REVERSE OSMOSIS FUNDAMENTALS

Reverse osmosis is a modification of the natural process known as osmosis. A French scientist first described the phenomenon of osmosis in 1748. This scientist noted that water spontaneously diffused through a pig bladder membrane into alcohol. Over 200 years later, researchers modified the process of osmosis and discovered and patented the process now known as reverse osmosis. The reverse osmosis process allows the purification of contaminated water by removing dissolved and suspended matter. Reverse osmosis has become a key process technology. Its use in industrial applications has had major advancements since the early 1960s.

reverse osmosis modules Reverse Osmosis (RO) uses a specialized membrane material that is selective about what it allows to pass through, and what it prevents from passing. The RO membranes pass water very easily because water has a small molecular size, but membranes stop many other contaminants from passing through. Only water and small non-ionized (or non-charged) molecules are allowed to pass through. RO has become very popular for water purification applications due to its ability to remove both suspended and dissolved impurities without the need for regenerant chemicals. Reverse Osmosis systems can be found today in a wide range of facilities: kitchens, hospitals, refineries, power plants, semiconductor manufacturing facilities, manned spacecraft, sailboats, etc.

Reverse Osmosis is a process that separates impurities from water by passing the water through a semipermeable membrane. The semipermeable membrane only allows very small atoms and groups of atoms such as water molecules, small organic molecules, and gases, to go through it. Hydrated ions, or ions that have been dissolved and are therefore surrounded by water molecules, cannot pass through the membrane. In order to understand reverse osmosis, the osmosis process needs to be understood. Once osmosis is understood, reverse osmosis can be clearly explained.

Osmosis and Its Cause

When two solutions with different dissolved mineral concentrations are separated by a semipermeable membrane, water flows from the less concentrated solution to the more concentrated solution. The water level rises on the more concentrated side. The dilution of the solution with the higher concentration is caused by the process called osmosis.

The simple definition of osmosis is the tendency of a fluid to pass through a somewhat porous membrane until the concentration on both sides is equal. It is better defined as follows: Osmosis is the migration of water molecules across a membrane caused by the attraction of the dipole moment of water molecules to ions and polar molecules on the other side of a membrane. Water molecules are attracted to ions on both sides of the membrane. The solution with a higher concentration has a greater number of dissolved ions, therefore, a greater number of water molecules are attracted to that solution. Ions, which are hydrated when in water, tend not to migrate through the membrane to equalize concentrations because of their hydrated size. Since the large hydrated ions do not migrate through the membrane, there is a net migration of water molecules through the membrane from the lower concentration solution to the side with higher concentration.

The result of this migration is that the water level rises on the higher concentration side, and decreases on the lower concentration side. This force driving molecules from one side of a membrane is called osmotic pressure. The direction of the osmotic pressure is always from the dilute solution to the concentrated solution. Osmotic pressure can be defined as the pressure and potential energy difference that exists between two solutions on either side of a semipermeable membrane. A rule of thumb for osmosis is that 1 psi of osmotic pressure is caused by every 100 ppm (mg/l) difference in total dissolved solids concentration (TDS).

If sufficient pressure is applied to the concentrated solution, we can actually reverse the direction of flow that would normally be caused by osmosis.

The driving pressure, and therefore the flow of water across the membrane, is reversed. The water begins flowing toward the solution with the lower concentration. This process is called Reverse Osmosis. Reverse osmosis is a process that forces water molecules to flow against a net osmotic pressure. This is accomplished by applying enough pressure on the high concentration side of a semipermeable membrane to reverse the net migration of water molecules. Thus, with adequate pressure, reverse osmosis can remove purified water from a sample containing higher concentrations of dissolved solids.

A very serious challenge for classical, staged RO operation is scaling due to the high concentrations of minerals that accumulate within RO elements. Concerns about scaling often far outweigh concerns over increased osmotic pressure. Numerous factors ranging from scaling to the required flow velocities on the feedwater side of an element influence what these flow rates should be for a given system.

The Reverse Osmosis process has both advantages and disadvantages. These must be weighed when deciding whether to use a Reverse Osmosis system. Some of the advantages of using an RO system are: 1) RO outperforms any filter on the market (even ultrafiltration) with respect to the size of particles rejected. RO is actually a "molecular filter". Its ability to reject dissolved substances depends on the hydrated size of the molecules or ions in the solution. Non-ionized gases and small organic molecules, which are small as a result of not being hydrated with water molecules, have a poor rejection rate. membrane separation methods such as ultrafiltration and conventional filtration. 2) RO removes up to 99.9% of the dissolved impurities in water without using regenerant chemicals.

Some of the disadvantages of using an RO system are: 1) The pump requires a significant amount of power to provide the driving pressure for the system. 2) Reverse Osmosis rejects a certain percentage of the feedwater as concentrated waste. Reject quantities typically vary from 15% to 40% of the water processed. This may be more waste water than an ion exchange system generates, depending on the system. In general, the higher the mineral content of the feedwater, the more advantageous RO is.

 

Membrane and Element Construction


The two primary membrane module configurations used for reverse osmosis applications are hollow fiber and spiral wound. Two other types of membrane configurations are tubular and plate and frame. These two have found acceptance in the food and dairy industry and in other special applications, but modules of these configurations have been less frequently used in water treatment applications.

HOLLOW FINE FIBER (HFF) MEMBRANES

This configuration uses membrane in the form of hollow fibers that have been extruded from cellulosic or polymeric material. The fiber is asymmetrical in structure and is as fine as a human hair. It typically measures about 42 micron (0.0016 inch) ID (inner Diameter) and 85 micron (0.0033 inch) OD (Outer Diameter). Millions of these fibers are formed into a bundle and folded in half to a length of approximately 120 cm (4 ft). A perforated plastic tube serving as a feedwater distributor is inserted in the center and extends the full length of the bundle. The bundle is wrapped, and both ends are epoxy sealed to form a sheet-like permeate tube end and a terminal end that prevents the feed stream from bypassing to the brine outlet. The hollow fiber membrane bundle, 10 cm to 20 cm (4 to 8 inches) in diameter, is contained in a cylindrical housing or shell that is approximately 137 cm (54 inches) long and 15 - 30 cm (6 - 12 inches) in diameter. The assembly is called a permeator. The pressurized feedwater enters the permeator feed end through the center distributor tube, passes through the tube wall, and flows radially around the fiber bundle toward the outer permeator pressure shell. Water permeates the outside wall of the fibers into the hollow core or fiber bore, through the bore to the tube sheet or product end of the fiber bundle, and exits through the product connection on the feed end of the permeator.

Concentration polarization is the ratio of the salt concentration in the membrane boundary layer to the salt concentration in the bulk stream (See Section 6.1.8-B for a more thorough discussion). The most common and serious problem resulting from concentration polarization is the increasing tendency for precipitation of sparingly soluble salts and the deposition of particulate matter on the membrane surface. Concentration polarization is worse under conditions of laminar flow. Laminar flow occurs when the velocity is so low that there is no turbulence in the water - the flow moves in "layers" with little or no mixing between layers.

Turbulence helps to mix the concentrated fluid at the membrane surface with the (relatively) dilute fluid in the bulk of the solution. An absence of turbulence allows the concentrated fluid at the membrane surface to become even more concentrated with respect to the bulk stream, causing concentration polarization. In a hollow fiber module, the permeate water flow per unit area of membrane is relatively low (because of the very high surface area of fibers) and may be laminar; therefore, concentration polarization is high at the membrane surface. Care must be taken to ensure that scaling and fouling of the membrane surface do not occur. The hollow fiber unit allows a large membrane area per unit volume of permeator, which results in compact systems. Hollow fiber permeators are available for brackish and sea water applications. Brackish water contains dissolved solids below about 15,000 parts per million. Hollow fine membranes are made of cellulose acetate blends and aramid (a proprietary polyamide type material in an anisotropic form). Hollow fiber membranes require feedwater with a lower concentration of suspended solids compared to the requirements of the spiral wound membranes. This is because of the very close packed fibers and tortuous feed flow inside the hollow fiber membranes. This is one of the reasons that hollow fiber modules are not as popular as spiral wound modules in the water treatment field.

 

SPIRAL WOUND MEMBRANES

In a spiral wound configuration, two flat sheets of membrane are separated with a permeate collector to form a leaf. This assembly is sealed on three sides, with the fourth side left open for permeate to exit. A feed/brine spacer material sheet is added to the leaf assembly. A number of these assemblies, or leaves, are wound around a central plastic permeate tube. This tube is perforated to collect the permeate from the multiple leaf assemblies. The typical industrial spiral wound membrane element is approximately 100 or 150 cm (40 or 60 inches) long and 20 cm (8 inches) in diameter.

The feed/brine flow through the element is a straight axial path from the feed end to the opposite brine end, running parallel to the membrane surface. The feed channel spacer induces turbulence and reduces concentration polarization. Manufacturers specify brine flow requirements to control concentration polarization by limiting recovery (or conversion) per element. Recovery is a function of the feed-brine path length. In order to operate at acceptable recoveries, spiral systems are usually staged with three to seven membrane elements connected in series in a pressure tube (or housing). The brine stream from the first element becomes the feed to the following element, and so on for each element within the pressure tube. The brine stream from the last element exits the pressure tube to waste or to feed another tube. Permeate from each element enters the permeate collector tube and exits the tube as a common permeate stream. A single pressure tube with six membrane elements connected in series can be operated at up to 50-percent recovery under normal design conditions. Each membrane element has a brine (or chevron) seal around the outside of the element at the feed end. The shape of the brine seal is designed to expand when the feedwater pressure pushes against it. When installed correctly, the brine seal prevents the feed/brine stream from bypassing the element.

 

Cellulose Acetate vs. Thin Film

The semipermeable membrane used for reverse osmosis systems consists of a thin film of polymeric material cast on a fabric support. The membrane must have high water permeability and a high degree of semipermeability. The rate of water transport must be much higher than the rate of transport of dissolved ions. The membrane must be stable over a wide range of pH and temperature and have good mechanical integrity. There are two major groups of polymeric materials that can be used to produce satisfactory reverse osmosis membranes: cellulose acetate (CA) and polyamide (PA). Membrane manufacturing, operating conditions and performance differ significantly for each group of polymeric material.

CELLULOSE ACETATE MEMBRANE

The original cellulose acetate membrane, developed in the late 1950s by Loeb and Sourirajan, was made from cellulose diacetate. Current CA membrane is usually made from a blend of cellulose diacetate and triacetate. The membrane is formed by casting a thin film of an acetone-based solution of cellulose acetate polymer with swelling additives onto a non-woven polyester fabric. During casting, the solvent is partially removed by evaporation. After the casting step, the membrane is immersed in a cold water bath, which removes the remaining acetone and other leachable compounds. Following the cold bath step, the membrane is annealed in a hot water bath at a temperature of 60 - 90°C. The annealing step improves the semipermeability of the membrane with a decrease of water transport and a significant decrease of salt passage. After processing, the cellulose membrane has an asymmetric structure with a dense surface layer of about 1000 - 2000 A (0.1 - 0.2 micron), which is responsible for the salt rejection property. The rest of the membrane film is spongy and porous and has high water permeability. Salt rejection and water flux of a cellulose acetate membrane can be controlled by variations in the temperature and duration of the annealing step. Water flux is defined as U.S. gallons of permeate produced per square foot of active membrane area per day (GFD). Using the metric system, this definition would be liters of permeate produced per square meter of active area per hour (LMH).

COMPOSITE POLYAMIDE MEMBRANES

Composite polyamide membranes are manufactured in two distinct steps. First, a polysulfone support layer is cast onto a non-woven polyester fabric. The polysulfone layer is very porous; it does not have the ability to separate water from dissolved ions. In a second, separate manufacturing step, a semipermeable membrane skin is formed on the polysulfone substrate by interfacial polymerization of monomers containing amine and carboxylic acid chloride functional groups. This manufacturing procedure enables independent optimization of the distinct properties of the membrane support and salt rejecting skin. The resulting composite membrane is characterized by a higher specific water flux (more water per psi of pressure) and lower salt passage (purer permeate water) than that of a cellulose acetate membrane. Polyamide composite membranes are stable over a wider pH range than cellulose acetate membranes. Polyamide membranes, however, are susceptible to oxidative degradation by free chlorine, while cellulose acetate membranes can tolerate limited levels of exposure to free chlorine. Compared to a polyamide membrane, the surface of cellulose acetate membrane is smooth and has little surface charge. Because of the neutral surface and tolerance to free chlorine, cellulose acetate membranes usually have a more stable performance (do not foul as quickly) than polyamide membranes in applications where the feedwater has a high fouling potential, such as with municipal effluent and surface water supplies.

Pressure Tube Construction

Spiral wound elements are placed in a structure called a pressure tube. Systems that use 8" x 40" elements typically have 6 elements per pressure tube. The pressure tubes are designed with two connections at each end. One of the connections is located at the center of the end cap and the other is to the side of this center connection. The middle, or center, connection is always the product outlet for vessels containing spiral wound elements. The offset connection on the end cap serves as a feedwater inlet on one side and a concentrate outlet on the other side of the vessel.

Terminology

Specific terms are used to identify certain aspects of the process of reverse osmosis. As water passes through the RO element, it separates into two flow paths. 1) One flow path contains water that has passed through the semipermeable membrane, removing 90 to 99.9% of the dissolved solids, and virtually all suspended solids. This water is called PERMEATE (water that has permeated the membrane) or PRODUCT water (water produced by the system). Either term can be used to refer to water that has passed through the membrane. 2) The second flow path is made up of feedwater that remains on the feedwater, or concentrate, side of the membrane. This concentration of the feedwater increases along the flow path, as water molecules pass through the membrane, leaving dissolved and suspended material behind. Water that takes this flow path is called CONCENTRATE (feedwater that has become concentrated) or REJECT (water that is being rejected).

Calculations

There are several terms used in the design of reverse osmosis systems that involve calculations. These calculations are normally performed by the RO membrane manufacturers' design software.

FLUX

Flux is a term that describes the amount of water produced per area of membrane per day. Flux is commonly measured as U.S. gallons of water produced per square foot of active membrane area per day (GFD). There is a relationship between the water flux and the rate of fouling on an RO unit. A high flux rate causes the membrane to foul faster. To achieve stable operation, the flux must be below some reasonable cutoff point. Different types of waters have different cutoff points.

SALT PASSAGE

Salt passage is defined as the ratio of concentration of salt on the permeate side of the membrane relative to the average feed concentration.

Applying the fundamental equations of water flow and salt flow illustrates some of the basic principles of RO membranes. For example, salt passage is an inverse function of pressure; salt passage increases as applied pressure decreases. This is because a reduction in pressure causes a decrease in permeate flow rate, and the salt flows at a constant rate through the membrane as its rate of flow is independent of pressure.

SALT REJECTION

Salt rejection is the opposite of salt passage. Water and salt have different mass transfer rates through a given membrane, creating the phenomena of salt rejection. No membrane is ideal in the sense that it absolutely rejects salts; rather, the different transport rates create an apparent rejection.

PERMEATE RECOVERY RATE (CONVERSION)

Permeate recovery is another important parameter in the design and operation of RO systems. The recovery, or conversion, rate of feedwater to product (permeate) is defined by R = 100% * (Qp/Qf) where R is the recovery rate (%), Qp is the product water flow rate, and Qf is the feedwater flow rate. The recovery rate affects salt passage and product flow. As the recovery rate increases, the salt concentration on the feed-brine side of the membrane increases, causing an increase in salt flow rate across the membrane.

The increased salt concentration in the feed-brine solution also increases the osmotic pressure, reducing the NDP, and consequently reducing the product water flow rate.

CONCENTRATION POLARIZATION

As water flows through and salts are rejected by the membrane, a boundary layer is formed near the membrane surface in which the salt concentration exceeds the salt concentration in the bulk solution. This increase of salt concentration is called concentration polarization. Concentration polarization reduces the actual product water flow rate and salt rejection below theoretical estimates.

The Concentration Polarization Factor (CPF), or Beta factor, can be defined as the ratio of salt concentration at the membrane surface (Cs) to bulk concentration of salt (Cb). CPF = Cs/Cb

Impurities In Water

The nature of a particular impurity dictates how it behaves in a reverse osmosis unit. Impurities in water can be broken into two main categories:

• Dissolved Impurities

• Suspended Colloids and Settleable Impurities Dissolved Impurities

The dissolved impurities can be further broken into two categories:

• Dissolved Solids or Liquids

• Dissolved Gases

 

Dissolved Solids or Liquids substances dissolve in water because of the attraction that exists between the dipole moment of water molecules and the charge present on molecules of the substance. When a substance or liquid dissolves, each of its molecules is surrounded by water molecules, or hydrated. Dissolved substances may or may not form charged species in water, but they always have water molecules closely associated with them due to charges and/or polar bonds. Dissolved solids are rejected by reverse osmosis membranes depending on the charge and solvated size of the particular dissolved species.

Dissolved Gases

Gases that dissolve in water exist as discrete gas molecules or as gas/water reaction products.

 

Non Water-Reactive Dissolved Gases

Gases that dissolve in water but do not form reaction products are weakly held by water molecules. Their dissolved concentration is dependent upon the partial pressure exerted by the same gas species above the gas/liquid interface. If the partial pressure above the liquid decreases, some of the dissolved gas molecules come out of solution to form small gas pockets (gas bubbles) within the liquid (or they diffuse out of the liquid).

Dissolved Water-Reactive Gases

Some gases, when dissolved, react with water to form ions. A water-reactive gas dissolves until the solution becomes saturated with ions formed by the reaction. Once this occurs, the solution can reach saturation with respect to non-reactive gas molecules. If non-reactive gas molecules are removed by degasification, reaction products (ions) recombine to form gas molecules. Carbon dioxide is a water-reactive gas; it reacts with water to form hydrogen and bicarbonate ions. The ions exist in equilibrium with CO2 molecules in solution. If CO2 molecules are removed from the solution, more are formed through the recombination of bicarbonate and hydrogen ions. Likewise, if bicarbonate and hydrogen ions are removed from the solution, more will be formed from the reaction of CO2 and water, to maintain the equilibrium. During reverse osmosis, carbon dioxide gas passes freely through the membrane, while bicarbonate ions are rejected. This selective permeability changes the ratio of the two species in the feedwater and results in a new equilibrium at a lower pH. The pH drop depends on the rejection of the bicarbonate. A higher rejection of bicarbonate produces a lower pH.

Suspended Colloids and Settleable Impurities

Settleable impurities are generally larger than 10 microns in size, and settle out of water over a period of time. Before the discovery of colloids it was widely believed that suspended solids could be filtered, and dissolved substances could not be. The distinction between dissolved and suspended solids has become difficult to distinguish due to an increased knowledge of colloids. Almost all particles in water carry a static charge on their surfaces. In most cases this charge is negative.

Suspended colloids are particles (groups of molecules) so small that they do not settle out of water. Their inability to settle stems from the surface charge just mentioned. Water molecules are attracted to these weak surface charges, which allows the particles to stay suspended indefinitely in solution. Suspended colloids range in size from .001 to 0.1 microns. The removal of colloidal particles usually requires coagulation to form larger particles which may then be removed by sedimentation and/or filtration.

Fouling and Scaling

Over time, membrane systems can become fouled with a wide range of materials such as colloids, organic matter and biological organisms. Fouling occurs because material in the feedwater that cannot pass through the membrane is forced onto the membrane surface by the flow of the water going through the membrane. If the "cross" flow (water that does not pass through the membrane) is not sufficient (is not turbulent), or if it is prevented from reaching the membrane (by deposits or a mesh spacer), the material from the feedwater is deposited on the membrane surface. Fouling increases with increasing flux rate (the flow of water through the membrane) and with decreasing feed flow (velocity). If left uncorrected, the accumulation of these foulants can cause a severe loss of performance in the system: pressure requirements increase to maintain flow, pressure drops increase, and salt rejection can suffer. If the system is not cleaned and continues to build up foulants, the elements may "telescope", or shear internally, causing the integrity of the membrane surface to be compromised and rendering the membrane irreversibly damaged.

Fouling tends to occur in membranes at the feed end of the system, where the flux rate is highest. Biological fouling can also occur due to the growth of algae or other biological contaminants in the membrane element. Although this type of fouling is caused by contamination rather than flow problems,the resulting blockage of the membrane is the same.

Scaling of the membrane surface occurs due to the precipitation of sparingly soluble salts. As water passes through the membrane, dissolved minerals from the feedwater become concentrated in the reject stream. If the concentrations of minerals in the reject stream exceed their solubility products, crystals will precipitate onto the membrane. Scaling occurs first in the last elements of an RO system because the feedwater is more concentrated near the end of the process.

The following is a list of some of the types of scale that may occur on the RO system membranes:

To begin exploring possible water treatment options, for any deserving project size, water analysis of the raw water and the required product water quality need to be submitted.

In order to obtain more information and/or request courtesy capacity plannings or case assessments, please sends an e-mail to answers@engineeringfundamentals.com. It is recommended to include a brief initial description of the case, ideally no more than one page, or project in progress / under study. The possibility of receiving and handling individual technical consultations has been foreseen but are limited as per availability. Alternatively make contact via telephone +1 917 720 3624 (Balestie & Balestie Ingenieros technical support).