Kademeli beslemeli biyolojik azot giderim sistemlerinin bilgisayar destekli tasarımı

Abstract

Gelişmiş ülkelerde azot-fosfor gibi besi maddelerine ilişkin deşarj standartları giderek sıküaştınlmaktadır. Avrupa BirliğTne üye olma hazırlığı içindeki Türkiye'nin de "Su Kalite Yönetimi" konusundaki bu gelişmelere karşı gerekli önlemleri şimdiden alma zorunluluğu vardır. Buradan hareketle, bu çalışmada, aüksulardan azot parametresinin giderilmesi amacıyla kademeli olarak beslenen nitrifikasyon-denitrifikasyon sistemlerine yönelik bir tasarım modeli ve bu modelin tam karışımlı, sürekli sistemler için çözümünü yapan bir bilgisayar programı geliştirilmiştir. Biyolojik azot gideriminin temellerine değinilerek kademeli beslemeli sistemlerle ilgili dünya literatürünün bir özeti çalışmanın basma eklenmiştir. Kademeli beslemeli sistemleri temsil etmek üzere seçilen dört reaktörlü - iki beslemeli ve ara reaktörlü - üç beslemeli nitrifikasyon-denitrifikasyon sistemlerinde herbir reaktör etrafinda, kolay ayrışabilir substrat, Ss, yavaş ayrışabilir substrat, Xs, amonyak, Snh ve nitrat, Sno parametreleri için kütle denge denklemleri içsel solunum modeli kullanılarak çıkartılmış ve daha sonra çıkışta istenen nitrat konsantrasyonuna göre sistemin temel tasarım parametreleri olan, çamur yaşı, 6x; debi fraksiyonu, i; hacimsel fraksiyonlar, a ve b; toplam hacim, V v.b. değerleri iteratif bir yaklaşımla elde edilmiştir. Model sonuçlan değerlendirildiğinde, kademeli beslemeli sistemlerle, diğer klasik biyolojik azot giderim sistemlerinde elde edilemeyecek düşük çıkış nitrat değerlerine ulaşılabildiği görülmüştür. Atıksu karakterizasyonuna ve istenen çıkış koşullarına bağlı olarak altı reaktörlü-üç beslemeli sistemlerin de kullanılmasıyla çıkışta çok daha düşük değerler elde edilebilir. Ayrıca kademeli beslemeli sistemlerde içsel nitrat geri devrine gerek kalmaması nedeniyle anoksik hacime verilen çözünmüş oksijen konsantrasyonu sadece çamur geri devri ile sınırlı kalacağından denitrifikasyon verimi artacaktır. Yine bu özellik öndenitrifikasyon sistemlerindeki yüksek geri devir pompaj maliyetlerinden kaçınılacağı için sisteme daha ekonomik olma özelliği kazandıracaktır. Kademeli beslemeli sistemlerin bir başka avantajı ise diğer sistemlere kıyasla daha düşük hacimlerde istenen çıkış verimine ulaşabilmeleridir. Konuyla ilgili gelecekte yapılması gerekli çalışmalar ise, evsel auksularda önemli konsantrasyonlara ulaşabilen yavaş ayrışan substrat, Xs'in hidroliz olma mekanizması üzerine yoğunlaşmak olmalıdır. Bunun dışında, ülkemiz için bir eksiklik olan pilot, tam ve laboratuar ölçekli çalışmalar modelin deneyim kazanarak güçlenmesini sağlayacaktır.

The growing demand for water resources has generated an equivalent need for effective water and wastewater management strategies. In all over the world, these driving forces have found support in the regulations that mandate the proper handling and treatment of wastewaters discharged to the aquatic environment. Application of processes to control wastewater discharges to the aquatic environment began around the turn of the century, generally addressing the removal of particulates and oxygen-demanding carbonaceous materials. The removal of other constituents did not receive substantive attention until the latter part of this century. The control of nitrogen has subsequently been identified as an important environmental activity, demonstrated by the adverse effects that excess levels of different forms of nitrogen have had on aquatic systems. Recently several biological nitrogen removal methods with a number of different reactor configurations have been used. Every day, the developed countries make the discharge standards of nitrogen and phosphorus more strict. Turkey as preparing herself to be a member of European Union has to seriously consider the concept of nutrient removal from wastewaters. In this context, this study defines a step feeding biological nitrogen removal system by using endogenous decay type of activated sludge model and develops a computer aided design approach. The step feeding configuration as defined in Figure 1., is basically a single sludge system and its kinetic evaluation should involve all the basic components and processes describing carbon oxidation, nitrification and denitrification. Its design relies on the selection of an appropriate sludge age for the autotrophic and the heterotrophic sludge biomass. The aerobic sludge age, 0xa is particularly important because complete nitrification is a prerequisite for an effective nitrogen removal which takes place within the total anoxic volume, Vd. Appropriate selection of the ratio of the anoxic volume to the total reactor volume, Vd/V and 0xa yields the total sludge age of the system: a °XA !--£ xui QRs Qw Figure 1. Schematic diagram of step feeding mtrification-denhrification system The total reactor volume should ensure complete removal of the biodegradable COD, C». Then the heterotrophic biomass, V Xh may be computed as follows: MXH=VXH = YHQ(CS0-CS) i+bHex 0, The amount of autotrophic biomass, Mxa, is a function of nitrification capacity, (Nox) and aerobic sludge age, (9xa ). MXA=VXA = It is postulated that all growth processes consume ammonia nitrogen, Snh as nitrogen source. In the aerated zones, this component is converted into nitrate nitrogen, Sno by means of nitrification. Furthermore, the total biodegradable organic nitrogen concentration in the influent, Cndo may be considered as a potential ammonia nitrogen source assuming that the ammonification of organic nitrogen is not rate limiting. In this context, the following expression defines the mass balance for nitrogen: Nox - Cmo + "Swo - Snh - Nx Nx may be defined on the basis of related kinetic expressions with the assumption that the heterotrophic activity is reduced in the anoxic zones by a correction factor. Different correction factors tig, tjh, t|E may be commonly defined to characterize respectively xiv growth, hydrolysis and decay under anoxic conditions. As no substantial proof has so far been provided in the literature to show that they are significantly different, a single correction factor, r| has been adopted in this study as shown in the following equations: Va+t,Vd=cV and H \ SO S ) Xx = OxB +*XE f*X K C0X) 1+cb Q The unit amount of nitrogen incorporated into biomass, Nx is set for selected values of Vd/V and 6x. For a desired effluent nitrate nitrogen concentration, Sno, Nox and the autotrophic biomass, V XA can be computed from expressions 3 and 5. Similarly, the following kinetic expressions may be derived for the particulate inert COD fractions, Xp andXj: ^xp=v^p=fEx^HexVXH Mn=VX^QXI(iex Slowly biodegradable particulate substrate, Xs accumulates in the system depending on 0x and hydrolysis rate ( KH ). Consequently amount of Xs can be calculated as follows: s \+KHex and, VXT = VX" + VXA + VXS + VXj + VXP An appropriate XT value, selected as the average total particulate COD in the system enables the calculation of the total and the anoxic reactor volumes, V and VD. The procedure so far described applies to all single sludge systems designed for nitrogen removal. Step feed reactor configuration, additionally involves three significant design parameters: the fraction of the influent flow rate fed to the first anoxic reactor, i; the xv volume ratio of the first anoxic reactor, a; and the volume ratio of the first aerobic reactor, b. The volume ratios, a and b are defined as follows: V V ' PI, r A\ a~v ~ V ¥D ' A The selection of the appropriate value for i, is most important for the optimum design of step feeding flow configuration, because the ammonia content of the wastewater fraction, 1-i, diverted to the 2nd aerobic reactor, aside from the relatively small portion incorporated into biomass, will be converted into nitrate and will leave the system as nitrate nitrogen. The latter may be considered as the effluent nitrate nitrogen concentration, Sno, provided that the 2nd anoxic reactor secures complete denitrification, virtually eliminating all nitrate When this balance is satisfied, the kinetic considerations, the following expressions may be derived to define Noxi and Ndk nitrogen (Sncs = 0). In this respect, the 2nd anoxic reactor is the key part of the whole system and should be designed to maintain the delicate balance between its denitrification potential, Ndr and tile nitrate nitrogen generated in the first aerobic reactor, Noxi: When this balance is satisfied, the amount of ammonia nitrogen nitrified in the second aerobic reactor, N0x2 will determine the effluent nitrate nitrogen concentration. On the basis of kinetic considerations, the following expressions may be derived to define Noxi andNcre: V 1 Xoxı=l>-ğXA(Y+ixB)MAi "^»-"'t^ 1- K, l rB Here it is shown that Noxi is a function of b and Sure in the first aerobic reactor. The level of Snh2 is set by the selected 0xa- Since b is computed to reach this level of Sum, the ammonia load entering the reactor determines the magnitude of nitrification. Therefore, Noxi basically relates to and increases with the flow fraction, i. The same way, Ndr varies with the heterotrophic specific growth rate at the third tank, Un3 and consequently with the (1-i) fraction. When the variation of Noxi, Ndr and Sno as a function of the wastewater flow fraction entering the system, i, for selected constant values of 6x and Vd/V is investigated in the second anoxic reactor, it is observed that, for low i values, the amount of available organic carbon in the 2nd anoxic reactor stays beyond the required level to remove the existing nitrate and consequently, Nik > Noxi, indicating an excess of wasted denitrification potential. Conversely, as the value of i increases, the amount of available organic carbon drops and becomes insufficient in removing the steadily increasing amount of nitrate entering the reactor. Therefore the lowest effluent nitrate concentration, Sno is obtained xvi when the denitrification potential is balanced with the amount of available nitrate nitrogen. This balance also yields the optimum value of i for a constant sludge age selected for the operation of the system. Sno cannot be reduced below a limit dictated by the selected 9x or VxJV. To lower this level is only possible by increasing the VdW ratio. It should be noted that the amount of nitrate recycled to the first anoxic tank by means of sludge return, Rs Q Sno determines the volume of this reactor or a. One of the attractive features of the step feeding configuration is the elimination of internal recycling and the sludge return is practically constant. Therefore a higher Vd/V ratio basically affects the volume of the 2nd anoxic reactor and increases Ndk. This lowers the achievable effluent nitrate nitrogen concentration, Sno and increases the optimum value of the flow fraction, i, or in other words, reduces the wastewater flow to be diverted as a step feed. The variation of Sno with i shows the previously explained trend, decreasing to a minimum value as the amount of nitrogen entering to the 2nd anoxic tank drops for higher i values, then increasing for insufficient Ndp2 levels. Finally, kinetic evaluations indicate that the selection of the volume ratio of the first aerobic reactor, b should best be in proportion to the ammonia nitrogen loads; thus b may be given the same value as i. It is concluded that, depending on the wastewater characterization and required effluent limits two fed-four reactor configuration may not be able to give the desired results. In this case, by using three fed-six reactor configuration, better effluent conditions can be obtained. Design procedure of such configuration has been also developed as similar to the previously mentioned two fed-four reactor configuration. There are a lot of advantages of step feeding systems. It is possible to reach very low effluent nitrate values that can never be obtained by predenitrification systems. Since the optimum carbon-nitrogen balance can be set by changing the flow and volume distributions, very important volumetric savings are possible. As the need of internal recirculation gets disappearing, because of the low oxygen transfer to the first anoxic reactor, the denitrification efficiency increases, moreover, the amount of pumping costs due to recirculation decreases. The multi-component activated sludge models - such as endogenous decay - emphasize the importance of wastewater characterization. In this study, the slowly biodegradable substrate, Xs has been assumed to be hydrolyzed with a first order reaction. There is a big gap at the literature about this topic. The hydrolysis mechanism of Xs still needs to be investigated to reach more realistic modeling approaches. Finally, especially in Turkey, more lab-scale, pilot-scale and full-scale experimental investigations have to be run as future works for step feeding systems.