The goal of processing by reverse osmosis (RO) is to remove undesirable contaminants from water in the most efficient way without fouling the membrane or otherwise damaging equipment. To manage this, system design usually includes pretreatment that either physically or chemically prepares the RO feed for processing. The type of pretreatment is defined by the characteristics of the source water and can involve processes such as microfiltration, carbon adsorption or ion exchange.
Pretreatment can also involve chemical additions, such as antiscalants, corrosion inhibitors or pH adjustments. Whatever the case, water quality testing is critical in determining a cost-effective system design. The same testing equipment used to design the system can also be used to monitor feedwater, permeate and concentrate after implementation to ensure processes are working optimally, as well as for equipment calibration and maintenance.
Characterizing the source water is the first step in determining what kind of system is viable. A total dissolved solids (TDS) measurement determines the type of membrane required. TDS does not measure colloidal or suspended solids, which are more likely to foul membranes, but describes the concentration of all ions completely dissolved or dissociated in solution. Alternatively, conductivity monitor/controllers and measurements can be used that are correlated to TDS levels. TDS, however, is the industry standard for dissolved solids measurement.
A portable TDS meter can be taken to the source to generate the most accurate results. Choose one that is standardized to the solution type you are testing. For example, natural waters, such as lakes, streams and rivers, are best correlated to the Myron L 442-3000 Natural Water Standard Solution. All readings should be compensated to 25˚C to establish a basis for comparison.
Beyond assessing system load, a TDS meter is useful for calibrating TDS monitor/controllers that alert to system failures and for checking permeate and concentrate ionic concentration. Comparing daily values and detecting trends can help predict the life of the membranes and is invaluable in troubleshooting problems with system performance, such as detecting scaling.
Controlling microbiological fouling is paramount for optimum system performance and for extending the life of membranes. Chlorine is the predominant sanitizer used. Membranes such as thin-film composite membranes, however, degrade when exposed to chlorine. This requires removal of chlorine from the feedwater, usually
by carbon adsorption or sodium bisulfite addition.
Checking free chlorine levels before the feed enters the RO unit, either by oxidation reduction potential (ORP) or free chlorine measurement, allows the operator to divert the stream and spare the equipment in the event of a chlorine breakthrough and to ensure adequate chlorine removal on a day-to-day basis to preserve membrane life.
Free chlorine can be monitored by ORP in systems where chlorine is the only sanitizer used. However, ORP gives the operator a better picture of all chemicals in solution that have oxidizing or reducing potential when chlorine is not used, such as bromine, chloramines, chlorine dioxide, peracetic acid, iodine or ozone.
ORP measurements can be positive or negative, indicating whether the solution is oxidizing or reducing, respectively. A very positive reading (for example, +900 mV) indicates a solution with a strong oxidizing potential, which effectively destroys bacteria and other organic matter, but also damages RO membranes. ORP greater than +300mV is generally considered undesirable for membranes. Check manufacturer’s specifications for tolerable ORP levels, if indicated.
An inline ORP monitor/controller can be placed ahead of the RO unit to automatically monitor for trends and breakthroughs, while a handheld meter can be used to spot-check and calibrate the monitor/controller and check permeate and concentrate ORP and free chlorine levels (which should effectively be zero).
Analyzing pH of the source water determines whether or not acid additions may be indicated to balance water according to the Langelier Saturation Index (LSI) and optimize performance of antiscalants, corrosion inhibitors and anti-foulants. This is because pH indicates the availability of hydrogen ions in solution, which affects the types and concentrations of chemical constituents in solution. For example, it is a well-known fact that the most effective species of free chlorine, hypochlorous acid, is at peak concentration at a pH of 6.5. All chemical additions will specify an optimum pH range within which they are most effective or effective at all. pH adjustments, typically acid injections intended to reduce the potential of the solution to scale, are made well ahead of the RO unit to ensure proper mixing and to avoid pH hotspots. A pH monitor/controller can be employed to automatically detect and divert solution with pH outside the range of tolerance for the RO unit. A handheld meter is used to spot-check and calibrate the monitor/controller as part of routine maintenance and to ensure uniform mixing.
Most antiscalants will specify a LSI maximum value below which the calcium carbonate remains suspended in solution or is unable to seed or react with seeds to form crystals. pH, alkalinity, hardness and temperature are the major factors in LSI. pH affects the carbonic acid equilibrium directly through the availability of the hydrogen ion and its interaction with bicarbonate and carbonate.
Practically, if pH is above 7, acid additions are used to shift the equilibrium to favor carbonic acid, which dissociates into water and carbon dioxide. At a pH of 6, 80% of the carbonate ion is converted to carbon dioxide gas, effectively lowering the concentration of the carbonate ion in solution. Calcium hardness and alkalinity account for the presence and propensity of calcium carbonate to form. Temperature also affects the solubility of calcium carbonate. Calcium carbonate is less soluble at higher temperatures, unlike other scalants. These measurements can be made separately, but some instruments can complete all measurements as part of an LSI function that generates an LSI value in situ.
The LSI is not an indication of the amount of scale or corrosion going on in a system at any given time, but rather a statement about the system’s tendency to scale calcium carbonate. In the absence of any standard method for determining the proclivity of other constituents in solution to scale, such as silicates, many chemical manufacturers and control systems develop their own proprietary methods based on solubility constants in a defined system. However, the LSI is still useful as a scaling indicator because calcium carbonate is present in most water.
Testing water for calcium hardness is helpful in determining whether or not ion exchange beds are indicated to soften the water before RO processing. Check hardness values directly after the softening process regularly to ensure it is working properly and to determine a regeneration schedule.
Silt Density Index (SDI) is measured to determine what kinds of filtration will be required to remove suspended solids that will foul an RO membrane. Generally, an SDI of less than 5 is required of the feedwater, which is equivalent to an NTU of less than 1. Processes such as microfiltration, ultrafiltration and media filtration are used to remove colloidal solids from the feed. Polyelectrolyte additions are sometimes used, and the effectiveness of these additions are pH dependent, which requires initial pH testing and adjustment to achieve the point of greatest efficacy.
Testing and monitoring pressure is a good way to evaluate system requirements and performance over time. The pressure differential between the feed and the concentrate stream should remain about constant. If this differential reading increases, investigate for scaling or rupture of membranes. It makes sense that if the pressure differential increases over the first stage in a two-stage system, the likely cause is either biological fouling or sediment. If the pressure differential increases over the second stage, the most likely cause is scaling by insoluble salts. This means that degradation in performance is likely due to the dissolved solids in the feed. Evaluating LSI and the factors that affect it can help remediate problems. An LSI calculator is useful in determining adjustments required to bring the system into balance. Also, monitoring TDS is the simplest method to determine performance.
Monitoring permeate rate is also important. A drop in transmembrane pressure usually indicates fouling on the concentrate side of the membrane. A common error made by new system operators is to increase the feed pressure to increase the permeate rate. This only increases the rate of fouling, further decreasing the permeate flow and potentially damaging the system. System maintenance is generally indicated if there is either a 10% to 15% drop in performance or permeate quality as measured by TDS.
If the system is designed correctly, monitoring pretreatment and treatment processes through water testing will ensure you deal with problems proactively, saving you time and money in the long run.