concentration o f 100 mg/L, and a N aC l concentration o f 1%, respectively. The m ain mechanism o f oil recovery enhancement w as found to be the im proved desorption o f PHCs from solid particles under ultrasonic irradiation. The study in Chapter 4 explores the ability o f advanced oxidation processes to degrade petroleum hydrocarbons in oily sludge. Three advanced oxidation methods were exam ined, including ultrasonic irradiation alone, Fenton’s reaction process alone, and the com bination o f ultrasonic irradiation and Fenton’s reaction. The highest reduction o f petroleum hydrocarbons was 43.1% under the combined process, indicating that the com bination o f ultrasound with Fenton’s reaction could be considered as an effective approach to rem ove petroleum hydrocarbons in oily sludge. The fractional analysis o f petroleum hydrocarbons indicated that the removal o f hydrocarbons in all three fractions was im proved by using the combined process. W hen evaluating the im pact o f individual factors on the com bined US/Fenton process, the best TPH degradation rate o f 88.1% was obtained w hen sludge content was set at 20 g/L, the H 2 (>2 /Fe2+ ratio w as set at 4:1 w ith US pow er at 60 W and US time for 5 minutes. The sludge content and ultrasonic treatm ent tim e were confirm ed to have significant impact on the degradation o f petroleum hydrocarbons. Limitations and future research In the study o f oil recovery technique, the individual impacts o f different factors on the combined treatment process were examined through a series o f laboratory experiments. 115 However, the interaction effects among these factors, the identification o f m ajor influencing factors, and the optimal combination o f these factors w ere not examined. This could be investigated through factorial experimental design m ethod in future studies. In addition, the recovered oil from the crude oil tank bottom sludge in this study was not a pure m ixture o f petroleum hydrocarbons, and m ay need further treatment. The value o f the recovered oil as fuel (such as asphaltene content, ash content, salt content, and heat o f com bustion) and the detailed cost/benefit o f the proposed treatm ent process were not analyzed, and should be examined in future studies. In summary, the com bined ultrasonic and freeze/thaw process could represent an environm entally friendly and econom ically competitive alternative for the effective treatment o f oily sludge waste from the petroleum industry, and is w orth o f further investigations. For the advanced oxidation techniques, the perform ance o f AOPs was exam ined by the reduction or degradation rate o f hydrocarbons and the results indicated that a large am ount o f hydrocarbons were oxidized. However, the presence o f oxidation products and interm ediates was not provided in this study. M ore evidence o f oxidation products m ight be necessary for the investigation o f com bining AOP on the oxidation o f various hydrocarbons at the same time. Furthermore, the combined ultrasound and Fenton process was proved to be an effective w ay to reduce contaminant concentration in oily sludge, but there is still some am ount o f hydrocarbons remained in samples after treatment. Further treatm ent o f such residue might be necessary to eliminate petroleum hydrocarbon pollution, such as the biorem ediation technologies. 116 REFERENCE Abramov, V. O., Myasnikov, S. K., & M ullakaev, M. S. (2009). Extraction o f bitum en, crude oil and its products from tar sand and contam inated sany soil under effect o f ultrasound. Ultrasonics Sonochemistry, 16, 408-416. Adewuyi, G. Y. (2005). Sonochemistry in environm ental remediation. 1 com binative and hybrid sonophotochemical oxidation processes for the treatm ent o f pollutants in water. Environtal Science & Technology, 39, 3409-3420. Adriana, U. S., & Nei, P. J. (2002). O ily sludge biotreatm ent. 9th A nnual International Petroleum Environmental Conference, (pp. 22-25). Albuquerque, NM . Al-Kdasi, A., Idris, A., Saed K., & Guan, C.T. (2004). Treatm ent of textile wastew ater by advanced oxidation processes-a review. Global Nest: the International Journal, 6 (3), 226-234. ASTM D1744, 1992. Standard test m ethod for determ ination o f water in liquid petroleum products by Karl Fischer reagent. A S T M International, West Conshohocken, PA. ASTM D5185, 2009. Standard test m ethod for determ ination o f additive elements, w ear metals, and contaminants in used lubricating oils and determination o f selected elements in base oils by Inductively Coupled Plasm a Atom ic Em ission Spectrometry (ICP-AES). A ST M International, West Conshohocken, PA. Avila-Chavez, M. A., Eustaquio-Rincon, R., Reza, J., & Trejo, A. (2007). Extraction o f hydrocarbons from crude oil tank bottom sludges using supercritical ethane. Separation Science and Technology, 42, 2327-2345. Banat, I. M., Makkar, R. S., & Cameotra, S. S. (2000). Potential commercial applications o f microbial surfactants. Applied M icrobiology and Biotechnology, 53, 495-508. 117 Benner, A. S., Devine G. K., M atveeva N. L, & Powell, H. D. (2000). The m issing organic molecules on Mars. Proceedings o f N ational Academ y o f Science, 97(6), 2425-2430. Bhattacharyya, J. K., & Shekdar, A. V. (2003). Treatm ent and disposal o f refinery sludges: Indian scenario. Waste M anagem ent & Research, 21, 249-261. Breitbach, M., Bathen, D., & Schmidt-Traub, H. (2003). Effect o f ultrasound on adsorption and desorption process. Industrial & Engineering Chemistry Research, 42, 5635-5646. Buyukkamaci, N., & Kucukselek, E. (2007). Im provem ent o f dewatering capacity o f a petrochemical sludge. Journal o f Hazardous M aterials, 144, 323-327. Buxton, G. V., Greenstock, C. L., Helman, W. P., & Ross, A. B. (1988). Critical review o f rate constants for reactions o f hydrated electrons, hydrogen atoms and hydroxyl radicals ( OH/ O -) in aqueous solutions. Journal o f Physical and Chemical Reference D ata, 17, 513-886. Calvo, C., Manzanera, M., & Silva-Castro, G. A. (2009). Application o f biorem ulsifiers in soil biorem ediation processes: Future prospects. Science o f the total environment, 407, 3634-3640. Cameotra, S. S., & Singh, P. (2008). Biorem ediation o f oil sludge using crude biosurfactants. International Biodeterioration & Biodegradation, 62, 274-280. Casero, D. S., Rubio, S., & Perez-Bendito D. (1997). Chem ical degradation o f aromatic amines by Fenton's reagent. Water Research, 31, 1985-1995. Castorena-Cortes, G., Roldan-Crrillo, T., & Zapata-Penasco, I. (2009). M icrocosm assays and taguchi experimental design for treatm ent o f oil sludge containing high concentration o f hydrocarbons. Bioresource Technology, 100, 5671-5677. CCM E 2000, (2000). Canada-W ide Standards for Petroleum H ydrocarbons (PHC) in Soil: 118 Scientific Rationale,” Canadian Council o f M inisters o f the Environment (CCM E). CCM E 2001, (2001). Reference method for the Canada w ide standard for petroleum hydrocarbons in soil - tier 1 method. Canadian Council o f M inisters o f the Environm ent (CCME). Cea, M., Jorquera, M., & Rubilar, O. (2010). Bioremediation o f soil contam inated w ith pentachlorophenol by anthracophyllum discolor and its effect on soil m icrobial community. Journal o f Hazardous M aterials, 181,315-323. Che, C. D., Wu, S L., Zhu, N. W., Shou, Z. Q., & Ye, Q. (2008). Study on the petroleum ether-extraction technique o f oil sludge. Journal o f Safety and Environment, 8, 56-58. Chen, L. L., Li, J. B., Ke, L., Zhang J., & Xiao H. N. (2011). Isolation o f biosurfactant producing microorganisms and their application to oily sludge rem ediation under low temperature conditions. International Symposium on Environmental Science and Technology (ISEST2011), pp 138-145, Doungguan, China, Science Press U SA Inc. Chand, R., Ince, H. N., Gogate, R. P., & Brem ner H. D. (2009). Phenol degradation using 20, 300 and 520 kH z ultrasonic reactors with hydrogen peroxide, ozone and zero valent metals. Separation and Purification Technology, 67, 103-109. Chen, G. H., & He, G. H. (2003). Separation o f w ater and oil from water-in-oil em ulsion by freeze/thaw method. Separation and Purification Technology, 31, 83-89. Cissokho, M., Boussour, S., Cordier, P., Bertin, H., & Hamon, G. (2009). Low salinity oil recovery on clayey sandstones: experimental study. International Sym posium o f the Society o f Core Analysts (SCA 2009-05), Noordwijk, N orw ay Collings, A. E , Farmer, A. D., & Gwan, P. B. (2006). Processing contam inated soils and sediments by high power ultrasound. M inerals Engineering, 19: 450-453. 119 Dewil, R., Baeyens, J., & Goutvrind, R. (2006). Ultrasonic treatm ent o f w aste activated sludge, Environmal Progress, 25, 121-128. Dukkanci, M., & Gunduz, G. (2006). Ultrasonic degradation o f oxalic acid in aqueous solutions. Ultrasonics Sonochemistry, 13,517—522. Ekp, M. A., & Udofia, U. S. (2008). Rate o f biodegradation o f crude oil by m icroorganism s isolated from oil. African Journal o f Biotechnology, 7, 4495-4499. Elektorowicz, M., & Habibi, S. (2005). Sustainable w aste management: recovery o f fuels from petroleum sludge. Cannadian Journal o f Civil Engineering, 32, 164-169. Elektorowicz, M., Habibi, S., & Chifrina, R. (2006). Effect o f electrical potential on the electro-dem ulsification o f oily sludge. Journal o f Colloid a n d Interface Science, 295, 535-541. Esplugas, S., Gimenez, J., Contreas, S., Pascual, E., & Rodriguez, M. (2002). Com parison o f different advanced oxidation processes for phenol degradation. Water Research, 36, 1034-1042. Feng, D., & Aldrich, C. (2000). Sonochem ical treatment o f simulated soil contam inated w ith diesel. Advances in Environmental Research, 4, 103-112. Ferrari, M. D., Neirotti, E., Albomoz, C., M ostazo, M. R. (1996). Biotreatm ent o f hydrocarbons from petroleum tank bottom sludges in soil slurries. Biotechnology Letters 18, 1241 - 1246. Franzetti, A., & Caredda, P. (2009). Potential applications o f surface active com pounds by Gordonia sp. Strain BS29 in soil rem ediation technologies. Chemosphere, 75, 801-807. Gaikwad, G. S., & Pandit, B., A. (2008). Ultrasound emulsification: effect o f ultrasonic and physicochemical properties on dispersed phase volum e and droplet size. Ultrasonics 120 Sonochemistry, 15(4), 554-563. G hazali, F. M., Ghazali, R. N. Z. A., Salleh, A. B., & Basri, M. (2004). B iodegradation o f hydrocarbons in soil by m icrobial consortium. International Biodeterioration & Biodegradation, 54, 61-67. Ghosh, S., & Coupland, J. N. (2008). Factors affecting the freeze-thaw stability o f emulsions. F ood Hydrocolloids, 22, 105-111. Gogate, R. P., & Pandit, B. A. (2004). A review o f im perative technologies for w astew ater treatment II: hybrid methods. Advances in Environmental Research, 8, 553-597. Gogate, P.R., Mujumdar, S., & Pandit, A.B. (2003). Sonochem ical reactors for w aste w ater treatment. Advances in Environm ental Research, 4, 283-299. Gopinath, R., Dalai, K. A., & Adjaye, J. (2006). Effects o f ultrasound treatm ent on the upgradation o f heavy gas oil. Energy & Fuels, 20, 271-277. Hamdaoui, O., & Naffrechoux, E. (2007). An investigation of the m echanism s o f ultrasonically enhanced desorption. A lC hE Journal, 53, 363-373. Hamdaoui, O., Naffrechoux, E., Sptil, J., & Fachinger, C. (2005). Ultrasonic desorption o f p-chlorophenol from granular activated carbon. Chemical Engineering Journal, 106, 153-161. Huesemann, M. H. (1994). Guidelines for land-treating petroleum hydrocarbon-polluted soils. Journal o f Soil Contamination, 3, 1-204. Jean, D.S., Lee, D. J., & Wu, J. C. S. (1999). Separation o f oil from oily sludge by freezing and thawing. Water Research, 33, 1756-1759. Kaushik, P., Kumar, A., Bhaskar, T., Sharma, Y. K., Tandon, D., & Goyal, H. B. (2012). Ultrasound cavitation technique for up-gradation o f vacuum residue. F uel Processing 121 Technology, 93, 73—77. Kim, Y. U., & Wang, M. C. (2003). Effect o f ultrasound on oil rem oval from soils. Ultrasonics, 41, 539-542. Kester, A. S. & Foster, J. W. (1963). Diterm inal oxidation o f long-chain alkanes by bacteria. Journal o f Bacteriology, 85, 859—869. Kuriakose, A. P., & M anjooran, S. K. B. (2001). Bitumenous paints from refinery sludge. Surface and Coatings Technology, 145, 132-138. Lager, A., Webb, K. J., & Black, C. J. (2006). Low salinity oil recovery-A n experim ental investigation. International Symposium o f the Society o f Core AnalystsfSC A 2006-36), Trondheim, Norway. Liang, J., Komarov, S., Hayashi N., & Kasai, E. (2007 a). Recent trends in the decom position o f chlorinated aromatic hydrocarbons by ultrasound irradiation and F enton’s reagent. Journal o f M aterial Cycles and Waste M anagement, 9, 47-55. Liang, J., Komarov, S., Hayashi, N., & Kasai, E. (2007 b). Improvement in sonochemical degradation o f 4-chlorophenol by com bined use o f Fenton-like reagents. Ultrasonics Sonochemistry, 14,201-207. Lin, C., & He, G. (2007). Freeze/thaw induced dem ulsiflcation o f water-in-oil em ulsions with loosely packed droplets. Separation and Purification Technology, 56, 175-183. Lin, J. J., Zhao, X. S., Liu, D., Yu, Z. G., Zhang, Y., & Xu, H. (2008). The decoloration and mineralization o f azo dye C. I. A cid Red 14 by sonochemical process: rate im provem ent via Fenton’ s reacations. Journal o f Hazardous Materials, 157, 541-546. Lu, M., Zhang, Z. Z., Qiao, W., Guan, Y. M., Xiao, M ., & Peng, C. (2010 a). Rem oval o f residual contaminants in petroleum -contam inated soil by Fenton-like oxidation. Journal 122 o f Hazardous Materials, 179, 604-611. Lu, M ., Zhang Z. Z., Qiao, W., & Wei, X. F. (2010 b). Rem ediation of petroleum-contaminated soil after com posting by sequential treatm ent w ith Fenton-like oxidation and biodegradation. Bioresource Technology, 101, 2106-2113. M achin-Ramirez, C., Okoh, A. I., M orales, D., & M ayolo-D eloisa, K. (2008). Slurry-phase biodegradation o f weathered oily sludge waste. Chemosphere, 70, 737-744. M ahamuni, N ., & Adewuyi, Y.G. (2010). Advanced oxidation processes (AOPs) involving ultrasound for waste w ater treatment: A review w ith emphasis on cost estimation. Ultrasonics Sonochemistry, 7, 990-1003. Mansano-Weiss, C., Cohen, H., & M eyerstein, D. (2002). Reactions o f peroxyl radicals with Fe(H20)62+. Journal o f Inorganic Biochemistry, 91, 199-204. Marin, J. A., Moreno, J. L., Hernandez, T., & Garcia, C. (2006). B iorem ediation by com posting o f heavy oil refinery sludge in semiarid conditions. Biodegradation, 17, 251— 261. Mansano-Weiss, C., Cohen, H., & M eyerstein, D. (2002). Reactions o f peroxyl radicals w ith Fe(HiO)62+. Journal o f Inorganic Biochemistry, 91:199—204 M asomboon, N., Ratanatamskul, C., & Lu, M. C. (2010). Chemical oxidation o f 2,6-dimethylaniline by electrochem ically generated F enton’s reagent. Journal o f H azardous Materials, 176, 92-98. Mason, T. J. (2007). Developments in ultrasound - Non-medical. Progress in Biophysics and M olecular Biology, 93, 166-175. M ason, T. J., & Lorimer, J. P. (2002). A pplied Sonochemistry: The Uses o f Pow er Ultrasound in Chemistry and Processing, Wiley-VCH, Weinheim. 123 M ohajerani, M., Mehrvar, M., & Ein-M ozaffari, F. (2010). Recent achievem ents in com bination o f ultrasonolysis and other advanced oxidation processes for w astew ater treatment. International Journal o f Chem ical Reactor Engineering, 8(R2), 1-78. Mulligan, N. C. (2005). Environmental applications o f biosurfactants. Environmental Pollution, 133, 183-198. Neppolian, B., Park, J., & Choi, H. (2004). Effect o f Fenton-like oxidation on enhanced ocidative degradation o f para-chlorobenzoic acid by ultrasonic irradiation. Ultrasonics Sonochemistry, 11, 273-279. Newman, A. P., Lorimer, J. P., Mason, T. J., & Hutt, K. R. (1997). A n investigation into ultrasonic treatment o f polluted solids. Ultrasonics Sonochemistry, 4, 153-156. Nitschke, M., Costa, S. G. V. A. O., & Contiero, J. (2005). Rhamnolipid surfactants: an update on the general aspects o f these rem arkable biom olecules. Biotechnology Progress, 21, 1593-1600. Noordman, W. H., & Janssen, D. B. (2002). Rham nolipid stim ulates uptake o f hydrophobic compounds by Pseudomonas aeruginosa. A pplied and Environmental M icrobiology, 68, 4502-4508. Nour, H. A., Sothilakshmi, R., & Nour, H. A. (2010). M icrow ave heating and separation o f water-in-oil emulsions: an experimental study. Intenrnational Journal o f Chemical Technology, 2, 1-11. Overcash, M. R., & Pal, D. (1979). Design o f landtreatm ent systems for industrial wastes: theory and practice. A nn Arbor Science, pp 152-219, A nn Arbor, M I. Pham, T. T. H., Tyagi, R. D., Brar, S. K., & Surampalli, R. Y. (2011). Effect o f ultrasonication and Fenton oxidation on biodegradation o f bis(2-ethylhexyl) phthalate (DEHP) in 124 w astewater sludge. Chemosphere, 82, 923-928. Pang, Y. L., Abdullah, Z. A., & Bhatia, S. (2011). Review on sonochemical m ethods in the presence o f catalysts and chemical additives for treatm ent o f organic pollutants in wasterwater. Desalination, 277, 1-14. Petrier, C., & Francony, A. (1997). Ultrasonic w aste-w ater treatment: incidence o f ultrasonic frequency on the rate o f phenol and carbon tetrachloride degradation. Ultrasonics Sonochemistry, 4, 295-300. Pignatello, J. J., Oliveros, E., & MacKay, A. (2006). Advanced oxidation processes for organic contaminant destruction based on the Fenton reaction and related chemistry. Critical Reviews in Environm ental Science and Technology, 36(1), 1-84. Rahman, K. S. M., Rahman, T. J., Kourkoutoas, Y., Petsas, I., M archant, R., & Banat, I. M. (2003). Enhanced biorem ediation o f n-alkane in petroleum sludge using bacterial consortium amended w ith rhamnolipids and micronutrients. Bioresource Technology, 90, 159-168. Rajakovic, V., & Skala, D. (2006). Separation o f water-in-oil emulsions by freeze/thaw m ethod and microwave radiation. Separation and Purification Technology, 49, 192-196. Ramaswamy, D., Kar, D. D., & De, S. (2007). A study on recovery o f oil from sludge containing oil using froth flotation. Journal o f Environmental M anagement, 85, 150-154. Reynolds, V. R., & Heuer, S. R. (1993). Process for the recovery o f oil from waste oil sludges. U.S. P a te n t: 5269906. Ron, E. Z., & Rosenberg, E. (2001). Natural roles o f biosurfactants. Environm ental Microiology, 3, 229-236. 125 Ronen, Z., Horvath-Gordon, M. & Bollag, J. M. (1994) Biological and chem ical mineralization o f pyridine. Environmental Toxicology and Chemistry, 13, 21-26. Schueller, B. S., & Yang, R. T. (2001). U ltrasound enhanced adsorption and desorption o f phenol on activated carbon and polym eric resin. Industrial & Engineering Chemistry Research, 40, 4912-491. Sedlak, L. D., & Andren, W. A. (1991). Oxidation o f chlorobenzene with Fenton’s reagent. Environmental Science & Technology, 25(4), 777-782. Shie, J., Chang, C., Lin, J., & Wu, C. (2000). Resources recovery o f oil sludge by pyrolysis: kinetics study. Journal o f Chemical Technology & Biotechnology, 75, 443-450. Sivasankar, T., & Moholkar, V. S. (2009) Physical insights into the sonochemical degradation o f recalcitrant organic pollutants w ith cavitation bubble hynamics. Ultrasonics Sonochemistry, 16,769-781. Stark, M. S., Wilkinson, J. J., Smith J. R. L., Alfadhl, A., & Pochopien A. B. (2011). Autoxidation o f branched alkanes in the liquid phase. Industrial & Engineering Chemistry Research, 50, 817—823. Sun, H. J., Sun, S. P., Sun, J. Y., Sun, R. Z., Qiao, L. P., Gua, H. Q., & Fan, M. H. (2007). Degradation o f azo dye acid black 1 using low concentration iron o f Fenton process facilitated by ultrasonic irradiation. Ultrasonics Sonochemistry, 14, 761-766. Suslick, S. K. (1990). Sonochemistry. Science, 247, 1439-1445. Suslick, K. S., Hammerton D. A., & Cline R. E. (1986). Sonochemical hot spot. Journal o f the American Chemical Society, 108(18), 5641-5642. Suri, R. P. S., Singh, T. S., & Aburi S. (2010). Influence o f alkalinity and salinity on the sonochemical degradation o f estrogen horm ones in aqueous solution. Environm ental 126 Science & Technology, 44, 1373-1379. SW-846 EPA (1993).Test methods for evaluating solid wastes: physical/chem ical methods, Publication 955-001-00000-1. United States Environmental Protection A gency (EPA), Office o f Solid Waste (OSW ), U.S. Governm ent Printing Office, Washington, DC. Taiwo, E. A., & Otolorin, J. A. (2009). Oil recovery from petroleum sludge by solvent extraction. Petroleum Science and Technology, 27, 836-844. Tekin, H., Bilkay, O., Ataberk, S. S., & Balta, H. T. (2006). U se o f Fenton oxidation to improve the biodegradability o f a pharmaceutical wastewater. Journal o f Hazardous Materials, 136, 258-265. Torres, A. R., Abdelmalek, F., Combet, E., Petrier, C., & Pulgarin, C. (2007 a). A comparative study o f ultrasonic cavitation and Fenton’s reagent for bisphenol A degradation in deionised and natural waters. Journal o f Hazardous M aterials, 16, 546-551. Torres, A. R., Petrier, C., & Combet, E. (2007 b). Bisphenol a mineralization by integrated ultrasound-UV-iron (II) treatment. Environmental Science & Technology, 41, 297-302. Torres, A. R, Mosteo, R., Petrier, C., & Pulgarin, C. (2009). Experimental design approach to the optimization o f ultrasonic degradation o f alachlor and enhancement o f treated w ater biodegradability. Ultrasonics Sonochemistry, 26, 425-430. Villa, R. D., Trovo, A. G., & N ogueira, F. R. (2008). Environmental im plications o f soil remediation using the Fenton process. Chemosphere, 71(1), 43-50. Virkutyte, J., Vickackaite, V., & Padarauskas, A. (2010). Sono-oxidation o f soils: degradation o f naphthalene by sono-Fenton-like process. Journal o f Soils and Sediments, 10, 526-536. Von Sonntag, C., & Schuchmann, H.P. Peroxyl radicals in aqueous solutions. (1997) In Peroxyl Radicals, Alfassi Z. B., ed., John W iley and Sons, N ew York, 173-234, 127 Walworth, J., Pond, A., & Snape, I. (2007). N itrogen requirem ents for m axim izing petroleum biorem ediation in a sub-Antarctic soil. C old Regions Science and Technology, 48, 84-91. Wang, Y. Q., Afsar, A. Md., Song, J. (2009). O ptim ization o f brazing conditions for OFHC Cu and ASTM A501 low carbon steel by Taguchi method. International Journal o f Precision Engineering and M anufacturing, 10, 97-104. Watts, J. R., & Stanton, C. P. (1999). M ineralization o f sorbed and NAPL-phase hexadecane by catalyzed hydrogen peroxide. Water Research, 33(6), 1405-1414. Xu, N., Wang, W. X., Hang, P. F., & Lu, X. P. (2009). Effects o f ultrasound on oily sludge deoiling. Journal o f Hazardous Materials, 171,914-917. Ye, G., Lu, X., Han, P., Peng, F., Wang, Y., & Shen, X. (2008). Application o f ultrasound on crude oil pretreatment. Chemical Engineering and Processing, 47, 2346-2350. Zubaidy, E. A. H., & Abouelnasr, D. M. (2010). Fuel recovery from waste oily sludge using solvent extraction. Process Safety and Environm ental Protection, 88, 318-326. 128