Silt Density Index (SDI) Kit - Manual

Advanced Reverse Osmosis Technology

Silt Density Index has become the accepted standard for assessing the suitability of membrane processes, particularly reverse osmosis or RO. The SDI value is merely a measure of the decline in filtration rate of a membrane filter under standard test conditions. In effect the test is an indirect "measure" of all suspended solids, bacterial, and colloidal matter in the water to be treated through an attempt to sacrificially plug a microporous cellulose ester membrane filter.

Accepta's manual Silt Density Index kit - In principle the measurement of SDI is easy. A simple apparatus is used although some industries may specify the type of membrane housing, hose connections and manufacture of filter paper for specific validation protocols. The procedure it is not quite as easy as it seems, but with practice is quite satisfactory. At the outset it is important, having carefully loaded the membrane filter, to ensure all air is bled from the inlet side of the filter housing. Once ready to commence a test the main concerns are:

Initially an operator has to deftly accomplish several operations simultaneously: start the flow; adjust the pressure; start the stopwatch; and suitably position a measuring cylinder for collecting the measured sample. Then feed temperature, pressure, sampling times and collected volumes must be monitored attentively.
The degree of concentration required to run a test is such that it is not the kind of operation that allows attention to avert to other site tasks at the same time.
At best it is reasonable to allow 30 minutes per test, and therefore it is unlikely that more than one value per shift will be logged.
The tediousness of carrying out many manual measurements means that there is a low incentive for using SDI to truly monitor feed quality and the performance of any pre-treatment. Therefore there is a temptation to overlook the value of SDI as a routine measurement, and only revert to it when there is a problem.
There is no scope for the logging and treatment of data other than by manual data collection and entry into a PC database or equivalent programme.

Given the constraints listed above it is not surprising that repetitive SDI determinations could be prone to operator error thus affecting the precision of the method.

The Silt Density Index (SDI) test is used to determine the fouling potential of water feeding a membrane filtration process such as a reverse osmosis (RO) system. This test is defined by its specific procedure (ASTM D-4189). The ASTM procedure should be referenced for a more detailed description of the procedure.

The nature of this test is such that it cannot be run in the laboratory. The SDI test should be run daily on the water entering the RO membranes after the cartridge filters. This frequency can be reduced to weekly once background data proves that less frequent sampling is sufficient. As such, a sample tap should be installed on the RO machine after the cartridge filters.

This test can also be run across vessels such as filters or clarifiers to see if they are doing the job expected of them. SDI tests on the raw supply water should be part of every feasibility study for an RO system and it is good to run one periodically during operation of the system to make sure changes haven't occurred.

It is recommended practice to keep a record of SDI values and filters to observe changes over time.

Principle of Measurement

The initial time (tI) is noted for a given quantity of the water to be tested, 1usually 500 ml., to pass through a new 47mm dia. 0.45 micron rated 2membrane filter under a constant pressure of 30 psig. Clean water passes at about 450 ml per minute depending upon the properties of the membrane used. The test continues uninterrupted and the same measurement repeated at time (tf) at intervals of 5 minutes 10 minutes, and 15 minutes. The SDI index is calculated as the percentage decay in flow per unit time with the 15 minute value taken as the test standard. Since time to pass a given volume of water is inversely proportional to flow the SDI may be expressed numerically as follows:

SDI15 = ((1 ? ti¬¬ / tf) x 100)/15 --------------- (i)

= (P30¬)%/15 ----------------------(ii)

Where: P30 is the percent plugging at 30 psig. or plugging factor
TABLE 1 gives values of SDI and (P30¬¬¬)% for various ratios of ti / tf.
The percent plugging should not exceed 75% for any 15 minute test (SDI = 5), and if it does the SDI value is calculated for the shorter time of 10 minutes, and if necessary, 5 minutes. For a manual test, If the water is so bad that the membrane filter exceeds 75% plugged in less than 5 minutes then the time to completely 3plug the membrane is noted (T), and a modified SDI estimated as:

MFI = 100/T

1. If ( ti ) is greater than 110% of time to pass 500ml of non-plugging water a smaller sample volume of 100 ml. should be chosen. A non-plugging water can be prepared by passing distilled water through an 0.2 absolute micron filter.

2. It is important to state the manufacturer and specification of the filter membrane since the results are specific to the filter paper used. It has been observed that results vary between membrane source and even between lot No's. from the same source. The 500 ml sample is based on a 47mm diameter filter, for a smaller diameter filter membrane the sample volume is reduced in direct proportion to the filter cross sectional area.

3. Complete plugging may be defined as occurring when the flow has declined to about one drop every 20 seconds.

Experience shows that the membrane fouling rate, and therefore the cleaning frequency, of RO modules will usually be within acceptable limits given an SDI15 value of not greater than 5 for spiral wound membranes, but for a hollow fibre type a value of 3 may be stipulated. Objective selection criteria are not altogether encouraged by the fact that membrane manufacturers' guarantees often imply stated feed SDI values are met. An SDI value greater than 5 does not necessarily preclude the selection of RO as a viable process since, unlike the SDI test itself, RO is a cross-flow separation process whereby not all the suspended solids / colloidal matter is presented to the membrane in a direction normal to its surface. For feed SDI above 10 some kind of RO pre-treatment is not unusual, and again SDI determinations can prove very useful in the assessment and selection of such treatments. For example multi-media filtration, with or without coagulant addition, and various microfiltration techniques. SDI can even be used to monitor post membrane treatments by using an SDI test based on a filter of appropriate absolute rating lower than the standard 0.45 microns.

Additional Information

For additional information please contact Reverse Osmosis Chemicals International on +44 (0) 161 877 2334 or e-mail sales@rochemicals.com.