Kuvvetli atıkların yukarı akışlı havasız filtrelerde arıtım kinetiği

Abstract

Havasız arıtım prosesleri yüksek miktarlarda organik madde içeren atıksuların arıtımda son yıllarda oldukça geniş kullanım sahası bulmuştur. Havalı proseslere kıyasla kuvvetli atıkların arıtımında daha ekonomik bir alternatif olduğu saptanan havasız arıtım proseslerinin düşük organik mad de içeren evsel suların da arıtımında kullanılması yönünde çalışmalar yo ğun bir şekilde devam etmektedir. Arıtılacak atıksuyun karakteristiğine bağımlı olarak uygulamaya konacak arıtım prosesinin seçimi bölge koşulla rı ve atıksu deşarj standartları da gözönünde tutularak ekonomik ve tek nolojik irdelemeleri gerektirmektedir. Endüstriyel uygulama yönünde geniş kullanım sahası bulmasına rağ men havasız arıtım proseslerinde substrat giderim kinetiği konusunda veri eksikliği bu sistemlerin tasarımı aşamasında darboğaz yaratmaktadır. Şim diye değin uygulamaya konan havasız arıtım tesislerinin işletilmesi süre cinde kazanılan pratik bilgi birikiminin değişken işletme koşullarına en tegrasyonu tecrübe sınırları ötesinde pek mümkün olamamaktadır. Bu durum da sistem karekteristiklerini kinetik esaslar ile bağdaştıran yaklaşımların benimsenmesi ve çalışmaların bu yönde yürütülmesi mevcut boşluğu doldura cak ve endüstriyel uygulama yönünde tasarım parametrelerinin daha sıhhat li ve güvenli esaslar üzerine oturmasına yardımcı olacaktır. Bu çalışma kapsamında kuvvetli atıkların yukarı akışlı havasız filtrelerde arıtımı kinetiği detaylı olarak incelenmiş ve endüstriyel uy gulama yönünde pratik kontrol alternatifleri tanımlanmıştır. Birinci bölümde çalışmanın önemi ve amacı vurgulanmış ve havasız arıtım proseslerinin uygulamaya yönelik teknik ve ekonomik boyutları ta nımlanmıştır. İkinci bölümde havasız proseslerin stokiometrisi, enerji bağıntı ları verilmiş ve çevresel faktörlerin proses verimi üzerindeki etkilerin den bahsedilmiştir. Bu bölümde ayrıca biyofilm kinetiği detaylı olarak verilmiş ve biyofilm kinetiğinin havasız proseslere uygulandığı model yak laşımlar izah edilmiştir. Deneysel çalışma düzeni ve çalışmalarda kullanılan sistemler üçün cü bölümde ayrıntılı olarak verilmiştir. Deneysel çalışmalarda karbon kaynağı olarak kullanılan melas için kinetik parametrelerin hesaplanma sında tam karışımlı kesikli reaktör kullanılmıştır. Öte yandan sürekli filtrasyon çalışmaları yukarı akışlı filtre sisteminde yürütülmüştür. Sürekli filtrasyon çalışmalarında 4 farklı substrat konsantrasyonunda (2500, 6500, 9500, 12500 mg KOI/lt) çalışılmış ve sistemde organik ve hidrolik yükleme değerleri sırasıyla 1.38-18.89 kgKOI/m3-gün ve 0.4-3.02 lt/lt-gün aralığında değiştirilmiştir. Sürekli filtre sisteminde pH, KOI, Uçucu asit, alkalinite dağılımları günlük olarak ölçülmüştür. Sistemden çıkan gaz sürekli olarak ölçülmüş ve gaz akımında CH4 % sinin % 55-73 ara lığında değiştiği saptanmıştır. Gaz akımı CH4 yüzdesinin artan substrat konsantrasyonu ve hidrolik kalış süresi ile arttığı deneysel olarak göz- IX lenmiştir. Sürekli filtrasyon çalışmalarında organik yüklemenin filtre hidroliği üzerindeki etkilerini araştırma yönünde değişik organik yükleme değerlerinde iz element (Lj.+) kullanılarak hidrolik çalışmalar gerçekleş tirilmiştir. Organikyükleme değerinin 2 kgKOI/m^-gün - 16 kgKOI/m^-gün aralığında sistem hidroliğinin "Dispersiyonlu Piston Akış" rejimine karşı geldiği deneysel olarak saptanmıştır. Dördüncü bölümde deneysel çalışmaların sonuçları verilmiş ve de ğerlendirilmiştir. Sürekli filtre sisteminde uçucu asitlerin ve biyokitle dağılımının homojenizasyonuna yönelik olarak uygulanan kademeli besleme koşullarında sistem veriminin işletme parametrelerine bağımlı olarak tüm akımın slttan beslenmesine kıyasla % 4-20 artış gösterdiği deneysel ola rak saptanmıştır. Sürekli filtre çalışmalarında sübstrat giderimine kar şın sistemde inert madde oluşumunun hesaplanmasına yönelik yöntemler ay rıntılı olarak verilmiştir. Melas besleme koşullarında sistemde oluşan inert madde miktarının giderilen KOI'nin % 6.7-13'ü aralığında, CH3COOH besleme koşullarında ise inert madde oluşumunun giderilen KOI'nin % 3-4.7 si aralığında değiştiği deneysel olarak saptanmıştır. Biyofilm kinetiğinin havasız filtre sistemine uygulanması aşama sında 2 model yaklaşım tanımlanmıştır. Havasız filtrenin tam karışımlı seri reaktörlerden teşekkül ettiğinin varsayıldığı birinci model yaklaşımda herbir kompartman için birinci, (1/2) ve sıfırıncı dereceden reaksiyon hız sabitleri hesaplanmıştır. Filtre sisteminin piston akışlı sabit yatak reaktörü olarak kabul edildiği ikinci model yaklaşımda da reaksiyon hız sabitleri hesap edil miştir. Her iki model yaklaşım sonucu sıvı faz sübstrat konsantrasyonunun 200-2300 mg/lt aralığında (1/2) dereceden reaksiyonun geçerli olduğu vur gulanmıştır. Değişik işletme koşullarında (1/2) dereceden reaksiyon hız sabitlerinin 20.0-160.0 (mg/lt)!'^ gün~l aralığında değiştiği saptanmış tır. Reaksiyon hız sabitlerinin organik yüklemenin fonksiyonu olarak de ğiştiği deneysel sonuçların bu yönde değerlendirilmesi ile ortaya konmuştur. Çalışmaların tüm sonuçları beşinci bölümde ayrıntılı olarak sunul muştur.

Anaerobic waste treatment is an important biological prosess that has long been used in the stabilization of municipal sludges. More recently there is an increasing interest in applying this process in the treatment of high and medium strength soluble and colloidal wastes. Depending upon the wastewater characteristics application of anaerobic processes have some advantages over aerobic ones. The cost of aeration increases with increasing concentration of organic matter and at some 5 gr COD/lt of substrate concentration, the system becomes oxygen limited. The only way to overcome the problem is to increase the hydraulic detention time ensuring sufficient oxygen transfer to the process. In the case of treatment of strong wastes (C0DS5 gr/lt) through anaerobic processes, elimination of costly aeration equipments and minimization of sludge removal facilities reduce the total cost of treatment. Furthermore CH4 gas obtained from anaerobic systems as a final product can be used as an alternative fuel supply. Thus the borderline between these two diffe rent processes should be carefully drawn through economical and technolo gical considerations. Conventional processes, contact processes, anaerobic filters and anaerobic sludge blanket systems are examples for anaerobic processes that have found wide application in waste treatment while expanded bed and fluidized bed anaerobic processes have received more attention espe cially in treatment of low-strength wastewaters in the last decade. Despite wide spread use of anaerobic treatment, optimum process performance is seldom achieved because a high degree of empiricism still prevails in design of such systems. As regard to the environmental and operational conditions in anaerobic treatment, degredation of organic material through the action of two different groups of bacteria follows such a complex pathway that creates difficulties to understand the infra structure of treatment mechanism. Thus, the evaluation of the rational basis for design and control is essential to the realization of full potential of anaerobic processes. There are basicly two different approaches to the design of anaerobic treatment processes. - The traditional approach by which years of experience is synthesized into permissible loading figures mainly appear in terms of hydraulic and organic loading values. Then the resulting degree of purification is expected. - The conceptual approach by which it is attempted to simulate the processes involved to such degree that the purification result can be predicted. Within the limits of experience, the traditional approach works well but it is somewhat dangerous to extrapolate the conditions beyond the general range of experience. The conceptual approach, on the other hand, strives to incorporate the main features of the process to increase the generality of design parameters for practical applications in a wide range. In the framework of this study treatment of strong wastes, in an upflow anaerobic filter was investigated in detail. The main objective of the study was the evaluation of such parameters that can be used for design and control of anaerobic biofilm reactors for industrial purposes. Modelling of the system through kinetical approaches applying conventional biofilm kinetics for anaerobic fixed film system was emphesized rather than determination of design parameters through traditional approaches. In the first chapter, the importance and the main objectives of the study are defined in detail. Global economical and technological concepts of anaerobic treatment are also discussed. In the second chapter, the basic principles of anaerobic processes are given. The effect of environmental parameters on process performance, the stochiometries of anaerobic reactions, energetics of anaerobic processes are discussed. Approaches for modelling of anaerobic fixed film systems through kinetical considerations in biofilm systems are emphasized. The results of extensive literature survey based an the studies on the treatment of strong wastes in anaerobic systems especially in anaerobic fixed film systems are discussed in detail. The traditional and conceptual approaches and basic design parameters for practical purposes are also defined. Experimental work plan is outlined within the third chapter. Molasses solutions which synteticaly prepared in laboratory conditions were used as carbon source through experimental studies. The nutrients (N,P) and trace elements such as Mg+2, Fe+2, Ni+2, Co+2, S~2 were added into feeding solutions in such amounts that proposed in literature. The ratio of COD/N/P was adjusted as to be 300/7/2. Kinetic parameters for molasses were determined using a completely mixed anaerobic suspended culture batch reactor with a volume of 1 liter. The diffusional resis tances were eliminated in batch reactor setting vigorous mixing condi tions in order to use Monod kinetics for determination of kinetic para meters. An anaerobic upflow filter was used through continious filtration experiments. The void volume and corresponding void fraction within the filter were 19.85 It and 0.44 respectively. Cylindrical PVC materials with 4 cm in diameter and 1 cm in height were packed into the reactor to encourage the biofilm build-up. The continious anaerobic filter was operated different organic and hydraulic loading values in the range of 1.38-18.89 kg C0D/m3-day and 0.4-3.02 lt/lt-day respectively. The substrate concentrations of feeding solutions were selected to be 2500, 6500, 9500, 12500 mg COD/lt in order to determine the effect of influent's substrate concentration on system's performance. The hydraulics of the continious filter was observed to be affect ed by organic loading applied on the system. The hydraulic regime of the filter was determined to follow 'Dispersed Plug Flow' pattern in the range of 2 kg COD/nP-day and 16 kg COD/m^-day with corresponding D/uL values of 0.10 and 0.23 respectively. Feeding solutions substrate concentrations pH, volatile acids, COD, alkalinity were measured daily while the gas volume was measured continiously using an electronic low gasmeter and a wet gasmeter connect ed each other in series. All of the experimental studies were carried out in a reference room in which the temperature was kept to be 35±2°C. Standart methods were used for analitical measurements. In the fourth chapter, the experimental results are presented and discussed. In view of the results obtained from completely mixed anaerobic batch reactor, the kinetic parameters for molasses were determined to be k = __ = 2.31 gun K = 650 mg/lt Considering Y values to be 0.1 mg UAK/mg COD as given in literature for combined cultures in anaerobic processes, the value for y can be calculated to be 0.23 giin"~l. m In continious filter studies, it was observed that removal of substrate takes place in the lower part of the filter and is associated with great amount of biomass in suspension. Depending on the operational conditions no treatment was observed beyond a certain height of the filter. The biogas obtained from the system was determined to be increas ing with organic load applied on the filter. As a general trend the composition of CH4 gas in gas stream was found to be increasing with increasing substrate concentration and hydraulic detention time. It was also measured that the CH4 composition in the gas phase was in the range of 55-73 % under various operational conditions. Extensive work was carried out for investigation of effects of step-loading application on system's performance. It was observed that step-loading has some advantages over the cases in which the substrate is totally fed from the bottom of the filter. The system's performance was increased by 4-20 % in terms of % removal of COD under step-loading con ditions. Furthermore volatile acid and COD distribution within the system was somewhat stabilized when step-loading was applied an the system. Inert material production relevant to substrate removal within the system was determined through kinetic studies. Thus we were able to determine the substrate concentrations in terms of biodegradable COD units at any point within the filter. Biodegradable COD values were used in kinetic rate expressions. It was determined that the inert material produced through biochemical reactions is about 6.7-13 % of COD removed - xııı - in the system. CH3COOH when used as substrate, the inert material produc tion was calculate to be 3-4.7 % of substrate removed in terms of COD. Regarding the experimental results, the inert material concentra tion in effluent was calculated by the following equation. Sl0(uA) = 0.085 (Ş S0(KOI) - 2500) Assuming the filter operating at sufficiently high detention time and the flow regime to be plug-flow the concentration of the biodegradable organics not converted to volatile acids in effluent can be considered to be negligible. Thus the total COD value of the effluent has been calculated simply by summing the values of inert material concentration and volatile acid concentration in terms of COD. S10(KOI) = S1Q( inert) + S1Q(uA) In the case of highly loaded conditions under which the concentra tion of organics can not be neglected in effluent, the total COD value of the effluent has been calculated either using the figures corresponding to % removal v.s. organic loading or substrate removal rate v.s. organic loading. The continious filter was modelled to be completely mixed reactors in series and a plug-flow reactor as a whole from kinetical point of view. Assuming zero order intrinsic reaction rate in partially penetrated bio- film, it was observed that the substrate removal rate follows 1/2 order of reaction with respect to the bulk liquid concentration of biodegradable substrate, in the range of 200-2300 mg C0D/lt. Beyond these limits zero order and first order removal rate constants were also evaluated. The rate constants calculated for different parts of the filter are given in the following: k (0-1) = 65 gr/lt-giin ^ /2(0-l) = 330-492 (mg/lt)1/2 gün"1 k^O-1) =27.5 gün"1 kQ(l-3) = 10 gr/lt-giin k1/2(l-3) = 33-156 (mg/lt)1/2 gün""1 k (1-3) = 6.89 gün"1 kQ(3-4) = 4 gr/lt-giin k1/2<3-4) = 47.9-115 (mg/lt)1/2 gün"1 kL(3-4) = 2.86 gün 1 kQ(4-5) = 3 gr/lt-gün k1/2(4-5) = 20-70 (mg/lt)1/2 gün"1 k (4-5) =2.66 gün"1 It was also determined that rate constants are highly effected by organic loading applied on the system. Thus, we concluded that the gas generation in the biofilm and hydrolic loading are both effective on substrate transfer mechanism into biofilm. It can be easily seen that the reaction rate constants within the part (0-1) of the filter were comparatively higher than those in the upper parts of the filter. It was concluded that relatively higher amount of biomass in suspension within the part (0-1) was the answer to the question. Using the kinetic parameters determined through experimental studies, it was calculated that the total biomass concentration within the part (0-1) was approximately 35000 mg/lt when maximum substrate removal rate was achieved. Considering the continious filter to be a plug flow fixed film reactor beyond the part (0-1), linear relations were observed between hydraulic detention time and square root of effluent COD concentrations in the range of 200-2200 mg/lt. under different operational conditions. Thus it was shown that (1/2) order of reaction could be applicable for anaerobic fixed film systems. The rate constants for (1/2) order of reaction were calculated to be changing between 30.4-160 (mg/lt)!' 2 day~l which were very similiar what were observed through completely mixed reactors in series model approach. It was also determined that the rate constants calculated for (1/2) order of reaction were increasing with increasing organic load applied on the part of the filter of interest. The organic load, when considered, is the multiplication of two components which are hydraulic loading and substrate concentration. Since hydraulic flow rate and concentration gradient both positively affect mass transfer into the biofilm, the results were considered to be well fitted on what was expected in this respect. Additionaly the gas released through the biofilm should increase the mass transfer into the biofilm as discussed in literature. As a result, extensive work has been carried out for evaluation of kinetics of strong waste treatment by anaerobic upflow filters. In addition to this some operation practices which can be well adapted for industrial applications were aimed to be given from engineering point of view. We believe that the present work will be useful for whom dealing with wastewater treatment aspects in fixed film systems as regard the methods and approaches defined within the framework of the study. The results are summarized in the fifth chapter.