
1.01 Introduction/Historic – Part I
Welcome to the ion exchange resins course, and let's start with a historical introduction.
The use of natural materials with ion exchange power for water treatment dates back to biblical times.
In 320 BC, Aristotle used an earthenware vessel containing a material with ion exchange properties to produce fresh water from seawater.
Long before synthetic ion exchange resins derived from polymers, the first ion exchangers were equipped with zeolites, which constitute a large group of minerals that contain a porous structure.
Zeolite is a word of Greek origin that means “boiling stone”, a phenomenon observed by the Swede Axell Cronstendt in 1756, who observed that after the rapid heating of a mineral, the stones began to jump as the water evaporated. Zeolites are minerals that have a porous structure, such as hydrated aluminium silicate, and that have an open structure that can accommodate a wide variety of positive ions, such as sodium, potassium, calcium, magnesium, and others.
In 1818, German chemist Johan Nepomuk von Fuchs discovered sodium zeolite by combining aluminium and sodium silicate.
The first studies involving cation exchange date back to 1850, by J. Thomas Way. In 1905, German chemist Robert Ganz discovered that zeolites could be used to soften hard water.
The first commercial zeolite was obtained by combining kaolin or kaolinite, sand, and ash. In 1934, with Adams & Holmes, there was a great advance in the area in which it was discovered that cation exchangers could be regenerated not only with salt but also with acid. This was the first step in the modern ion exchange process.
Later work led to the development of sulfonated styrene resins and the polymerization of styrene and divinylbenzene.
In 1950, the definitive milestone was reached: based on the research of Gaetano Frank D'Lelio, in 1945, Kunin & Myers developed and patented the process involving sulfonation, crosslinking, and polystyrene.
These advances were followed by McBurney, who around 1952 developed the styrene divinyl benzene copolymer, which was the basis for the definitive establishment of ion exchange technology.
There is academic interest in whether the ion exchange process is an absorption or an adsorption process. "AD" is a process that occurs at an interface or surface involving the interaction of ions. "AB" absorption involves the incorporation of a substance in one state that attracts another in another state, such as solids attracting liquids throughout their entire structure. So the phenomenon of ion exchange can be placed in both categories and can therefore be more safely considered as an sorption process, a term that avoids controversy among those most devoted to the prefixes "AD" and "AB".
Finally, what are ion exchange resins, and what is this process for? The ion exchange technology used for water treatment is a process in which a granular filtering medium composed of microspheres varying between 300 and 1200 microns, composed of polymers and electrically charged, aims to attract certain types of ions to itself, depending on the constitution of the end of the polymer chain.
The process normally takes place inside a vessel, usually cylindrical, and as the water passes through the resin, ion exchange takes place when certain undesirable ions are removed and replaced by more desirable ones.
1.02 – Introduction - Part II
The ions that are attracted by the resin and those that are released by it depend on the type of resin and the type of regenerant. The regeneration involves a chemical support solution that, when applied, allows the resin to be exhausted by the exchange process. Now no longer able to attract new ions, it returns to its initial condition to restart a new exchange cycle, which means that this process is reversible.
Well, how do these ionic exchanges and interactions between resin and filter medium ions occur?
One of the theories is the "Crystalline Network" theory proposed by Paulling and Bragg. A sodium chloride crystal contains sodium ions and chloride ions. Each ion in the crystal is surrounded by a fixed number of oppositely charged ions and subject to certain "Coulombic" attraction forces, whereby channels on the crystal surface experience less attractive force than crystals further below the surface.
When they are in contact with a polar medium, such as water, the forces of attraction around the crystal decrease to such an extent that an exchange of this ion for another becomes possible.
The ease with which surface ions will be replaced by other ions will depend on the nature of the Coulombic forces around the crystal, the concentration of the ion to be exchanged, the electrical charge of the exchange ion, the size of the ions, the accessibility of the lattice formed by the ions and the effects of solubility.
The ion exchange mechanism in resins is analogous to the ion exchange in crystals. Resins can be considered high-molecular-weight and insoluble polymeric electrolytes. The ion exchange capacity of the resins is due to the functional groups at the end of the polymeric chain, such as sulfonic, carboxylic, and phenolic.
Another theory that explains ion interactions is the double layer theory, which was proposed by Helmholtz in 1879, revised by Gouy-Chapman in 1910, and also realized by Stern in 1924.
This theory argues that those present in a diffuse layer of colloids do not behave as if there were a defined division between the layers, but rather the existence of a diffuse layer that remains in equilibrium and that is dependent on concentration and pH.
Although there are other theories that explain ion interactions, they are all similar in one respect: that the exchange of ions must satisfy the law of electroneutrality.
The law or principle of neutrality was proposed by Linus Pauling in 1948 and later revised. The principle states that the electrons in a molecule are distributed in such a way that the charges on the atoms should converge to zero as much as possible.
The differences between the theories are related to the position and origin of the exchange sites. In all cases, the exchange sites are formed essentially by a non-diffuse ionic cluster formed by an electrostatic bond, whose ion can be replaced depending on the binding strength.
The theories of bonds involving ion exchange can become complex, but deepening these theories is of more interest to students, notoriously postgraduate and master's students.
For engineers and technicians more focused on designing ion exchange, softeners, and demineralization systems, it is enough to know that there is an exchange interaction between oppositely charged ions and that this phenomenon provides what we want, that is, the removal of undesirable ions by other less undesirable ones.
This is the essence of ion exchange.
1.03 Resins physical aspects
Okay, talking a little more now about how the resins are presented and how the resin is shaped. The ion exchange resins are microspheres and have a diameter varying between 300 and 1200 microns in the times part; this diameter is a variation for non-uniform resins. For unformed resins, the evaluation is much lower and is in the range of 600 microns, with some variation for more or less. The nomenclature of the resins is designated by three letters, which will be explained later and refers to the international nomenclature. Here are some examples of resin types:
strongly acidic and uniform resin
- then here is another type of strongly acidic resin; you can see the uniformity of the diameter of the particles.
Strongly acidic resin with high crosslinking, we will explain what crosslinking is in the following classes:
Strongly acidic resin does not shape; you can see smaller and larger spheres distributed.
uniform, strongly basic resin, with a lighter and more silvery color,
strong basic resin with high crosslinking, which is basically a better-quality resin,
Strongly basic resin common grade; you notice the difference in particle diameter,
high-quality weakly basic resin of the uniform type, suitable for softening water with a high organic load,
Another example of a weakly basic resin,
Here is an example of mixed resin, which is a mixture of strongly acidic and strongly basic resins.
Another example of mixed resins: let's explain later what mixed resins are for.
Microscope image of the surface of the weakly acidic resin, and here is a scanning electron microscope image of the surface of a strongly acidic resin.
In this image, the resin is already accommodated inside the tank; the view from above looks like this: a strongly acidic resin with a bright golden amber color.
This table shows the variation in the diameter of the resins, starting with the smallest particle at 250 microns and ranging up to 2,000 microns. In practice, however, commercially, the values vary between 300 and 1200 microns, this being the micron for non-uniform resins.
For uniform ones, the granulometry is concentrated in the range between 500 and 700 microns, with the vast majority of them around 600 microns with 50 microns more or less.
The most common commercial presentation of resins is in 25-liter bags or 141.5-liter barrels.
Suppliers usually quote per liter of resin, with prices often quoted in dollars.
With regard to suppliers, a specific class on the market will be given later on, when the main brands of resin available on the market will be presented.
Below, we will present images of the resins that were obtained through observation under a microscope. The device used here was a Nikon Labophot with 40x magnification. Firstly, as a highly acidic, non-uniform resin, you can see the large variation in the diameter of the spheres;
Another image:
It's a new resin, so the spheres are clean.
Here an uniform cationic resin, it is noticed that there is practically no variation in diameter;
Here is footage of this resin; you can see some variation in diameter, but in general, the resin maintains a diameter of around 600 microns on average.
Here is another type of resin from another brand, a new brand on the market. it has a color that is not yellow, let's say a more silvery color, despite being a strong cationic resin. It looks like a more faded yellow, pulling with silver, which is characteristic of this brand.
In this image, it is a strong cationic resin; its manufacturer promises to remove large amounts of iron. It can be seen in the conformation of the internal structure, which is different from that of common resins. This resin model is not classified as a uniform resin because the diameter variations in its spheres are clearly perceived. But it is a resin considered to be of high quality and has the specific function of removing metals, according to the manufacturer.
We will comment on alternatives for removing metals from water in later classes.
We now move on to examples of used resins that were removed from softeners that had already been in use for some years, and the dirt and contamination of this material can be seen.
Here in this image of a uniform resin, you can see the accumulation of material already occurring; there is a little fouling, and this is an aspect of the resin after a few years of use.
In this way, the ion exchange capacity is progressively lost, so a microscope analysis also allows for assessing the condition of the resin and whether or not it is time to change it. These images then show that the resins have exhausted their ion exchange power and that it is necessary to replace them with new resins.
Microscopy of ion exchange resins will be covered in more detail in a specific class later on.
1.04 Chemical structure
Let's talk now about the chemical formula of resin. The chemical formula is basically a polymer of benzene rings and a radical, which can vary but is basically a chain of benzene rings interconnected to the polymer.
Another image of a carbon chain, benzene rings, and a spike represents the exchange site.
The tip can have several conformations and is the part that gives the characteristics of the type of resin.
Here is the most common form of cationic resin: it has an SO3H or SO3Na group, which is even more common. This is where the exchange takes place. The hydrogen, which could also be sodium, will leave calcium, magnesium, or another ion that you want to remove.
The production of resins is basically carried out in two ways: one through the reaction of methacrylic acid with divinylbenzene.
The other way is by combining styrene with divinylbenzene, followed by sulfonation. Here, in the case of the production of the cationic resin, we have the carbon chain, a benzene ring, and the SO3H group. So this grouping here is a common find. So we saw in a generic way how the resin is produced.
Let's now separate the types.
The strongly acidic SAC resin in which the reaction is carried out with hot sulfuric acid, after which the resin is washed to remove excess acid, is a process that has to be conducted with total control because this addition of acid causes the grains to swell in the resin and, if not controlled, can damage the resin.
Then it is produced in the H+ cycle, carrying the acid cycle, so if you want to produce a resin for softening, another step must be added to convert the H+ cycle to the Na+ cycle. We then move on to strongly basic SBA resins. Its production is more complex, more time-consuming, and requires more expensive chemicals. Activation begins with the polymer undergoing a reaction with chlorine methyl ether, which then produces the polymer with the final cluster having CH2Cl. So the first step is chlorine methylation. Next, the chloride group at the end of the chain is bonded to carbon, but it is not ionized, so until this step, there is no charge to be exchanged. This is where the second stage comes in, in which it is reacted with trimethylamine, producing the SBA type 1 resin, or else with dimethyl ethanol amine, producing the SBA type 2 resin. Type 1 SBA resin is strongly basic, and Type 2 SBA resin is intermediate between strongly basic and weakly basic. We then moved on to the weakly basic WBA resin. The same radical is used, which is chlorine methyl styrene, but instead of reacting with dimethyl ethanol amine, it will react with diethyl amine, thus producing the WBA resin, which has one carbon less at the end of the chain. We then arrive at WAC, or weakly acidic resins, which are formed from the polymerization of acrylonitrile or methyl acrylate, compounds that are hydrolyzed. In the case of acrylonitrile, it is hydrolyzed with sulfuric acid, while the acrylate polymer is hydrolyzed with soda. The end product of these two compounds is a carboxylic acid; the WAC resin cluster is a carboxylic acid.
The details of the reactions and more information about chemical reactions to produce resins are a little more complicated because the companies that manufacture them have, in fact, an industrial secret, so it is more complicated to scrutinize further technical information about the reactions, for it is a process well kept under lock and key by the multinational companies that produce the resins and own this technology, in addition to which they compete a lot with each other. But at least we can get a basic idea of the reactions.
1.05 Milliequivalent-gram concept
One of the most used concepts in calculations and sizing of ion exchange resin systems is the concept of milli-equivalent gram or equivalent gram, which defines the capacity of the resin for design calculations of softeners.
Here in the datasheet of a resin, among other information, the theoretical exchange capacity appears, in this case, 2 equivalents per liter.
But what is the milli-equivalent? You can express milliequivalent per ml or equivalent per liter, which is the same thing.
And how to calculate the milli-equivalent?
First, I have to calculate the atomic weight. For example, I want to calculate the gram equivalent or milli-equivalent of calcium.
Calcium has an atomic number of 20 and an atomic mass of 40. As it is bivalent, that is, calcium +2, then I have to divide 40, which is the atomic mass, by 2, which is its valence, resulting in 20. So the equivalent of calcium is 20 g; the milli-equivalent is obtained by dividing the result by 1000, so I will have 0.02 g, which is 20 mg.
One mile-equivalent of calcium is 20 mg.
To better establish the concept, we have a water analysis that shows 100 mg per liter of calcium in the CaCO3 format, which is very common in laboratory analysis. It is difficult to always see total hardness in calcium analysis.
The molecular weight of calcium carbonate is 100 grams; the molecular weight of calcium is 40 grams, so it gives a ratio of 2.5, which then represents that my calcium hardness is 40 mg/l of calcium in calcium format; if we have 1 meq of calcium, it is 20 mg , we saw before, so if we have 40 mg, we have 2 meq/l of calcium.
This unit is important to get used to using in meq or milli-equivalent per liter because it avoids this kind of confusion:
Is it in CaCO3?
Is it calcium?
Is it in mg/l?
The advantage of the meq unit is that it is standard and avoids this type of error.
Therefore, it is recommended to get used to ion exchange from now on, converting everything to milli-equivalents per liter (meq).
Some calculation and sizing software already has various formats, and you can feed the data in any format.
So, there will be no problem with that, but it is recommended to use a meq format and let the program do the necessary conversions.
The milli-equivalent concept also helps explain the issue of valences and how this affects ion exchange.
It is not difficult to understand that calcium, which has a +2 valence, will occupy twice as many sites in the resin when it passes through it, which contains sodium as the species at the end of the chain, and will therefore displace 2 sodiums.
Therefore, meq avoids calculation errors precisely because it already considers the valence of the structure, and in ionic exchange, valency is a very important concept.
A trivalent ion will occupy three times the resin site as a monovalent one, and this will drastically affect the exchange capacity.
The concept of valency will be detailed later.
1.06 Water hardness
One of the most common applications of ion exchange resins is the removal of calcium and magnesium from water, also known as hardness.
Water hardness is a property related to the concentration of ions of certain minerals dissolved in it; more specifically, hard water contains mainly divalent salts and relatively high concentrations.
Water hardness is predominantly caused by the presence of calcium and magnesium salts so these are the main ions taken into account.
Eventually, we will also have zinc, strontium, iron, manganese, and aluminum, among other metals that may be taken into account when measuring hardness.
Concepts of temporary and permanent hardness:
Temporary hardness is caused by calcium and magnesium salts bound to carbonate and bicarbonate ions.
These metals are, by definition, alkaline earth metals, and when they combine with bicarbonate and carbonate ions, they form calcium bicarbonate, calcium carbonate, magnesium bicarbonate, and magnesium carbonate. These substances can be removed by boiling.
Permanent hardness is the same as cations but linked to strong anions, such as sulfate and chloride.
So we have calcium sulfate, calcium chloride, magnesium sulfate, and magnesium chloride.
Therefore, this hardness is what causes the accumulation of material deposited in pipes (incrustation), and it is a very common problem not only in industrial equipment but also in water pipes in condominiums and residences.
Calcium and magnesium removal is one of the most common applications involving ion exchange technology, commonly referred to as softening.
A question that arises and will become evident later on when we study the companies' ion exchange sizing software, concerns the interpretation of the water analysis for the hardness parameter.
The software does not have the option of entering total hardness or total alkalinity. You actually enter the calcium and magnesium readings separately.
For alkalinity, you will enter with bicarbonate and carbonate.
But what can be done in this case, and how does this detail influence the calculation?
So let's start with hardness. Here, the ideal is that, for any situation, you ask the analysis laboratory to measure calcium and magnesium separately. Still, having only the total hardness reading is possible to make the calculation, and let's see how this is possible.
You can estimate if you only have the total hardness measurement.
Considering as an example: 100 ppm of total hardness, considering 80 ppm of calcium and 20 ppm of magnesium, is an estimate and a very rounded calculation, but a good approximation and is usually an average of the most common water analysis.
But what will this influence in the calculation? Very little indeed. Note the selectivity and separation factors, a subject that will be dealt with in detail later: if the selectivity of calcium is 1.90 and magnesium is 1.67, you can say that these two substances are more or less attracted by the same intensity or very little difference in intensity by the sites of the cationic resin.
Testing in practice with sizing software did not result in any difference. If you add 2.0 meq of calcium and 0.0 meq of magnesium, or 1.5 meq of calcium and 0.5 meq of magnesium, or even 1.0 meq of calcium and 1.0 meq of magnesium, always totaling 2.0, you can see that for the calculation it will not influence anything.
So it's not something to worry about, but if you want to do it, ask the laboratory to detail these cations. It is possible that some software could result in some small difference, but it is more of a calculation precision.
1.07 Water Alkalinity
We will now move on to the concept of alkalinity.
Alkalinity for sanitation is not a substance that presents a risk to health and, therefore, has less health interest.
It is important as a buffering effect in effluent treatment, particularly within the anaerobic reactor.
It plays a fundamental role in flocculation reactions since the PAC, or aluminum salt, needs alkalinity to form aluminum hydroxide.
In this process, if the alkalinity is low, it will have to be added, and this explains why in flocculation, the flocculant plus the alkalizing agent are dosed at the same time.
And it is also of some importance in swimming pools, where it is recommended to keep its content between 80 and 120 ppm, since below 80 ppm the pH can suffer great variations and the pool becomes unstable, and above 120 ppm, or when it is very high, can cause incrustation or change the turbidity of the water.
Alkalinity is easier to determine than hardness.
Here is a graph that relates the pH to the two most common substances of alkalinity: carbonate and bicarbonate, remembering only that we have the types of alkalinity: if the pH is greater than 9.4 we have a predominance of hydroxide and carbonate; If the pH is between 8.3 and 9.4, we have a predominance of carbonate and bicarbonate; if the pH is between 4.4 and 8.4, we have a predominance of bicarbonate.
So back to the graph.
The graph shows us two curves, one representing the amount of bicarbonate (HCO3) and another representing the amount of carbonate (CO3).
Observing the pH range between 6.5 and 9.5, the predominance of bicarbonates is noted.
At a more acidic pH, we have a predominance of carbonates and a decrease in the amount of bicarbonates.
At a higher pH, above 10, the condition observed at a more acidic pH is repeated: the carbonate again predominates.
Showing a practical example: if you have a pH of 7.5, look at the graph, you will have a bicarbonate concentration of 0.87. Let's say you have 100 ppm of alkalinity; 87 ppm will correspond to bicarbonate and 13 ppm of carbonate.
In another example, if you have a pH of 8.3, it will correspond to 100% bicarbonate.
It means that from the graph, you can estimate these chemical species because they depend on the pH of the water.
Remembering that it is always recommended to have a more exact calculation, ask the analysis laboratory to perform bicarbonate and carbonate alkalinities.
This example is given just in case you don't have the information, i.e., if the water analysis only specifies total alkalinity.
Finally, the hardness and alkalinity ratio:
- When the total hardness is greater than the total alkalinity, the water contains carbonate and bicarbonate hardness.
When the total hardness is equal to the total alkalinity, the water contains only carbonate hardness; when the total hardness is less than the total alkalinity, the water contains only carbonate hardness, and the excess alkalinity corresponds to sodium, potassium, carbonate, and bicarbonate.
1.08 What to analyze in the water?
The first step for a project involving ion exchange resin, such as softening, removal of specific ions, or demineralization, is the analysis of the water to be treated.
But what should I analyze in the water?
First, it is necessary to define the objective of the treatment. What does your project consist of? Is it a softening project, specific anion removal such as NO3-, silica removal, or demineralization? Is it an anion removal project?
In the case of softening, the main parameters are: calcium, magnesium, sodium, iron, and manganese.
Rarely a water analysis shows calcium and magnesium separately, so you can estimate these contaminants by reading the total hardness (as seen earlier).
Secondary parameters would be potassium, strontium, barium, and the ammonium ion.
For the removal of anions, the main parameters are nitrate, chloride, and sulfate. Normally, when it is desired to remove anions, such as nitrate, the sulfate ion becomes an important contaminant and prevails due to its selectivity, being removed before and with priority to the nitrate. Therefore, when you want to remove nitrate, you must also analyze sulfate.
The secondary parameters for anions would be bromine, fluorine, bicarbonate, and phosphate. All of them are in the companies' software, so with these parameters, the software will be fed with all the data, thus producing a more correct calculation, but we'll see that later.
In the case of silica, it is a little more complicated because silica in water is usually in the form of a colloid (colloidal silica), and, being in colloidal form, the resin does not have the ability to attract this silica format because it has no electrical charge.
Therefore, for silica, the recommended technology is membranes; even so, we will show a topic about silica later on.
For demineralization, all the main ions (anions and cations) must be analyzed, and it is highly recommended that all secondary ions are also analyzed, even more so if you want to have ultra-pure water with very low electrical conductivity.
Last but not least is the analysis of the organic matter parameter, which is rarely included in water analysis but is one of the most important. This is the "organic matter per consumed oxygen" parameter, which practically all the more structured analysis laboratories use, and it is extremely important even to assess the eventual need for pre-treatment in the water since all the water that enters a vase containing ion exchange resin must be previously treated with filtration and, depending on the conditions of the raw water, undergo other pre-treatments.
Finally, if you are going to do a softening project, you don't have to worry about the anions because the sizing software itself will perform a load balance based on the cation data feed. However, if you intend to carry out demineralization or alkalinity removal, which we will see later in a video on alkalinity removal with resin WAC, then it would be interesting to request an alkalinity analysis as well. In this case, you can have only the alkalinity analysis and make the estimate of carbonate and bicarbonate (already explained earlier) or do the analysis of alkalinity to bicarbonate and alkalinity to carbonate.
In “The Nalco Water Handbook,” several tables of water analysis parameters are presented, considering each process:
WAC + SAC with degasser (dealkalization with caustic adjustment);
SAC + SBA (chloride dealkalizer);
mixed bed for dealkalization;
Demineralization: twin bed with degasser
SAC for softening—this will be more detailed.
According to Nalco, cations should be calcium, magnesium, sodium, and potassium. For anions: bicarbonate, chloride, sulfate, and nitrate.
Other parameters are alkalinity, carbon dioxide, silica reactive (not colloidal), conductivity, and pH.
Note that in the case of softening, there is an increase in the amount of sodium in the treated water, which can become a problem depending on the type of use the water will have.
Especially if the water already contains sodium in an appreciable amount before it undergoes softening.
This will be the subject of an extra class, which will detail the limitations of the softening of water intended for human consumption.
The more information and parameters you have, the better, so also take into account the cost of water analysis in the project.
1.09 Resin types
The types of resins can be divided into four groups. It is important to get used to the international nomenclature where resin types are designated by a 3-letter abbreviation:
SAC, which means Strong Acid Cation, or strongly acidic;
SBA, which means Strong Basic Anion, or strongly basic;
WAC, which means Weak Acid Cation, or weakly acidic;
WBA, which stands for Weak Basic Anion, or weakly basic.
Here is a chart showing the four main types of resins and their properties and characteristics. Let's comment on some of these features.
The SAC is perhaps the most used and most popular type among the four types; its chemical structure is a polyvinyl, therefore a polymer, and it has the two main radicals of the sulfonic acid type:
SO3H (acid regeneration cycle) and SO3Na (sodic cycle or regeneration salt cycle)
The acidic cycle is mostly used in demineralization projects, and the sodium or saline cycle is more used for removing calcium and magnesium hardness in softeners.
SBA is the same polymer, but at the end of the chain it has a quaternary ammonia-type grouping and may have an OH- as a kind of exchange at the end of the chain (when regeneration is with soda—demineralization) or else a Cl- (when regeneration is with salt—removal of nitrates, sulfates, and chlorates).
WAC has as a chemical structure the derivative of carboxylic acid or carboxylate as a radical, and its application is the removal of calcium and magnesium hardness in waters that contain a lot of alkalinity.
The WBA has a grouping similar to the previous one, but here the radical, instead of having three methyls, only has two. Application: removal of chloride, sulfate, and nitrate from substances derived from strong acids.
Here is another table listing the types of resin, already showing some important information, such as the ion exchange capacity, which will be discussed in detail later on.
It is noticed that the strongly acidic resin has a capacity of around 1.7 to 2.1 milliequivalents per milliliter of resin, or meq/ml.
We saw before that what was already the definition of resin capacity, in which 1 ml of resin has the capacity to remove from 1.7 to 2.1 meq of the contaminant that is intended to be removed from the water.
Comparing the strongly acidic with the strongly basic, we realize that the strongly basic has a much lower capacity, which is normal. It is much more difficult to remove anions than cations, and this is a characteristic of ion exchange technology.
To better fix the concept of radicals in the four types of resin:
The strongly basic has a quaternary ammonium (Q)-type radical.
The weakly basic has a dimethyl-amino-ethyl, or DEAE-type, radical.
The strongly acidic has a sulfopropyl (SP)-type radical.
The weakly acidic has a radical derived from carboxylic acid, carboxymethyl (CM).
Here is another author who presents a slightly different, but very similar, nomenclature.
Strongly acidic has the sulfonate group. The removed ions are: calcium, magnesium, radium, barium, and lead. The regenerant can be either acid or salt. The pH range is very wide.
The weakly acidic has the carboxylate radical; here, the removed species are practically the same as the SAC but more indicated in water with a high alkalinity content or else as one of the demineralization steps of more sophisticated processes. The recommended pH here is above 7.
The strongly basic resin's radical is quaternary ammonium. The species involved here are nitrate, sulfate, chlorate, arsenite, and selenite.
Regeneration can be done either with soda or with salt. It can also be used in a high-pH range.
The weakly basic have the tertiary ammonia radical (one less methyl). The ions involved are practically the same; however, the regeneration here is with soda or calcium hydroxide.
Application pH: below 6.
So those are the four main types of ion exchange resins.
1.10 Total ion exchange capacity
The total or theoretical ion exchange capacity is determined by the resin manufacturer; as an example here of the SAC Amberlite IR122Na resin (Dow/Dupont), it can be seen in the technical specification that the total exchange capacity appears. The value reported by the manufacturer is 2.1 meq/ml. However, we will comment and detail later that, in practice, this value is lower, but this is a value of the exchange capacity of the resin.
This value can only be obtained in very specific circumstances, which in practice can never be achieved. We would have to have a perfectly distributed bed, hydraulic distributors very well constructed and in large numbers to promote the perfect distribution of water throughout the resin bed, avoiding preferential paths (channeling), contact should be slow and without deviations and across the entire surface of the resin, among other very specific process conditions that can only be obtained in the laboratory. The capacity reported by the manufacturer, therefore, in addition to being theoretical, can be considered the maximum amount of ion exchange that the resin can provide.
Here is the resin model SAC C100 from Purolite, which is equivalent to this one. It has also been informed by the manufacturer that it has a capacity of 2 equivalents per liter. It is what is expected of a strong cationic resin—something around this value, approximately 2.0 meq/ml.
Here, in the evaluation of another author, in this case the famous book "Chemical Engineer's Manual" by Perry, the strong cationic resin appears, and how much the ion exchange capacity varies: From 0.4 to practically 2.3, this is informed by the manufacturer, as Total or Theoretical capacity, so the strongly basic resin also ranges from 0.4 to 2.5 meq/ml, remembering once again that this is the total (theoretical) exchange capacity.
For SBA resins, the ion exchange power is lower, ranging from 0.4 to 1.5 meq/ml.
In later chapters, we will go into greater detail about the issue of ion exchange capacity and the ratio between total or theoretical capacities (informed by the manufacturers) and operational, or real capacities.
The ratio between the total capacity and the operational capacity of an ion exchange resin can be seen in this graph.
The red line represents the water hardness behavior throughout the exchange cycle.
In the first stage, the hardness drops drastically, an effect provided by the resin at the beginning of the cycle and, therefore, with all its exchange sites available.
In the second stage, the hardness remains low because most of the resin exchange sites are still available.
Throughout the third stage, the amount of hardness progressively increases; this is when the exchange sites are becoming exhausted (mass transfer zone). At the end of this stage, the hardness practically returns to the levels prior to the beginning of the cycle; it is time to regenerate.
The area in red represents the amount of hardness removed considering the total capacity of the resin.
In practice, however, what is observed is that this value is much lower, somewhere between 55 and 65% of the total capacity, which is configured as the operational or real capacity and is represented in the graph by the square area in green color.
The ratio between these capacities depends on numerous factors and the type of resin involved; we will see more about ion exchange capacity in later classes.
1.11 Resin datasheet interpretation – Part I
Now let's talk about the resin manufacturers' data sheets to better interpret this document that details the technical specifications of ion exchange resins.
As a practical example, let's start with Purolite's C100E model, which is a strongly acidic resin. Here we have several technical details, and we will detail each one of them:
Polymeric structure: reticulated polystyrene gel with divinylbenzene is the main chain of the polymer from which the resin is manufactured;
Appearance: spheres, which is a standard for all resins;
Functional group: sulfonic acid, as it is a strong cationic acid; if it were another type of resin, another functional group would replace the sulfonic group;
Ionic form: it is a resin that is regenerated with salt, so it is called the sodium cycle, and the functional group is Na+ at the end of the chain;
Total Capacity: 1.9 equivalent per litre (already commented but will be better detailed soon);
Moisture retention: 46 to 50%: the resin already comes with this characteristic of internal humidity, that is, it is a standard, and the range is something around this percentage for all resins;
Particle Size Range: from 300 to 1200 microns; we already saw this at the beginning of the training; this resin is not uniform; it is a more common type, and therefore the variation is great in the size of the spheres, with the small ones around 300 microns to the largest ones above 1,000 microns, which is typical of common resins;
Still, in relation to size, the number of spheres smaller than 300 microns is a maximum of 1%. Values smaller than 300 microns can cause these grains to escape through vessel nozzles;
- Coefficient of uniformity 1.7: for this coefficient, the ideal situation is that the closer to 1, the better, which is a characteristic of high-grade or uniform resins; the best-quality Purolite PPC100 uniform resin has a particle size range of 650 microns with plus or minus 50 micron tolerance; these types of resin usually have a coefficient of uniformity ranging between 1.0 and 1.2;
Reversible swelling: a slightly more technical and less relevant concept; when the resin is exchanging ions, that is, when it is in service, it retains moisture, but after regeneration it returns to normal, returning this level of moisture retention; it can then be said that the resin “swells” when retaining water, and this swelling is around 10%;
Other, less relevant information may appear, depending on the brand and type of product.
Continuing, let's now analyze a resin from the Dupont brand. This is Amberlite IR 120 Na+, strongly acidic type.
The Datasheet presents a lot of information and we can start with;
- Physical properties: “amber” which means “golden color” and form of microspheres;
- Matrix or main exchange radical: styrene divinylbenzene copolymer (standard for this type of resin);
- Functional group: sulfonic acid, as it is a strongly acidic resin, already seen in a previous class;
- Ionic form: uses salt as a regenerant because it is sodium cycle;
- Total exchange capacity: 2.0 equivalent per liter; this value is standard for all manufacturers for this strong cationic resin grade;
- Water retention capacity: 45 to 50%
1.12 Resin datasheet interpretation – Part II
Coefficient of uniformity: 1.9; it is noteworthy that the manufacturer informs 1.9 because at the same time we noticed that the size varies between 600 and 800 microns, which suggests an inconsistency between these data; however, it is known that this resin grade is the most common type, so it can be considered that, in practice, the variation in the size of the microspheres can be greater;
Maximum moisture retention or swelling: < 11%; that is, the difference in moisture retained by the resin in the process in relation to the regeneration stage with salt.
Maximum operating temperature: 135 degrees Celsius or 275 °F;
Minimum bed depth: 700 mm. This is interesting data because it already provides information that can be used in the design of the exchanger vessel. In other words, a very wide and low vase is not recommended, but a narrower and taller vase is preferred. In practice, this is already met when fibreglass-reinforced polyester resin (PRFV) tanks are used, which are becoming more and more common.
The recommended service rate is 5 to 40 BV/h. Later on, the concept of service flow rate will be detailed.
Regeneration: We won't go into details yet about regeneration because there's a whole block of training dedicated to it. But we can quickly show you here that with salt, the recommended level is 80 to 250g; we will see later that the index of 150 grams of salt for each liter of resin contained inside the vessel is used a lot (usual concentration), and the recommended concentration is 10%; all this will be shown and calculated later with details and examples;
When sulfuric acid is used, the recommended level is from 60 to 240 grams per liter of resin, in a concentration of 0.7 to 6%.
When hydrochloric acid is used, the level is from 50 to 150 grams per litre of resin, in a concentration of 1 to 5%.
- The minimum time contact between the regenerant and the resin is 30 minutes, and therefore one more parameter that can be used later in the project;
Other less relevant information, which will also be detailed in the regeneration block.
Finally, let's quickly comment on some hydraulic characteristics.
First, the backwash: usually, in the vast majority of systems, we work with linear velocity rates around 15 m/h because we work with a flow rate in the range of 3 to 10 m3/h and a cross-sectional area of the vessel between 0.3 and 0.6 m2. At this rate, we will have a bed expansion of approximately 50% (for an operating temperature of around 20°C), hence the care when designing the vessel.
Secondly, pressure drop: it is important to note that pressure drop does not depend only on the resin bed, which in this case represents around 0.2 bar/m of bed, considering the same linear velocity.
It also depends on the inlet and outlet piping of the vessel (bends and gauge), the shape and size of nozzles, and encrustations.
In practice and after observing the behaviour of the vessels in operation, it is possible to estimate a head loss of around ½ bar (or 5 mWC) for the entire vessel; this is an approximation that can be used in projects involving both filters and softeners.
2.01 Case 1: Calcium removal
Now let's take a break from theory and move on to a practical calculation.
This calculation is theoretical and serves to establish the concepts presented so far, such as milliequivalents, unit conversion, and an estimate of what would be a calculation of the volume of resin required for a softener.
In this “case”, our objective is to remove 230 mg per liter of CaCO3 of total hardness from well-maintained water, where the service flow is 8,000 liters per hour for 10 hours daily, and the resin used is Amberlite IR 120 sodium cycle. Regeneration should occur within two days.
What volume of resin is required for this process?
Let's consider for theoretical purposes and for a better understanding that, in this specific case, we only have calcium in the water as a contaminating species. We know that in practice this is very unusual; usually, calcium is accompanied by magnesium, iron, and other salts. However, for didactic purposes, we will only consider the presence of calcium.
Let's first calculate the amount of calcium, that is, how much calcium we have in the sample in milligrams.
We have 230 mg of calcium as CaCO3, which is how it's reported from the lab, so I need to do a conversion.
We have that the molecular weight of calcium is 40 and the molecular weight of calcium carbonate is 100, so making the correlation, this division has a ratio of two and a half.
If in the sample we have 230 mg of calcium as CaCO3, divided by 2 and a half, we will then obtain 92 mg of calcium as calcium itself.
So, the first stage of our calculation is to understand the units and not confuse the results presented in CaCO3, calcium, etc.
We have to be very careful about this issue of units.
But then we have 92 mg of calcium in our sample.
Now let's calculate the number of milliequivalents of calcium in the sample. We know that a calcium equivalent is the molecular weight of calcium divided by two; we've seen this before.
So 40 divided by 2 means that the milliequivalent of calcium is 20 mg.
So, we have:
1 milliequivalent (or meq) of calcium makes 20 mg.
"x" makes 92 mg.
"x" results in 4.6 meq of calcium.
We made a point of showing the calculation in this format because this way it is easier to understand than presenting a ready-made formula. Thus, the deduction facilitates the understanding of how to arrive at the result.
Our campaign provides for a regeneration every two days. Therefore, if we have 8,000 liters per hour every 10 hours (a day's production), we have 80,000 liters per day. As we can only regenerate every two days, we have a production of 160,000 liters per campaign.
The concept of the ion exchange campaign is the time the resin works before being regenerated, so the process will work for two days until it is interrupted to regenerate and start a new cycle.
So we have to:
4.6 milliequivalents of calcium in 1 liter of water
In "y,” we have 160,000 liters of water.
y = 736,000 meq Ca+2
1 ml of resin accounts for 2.1 meq Ca+2 (manufacturer's data).
In "z,” we have 736,000 meq Ca+2.
z = 350,476 ml ~ 350 liters of resin
As the commercial presentation of resins is usually in 25-liter bags, we will need 14 bags.
Let us now understand why this calculation is wrong.
Not considered in the calculation:
- Selectivity;
Separation factors;
These are very important concepts, and we will present them in detail later in the course.
The level of salt regeneration was also not taken into account, which also impacts the resin removal capacity.
The difference between the theoretical or total capacity of the resin and the respective operational capacity was not considered; this will also be discussed throughout the training.
- The presence of other ions: calcium never comes alone; hardness is always accompanied, in addition to calcium, magnesium, and eventually metals, such as iron and manganese, in addition to other less relevant cations;
Other parameters that can change the result, such as vessel hydraulics, flow velocity, and filtration rate, among others.
So this calculation is wrong because we will see later that in practice the resin volume calculation would give a much higher value than the one presented, and we will see why.
But it is important to present it this way, to present a preliminary theoretical estimate. Also, it's important to learn where these calculations come from because they are important to fix the concepts related to the subject.
Later, we will redo the calculation, expanding our knowledge, and we will see why the theoretical capacity reported in the resin datasheet should not be considered a calculation parameter but only as basic technical data related to the type of resin.
2.02 Case 2: Nitrate removal
In this second "case,” we are going to simulate nitrate removal using strongly basic resins, but doing it as if it were a simulation of demineralization.
We are going to present the manual calculation, as in the previous “case”, not being used as the resource of the software of the companies, which will have an entire section of the course dedicated to this.
In this simulation, we will not only remove nitrate but all anions in the sample.
So let's go to Case:
Water needs nitrate removal, and it was decided to use ion exchange technology. The most important ions are listed in the following table. The average daily production is around 100 cubic meters, with a flow rate of 10 cubic meters per hour for 10 hours of operation.
We will therefore consider here a campaign of 100,000 liters.
The resin options to treat this water are:
Dupont Amberlite HPR 4200 cycle caustic soda
Purolite A400 cycle caustic soda
Estimate the minimum resin required in each campaign, assuming the nitrate is completely removed.
Here is the ionic balance chart, where the cations appear on the left side, which is not of interest to us and is shown here only to be able to balance the values of all the ions present in the sample. What matters to us here are the anions.
We have chloride, sulfate, bicarbonate, and nitrate, each in the proportion of 0.5 milliequivalents per liter.
Nitrate is the object of our case, the species we really want to remove.
So in total, we have 2.0 milliequivalents per liter, adding up all the anions.
The two resin options we have are Purolite A400, which has a total capacity of 1.3 equivalents per liter which is similar to the Dupont HPR 4200 model, which also has 1.3 equivalents per liter.
So these resins are basically equivalent in terms of capacity; therefore, they are products that are not exactly the same but that can be compared.
If we have a campaign of 100,000 liters times 2.0 meq, which is the sum of the four contaminating anions (0.5 of each), we obtain a total of 200,000 meq of total anions, or 200 equivalents.
According to the manufacturer, 1 liter of resin is enough for 1.3 equivalents, so how many liters of resin (x) will I need to remove 200 equivalents?
Making this calculation, we reached 154 liters of resin for each campaign. It means that each day could, in theory, produce 100,000 liters of water, interrupt the process, regenerate it, and then restart a new cycle.
This calculation can be considered a “back-off-the-envelope”, which is when the first estimate for the amount of resin is made.
In practice, however, we will see that this number is much higher and explain why.
This case is an example adapted from the book “Water Treatment Principles and Design” by MWH, and it even quotes the expression "back off the-envelope” referring to the approximate calculation of the amount of resin.
This estimate is based on the information about total capacity provided in the product DataSheet; therefore, it is speculative, or yet, notional data.
But why is this calculation unreliable?
For the same reasons explained in the previous case: the selectivity, the separation factors, the level of regeneration, and the ratio between the theoretical or total capacity and the operational capacity of the resin were not taken into account, and also because the resin is not specific or selective for nitrate removal, and yes, a generic resin for anion removal.
So the calculation is not correct and can result in wrong sizing in ion exchange equipment projects, but it serves to give a very approximate idea and an estimate of the amount of material that will be needed.
Generally speaking, an anionic resin removes all anions, not just nitrate. Although there may be more suitable resins for this application (selective ones), there is no resin that removes only a certain anion but the vast majority of negatively charged ionic species.
Analogously, this occurs with cationic resins.
Finally, we can ask the following question: Why is caustic soda used and not salt as a regenerating agent?
From the Selectivity Coefficient table, the target of the next training block, it is observed that the selectivity of anions is greater for chromate, selenate, iodide, and nitrate, in that order.
With chloride in the middle of the table, it would not be able to remove sulfate and bicarbonate because they are higher on the list, have a lower selectivity, and can only be removed with caustic soda, not to mention that even chloride could not be removed because salt contains chloride.
But when we also consider the separation factors, we realize that sulfate prevails over all these ions, being preferentially removed, followed by nitrate, chloride, and finally, bicarbonate.
This is because its α (alpha) is greater, which directly interferes with which ions are removed more easily or in the first place, with priority.
Here it is already possible to understand why, when we want to remove nitrate from water, we also need to consider the presence of sulfate and other ions that prevail over the target of what we really want to remove.
In other words, caustic soda will always be more efficient as a regenerating agent (for soda cycle resins) than NaCl, although it is a product that requires more care in handling.
Likewise, the acid will always be more efficient than the salt when the objective is to remove cations because its selectivity takes precedence over other cations and only the H+ species has the capacity to remove Na+ itself, which would be necessary for the total removal of ions. This is the principle of demineralization.
2.03 Service flow rate BV/h
One of the most important parameters for the design of ion exchange systems is the service rate, also known as volumetric flow.
This rate represents a flow through a certain volume of resin.
So here's the service charge calculation: it's a flow rate divided by resin volume. Manufacturers and designers recommend a value between eight and forty.
Ahead is a practical example: A column that operates at a flow rate of 229 cubic meters per hour has a radius of 1.83 meters and a bed depth of 1 meter. Calculate the service rate (BV/h, or S.F.R.), which is the tag used to designate the service rate, and check if it is within the ideal range.
The first step is to calculate the resin volume (BV).
Considering that the tank has a cylindrical section:
The area of the cylindrical section times the resin bed height totalizes 10.5 cubic meters of resin.
Thus, BV/h is flow divided by the volume of resin, that is, 229 cubic meters per hour divided by 10.5 cubic meters of resin, and we obtain an approximate value of 22 BV/h.
So, this calculation is within the recommended range and attests that the proportion of flow by volume of resin is well-dimensioned.
2.04 Linear flowrate or Linear velocity
Another concept is that of linear velocity, which relates the water flow during service to the tank section area.
The SFR, or BV/h, seen earlier, is calculated by dividing the water flow rate by the volume of resin.
The linear velocity is the water flow divided by the section area of the tank.
In the example, we have the resin tank and the water flow passing through it, but now let's relate this flow to the tank's cross-sectional area. Dividing a flow rate, which is in cubic meters per hour, by an area, which is in square meters, we have the result in meters per hour, which is a unit of velocity.
So using the very same example as above: A softening column with a flow rate of 229 cubic meters per hour, a tank 3.66 m in diameter, and a resin bed depth of one meter: Calculate the linear velocity.
The area of the section is 10.5 m2.
Linear velocity is obtained by dividing the flow by the area.
We get 22 meters per hour.
In this Dow/Dupont datasheet, a linear flow rate of 24 meters per hour appears, which may vary according to the project and the resin manufacturer's recommendation, but remember that at higher speeds the pressure drop will also be greater.
The conclusion that can be drawn is that, at very low speeds, there will be a reduction in the efficiency of the process, and on the contrary, very high speeds can cause high head loss. When using FRP (Fiberglass Reinforced Plastics) tanks, this ratio between the volume and the section area of the tank is already considered by the manufacturers, and it is around 1.45.
Although some manufacturers cite possible linear velocity variations between 12 and 50 m/h, in practice values above 20 m/h are uncommon, either due to the limitation of the booster pumps, the loss of pressure in the hydraulic systems, or the resin bed itself (head loss).
Very high linear velocities can compromise the efficiency of the ion exchange: water passing through the resin too quickly will face greater difficulty in exchanging its ions with the resin's ions; a minimum contact time is required for ion exchange to occur.
One possibility would be to use narrower and taller tanks, in which the same volume of resin would have a higher bed height.
The difference in exchange efficiency would not be much different, considering that, in a taller and narrower tank, the water flow would be higher, even if the path is longer due to the higher resin bed.
However, it is observed that in these cases there is an increase in pressure due to the friction of the water with the surface of the resin; according to the Darcy-Weisbach equation, the fluid velocity contributes exponentially to the head loss.
As mentioned before, when using tanks in FRP, this concern will be less important because it has already been considered by the manufacturers.
For compact systems, in tanks of up to 1000 liters, the linear velocity will be around the range of 15 to 20 m/h in the vast majority of projects, this is what works in practice.
2-05 Selectivity
Ion-exchange resins have a certain affinity or preference for certain ions in an aqueous solution. This affinity is called selectivity.
As you can see in the table, several positive and negative ions are listed with a number next to them. This number is your selectivity coefficient.
The cations are listed on the left side of the chart, and we have the highest selectivity at the bottom: radium, barium, silver 2, strontium, calcium, and so on.
Analogously, on the right side of the table, the anions that have the greatest selectivity are located at the bottom of the table:
Chromate, selenate, iodide, and nitrate.
In general, selectivity is greater for ions with greater charge; therefore, ions with more charge occupy more sites on the resin and, thus, are more attracted to it.
The thorium and aluminum cations, for example, are more attracted to each other than the others.
Likewise, more electronegative anions are attracted, such as chromate and phosphate.
In general, the rule for both positive and negative ions is that the greater the charge, the greater the force of attraction exerted by the resins.
Cátions: Th+4 > Al+3 > Ca+2 > Na+
Ânions: CrO4-2 > PO4-2 > SO4-2 > I- > NO3- > Br- > Cl- > F-
We will now introduce the concepts of equilibrium in ion exchange and the coefficient of selectivity.
Let's take a look at the basic formula of ion exchange, the generic formula, in which the radical (R-) appears here, which represents the resin itself, grouped with sodium (this resin is the sodium cycle) and calcium (ion originating from water and species to be exchanged for sodium).
In the right side of the equation, the species appears with calcium already retained in the structure and releasing sodium.
So this is a very generic expression of what the chemical representation of ion exchange would be.
With these data, we are now going to calculate the coefficient (K), which is the selectivity coefficient.
This was efficient in relating the concentrations of calcium, which is the contaminant, and sodium, which is being exchanged by the resin in the aqueous phases and on the surface of the resin.
We generally have to:
Cjn x qi
Kij = ------------------
qjn x Ci
Where:
Species “i”: Ca+2 (also called “counter-ion” )
Species “j”: Na+ (also called “presaturating ion” )
n = valence of the counterion
Cj = CNa+ = Concentration of the pre-saturating ion in the aqueous phase
qi = qCa+2 = Counterion concentration in the resin phase
qj = qNa+ = Concentration of the pre-saturating ion in the resinous phase
Ci = CCa+2 = Counterion concentration in the aqueous phase
qCa+2 x (CNa+)2
KCa+2 Na+ = --------------------
CCa+2 x (qNa+)2
Likewise, nitrate removal follows the same system. The anion resin, where we now have the positive radical (the anion resin has a substrate with a positive charge), is associated with chloride, a state in which it is at the beginning of the cycle. Next, the radical receives the nitrate anion (contaminant species). In the right part of the formula, the resin already retains the nitrate and releases the chloride. So, in the aqueous phase and the resinous phase, we will have different concentrations of all these substances. Grouping the concentrations of these species allows calculating the coefficient (K), which is nothing more than the selectivity coefficient, which relates the concentrations of nitrate and chloride in both the aqueous and resinous phases.
2-06 Separation factors
Equilibrium can be expressed in terms of equivalent fractions instead of concentration.
The binary separation factor alfa i j is a measure of the preference for one ion over another one, during ion exchange and can be expressed as:
alfa i j = (Yi times Xj) divided by Xi times Yj.
The equivalent fraction in the aqueous phase is calculated from:
Xi = Ci divided by CT .
Xj = Cj divided by CT.
where,
CT = total aqueous ion concentration, equivalent by liter.
Ci = aqueous-phase concentration of counterion.
Cj = aqueous-phase concentration of presaturant ion.
The equivalent fraction in the resin phase is expressed as follows:
Y, i = qi divided by qT.
Y, j = qj divided by qT.
Where,
qi = resin-phase concentration of counterion, equivalent by liter.
qj = resin-phase concentration of pre saturant ion.
qT = total exchange capacity of the resin.
For the special case of monovalent ion exchange with a monovalent pre saturant ion, the separation factor is constant and equal to the apparent equilibrium constant:
alfa i j = K i j = (qi times Cj) / (Ci times qj).
For multivalent ion (i) exchange with a resin having a monovalent pre saturant ion (j), the separation factor depends on concentration (proposed by Harland 1994):
alfa i j = K i j times (qj / Cj) raised to (z-1).
It was seen that the alpha is variable and depends upon the concentrations of the species in the aqueous and resin phases, in addition to the charges of the ions.
We have also seen that the separation factor of a monovalent species is equal to the selectivity coefficient of this species.
Alpha i j = K i j (for monovalent ions, such as nitrate).
It is noticed that the order of the ions is different from the table of selectivity coefficients, although they follow the same coherence, and in a way, they can be compared.
It is important to note that the Separation Factor may not be a constant but rather is influenced by various factors:
- Exchangeable ions (size and charge);
- Properties of the resins, including particle size;
- Degree of crosslinking;
- Capacity and type of functional groups occupying the exchange sites;
- Water matrix, which includes total concentration, type and quantity of organic compounds present in solution;
- Reaction period;
- Temperature.
Because Separation Factors can be influenced by several factors, they are usually determined by performing an equilibrium experiment called a Binary Isotherm. A Binary Isotherm involves performing a batch equilibrium experiment for a binary system. Both binary components and isotherms are performed by Clifford (1999) and produced in Table 16-7, already seen.
Separation factors for commercially avaliable SAC and SBA exchange resins are given in Table 16-7. Based on the definition of first equation seen (alfa i j = (Yi times Xj) divided by Xi times Yj), a separation factor greater than 1 means that ion “i” is preferred over ion “j”.
Ex.: alpha NO3- Cl- = 3.2 , expressed in equivalents, at equal aqueous-phase concentration, nitrate is preferred over chloride by 3.2 to 1.0.
The magnitude of the separation factors is different for WAC and WBA resins from those shown in the Table 16-7 for SAC and SBA resins.
Clifford (1999) provides a detailed experimental procedure and example for determining separation factors.
The concept involving separation and Selectivity factors can be a little tricky, but this will become clearer in the examples presented ahead.
For the time being, it is enough to know that the chemical species influence each other in different ways, and depending on the presence and concentration of a contaminant, this can greatly influence the removal of the target species that you effectively want to remove.
2-07 Cross-linking
We begin this chapter by reviewing some important concepts.
Monomer: a molecule (mostly organic) that can react together with other monomer molecules to form a larger polymer chain or three-dimensional network in a process called polymerization;
Polymers are species consisting of very large molecules called macromolecules, composed of many repeating subunits.
Polymerization is the process of combining many monomers into a covalently bonded chain or network.
Copolymer: a polymer derived from more than one species of monomer. The polymerization of monomers into copolymers is called copolymerization.
We will now go into more detail about the cross-linking: A cross-link is a bond or a short sequence of bonds that link one polymer chain to another. These links may take the form of covalent bonds or ionic bonds. In polymer chemistry, "cross-linking" usually refers to the use of cross-links to promote a change in the polymers' physical properties.
Ion-exchange polymeric resin is composed of a three-dimensional, cross-linked polymer matrix that contains covalently bonded functional groups with fixed ionic charges.
Vinyl polymers (typically polystyrene and polyacrylic) are used for the resin matrix backbone.
Divinylbenzene (DVB) is used to crosslink the polymer backbone. The level of DVB (cross-linking) a resin has determines its tightness. Higher cross-linked resins have lower moisture content, meaning more plastic and higher capacity.
Resins with more DVB are designed for higher-temperature applications (~250°F+) and highly oxidative environments. A DVB level of 15–16% is indicated for extreme conditions for residential or commercial applications (~150°F) with high chlorine.
More DVB means more plastic and less water. But what does 10% DVB cross-linking really mean?
In theory, every tenth unit of styrene in the backbone is a DVB. Some manufacturers define a 10 percent crosslink as the amount of DVB added to a batch as a percentage of styrene monomer.
According to Robert Kunin in his 1972 book “Ion Exchange Resins" , cross-linking can be considered a swelling phenomenon.
At low degrees of cross-linking, the resin is capable of swelling considerably, and the fixed ion concentrations that result are low.
On the other hand, at high degrees of cross-linking, the swelling of the resin is minor and the fixed ion concentration is high, resulting in a greater degree of ion-ion interactions.
Detailing a little more moisture retention, also called “swelling”.
In this particular case, it is 46 to 50%, which shows that this resin is more or less in the average of this parameter.
In another model, we have a similar value, between 45 and 50%, so this moisture value is a strongly acidic resin.
This humidity is intrinsic to the resin; that is, it is the humidity that results from the production process.
In the table, around 46% to 50% on average, we have the related theoretical capacity of 2 milligrams per millilitre, and it can be seen that the cross-linking percentage is 8 percent.
But is 8% too much? Is it too little?
Well, cross-linking is basically the ratio of how much moisture there is in the resin in relation to what polymeric material it has available for ion exchange.
So it's a value that revolves around 8 to 10% in most cases.
The production of resins is an industrial secret, but it is known that they are obtained through chemical reactions between styrene and divinyl benzene, and cross-links can be formed by chemical reactions initiated by heat, pressure, a change in pH, or radiation.
For example, mixing an uncured or partially cured resin with specific chemicals called a cross-linking reagent results in a chemical reaction that forms cross-linking.
Cross-linking can also be induced in materials that are normally thermoplastics through exposure to a radiation source: an electron beam, gamma radiation, or ultraviolet light.
A better-quality resin usually has a higher percentage of divinylbenzene than that of styrene because divinylbenzene is a more expensive material; it costs up to five times more.
In this other table, we can check the selectivity related to cross-linking.
On average, the most common resins available on the market have a cross-linking of around 8%.
The removal capacities of each cation considering this crosslinking are shown here.
Better resins will have greater cross-linking and will have proportionally more divinyl benzene in their chemical composition, resulting in a greater removal capacity and greater added value.
2.06 Coefficients “α” and “K”
Equilibrium in ion exchange, which depends on coefficients, allows estimating a more precise amount of resin in an ion exchange system.
The “K” and “alpha” coefficients can be obtained from tables.
“K” represents the selectivity of the resins, and “alpha” represents the separation factor.
The interaction between the resin and water ions presents a complex dynamic; however, it is known that the efficiency in the removal of certain ions depends on the influence of others, and each chemical species influences more or less the general dynamics of the ion exchange.
The use of separation factors allows for accurately estimating the amount of resin for a given task, but the calculation becomes more complex.
It will be demonstrated how this works, and examples will be presented, although calculation software already considers this condition.
We will not detail mathematical deductions, but we will present a basis that allows obtaining both the selectivity coefficients and the separation factors.
The “K” coefficient can be obtained by calculating the formula presented here, which correlates the concentrations of ions in the aqueous and resin phases.
This relation takes into account that a certain ion is also called a “counter-ion”, and represents an entity that is intended to remove from water.
The “presaturant ion” is the entity that is added to the resin and which will effectively change places with the “counter-ion”.
The calculation of the K coefficient can be obtained by using the basic formula of the ion exchange process, here represented generically by a sodium cycle resin (presaturant ion) and the species to be removed, in this case, calcium (counter ion). In this way, the K of calcium over sodium is obtained.
Analogously, for nitrate, the K of nitrate over chloride is obtained.
The “alpha” coefficient is obtained through a series of deductions and formulas, already seen in a previous class. An example of the “K”, the relation between the ions, and the “alpha” for each component can be obtained by formulas 16–30.
The concentrations of each species are given by the formula 16–20.
The fractions with which the capacity of the resin is divided between the components are answered by formulas 16–21 and represent the balancing effect produced by the concepts of selectivity and separation factors.
We will present a practical example ahead: how and how much certain ions interfere with the removal of the target species.
Calculation of the alpha coefficient:
The alpha coefficient always has a superscript “i” (counter ion) and a subscript “j” (pre-saturating ion).
The calculation of ions participating in the solution can be obtained, as already seen, by the follow equation:
alpha, i, k, = alpha, i,j, plus alpha,j,k.
For a binary system, the alpha of nitrate over chloride is taken directly from Table 16.7. Alpha nitrate over chloride = 3.2.
The alpha of chloride over nitrate is obtained by calculating (1 divided by x). Chloride alpha over nitrate = 0.3125 .
In the case of multicomponent systems, the fractions into which the resin is divided acting on each component, also called the concentration of the counterion in the resin phase, can be treated by the formula q,i.
The alpha coefficient follows the same systematics as the binary system. In the following example, the objective is to calculate all the alphas of the chemical species nitrate, chloride, and sulfate.
Here, the pre-saturating ion will be chloride.
The alphas of the nitrate, sulfate, and chloride counterions are taken directly from Table 16.7.
From equation 16.30, all the other alphas are calculated, as shown in this slide. The newly computed alpha values will be used in a practical example of the resin needed for a three-component system, with a case presentation coming soon.
Comparative calculations will be shown for systems composed of only one component, binary, or multicomponent.
2.07 Anderson proposal
In ion exchange technology, the affinity of exchanges is related to charge and size. The higher the valence, the greater the affinity, and the smaller the effective size, the greater the affinity. For a given sense of similar ions, there is a general order of affinity for the exchanger. For synthetic resin exchangers, the relative affinities of common ions increase, as shown in the following table:
The design of ion exchange units is based on ion exchange equilibria. The generalized reaction equation for the exchange of ions “A” and a cation exchange resin can be expressed as shown:
The selectivity constant depends upon the valence, nature, and concentration of the ion in the solution. It is generally determined in the laboratory for specific conditions measured by:
Anderson (1975) rearranged Eq. (6.107) using a monovalent/monovalent exchange reaction with concentration units to an equivalent fraction as follows:
Let's pay more attention to these last two expressions.
First, for monovalent ions:
Let's identify each component:
B = Specimen to remove
A = Other ions
_
XB+ = Fraction of the species to remove, related to the total capacity of the resin. Usually, the factor you want to determine;
XB+ = CB/CT Ratio between the species to remove and all species present in the sample;
KBA = Average selectivity of other contaminants.
KBA1 = KBNa / KA1Na (cycle sodium)
OR
KBA1 = KBCl / KA1Cl (cycle chloride)
KBA2 = KBNa / KA2Na (cycle sodium)
OR
KBA2 = KBCl / KA2Cl (cycle chloride)
T,
KBA1 + KBA2 + ...
KBA = --------------------------
nr. species
Secondly, for bivalent ions:
In the case of the presence of four ions, for example, sulfate, bicarbonate, chloride, and nitrate, our main ion will be sulfate because it is bivalent and preferable over others, which in turn will reduce to a single monovalent species.
After finding the percentage of sulfate, that is, the percentage of resin sites reserved for this species, a new calculation will be performed, now as if there were only three entities.
As there are three monovalent ions, we will consider the most important ion as the target of the calculation, usually for anions, the nitrate.
Let's identify each component:
_
XBn+ = Fraction of the species to be removed, related to the total capacity of the resin. Usually, the factor you want to determine;
n = Valence
XBn+ = Ratio between the species to be removed and all species present in the sample
KBA = Average selectivity of other contaminants
_
C = Total resin capacity, in meq ou eq
C = Σ cations or Σ anions, in meq ou eq (keep same unity as previous)
So far, the theory and deductions of some chemical equilibrium equations involving ion exchange have been shown for Anderson's proposal, presented in 1975. Further on, detailed examples and case studies will be presented for further understanding.
1.06 Iron and manganese removal – Part I
Much is said about the removal of iron and manganese with resins.
This type of contaminant is very common in groundwater, and resins are increasingly being used for its removal.
However, let's make some considerations.
The strong cationic resin not only removes iron and manganese but also any positive ions; we'll talk about the selectivity issue later.
The SAC resin does remove iron and manganese, but these metals are ions that do not have good selectivity, but an average one.
So other ions are removed in priority before iron and manganese.
A manufacturer produces a resin that promises to remove up to 10 mg/l of iron from water.
In the tests, it was observed that water with about 5 mg/l of iron in groundwater was without any iron after the softening process.
There are several requirements for this water: it cannot have organic matter, and if it does, it has to be low, and there is no way to guarantee how long this resin will last.
The tests were conducted for six months, during which the resin worked perfectly, but after that, the well was closed and the tests were ended.
It is necessary to evaluate with the manufacturer the durability of the resin because one of the criticisms that is made in relation to the removal of iron from water with ion exchange resin is the cost.
The cost of the resin is much higher than the cost of removing metals in another way, with pre-oxidation and filtration.
An assessment should be made of whether it is economically feasible to remove iron with resin.
A German study related the effect of iron on the resin not as a species to be treated but as a contaminant.
They did the tests, in which 0.2 mg/l of iron reduced the capacity of the resin by 6%.
Between 0.5 and 4 mg per liter of iron, the reduction in capacity amounted to about 13 percent.
And with an iron content of 8 mg/l, the resin capacity reduction was 16%.
It is necessary to evaluate waters with iron and manganese and how these contaminants can greatly reduce the useful life of the resin.
It is recommended, therefore, that one of the options for using the resin to remove iron and manganese is a pre-treatment with oxidation followed by filtration.
Oxidation can be done with hydrogen peroxide, chlorine, ozone, chlorine dioxide, or aeration.
According to the book “Water Treatment/MWH”, doses above 5 mg/liter of chlorine have been effective in the oxidation of Fe+2 as well as organic iron. However, high doses of chlorine can compromise the potability of the water in view of the probable formation of disinfection by-products (DBP). Fe+2 oxidation requires 0.64 mg chlorine per mg iron, while Mn+2 oxidation requires 1.29 mg chlorine per mg manganese. Regarding the pH, the recommended range is between 7.5 and 8.0, but for manganese, the pH should be higher, around 9.5. For contact time: between 15 and 30 minutes for iron and between 2 and 3 hours for manganese.
Filtration can be carried out using layers of activated and anthracite charbon, or even specific zeolites.
We still have to consider the issues of iron +2 and iron +3.
Iron 2 is iron dissolved in well water, which, in contact with an oxidizing species such as oxygen or chlorine, will transform into iron 3.
Iron three ends up producing fouling, which is an incrustation on the resin surface, which causes a huge reduction in the exchange capacity, and finally the need to carry out an acid cleaning on the resin to reconstitute it.
This involves cleaning with hydrochloric acid heated to 140 to 160 degrees Fahrenheit, leaving it in contact for practically half a day.
That is, if it is not necessary, in the most extreme case, to completely change the resin.
We'll talk more about fouling on another topic, but as we've seen, one of the possibilities for resin contamination is the iron itself.
1.07 Iron and manganese removal – Part II
In the book “Iron and Manganse Removal Handbook” by Elmer O. Sommerfeld, ion exchange technology is not even mentioned. It confirms the most suitable technologies for this purpose, such as: pre-treatment with oxidants, ozone, chlorinated derivatives and hydrogen peroxide, filtration and also mentions alternative technologies:
- Zeolite softening: this is what most resembles ion exchange, especially because it uses salt as a regenerant. Zeolite can remove hardness as well as metals as long as they remain in the ionic state. In this process, no oxidant is used because the metals will precipitate in the form of oxides, covering the surface of the zeolite.
- Lime/soda ash process: adding soda ash to a pH of ~10 is suitable for removing calcium, but if it is to precipitate ferrous and manganous ions, it must be raised to 11 – the metals will precipitate as hydroxides. The precipitate is removed together with calcium carbonate. It is a process that requires many chemical specialties, and is therefore not economical.
- Sequestrants: This process, which mainly uses sodium silicate and orthopolyphosphate together with chlorine as an oxidant. The sequestrant has the property of sequestering colloidal iron ions, that is, oxidized from ferrous to ferric, and therefore, the oxidant is dosed before the sequestrant. Orthopolyphosphate works better at lower pH. The hardness of the water directly affects the efficiency of the process. Unlike iron, manganese takes longer to oxidize. It is also an expensive process, due to the high cost of sequestrant agents.
The use of ion exchange resins for iron removal can be considered when the amount of iron is small (< 1.0 ppm) and it is also desired to remove hardness, with iron entering as more of a contaminant and not as a target species to be removed. Most applications do not even mention iron as one of the cations in the ionic balance table.
In the book “Water Treatment Principles and Design” this subject is also covered. It is stated that ion exchange resins can be used to remove quantities of less than 0.5 g/l of iron and manganese from groundwater. Normally, sodium cycle SAC resin is used.
Separation factors negatively influence removal, because these metals are not privileged in the order of removal preference, both below calcium. Iron is a little better, ahead of magnesium. This means that, in water with low hardness, these metals can be removed, however the frequency of resin regeneration must be greater than that carried out in water with the same hardness but free of metals.
It is important to highlight that only metals in ionic format will be removed by ion exchange: for those complexed by phosphate derivatives or oxidized by peroxide, ozone or air, the SAC resin will have no effect. However, the use of SBA resin is mentioned to remove complexed or oxidized iron, under specific conditions.
When the iron content is high and this is the target species for removal, the recommendation is to opt for technologies other than ion exchange.
Hello!
My name is Carlos Maffessoni. I am a chemical engineer at IG Engenharia, and I am pleased to be here to present the Ion Exchange Resins course, which aims to treat water for human or industrial consumption.
The success of the course launched in 2021 motivated us to expand and review all topics, which is even better and more complete now. The presentation has improved both image and sound quality. We counted around 80 students until February/2025 on another platform.
The training is extensive and covers everything from historical aspects of ion exchange resins to a very detailed theoretical basis – including the concept of separation factors, the various configurations of the four types of resins available on the market, and what engineers and technicians are most looking for: sizing projects, manually and with software.
The course is divided into seven chapters, and we will briefly comment on each of them.
In the first chapter, we will cover the basic concepts necessary to understand this technology. Main topics covered: historical review, chemical structure and types of resins, units, water contaminants that can be removed with this technology, and the first simple examples of softener sizing calculations.
In the second chapter, the concepts are expanded, and we will see what “Service Flow Rate”, “Linear Flowrate”, "Selectivity" and "Cross-linking" represent. The theory of Selectivity and Separation Factors will also be presented, which explains why very simplified calculations can result in large sizing errors, simply because the water contains a more selective contaminant than the species to be removed. Finally, we will learn how to calculate dimensional systems using another calculation method, presented by Anderson in 1975.
In the third chapter, the enemies of resins will be presented and what type of damage they cause to them, as well as "Preferred Paths" or "Channeling" and the applicability of WAC-type resins for alkalinity removal.
The main theme of the chapter is the relation between the theoretical or nominal capacities and the real or operational capacity of the resins, and how much this affects the sizing calculation, especially the manual one.
Nitrate removal will be extensively detailed in chapter four, as it is an important contaminant rarely detailed in the literature and even less in sizing software. We will see how much a particular contaminant, in this case sulfate, alters the nitrate removal capacity of anionic resins. Anderson's technique will be used, and a study case will be carried out.
In the fifth chapter, the regeneration stage will be detailed, presenting everything from water suitable for regeneration, through types and concentrations of regenerants, and a practical calculation will be presented containing all the information necessary to adjust your process correctly.
Subjects little or never covered in courses and literature will be mentioned in chapter six. Hydraulic details of softening and demineralization systems, mixed beds, unit conversion, and even the current market situation will be presented. Limitation of softening will also be taught, because when you remove one ion, you add another, and this needs to be considered. The resins will also be presented as they are, with images and films obtained through microscope observations.
Finally, in the chapter 7, the most desired subject for engineers, technicians, and those on the front line of projects and assembly of ion exchange systems, but which can also be of great use for students and master's students: ion exchange system projects, contemplating the sizing of the amount of resins required for a given type of water at a given flow rate. Hydraulic details of real projects. The influence of the degassing stage on demineralization projects. Comparison of results using similar resins from different brands. Alkalinity removal. The various combinations of vessels for demineralization, using the four types of resins available on the market, among other important issues that rarely are commented on even in specialized literature.
A complete theoretical-practical training, useful for engineers, technicians, students, teachers, and people who work directly or indirectly with ion exchange resins, softeners, and demineralization systems, both for water intended for human consumption and for industrial use.
Thank you very much, success!