• 상하수도학회지
• /
• 제22권4호
• /
• pp.437-444
• /
• 2008

Many other countries have investigated the new backwash water treatment process to save the existing water resource. There are various methods for reusing backwash water, but the membrane system has received the most interest for its efficiency. The objective of this study was to certify the application of membrane filtration system for the backwash water treatment. The experiment equipment was composed of Lab scale tubular membrane filtration system. Generally, cross-flow operation mode is used in the tubular membrane system but cross-flow operation mode demands high electric cost mainly for the pump energy. So to cut off electric cost, dead-end operation mode was used in this experiment. Filtration and bleed operation cycle was used in this membrane system. Backwash water was concentrated during the filtration process and when backwash water reached our target suspended solid concentration, it was discharged from this system. For efficient operation of filtration and bleed, mathematical matrix was drawn up and with this matrix we could simulate various sets of filtration and bleed time.

• 멤브레인
• /
• 제26권4호
• /
• pp.266-272
• /
• 2016

전량여과(dead-end filtration)를 사용하여 계란 난백용액으로부터 라이소자임 효소의 분리정제에 대한 연구를 수행하였다. 전량여과 시 사용된 한외여과 막의 분획분자량과 난백용액의 pH 변화에 따른 라이소자임의 난백에 존재하는 다른 종류의 단백질인 오발부민과 콘알부민에 대한 막을 통한 투과선택도(permselectivity)를 측정하여 투과선택도가 가장 큰 값을 나타내는 전량여과의 최적 운전조건을 결정하였다. 이 결과 난백용액으로 라이소자임을 효과적으로 투과 분리시키기 위한 최적 운전조건은 한외여과 막의 분획분자량 30 kDa, 난백용액의 pH 11이었다. 최적 운전조건에서 난백 내 총 단백질에 대한 라이소자임의 투과선택도가 약 83의 가장 큰 값을 나타내었다.

• 멤브레인
• /
• 제9권1호
• /
• pp.36-42
• /
• 1999

본 연구에서는 Si 미립자를 함유한 반도체 세정폐수의 관형막을 이용한 한외여과특성을 검토하였다. 관형막의 시간변화에 따른 투과유속의 감소현상은 막표면에 형성된 케익층의 증가 및 기공막힘에 기인하며, cross flow는 케익여과에 의한 막오염 형태를 보였으나 dead-end flow는 기공막힘과 케익여과에 의한 혼합형태를 보였다. Cross Flow의 케익저항의 크기는 3.16$\times$$10^{12}$ ~4.34$\times$$10^{12}$ $m^{-1}$ 였고, dead-end flow 는 6.6 $\times$$10^{12}$ ~12.19$\times$$10^{12}$ $m^{-1}$였다. 운전초기의 흐름형태에 따른 투과유속은 cross flow 가 dead-end flow 의 약 7 배였다. Cross flow 투과유속은 약 42 $\ell$/$m^2$ hr, 용질배제율은 약 96 % 였으며, 분리막공정을 거친 투과수 중의 Si 입자의 평균크기는 20nm였다.

• Membrane and Water Treatment
• /
• 제4권3호
• /
• pp.157-174
• /
• 2013

The present study is devoted to the validation of a new method for in line electrokinetic characterisation of deposits on membrane surfaces. This method is based upon simultaneous measurements of transversal streaming potential and permeates flux at constant pressure before and during the deposit formation. Dead-end filtration experiments were conducted with negative flat membranes forming a narrow slit channel, negative hollow fiber membranes and mono-dispersed suspensions of (negatively charged) polystyrene latex and (positively charged) melamine particles at various concentrations. It was observed that the overall streaming potential coefficient increased in absolute value with the deposited latex quantity, whereas it decreased and changed of sign during the filtration of melamine suspensions. By considering a resistance-in-series model, the streaming potential coefficient of the single deposit ($SP_d$) was deduced from the electrokinetic and hydraulic measurements. The independence of $SP_d$ with respect to growth kinetics validates the measurement method and the reliability of the proposed procedure for calculating $SP_d$. It was found that $SP_d$ levelled off much more quickly when filtration was performed through the slit channel. This different behaviour could result from a non-uniform distribution of the deposit thickness along the membrane given that the position of measuring electrodes is different between the two cells.

• 한국대기환경학회:학술대회논문집
• /
• 한국대기환경학회 2003년도 춘계학술대회 논문집
• /
• pp.405-406
• /
• 2003

분진여과시 형성되는 분진 케이크는 궁극적인 여과 성능을 좌우하며, 필터의 구조, 분진 입자의 형태나 크기, 분진농도, 여과속도 등에 의해 영향을 받는다 특히 필터의 표면 구조나 기공 크기는 여과 초기 단계의 케이크 층의 구조를 결정하며, 연속적인 여과시 초기 케이크 층은 다음에 쌓이는 케이크 형성과 비저항에 영향을 주므로 필터 medium 구조 또한 분진여과 과정에서 중요한 영향 요인이 된다. 본 연구에서는 필터 medium의 구조가 서로 다른 고온가스정화용 복합 세라믹 필터, metal fiber mat, 스테인레스 필터를 이용하여 분진농도와 여과속도를 변수로 하여 분진 여과 실험을 함으로써 각각의 필터에 대한 케이크 비저항과 기공율을 실험과 이론식으로부터 추정하였다. (중략)

• 상하수도학회지
• /
• 제21권6호
• /
• pp.717-726
• /
• 2007

Fully-grown flocs in a mixing tank of membrane filtration with dead-end membrane are ruptured while passing through a pump and the ruptured flocs are aggregated again in a Membrane-Feed-Pipe (MFP). To look at more details, this study tries to relate the reaggregation to a parameter of mixing intensity in MFP, i.e., G-value. The G-value is a function of Reynolds number, pipe diameter, friction factor and average velocity in MFP. To deal with polydispersity condition, we develop a representative particle size called in this study EDPD (Effective Diameter for Polydispersity condition in Dead-end filtration). The experimental results show that as the G-value increases, the EDPD decreases and also the cake resistance increases. Through comparison between EDPD and cake resistance, these results show that cake resistances are controlled by reaggregation phenomenon in MFP. The effect of detention time in MFP, however, does not affect the reaggregation of the broken flocs as G-values are increased.

• 한국막학회:학술대회논문집
• /
• 한국막학회 1999년도 춘계 총회 및 학술발표회
• /
• pp.83-85
• /
• 1999

• 멤브레인
• /
• 제29권3호
• /
• pp.147-154
• /
• 2019

BSA 단백질 용액의 전량 막여과에서 상용(commercial) 막모듈의 경사각 변화에 따라 발생된 자연대류 불안정 흐름(NCIF)의 투과 플럭스 증가 효과를 측정한 결과, 막모듈의 경사각이 $0^{\circ}$에서 $180^{\circ}$로 증가하면 NCIF 발생이 커져 플럭스가 증가하였다. 그러나 상용 막모듈은 모듈 내에 존재하는 공기층을 완전히 제거해야 NCIF를 발생시킬 수 있다. 본 연구에서 설계 제작된 custom design 막모듈은 crossward 방향($90^{\circ}$)으로 막투과가 이루어져 모듈 상부에 공기층이 존재하더라도 항상 NCIF가 발생된다. Custom design 막모듈에서 BSA와 dextran 용액의 막여과 실험을 수행한 결과, NCIF의 발생으로 2시간 조작 시 BSA 용액의 경우 약 3.8배, dextran 용액의 경우 약 1.8배까지 투과 플럭스가 증가하였다. 또한 BSA 용액을 대상으로 한 20시간의 조작에서도 NCIF의 발생이 지속되어 플럭스가 약 7.5배까지 증가하였다. 본 연구에서 설계된 막모듈은 항상 NCIF가 발생되므로 막오염 형성 억제에 따른 투과 플럭스 증가를 기대할 수 있어 전량여과 막모듈로서의 활용성이 있음을 확인하였다.

• Membrane and Water Treatment
• /
• 제7권4호
• /
• pp.313-325
• /
• 2016

South Africa recognises piped water as the main source of safe drinking water supply. Remote areas do not have access to this resource and they rely solely on surface water for survival, which exposes them to waterborne diseases. Interim point of use solutions are not practiced due to their laboriousness and alteration of the taste. Bio-ultra low pressure driven membrane system has been noted to be able to produce stable fluxes after one week of operation; however, there is limited literature on South African waters. This study was conducted on three rivers namely; Umgeni, Umbilo and Tugela. Three laboratory systems were setup to evaluate the performance of the technology in terms of producing stable fluxes and water that is compliant with the WHO 2008 drinking water guideline with regards to turbidity, total coliforms and E.coli. The obtained flux rate trends were similar to those noted in literature where they are referred to as stable fluxes. However, when further comparing the obtained fluxes to the normal dead-end filtration curve, it was noted that both the Umbilo and Tugela Rivers responded similarly to a normal dead-end filtration curve. The Umgeni River was noted to produce flux rates which were higher than those obtainable under normal dead-end. It can be concluded that there was no stabilisation of flux noted. However, feed water with low E.coli and turbidity concentrations enhances the flux rates. The technology was noted to produce water of less than 1 NTU and 100% removal efficiency for E.coli and total coliforms.

• Membrane and Water Treatment
• /
• 제1권1호
• /
• pp.61-74
• /
• 2010

This paper will integrate models at different levels (from filtration, backwashing to chemical cleaning and membrane lifetime) that can be used to minimize overall operating costs of a dead-end ultra filtration process that is used for the purification of surface water. Integration of the models leads to a multi-level optimization problem (at different levels different objectives should be reached). This problem is solved as a MINLP. Systematic modelling and optimization of membrane systems is not extensively discussed in the scientific literature. In this paper the first steps are taken in the formulation of proper models and the use of systems engineering tools to come to real optimal operating conditions. The optimized variables are used to calculate fouling profiles which can subsequently be used as inputs for a control system that actually enforces the profiles to a real pilot plant.