Removal of odor emissions from food fermentation and petrochemical production processes with using biological treatment methods

Laburpena

Up to date, air pollution, which includes contamination of both indoor and outdoor air, has received less attention compared with other forms of pollution such as water or soil pollution. Howewer, the increasing public concern on environmental challenges and human wellbeing has driven the attention of the public administration to air pollution management. Although some air pollutants are emitted naturally from volcanoes or forest fires, most sources are originated from human activities (transportartion, industrial activity, waste management, etc.). Thus, abatement and minimization of anthrophogenic air pollution have become one of the main challenges in air pollution management in this XXI century. Nowadays malodorous emissions are considered a part of the main components of air pollution. Although odor threshold levels are tightly dependent on human sensation, and in some cases malodours can not be detected even at trace concentrations, malodorous emissions can contain toxic materials for humans and the environmental even if concentrations remain below their odour thresholds. Food and petrochemical industries are major sources of pollutant emissions to atmosphere. Most of these emissions can cause odor nuisanse in the surrounding residental areas of industries, while long-term exposure to these emissions result on symptoms such as emotional stresses. At this point, an accurante characterization of malodorous emissions is crucial for a proper and effective odor abatement in the industrial sector. However, there is a lack of comparative data assessing the proper characterization of volatile organic compounds (VOCs) and malodous compounds in literature. As a result of the sensorial impact of malodorous VOC, the characterization of emissions by instrumental techniques such as gas chromatography coupled mass spectrometry does not typically provide a sufficient understanding of the direct impact of air pollution on humans. In this context, sensorial techniques such as dynamic olfactometry, based on the use of the human nose as a sensor, allow to determine the odor threshold and the concentration/intensity of odor emissions, which ultimately characterize the direct impact of malodous VOCs on people. The first study of this thesis consisted of a case study carried out for bakery yeast fermentation process. The emissions of this process were analyzed with instrumental and dynamic olfactometry methods to characterize malodorous VOCs. The emissions were treated in a pilot scale biofilter. Instrumental analyses were performed to determine the chemical composition of waste gas emissions for a gas flowrate of 6500 m3 h-1. Ethanol represented the main VOC of the process with a peak hour concentration of 764 mg m-3, followed by acetaldehyde and acetone as 331 and 65 mg m-3, respectively. Moreover, only trace propanol levels were recorded only at peak hours in a 17-hour fermentation cycle. Dynamic olfactometry analyses were also performed to quantify the sensory impact of malodorous VOC emissions. The odor concentration was recorded in terms of European Odor Unit (OUE) as 39725 OUE m-3, which was almost 40 times higher than the regulatory limits in Turkey. At this point, a biofilter was selected as a suitable alternative for the abatement of dilute VOC emissions (< 2000 mg VOC m–3) at waste gas treatment capacities of the process. Thus, malodorus emissions from the fermentation process were treated in a 46 L biofilter packed with raschig rings with a surface area 400 m2 m-3 at an empty bed residence time (EBRT) of 80 s with ethanol concentrations 200, 400 and 700 mg m-3 with an almost complete biodegradation regardless of the ethanol concentration. On the other hand, ternary waste gas streams containg acetaldehyde (~ 300 mg m-3), acetone (70 mg m-3) and ethanol (700 mg m-3) were treated with removal efficiencies higher than 90 % for ethanol and acetaldehyde at EBRT of 53-80 seconds, while only a partial acetone removal which reached to its maxium with 70 % of RE was observed. An improvement in the abatement of acetone and in the VOC mass transfer in order to decrease the operational EBRT will be required to achieve a proper process optimization. Petrochemical industry was also analysed as a model sector as a result of its major contribution to the emissions of various sulfur based organic compounds such as ethanethiol (ET) and also human and environmental toxic compounds such as benzene, toluene, ethylbenzene, xylene (BTEX), which are used as reagents for the synthesis of multiple C-based products. To date, petrochemical emissions are well characterized but there is a lack of studies in literature assessing the abatement of O2-free petrochemical industry emissions. A new generation of O2-free biological abatement alternatives capable of coping with the potential explosion risks of petrochemical emissions is required. At this point of concept, ET emissions which are commonly present in petrochemical industry, were investigated due to its low odor threshold level as 0.7 μg L-1 and its flammable character at extremely low concentrations. The use of an anoxic bio-scrubber for ET biotreatment enables the integration of waste gas treatment and wastewater treatment. The nitrification step uses NO3- as an electron acceptor source to oxidize ET under a specific ET/ NO3- ratio (0.74 in theory) to the main end product of Sº, while during denitrification NO3- is converted to N2 gas. In order to optimize process design and operation for ET emissions removal, anoxic bioscrubber operation was examined as a function of inlet concentration (150, 350, 850, 1450 mg m-3), EBRT (30, 60, 90,120 s) and spray density of irrigation (0.12, 0.18, 0.23, 0.30, 0.45 m3 m-2 h-1). According to the results, the best operation conditions were achieved at an inlet concentration of 150 mg m-3, a spray density of 0.23 m3 m-2 h-1and an EBRT of 90 s. Under these conditions, an average RE of 91% and an elimination capacity (EC) of 24.74 g m-3 h-1 were recorded, while Sº was obtained as end product rather than SO4-2. Average experimental yield values close to the YET/NO3- theoretical value of 0.74 were observed. Apart from ET emissions in petrochemical industry, BTEX emissions widely used in the sector present a major impact due to their cancinogenic effects. Nevertheless, presence of multiple BTEX compounds and their interactions during the biodegradation process are still poorly understood. Similarlly, kinetic parameter estimation is of key relevance for a correct dimensioning of O2 free VOC abatement bioreactors. Anoxic mineralization of BTEX represents a promising alternative for the abatement of these VOCs from O2-deprived emissions. However, the kinetics of anoxic BTEX biodegradation and the interactions underlying the treatment of BTEX mixtures are still unknown. An activated sludge inoculum was used for the anoxic abatement of single, dual and quaternary BTEX mixtures, being acclimated prior performing the biodegradation kinetic tests. The Monod model and a Modified Gompertz model were then used for the estimation of the biodegradation kinetic parameters. Results showed that both toluene and ethylbenzene are readily biodegradable under anoxic conditions, whereas the accumulation of toxic metabolites resulted in partial xylene and benzene degradation when present either as single components or in mixtures. Moreover, the supplementation of an additional pollutant always resulted in an inhibitory competition, with xylene inducing the highest degree of inhibition. The Modified Gompertz model provided an accurate fitting for the experimental data for single and dual substrate experiments, satisfactorily representing the antagonistic pollutant interactions. Finally, microbial analysis suggested that the degradation of the most biodegradable compounds required a lower microbial specialization and diversity, while the presence of the recalcitrant compounds resulted in the selection of a specific group of microorganisms. The continuous biodegradation of BTEX (benzene, toluene, ethylbenzene and xylene) using nitrate as the electron acceptor is of key interest to reuse the residual gas for inertization purposes. However, the biological mineralization of BTEX is often limited by their recalcitrant nature and the toxicity of the secondary metabolites produced. The potential of an anoxic biotrickling filter for the treatment of a model O2-free BTEX-laden emission at inlet individual concentrations of ~ 700 mg m− 3 was here evaluated. A UV oxidation step was also tested both in the recycling liquid and in the inlet gas emission prior to biofiltration. Removal efficiencies > 90% were achieved for both toluene and ethylbenzene, corresponding to ECs of 1.4 ± 0.2 g m− 3 h− 1 and 1.5 ± 0.3 g m− 3 h− 1, respectively, while ~ 45% of xylene (EC = 0.6 ± 0.1 g m− 3 h− 1) was removed at a liquid recycling rate of 2 m h− 1. Benzene biodegradation was however limited by the accumulation of toxic metabolites in the liquid phase. The oxidation of these intermediates in the recycling liquid by UV photolysis boosted benzene abatement, achieving an average EC of 0.5 ± 0.2 g m− 3 h− 1 and removals of ~ 40 %. However, the implementation of UV oxidation as a pretreatment step in the inlet gas emission resulted in the deterioration of the BTEX biodegradation capacity of the biotrickling filter. Finally, a high bacterial diversity was observed throughout the entire experiment, the predominant phyla being Proteobacteria and Deinococcus-thermus. Another case study was carried out to remove trimethylamine emissions from a chemical production facility. For this purpose, the performance of a bacterial bioreactor and an algal-bacterial photobioreactor was comparatively evaluated. Thus, a bacterial Bubble Column Photobioreactor (BCB) was operated with a synthetic TMA laden emission at ~ 500 mg m-3 at an EBRT of 2 min, which resulted in 78 % of TMA oxidation. The operation of the BCB with an algal-bacterial consortium increased TMA removal by 10 % at an EBRT of 2 min, and supported a decrease in the EBRT down to 1.5 min and 1 min without a detrimental effect on TMA biodegradation (REs of 98 ± 2 % and 94 ± 6 %, respectively).