Ultrafiltration (UF) rapidly is becoming a common and practical method of preparing pure water that is free of particulate matter. One of the major concerns associated with membrane filtration is membrane integrity or the intrusion through a membrane barrier resulting in a tear or puncture of the membrane. The loss of integrity allows contaminants such as particles or, more seriously, pathogens to freely pass though the membrane. The result is decreased water quality and increased health risks when the filtered water is consumed. It is critical to have a means of monitoring membrane integrity to ensure membrane integrity during filtration processes.
There are several available methods for monitoring membrane integrity. All have benefits and drawbacks. One method in particular, analysis for turbidity, provides information regarding the integrity of membranes.
Turbidity is a simple qualitative method indicating the presence of particles in a sample. Turbidimetric analysis has a solid proven scientific background and has been used for generations in all aspects of water quality monitoring. Turbidity analysis is applied easily to the monitoring of UF processes for particle intrusion.
What is turbidity?
Specifically, turbidity is the resultant light scatter that occurs when light interacts with particles in a sample. In the absence of particles, light will travel uninhibited in a straight line. If particles are present, some of this light will be scattered in all directions. The more particles that are present, the greater the amount of light scattered. Turbidity is an optical property exhibited by a sample and, in its truest form, is a qualitative measurement. Through the use of standardized instrumentation and calibration standards, quantative aspects of turbidity also can be applied.
Light scatter and the resulting turbidity measurement is affected not only by the amount of particles present but also by their size, shape and composition. For example, light scattering increases as the difference in the refractive index between the particle and water increases. Larger particles tend to scatter long wavelength light more readily than short wavelength light. Conversely, smaller particles tend to scatter shorter wavelengths of light better than long wavelengths. Shape and surface topography have an effect on light scatter. Spherical particles will scatter light differently than rod-shaped particles. The color of particles also affects light scatter due to their light absorption properties.
The key to turbidity analysis is that light scatter results from an array of characteristics and is a collective measure of all the material present in a water sample. Through optimization of instrumentation, turbidity can be used to detect particles in the lowest concentrations and, therefore, to detect a breakdown in membrane integrity.
Turbidity measurement is based on the detection of light scatter. The detection of scatter must be optimized, for light will scatter more effectively in one direction as opposed to another direction. Current instrumentation uses specific design criteria for detection, which are summarized below.
1. The incident light must be columinated. As a columinated incident beam of light enters a sample, particles that are present will scatter the light in directions other than that of the incident light.
2. The detection of the scattered light is set at 90 degrees relative to the centerline of the incident light beam. This angle was chosen because it effectively detects light scatter from particles that are in the size range of 0.1 to 1.0 µm. Coincidently, this also is the particle size filtered out using UF.
3. The optimal wavelength of incident light is between 400 and 600 nm. However, incident light wavelengths up to 900 nm have been demonstrated to be effectively scattered by particles in the 0.1 to 1.0 µm size range.
4. The path-length of the sample cell should not exceed 100 mm. This allows for a broad dynamic range of measurement of turbidity.
5. The spectral response of the detectors should encompass the wavelengths of the incident light. Specifically, a detection system that is sensitive to shorter (400–700 nm) wavelengths of light will be able to detect light scatter from smaller particles in lower numbers.
Instrumentation that is designed around these criteria are referred to as turbidimeters or, specifically, as nephelometers (indicating light scatter at 90 degrees from the incident light path). Figure 1 illustrates the basic concept of a nephelometer.
Since turbidity is such a sensitive tool for the detection of particles in a sample, sampling and measuring techniques are critical in ensuring the analysis is accurate. Samples that are produced by UF would be classified as low turbidity samples and typically have a turbidity less than 0.100 Nephelometric Turbidity Units (NTU). Such low-level samples can be analyzed accurately and consistently only if specific criteria are addressed during analysis. These issues are addressed below.
1. Samples should be analyzed as soon after collection as possible. Over time, the particle size distribution can change.
2. Sample cells must be clean inside and out. The use of strong acid washes coupled with an ultrasonic bath is appropriate for measuring turbidity samples below 0.3 NTU. An ultrafiltered water source should be available for final rinsing of cleaned cells. After rinsing, cells should be capped to prevent particulate contamination.
3. The outside surface of sample cells should be polished using silicone oil. The silicone oil will fill in minor scratches, which will otherwise scatter light and lead to positive measurement error.
4. Sample cells should be indexed and then placed in the sample cell holder at the same orientation each time a measurement is taken.
5. Nephelometers should be kept in a clean environment. Dust will contaminate any optical surface in the instrument and result in positive error.
6. Details on calibration, maintenance and verification as explained in instrument manuals should be followed strictly. This will ensure the instruments remain in top measurement performance over time.
Turbidity measurement can be achieved using a wide array of turbidimeters. Instruments range from on-line (or process) in design to highly accurate laboratory instruments to portable batch instruments. The newest process instruments now are designed using laser-light technology, which has optimized particle event detection. This new generation of instrumentation, referred to as laser nephelometers, displays particle sensitivity that normally only is detectable using sub-micron particle counters and other high-end instrumentation. However, unlike most high-end particle detection instruments, laser nephelometers still possess simple calibration and maintenance schedules that normally are associated with turbidity instruments. Thus, the analyst now has the ability to apply high-end instrumentation to processes such as UF with relatively low cost associated with the instrument, maintenance, calibration and verification.
When used properly, all turbidity instruments should complement each other to within their performance specifications. The key to achieving this agreement is through proper calibration followed by verification in the measurement range of interest. This is true especially in low-level turbidity measurement, where contamination is the major cause of deviation between instrumentation and problems in the preparation of low-level standards. Fortunately, instrument manufacturers have optimized calibration procedures that easily and consistently produce accurate calibrations suitable for low-level measurement. Further, the development of stable calibration standards such as stabilized formazin has enhanced the accuracy of the turbidity calibration and verification processes. Recent development of low-level turbidity standards now provides accurate verification standards down to turbidity levels as low as 0.05 NTU.
The final and most critical issue to be addressed when performing low-level turbidity measurements is verification. After calibration, the use of secondary or verification standards can affirm at any time—as often as needed—that the instrument is performing properly.
This confirms that the instrument is properly maintained and cleaned and that the ensuing measurements being read are indeed meaningful and accurate. Verification is an underrated stamp of approval that confirms the accuracy and reliability of low-level measurement. It is the ultimate proof of the effectiveness of the UF processes when applied to a turbidity-monitoring program.