Substrate and nitrogen conversions in rotating biological filters | |
| Author | Ely Anthony Rosales Quano |
| Call Number | AIT Diss. no.D10 |
| Subject(s) | Filters and filtration Sewage--Purification--Filtration |
| Note | A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Engineering |
| Publisher | Asian Institute of Technology |
| Abstract | Studies were conducted on a rotating biological filter unit con- structed like an Imhoff tank with rotating contact media incorporated in the sedimentation chamber, in this case called the aerobic chamber. Rotation of the contact media kept the upper chamber aerobic and alternately exposed the microbial slimes on the media to the atmosphere and to the wastewater. The sludge hopper under the aerobic chamber served as an anaerobic digester for the biological solids sloughing off the media in the aerobic chamber. A number of design criteria for rotating disc filters have been developed empirically for sewage and bottling plant wastewater. These empirical equations c0uld not be applied readily to changes in the concentration and type of wastewater, changes in the geometric shape of the contact media and other operating variables, without repeating a series of experiments involving combinations of the operating variables. The present research attempted to formulate the basic kinetic equations for substrate degradation and nutrient conversion, oxygen mass transfer and consumption, and mass balance of materials entering and leaving the different parts of the system. With these fundamental equations, the design of rotating biological filter units can be approached rationally after the kinetic constants and mass transfer coefficients have been determined. The optimal operating conditions for any particular situation can be readily estimated mathematically using these The substrate concentratiOns at a particular stage in the aerobic chamber of the rotating biological filter can be expressed as: Lm = L (1 +1<0 AN 9) m-l , . . . where, L is the substrate concentration in the prev1ous stage, mg/l of COD, Lm is the substrate concentration in the mth stage, mg/l of COD, A/V is the areal density of media in the aerobic chamber, K0 is the kinetic constant, and 9 is the detention time. The conversion of organic nitrogen to ammonia nitrogen can be represented by: K A/VQNmLm m-l m _ 1 0 || 2 N 0 ° K2+Nmwhere, Nom is the organic nitrogen concentration at the mth stage, ms/l. Nam-1 is the organic nitrogen concentration at the previous stage, mg/l, and- KlaKQ are the kinetic constants. If the organic nitrogen concentration is very small, the denominator approaches a constant Value so that the organic nitrogen conversion equation can be expressed as: _ m m — K1 A/V NO L A similar phenomenon was observed if the substrate concentration to organic nitrogen concentration ratio was very high. The ammonia nitrogen conversion can be expressed as: N m-l N m K A/V N m Lm a ' a _ 1 o m m e — -——m + K3 Na L (A/V)+K4 K2 + N o where, Nam is the ammonia nitrogen concentration at the mth stage, 1 mg/l. Nam" is the ammonia concentration at the previous stage, mg/l, and K3 is the kinetic constant. The oxidized nitrogen concentration can be expressed as: Nam-1 - Nom m m —=—K4+K5N LA/V g o where, No is the oxidized nitrogen concentration at the mth stage, mg/l, Nam—l is the oxidized nitrogen concentration at the previous stage, mg/l, and K4,K5 are the kinetic constants. Dimensional analysis was applied to all the factors that affect oxygen transfer to the liquid bulk. It was found that the overall Oxygen mass transfer coefficient can be expressed as: KLV/At _ A a b 'r—EEf—' - 01 ( 3;) (NRe)where, m ,b are constants V is the volume of the tank, m3, At is the surface area of the tank, m2, DL is the diffusivity of oxygen in water, ma/sec, A is the projected area of the contact media on the tank surface, m2, and KL is the overall oxygen mass transfer coefficient, m/sec.fundamental The oxygen transfer efficiency for three types of substrates used; soft~ drink bottling plant wastewater, domestic sewage from AIT campus and synthetic sewage were determined by running the unit without any slime on the contact media. The values were graphed for various ranges of Substrate concentration. The weights of oxygen required to remove unit weight of COD for the different wastewaters used were determined by a respirometric study. It was found that for the bottling plant wastewater the oxygen required per gram of COD removed was 0.56 grams, for AIT sewage 0.46 grams and for synthetic sewage 0.48 grams. The oxygen transferred into the slime layer as predicted using the diffusion equation: 2 K (Lm l _ Lm) D d O _ s _ I; a ———- _ _c___________. _ o ax? 9 r ere, is e rams o ox en re uire er ram 0 remove , w KS ' t ' d f COD d is the oxygen transfer efficiency, and is the oxygen concentration. 0Q The total oxygen mass transfer into the System is: m-l m _ —————————————— V O 6 O = - % \/ 20S 0r DL 0 - KLaa (i: (05-0) + 0r where, OH1 is the oxygen concentration in the liquid bulk at stage m, mg/l, Om_1 is the oxygen concentration in the liquid bulk at the previOus stage, mg/l, Or is oxygen uptake rate, and 03 is oxygen concentration at saturation, mg/l It was noticed that the limiting stages of the oxygen mass transfer was the transfer of oxygen into the liquid bulk and into the slime from the atmosphere. The transfer of oxygen from the liquid bulk to the slime layer was found to be fast enough, that is non-rate limiting. AlthOugh Substrate degradation takes place in the slime layer, dissolved oxygen in the liquid bulk exerts a great influence in supplying the required oxygen. (vi)The kinetic equations for substrate and nitrogen conversion in the aerobic chamber of rotating biological filters were verified using a total of 26 sets of experimental data. It was found that except for 3 sets, the magnitude of the values predicted by the kinetic equations were very close to the experimental values. Although the percentage errors were high at the effluent end of the units, this statistic is rather an unreliable indicator of the accuracy of the results, as the errors in the determination of the influent concentration are compounded as the iteration proceeds, and are thus highest when the magnitude of the levels are very small. The substrate degradation equation applying to the aerobic chamber was optimized using geometric programming. It was found that for optimum conditions, the product of the areal loading and the detention time must be equal at each stage. The number of stages is recommended to be at least fOur to prevent short circuiting, although this could be lowered to two or three if the required percentage removal of substrate is lower than 90%. Experiments were carried out to determine the ammonia nitrogen and substrate conversions and gas production in the anaerobic compartment of a rotating biological filter unit located underneath the aerobic compartment, as in an Imhoff Tank. It was found that in the absence of any mixing, most of the volatile solids were raised into the soum layer after three days, which gave a low rate of gas production. The total nitrogen to COD ratio in the anaerobic liquor was found to be 50% higher than in the aerobic chamber. The recycle of nitrogen into the aerobic chamber is controlled by displacement of anaerobic liquor due to sludge produced in the aerobic chamber.equations |
| Year | 1974 |
| Type | Dissertation |
| School | AIT Publication (Year <=1978) |
| Department | Other Field of Studies (No Department) |
| Academic Program/FoS | Dissertation (D) (Year <=1978) |
| Chairperson(s) | Pescod, M.B. |
| Degree | Thesis (Ph.D.) - Asian Institute of Technology, 1974 |