Optimising coagulation and ozone pre-treatments and comparing the efficacy of differently pre-treated grain distillery wastewaters in an upflow anaerobic sludge blanket (UASB) reactor

Robertson, Louise (2014-04)

Thesis (MScFoodSc)--Stellenbosch University, 2014.

Thesis

ENGLISH ABSTRACT: Large volumes of high strength wastewater are generated annually by distilleries. Treatment of this wastewater is essential to increase its reuse potential. Anaerobic digestion (AD) has been widely applied as high strength wastewater treatment. Additionally, grain distillery wastewater (GDWW) is high in fats, oils and grease (FOG) which can cause problems during AD. Pre-treatment is therefore often required to make the AD process more efficient. The overall objective of this study was to investigate the operational efficiency of upflow anaerobic sludge blanket (UASB) reactors while treating GDWW as substrate. The first aim was to investigate the pre-treatment of the GDWW specifically for the removal of FOG by evaluating the effect of pH adaption, coagulant concentration and ozone (O3) dosages on the FOG reduction efficiency. Secondly, the effect of two different pre-treatments (only coagulant and coagulant and ozone) on the subsequent UASB treatment step was investigated. The pH of raw GDWW (pH 3.4) was adapted to three different pH values (5.0, 6.0, and 7.0) and the coagulant, aluminium chlorohydrate (ACH) (140 mg.L-1), was added. To make the process more economically viable, the lowering the coagulant concentration (to 100 mgACH.L-1) was also investigated. Optimal reductions for chemical oxygen demand (COD) (33.2% ± 4.93), total suspended solids (TSS) (91.9% ± 1.73) and FOG (84.1% ± 1.98) were, however, achieved at a higher coagulant concentration of 140 mgACH.L-1, and at a wastewater pH of 7.0. The effect of additional ozone treatment was also investigated. Maximum reductions for the ozone treatment were obtained at 100 mgO3.L-1 for COD (3.6% ± 4.08), and at 900 mgO3.L-1 for TSS (27.7% ± 5.58) and FOG (23.9% ± 1.83). The ozone treatment was most efficient for FOG reduction (in terms of mg FOG reduced per mg ozone) at 100 mgO3.L-1. An ozone dosage of 300 mgO3.L-1, was decided on based on economic feasibility, findings in literature on toxicity of ozone and the potential degradation of recalcitrant compounds at this dosage. The final pre-treatment thus included pH adaption to 7.0, coagulant dosage of 140 mgACH.L-1, and an ozone dosage of 300 mgO3.L-1. The second part of this study involved the operation of two 2 L laboratory scale UASB reactors for 277 days. The substrate of the first reactor contained GDWW that had only undergone coagulant pre-treatment (Rcontrol), while the substrate of the second UASB reactor consisted of GDWW that had undergone coagulant and ozone pre-treatment (Rozone). Both reactors treated the pre-treated GDWW successfully at ca. 9 kgCOD.m-3d-1. COD reductions of 96% for Rcontrol and 93% for Rozone, were achieved. FOG reductions (%) showed variations throughout the study and maximum reductions of 88% and 92% were achieved for Rcontrol and Rozone, respectively. The Rozone produced more biogas, but the methane content was similar for both reactors. The additional ozone pre-treatment did not show any added benefits to the reactor performance results. UASB granule washout in Rcontrol did, however, suggest possible toxicity of unsaturated fatty acids present in the non-ozonated substrate. The feasibility of FOG removal was shown as both reactors successfully treated pre-treated GDWW. Ozonation, after a coagulant dosage, which resulted in further reduction in the FOG content of the GDWW, is thus not essential to ensure the success of an anaerobic digestion step. Ozonation of the pre-treated GDWW could, however, be beneficial to gas production and the efficiency of a tertiary biological process.

AFRIKAANSE OPSOMMING: Groot volumes hoë-sterkte afloopwater word jaarliks deur distilleerderye opgelewer. Die behandeling van hierdie afloopwater is noodsaaklik om die hergebruiksmoontlikheid daarvan te verhoog. Die toepassing van anaërobiese vertering (AV) is wydverspreid in hoë-sterkte afloopwaterbehandeling. Graandistillerings-afloopwater (GDAW) is boonop hoog in vette, olies en ghries (VOG), wat probleme kan veroorsaak tydens AV. Voorbehandeling word dus dikwels verlang om die AV meer doeltreffend te maak. Die oorhoofse doelstelling van hierdie studie was om die doeltreffendheid van opvloei-anaërobieseslykkombers- (OAS-) reaktore wat GDAW as substraat behandel, te ondersoek. Die eerste mikpunt was om die voorbehandeling van die GDAW, te ondersoek vir die verwydering van VOG. Dit is uitgevoer deur die uitwerking van pH aanpassing, koagulantkonsentrasie en osoon(O3)dosis op VOG vermindering te evalueer. Tweedens is die uitwerking van twee verskillende voorbehandelings (slegs koagulant asook koagulant en osoonbehandeling) op die opvolgende OAS-behandelingsstap ondersoek. Die pH van rou GDAW (pH 3.4) is aangepas tot drie verskillende pH waardes (5.0, 6.0 en 7.0) en die koagulant, aluminium-chlorohidraat (ACH), is bygevoeg (140 mg.L-1). Om die proses meer ekonomies uitvoerbaar te maak is ‘n verlaagde koagulantkonsentrasie (verlaag tot 100 mgACH.L-1) ook ondersoek. Die optimale afnames vir chemiese suurstofvereiste (CSV) (33.2% ± 4.93), totale oplosbare vastestowwe (TOV) (91.9% ± 1.73) en VOG (84.1% ± 1.98) is egter bereik teen ‘n hoër koagulant konsentrasie van 140 mgACH.L-1, en teen ‘n afloopwater-pH van 7.0. Die uitwerking van ‘n bykomende osoonbehandeling is ook ondersoek. Die hoogste afnames tydens die osoonbehandeling is bereik teen 100 mgO3.L-1 vir CSV (3.6% ± 4.08), en teen 900 mgO3.L-1 vir TOV (27.7% ± 5.58) en VOG (23.9% ± 1.83). Die osoonbehandeling was mees doeltreffend vir VOG (in terme van mg VOG verwyder per mg osoon) teen 100 mgO3.L-1. Daar is besluit op ‘n van 300 mgO3.L-1, as gevolg van die ekonomiese uitvoerbaarheid, bevindinge in literatuur vir die toksisiteit van osoon, en die moontlike afbraak van moeilik-afbreekbare komponente teen hierdie dosis. Die finale voorbehandeling het dus bestaan uit ‘n aanpassing van die afloopwater-pH na 7.0, ‘n koagulantdosis van 140 mgACH.L-1, en ‘n osoondosis van 300 mgO3.L-1. Tydens die tweede gedeelte van hierdie studie is twee 2 L laboratoriumskaal OAS reaktore bedryf vir 277 dae. Die substraat van die eerste reaktor het GDAW bevat wat slegs ‘n koagulant-voorbehandeling ondergaan het (Rkontrole), terwyl die substraat van die tweede OAS-reaktor GDAW bevat het wat koagulant- en osoon-voorbehandeling ondergaan het (Rosoon). Beide reaktore het die voorbehandelde-GDAW suksesvol behandel teen ‘n organieseladingstempo van ca. 9 kgCSV.m-3d-1. Afnames in CSV van 96% vir Rkontrole en 93% vir Rosoon, is bereik. Tydens die studie is deurlopende wisseling in VOG verwydering (%) waargeneem, en die hoogste verwyderings wat bereik is, is onderskeidelik 88% en 92% vir Rkontrole en Rosoon. Die Rosoon het meer biogas geproduseer, maar die metaanpersentasie was soortgelyk vir beide reaktore. Die osoon-voorbehandeling het nie enige toegevoegde voordele getoon in die reaktorprestasie resultate nie. Die uitwas van die OAS-granules vir die Rkontrole het egter moontlike toksisiteit van onversadigdevetsure aangedui, wat moontlik teenwoordig kon wees in die nie-geösoneerde substraat. Die uitvoerbaarheid van VOG verwydering is getoon aangesien beide reaktore voorbehandelde GDAW suksesvol behandel het. Osonering, wat verdere afname in die VOG inhoud van GDAW teweeggebring het (na ‘n koagulant dosis), is dus nie noodsaaklik vir die suksesvolle verloop van ‘n anaërobiese-verteringsstap nie. Osonering van die voorbehandelde GDAW kan egter voordelig wees vir gasvorming, en kan ook verder die doeltreffendheid van ‘n tersiêre biologiese behandeling verhoog.

Please refer to this item in SUNScholar by using the following persistent URL: http://hdl.handle.net/10019.1/86286
This item appears in the following collections: