Full Description

The formation of micro-bubbles in immersed membrane processes was examined toevaluate Silt Density Index (SDI) interferences. Water produced by ultrafiltration (UF)membrane processes is well suited to feed reverse osmosis (RO) systems and generallyexhibits silt density indices of approximately 2. In some instances, the formation ofmicro-bubbles in immersed membrane systems can lead to higher SDI values as a resultof bubbles blocking pores within the 0.45m SDI filter paper. These higher SDI valuesdo not correspond to higher fouling rates for downstream RO processes since micro-bubblesdo not adversely affect RO filtration. In this study, several methods were used to prove thatthe presence of micro-bubbles can interfere with SDI measurements.The first method involved filtering UF permeate through an SDI filter paper and subsequentlyvacuum drying the filter paper to examine the permeability that was recovered. Thepermeability recovery should give an indication of the amount of original permeabilityreduction that was associated with the presence of micro-bubbles.The second method involved collecting SDI values measured on two types of UFpermeate. The same feed water was sent to both an immersed UF system (where avacuum is used to drive the process) and a pressurized UF system. The results using thesame feed water and same membrane showed average SDI values that were 50% higherfor the vacuum driven system where it is expected that micro-bubbles are present.A third method repeated the previous test, but with the raw feed being continuouslyaerated. Again the results showed significantly higher SDI values for the permeateassociated with the vacuum system. The SDI values were much higher for both systemscompared with the non-aerated tests. A fourth method involved conducting multiple SDI tests on RO permeate and comparingthe results with SDI values found on the same RO permeate that was pressurized while incontact with air and then depressurized to form micro-bubbles. The RO permeate thatwas loaded with micro-bubbles showed SDI values that were significantly higher.A final demonstration is presented where a high pressure pump was used to re-dissolvemicro-bubbles present in UF permeate. SDI values obtained from UF permeate werecompared with the SDI values from the same UF permeate where the micro-bubbleconcentrations were reduced by sampling after the UF permeate was pressurized to~6000 kPa (900 psi). The average SDI values of the UF permeate were significantlyhigher than those obtained after the water was pressurized.The data that is shown in this paper presents irrefutable evidence that the presence ofmicro-bubbles in a solution causes elevated SDI values. It also points to a practicalsolution for reducing SDI values by taking SDI samples downstream of a high pressurepump. In most cases SDI values are only of interest when an RO system is presentdownstream of a UF system, therefore, a high pressure pump is usually available toprovide this pressure. Includes 2 references, figures.

Product Details

Edition: Vol. - No. Published: 03/01/2007 Number of Pages: 11File Size: 1 file , 340 KB