No.

1 September 2003

Silt Density Index (SDI) : Measurement And Significance In Membrane Applications

epleting supplies of fresh water, ground water contamination and the rising cost of water have compelled industries to increasingly turn to desalination of brackish/sea water. The recent years have also witnessed a phenomenal increase in water recycle plants with a terminal reverse osmosis (RO) system for removal of dissolved solids. Water is recovered from treated industrial effluent as well as secondary treated municipal sewage. The presence of finely divided particles in feed water to an RO system can cause serious problems; as they are too small to be retained by conventional pretreatment systems they find their way to the RO membrane. The function of an RO membrane is to concentrate dissolved solids on one side of the membrane and allow water molecules to permeate through the other side. Suspended, colloidal particles are larger than dissolved solids and are also rejected by the membrane. These fine undissolved suspended particles tend to settle on the surface and, in the process, foul the membrane. The net result is an increase in feed pressure, reduced flow and high conductivity of permeate. Spiral wound and hollow fine fibre membranes are the most vulnerable and once fouled, cleaning of the element is often difficult.

Silt Density Index

Silt Density Index (SDI) is an empirical test developed for measuring the rate of fouling of nano filtration (NF) or RO membranes. It represents the potential for fouling of the membranes by finely suspended particles present in feed water to the membranes. There is no direct correlation between turbidity of a water stream and its SDI. Experience has shown that water with a very low turbidity (< 1 NTU) may have a high SDI value. Deep tubewell waters, as a rule, are free from suspended solids and exhibit a low turbidity value/SDI. On the other hand, surface water invariably contains high levels of fine suspended solids and therefore exhibits high turbidity as well as high SDI, calling for extensive pretreatment to achieve acceptable values of SDI and turbidity in feed water. The measurement, its significance and the interpretation of the results of SDI tests are described below. The equipment used and the procedure (ASTM standard test method D 4189-82) are also given.

Measuring SDI
Apparatus
The SDI assembly is schematically described in Figure 1. All wetted parts should be made of high quality stainless steel or plastic to prevent contamination by corrosive products. Do not use reactive materials such as carbon steel or cast iron. Suitable filter holders, designed to withstand an operating gauge pressure of 350 kPa (50 psi) can be obtained from suppliers of membrane filters.

Other components required for the test are as follows: Membrane filter, 47 mm in diameter, gridded, and with a mean pore size in the range of 0.45 0.02 micron, inclusive. Use only filters that are packaged in the same orientation. Graduated cylinder, 500 ml capacity. Stop watch, graduated in hundredths of a minute. Thermometer, liquid-in-glass, suitable for measuring the temperature of the water sample, capable of being read within 1C.

and 500 ml samples. Repeat again after 10 and 15 minute intervals. 9. If the time required to obtain a 100 ml sample is greater than 60 seconds, pluggage will be about 90%, and it is not necessary to continue the test. 10. Measure the water temperature again to ensure that it did not vary by more than 1C from the initial temperature. 11. After completing the test and disconnecting the apparatus, the filter paper may be saved in a plastic bag for future reference. Note: For best results Use dull tweezers when positioning the filter to prevent puncturing the filter. Ensure that the O-ring is clean and in good condition, and is properly positioned. Avoid touching the filter with fingers. Flush the apparatus to remove any contaminants that may be in it.

Test Procedure
The procedure is based on the time required to filter a volume of feed water through a 0.45 micron filter paper at a feed pressure of 30 psig at start and then after 5, 10 and 15 minutes of continuous filtration. If the test is limited to only 5 to 10 minutes reading, due to the plugging of the filter pad one can expect a high level of plugging of the membrane. 1. Assemble the test equipment as per Figure 1. 2. Locate a sample tap on the feed water piping and install the test equipment. 3. Adjust the pressure regulator to 30 psi with a filter pad installed. Use a fresh filter for the actual test. 4. Take the temperature of the feed water. The temperature should not vary more than 1C between the start and end of the test. 5. Bleed any entrained air in the filter holder. Depending on the model of the filter holder, either open the bleed valve, or loosen the filter holder while cracking the ball valve. Then close the bleed valve or filter holder. 6. Place the 500 ml graduated cylinder under the filter to measure the amount of water that passes through the filter. 7. Open the ball valve fully, and measure the time required to collect 100 ml and 500 ml from the time the ball valve is opened. Record these times, leaving the valve open and letting the flow continue. 8. After 5 minutes, repeat the time measurement required to collect 100 ml

Calculations
SDI = P30 /Tt = 100*(1- Ti/Tf) / Tt Where SDI = Silt Density Index P30 = % pluggage at 30 psi feed pressure ** Tt = Total test time in minutes (usually 15 minutes, but may be less if 75% pluggage** occurs in less than 15 minutes). Ti = Initial time in seconds required to obtain sample. Tf = Time required to obtain sample after 15 minutes (or less). Note: * Time to collect 500 ml should be approximately 5 times greater than the time to collect 100 ml. If 500 ml collection time is much greater than 5X, SDI should be calculated using 100 ml collection times. ** For accurate SDI measurements, P30 should not exceed 70%. If P30 exceeds this value, re-run the test and obtain Tf at a shorter time (T).

Limitations
The test, although popular, is at best an empirical method and its interpretation calls for some expertise on the part of the person carrying out the test. Feed pressure must always be maintained at around 30 psi as any variation in feed pressure during the run can lead to erroneous results. SDI may vary with the make of the filter used; hence it is suggested that filter papers of the same make be used for each study. SDI is also a function of the water temperature and values obtained at different temperatures may not necessarily be comparable. Temperature should not vary more than 1% C between the start and end of the run.

Interpretation of Results
SDI Value
1. Value higher than 6: Indicative of very high potential for fouling. Do not attempt to continue passing the feed through the membranes till the pretreatment is set right and a value lower than 4 is obtained. 2. SDI Profile: SDI reduction across various equipment like multigrade (MGF), activated (ACF) and cartridge (CF) filter should be checked regularly. The following conclusions can be drawn from the results: i) MGF: SDI reduction across the MGF should be maximum. If no significant SDI reduction is seen across the MGF, this indicates that the size of the particles in the raw water is less than 25 m. Hence, coagulants/flocculants should be added to increase the size of the particles and improve the removal efficiency of the MGF. ii) CF: SDI reduction across the CF should be minimum. If a significant reduction in SDI is seen across the CF, it will foul very frequently. The remedy is to improve the particle removal efficiency of the MGF. 3. Colour of filter paper used for SDI test: i Slightly yellow The possible problem could be presence of iron and/or organics in feed water. Check the source for presence of iron and organics in the water source. Also check for possible corrosion in the lines. Positive indication of presence of iron in feed water. Take corrective steps for removal of iron. Indicates presence of carbon fines. Check the carbon filter, remove carbon fines. Check the cartridge filter and the filter elements. If the colour dissolves in the acid, it indicates the presence of manganese in the water.

ii iii iv v

Reddish brown Gray Particles present on the filter paper Dark Black

Significance and Use of SDI

1. SDI can serve as a useful indication of the fouling potential due to undissolved particles of the feed water to NF and RO membranes. 2. SDI can be used to ascertain the effectiveness of the pretreatment processes like clarification and filtration.

Pretreatment Options
Membrane manufacturers and system designers insist that the value for SDI (15 min) of feed water to membranes should be as low as possible and in any case should not exceed 4.0, to minimise fouling of membranes. The index is also used to evaluate the effectiveness of various pretreatment processes like clarification and filtration. Experience has shown that conventional pretreatment plants using clarification and filtration through granular media are often inadequate in reducing the turbidity and SDI of feed water streams. Industries now rely more on the use of membrane processes like ultrafiltration especially when dealing with industrial effluents or secondary treated municipal sewage as feed water to RO systems for water recovery. Ultrafiltration systems have already replaced the conventional pretreatment plants at several industrial installations like Madras Fertilisers Ltd., Chennai and Sundaram Fasteners, Madurai.

The following graph shows SDI reduction through ultrafiltration:

654-

SDI15

32100.0

10.0

20.0

30.0

40.0

50.0

60.0

70.0

80.0

90.0 100.0 90.1% 98.4%

Percentage of SDI measurements

Figure 2

SDI reduction of less than 3.0 is achieved in 98.4% of the cases. SDI reduction of less than 2.0 is achieved in 90% of the cases.

ilt Density Index has come to stay as one of the most important feed water limiting parameters that must be considered both during design and commissioning of water treatment/ recycle plants based on either RO or NF membranes to minimise operational problems that can occur due to fouling of membranes by finely divided particles present in feed water. A carefully designed pretreatment plant aimed at reducing the SDI to as low a value as possible will go a long way in troublefree operation of the plant.

Ion Exchange (India) Ltd., Tiecicon House, Dr. E. Moses Road, Mahalaxmi, Mumbai 400 011, India. Phone : 91-22-2493 9520/23/25/31 Fax : 91-22-2493 8737 E-mail:ho.techno@ionexchange.co.in Website : www.ionindia.com

09033K