Hibrit mikrofiltrasyon proses ile sulardan nikel gideriminde akı düşüşünün istatistiksel yöntemlerle incelenmesi

The investigation of flux decline by statistical methods in the nickel removal from waters using hybrid microfiltration process

Özet

Son yıllarda, çözünmüş organik ve inorganik safsızlıkların sulardan yenilikçi hibrit membran proseslerle giderimi yaygınlık kazanmıştır. Bu proseslerden membran-toz aktif karbon, membran biyoreaktör, vakum sürücülü membran-flotasyon, membran-ozonlama ve membran-elektrokoagülasyon prosesleri, özellikle dikkat çekici olanlarıdır. Membran proseslerin uygulamasında temel kriterlerin başında akı azalması gelmekte, buna sebep olan faktörler belirlenerek prosesin daha yüksek akı değerlerinde işletilebilmesi amaçlanmaktadır. Bu noktadan hareketle bu çalışmada, yüzey aktif madde destekli toz aktif karbon/çapraz akış mikrofiltrasyon hibrit prosesi kullanılarak sulu ortamdan nikel iyonlarının giderimi incelenmiştir. Proseste, membran türü ve gözenek boyutu, adsorban türü ve Yüzey Aktif Madde (YAM) türü olmak üzere 4 değişken için deneysel tasarıma göre gerçekleştirilmiş çalışmalar kapsamında, membran kirlenmesi ve akı azalması olayları analiz edilmiştir. Kek tabakasının parçacık çapı ve porozitesine, keke katılan YAM ve Toz Aktif Karbon (TAK) miktarlarına ve membran gözenek boyutuna bağlı olarak membran üzerinde dinamik bir kek tabakasının oluştuğu belirlenmiştir. Membrandaki kirlenmenin, membran üzerinde tutunan kütle ile orantılı olarak değişmediği ve keke katılan YAM miktarı ile TAK miktarı arasında ters orantılı bir ilişki bulunduğu tespit edilmiştir. Akı azalmasında etkili mekanizmanın, kısmen membranın iç kısımlarında ve özellikle membran üzerindeki kek tabakasında tutunan YAM miktarı ile ilişkili olduğu saptanmıştır. YAM türü, akı azalması üzerine en etkili parametre olarak belirlenmiştir. Azalan YAM miktarı ile kek tabakasındaki TAK miktarının arttığı, bunun da, proseste daha yüksek akı elde edilmesine imkân sağladığı saptanmıştır.

Abstract

Microfiltration (MF) and ultrafiltration (UF) processes are especially preferred by reason of low pressure requirement and flexible operation in various water and wastewater treatment applications. Despite the high efficiency achieved in solid-liquid separation, these membranes can not remove the soluble organics and inorganic impurities such as heavy metals. Nevertheless, both processes can be effectively used to remove soluble materials from wastewaters as a hybrid process combined with conventional treatment methods. Hybrid membrane processes compared with traditional membrane processes provide some advantages such as high quantity of treated wastewater, high removal efficiency, fouling control, low energy consumption and lower back-washing time. Among these processes, surfactant enhanced powdered activated carbon (PAC)/crossflow microfiltration (CFMF) hybrid process could be used as a promising technology to remove heavy metals from water environment. In this process, the adsorption capacity for metal ions on surfactant modified PAC could be higher than that for untreated PAC. The dispersed activated carbon would than be easily removed with lower energy consumption at lower pressures using MF process. The purpose of the study was to explore the flux decline of surfactant enhanced PAC/CFMF hybrid process in the removal of nickel ions from aqueous solution. For this aim, the influence of systemcomponent variables related to membrane material and pore size, adsorbent and surfactant types as input variables was investigated. Because surfactantadded hybrid PAC/CFMF process inherently involves a great number of independent variables which may affect the flux decline, the experiments were carried out at the base of the Taguchi experimental design. In this design, while membrane pore size was chosen as a two-level factor, the other system- component variables were embodied into the design at three levels by means of using “Idle Column Method”. Process variables such as process time, crossflow velocity, transmembrane pressure, and characteristics of feed solution were chosen as constant levels for all experiments. To elaborately assess the flux decline, the modified fouling index (MFI), specific cake resistance ($alpha$) and total dried solid mass of cake per unit membrane area ($omega$) as membrane fouling parameters and, non-steady state flux (J(t)) and steady state flux (J*) values were taken into account. To interpret the influence of any input variable on membrane fouling and flux decline, analysis of variance (ANOVA) statistical technique was used. J(t) values exhibited different variations with time according to the each surfactant type. At the beginning of the process, especially in first 8-10 seconds, a considerable decrease in flux appeared for sodium dodecyl sulfate (SDS). The highest J(t) values throughout the process were obtained by means of 1-hexadecanesulfonic acid sodium salt (HDSA). The optimum combinations of membrane, adsorbent, pore size, and surfactant types for the least of mass deposition and membrane fouling were established as “cellulose acetate, 89440, 0.45 $mu m$ and SDS” and “cellulose nitrate, C9157, 0.45 $mu m$ and HDSA”, respectively. On the other hand, for the highest steady state flux, the optimum combination was found as “mixed cellulose ester, C9157, 0.2 $mu m$ and HDSA”. This conclusion put forward that, a dynamic cake layer occurred on membrane considering various experimental conditions. Membrane fouling comprised a secondary membrane layer which formed by the surfactants micelles both on the membrane surface and within the membrane pores in addition to the PAC layer on membrane surface. According to the membrane fouling, membrane, adsorbent and surfactant types were determined as important variables, while membrane pore size performed the lowest effect. Effective parameters on flux decline were seen as surfactant type and membrane pore size owing to the blocking of membrane pores with free surfactant aggregates with or without nickel. While membrane type has no effect on steady-state flux, adsorbent type exhibited very little influence. As concluding remarks, with regard to the total influence of each variable on membrane fouling and flux decline, it can be said that surfactant type has the greatest total influence compared with the others. Adsorbent type and membrane pore size come thereafter. The influence of membrane type is the lowest compared with other three system-component variables.