Although the basic scientific principles behind membrane technology had been developed by the 1950s, it was not until the 1970s that crossflow membrane technology, in the form of UF and RO, began to be recognized as an efficient, economical and reliable separation process. RO was first used primarily for desalting brackish and sea water. In the last six to eight years the technology has gained industry acceptance as a viable water treatment option for many different fluid separation applications. Low operating costs and the ability to remove organic contaminants and 95-99% of inorganic salts with minimal chemical requirements make RO an attractive technology for many industrial applications.
Introduction to Membrane Technology
To understand membrane technology, it is important to understand how crossflow membranes work. In simple terms, a crossflow filtration system separates an influent stream into two effluent streams known as the permeate and the concentrate. The permeate is the portion of the fluid that has passed through the semipermeable membrane. The concentrate stream contains the constituents that have been rejected by the membrane.
An inherent design advantage of crossflow filtration is its ability to operate continuously in a self-cleaning mode. The membrane system is self-cleaning because suspended solids and rejected solutes are constantly being swept away from its surface. In contrast, conventional filtration is traditionally performed under 'normal' or 'perpendicular' flow conditions, where the entire influent stream flows through the filter, except for large particles which are trapped by the filtration media. Typical equipment used for normal flow filtration includes multimedia filters, sand filters, bag filters and cartridge filters. This type of filtration can also be used as pretreatment to a crossflow membrane system to remove large particles and suspended solids.
Crossflow membrane filtration is generally divided into four groups: RO, NF, UF and microfiltration (MF). The filtration spectrum (Figure 1) displays the various filtration groups based upon the size of the particles they can remove. RO membranes have the smallest pore structure, with pore diameter ranging from approximately 5-15 angstroms. The extremely small size of RO pores allows only the smallest organic molecules and uncharged solutes to pass through the semipermeable membrane along with the water. Greater than 95-99% of inorganic salts and charged organics will also be rejected by the membrane due to charge repulsion established at the membrane surface.
Nanofiltration is typically referred to as 'loose' RO due to its larger membrane pore structure, which allows more salt passage through the membrane. Because it can operate at much lower pressures, and passes some of the inorganic salts, NF is used in applications where high organic removal and moderate inorganic removal are desired. An advantage of NF over RO is that NF can typically operate at higher recoveries, thereby conserving total water usage due to a lower concentrate stream flow rate.