Introduction to chemical engineering/chemical processes
Chemical processes/chemical engineering
Ah! Chemical engineering. Universal engineering. Look around you. Can find three objects that haven’t required a contribution by chemical engineers? Probably not. Clothes, food, pharmaceuticals, paints, paper, automobile tires, plastics, even paper clips—they all have constituents that were developed, designed, and manufactured by chemical engineers.
There’s something for everybody in chemical engineering. If you like the thrill of accidental discovery, it’s got it. Super balls, nylon, cellophane, Bakelite, NutraSweet were all “discovered” , not predicted nor anticipated. Do you want to save the lives of a half billion people each year? Be a chemical engineer and contribute to the fertilizer industry that helps enhance food production. On average, according to the International Fertilizer Industry Assn, fertilizer increases crop yields by 40—50%. Who won World War II for the allies? You guessed it. Chemical engineers. (Others helped.) The development of synthetic rubber, production of penicillin, and manufacture of fissionable materials for the atomic bomb tipped the scales in this conflict. Protecting the environment, devising synthetic drugs for the human body, developing new materials for computers and space travel are all within the purview of chemical engineers. No wonder they call it universal engineering. . .
Chemical engineering is not just applied chemistry. It is true that chemical engineers produce chemical compounds and materials. But that's a vastly oversimplified view of what really goes on. In most cases the chemistry part is the simple part. The real problem that chemical engineers deal with is how to produce these chemicals at low cost, with the least-hazardous by-products, and in vast quantities. A chemist mixes a few chemicals in a test tube, shakes it up, and, voilà, a new chemical compound. A chemical engineer feeds two chemicals into a 10,000 liter, continuous process reactor at a rate of 1000 liters/second and hopes that the resultant product will have a satisfactory yield and won't explode in the process. The problems addressed by chemists and chemical engineers are vastly different.
Chemicals can be corrosive or toxic. Chemicals can be solid, liquid, or gas. Chemical reactions can generate heat or require heat. Chemical reactions require controlled mixing. And the materials are to be handled in very large quantities. Who's going to design the apparatus to deal with all this? It's the chemical engineer. Of course a chemical engineer must know chemistry. But he also must know fluid dynamics, heat transfer, materials science, and feedback control systems. And we haven't even mentioned bio-chemical applications. The chemical engineer is often called the universal engineer, because he must know something about almost every branch of engineering. And paychecks reflect that. Traditionally chemical engineers command the highest salaries in engineering.
Where did “chemical engineering” come from? Originally chemical production was the purview of mechanical engineers and applied chemists. In the early 1800s an important chemical, soda ash (alkali), was being produced by the Le Blanc process which released hydrochloric acid, sulfur, and chlorine gas—all nasty stuff—into the atmosphere. At that time, hazardous waste was such a problem in England that the government dispatched teams of alkali inspectors to review chemical plants. One of them, George Davis, in 1887, decided to organize his years of experience into a series of 12 lectures on the engineering of chemical processes. That was the first effort at defining a new field. Then in the next year, Lewis Norton, a chemistry professor at MIT, initiated the first four-year course "Course X". Chemical engineering as a discipline was born. So, the academic element of chemical engineering is just over 100 years old.
Much younger than that is the explosive new field of bio-chemical engineering. This is the arena of genetically-altered agricultural products, synthetic drugs, and transdermal drug delivery systems. Who knows what is next.
The basic materials
Before we begin with what chemical engineering is all about, we should first say a little bit about the raw materials: atoms and molecules. These are the building blocks of everything. What chemists and chemical engineers are interested in are the characteristics of these atoms and molecules individually and collectively.
Individually they have properties like mass, momentum, geometrical form, and electrical polarity. In proximity with other molecules or atoms, they have properties like chemical reactivity. And in large numbers, they exhibit bulk properties, e.g., viscosity, electrical conductivity, heat capacity. And, even though these atoms and molecules are discrete entities, there are so many of them in the tiniest speck of material, that the material can often be treated as a continuum, i.e., a homogeneous substance.
Molecules can be relatively small and simple, like ammonia (NH3), sulfuric acid (H2SO4), ethylene (C2H4), and sugar (C12H22O11). Or they can be very large, as with polymers. What are polymers? They are very long molecules consisting of a string of like monomers. A monomer is a simple molecule that can be linked to another like molecule. Two monomers bolted together is a dimer. Three bolted together is a trimer. Many bolted together is a polymer. Ethylene is an example of a monomer. As a monomer, its formula is written as [-CH2-CH2]. Two ethylene monomers bolted together is denoted [-CH2-CH2]2 . Many bolted together is denoted [-CH2-CH2]n , where n can be a million. This is polyethylene.
Polymers, in spite of their size, are very simple--merely very long chains of the same monomer:. But, in some cases, especially biological ones, molecules can be both very large and very complex. One example is the protein insulin which controls sugar levels in the body. Its chemical composition is C257H383N65O77S6. And that’s one of the simpler biological compounds.
How one works with these molecules—produces them, extracts them, modifies them, isolates them is almost unique to each molecule. That’s what makes chemical processing so interesting. In some cases, molecules can be broken into smaller constituents by thermo-physical means—for example “cracking” crude oil into oil, gasoline, and tar. In other cases, molecules are produced and extracted by biological means. Synthetic insulin is produced by extracting protein from specifically-modified e-coli bacteria.
Sample chemical processes
Let’s talk about some specific chemical processes. How’s about beginning with the production of synthetic insulin? How is it produced? First, understand that insulin is a natural protein that is produced in that most-sophisticated-of-all chemical processing factories: the human body. It controls sugar levels within the body. So, when there is a problem with natural insulin production in the body, it is manifest as a sickness called diabetes. For many years, insulin was separated and purified from the pancreases of cows and pigs. But this animal insulin was not identical to human insulin. It provided some of the sugar-controlling properties, but being a foreign substance, the body produced anti-bodies which tended to neutralize its actions. Human insulin would be better, but how to produce it?
Researchers discovered a way. They insert the human insulin gene into an Escherrichia coli (E. coli) bacterial cell to make it a part of the cell’s genetic makeup. The insulin gene is then able to command the cell to produce insulin. There it is, a bacterial chemical processing plant. And as the cell divides, each new generation of cells would contain that human insulin gene thus creating additional factories. It sounds simple, but it’s really not. To begin with, a plasmid (a circular DNA molecule) has to be extracted from a single cell. Then, special enzymes are added to “open” the plasmid so that the human insulin gene can be inserted. Then, additional enzymes are added to reclose the plasmid. Thus the researchers create a “genetically engineered” molecule that can induce insulin production. But, the task is still not completed. At this point, they must insert this recombined plasmid into another weakened E. coli cell so that it’s not naturally rejected by the cell’s immune system. Only then can the desired product be produced.
At each step along the way--extracting the plasmid, cutting and splicing genetic material, reinserting recombined plasmids—there is a call for chemical engineering. How to do it, and how to do it on a large scale at low cost. An interesting set of problems. And we haven’t even talked about extracting and purifying the insulin generated within the bacterium.
Another example of a really interesting chemical process that has enormous implications for the computer industry is one that involves co-polymers. It’s a method for producing a material that can store a terabit (1012 bits) of information in one square inch. Imagine ten 10-gigabyte disk drives crammed into an area the size of a postage stamp! This technique involves the production of a special configuration of block co-polymers. Recall, a polymer is a very long chain of simple monomers. A co-polymer is a chain of two polymers linked together. In this case the co-polymer consists of polystyrene and polymethylmethacrylate. Just call them PS and PMMA. If this co-polymer resides in the solvent toluene and is spin-cast onto an electrically-conducting plate, the co-polymer will self-organize itself into a uniform matrix of PMMA cylinders surrounded by PS. Then, if this coating is heated and subjected to an electric field, these cylinders will orient themselves in the direction of the electric field. The PMMA cylinders have diameters of 14 nanometers and will be spaced approximately 24 nanometers apart. How many cylinders per square inch is that? A very large number. Now the interesting part. Acetic acid can be used to dissolve the PMMA cylinders. The remaining holes can then be filled electrochemically with a magnetic material, for example, cobalt. This produces an ordered array of nanowires, each able to store a bit of information via its magnetic polarity.
How many chemical engineering processes were involved with this technique? Producing the co-polymer, mixing it with toluene to give it the right viscosity for spin-casting, heating under an electric field, dissolving away the PMMA cylinders, electro-depositing a metal in the holes. And each of these processes requires its own elaborate technique. All for a square inch of unusual material. This process is still in the research stage, but it won’t be long before it’s in production. This is chemical engineering.
Suppose you don’t like doing stuff at the atomic or molecular level. Suppose you want to work with huge production reactors—like those that produce H2SO4 (sulfuric acid). What might be interesting in these processes? Maybe money. World sulfuric acid consumption in 1997 was about 160 million metric tons at a cost of approximately $8 billion. In its simplest form, we produce sulfuric acid through the reaction 2S+3O2+2H2O(2H2SO4 , i.e., 2 atoms sulfur plus 3 molecules of oxygen plus 2 molecules of water yields 2 molecules of sulfuric acid. If the process for effecting this reaction could be improved by just one percent, the cost savings would be $80 million per year, every year. Maybe improved mixing; maybe better process control; maybe better reactor design—any of these things could do the trick. $80 million per year is a reward worth striving for.
One can think of chemical engineering as implementing two basic processes: reactions and separations. The efficiency of these processes involves transport --flow, mixing, diffusion--and thermodynamics. And, to quantify and model these processes, one uses the principles of material and energy balance.
Reactions create new molecules out of two or more constituent components. In our insulin example above, a bacterium was a natural reactor using materials within its cell to create the new molecule insulin. That’s what chemistry is all about. Separations attempt to isolate a substance which is contained in a mixture of other ingredients. In the above examples of chemical engineering processing, we’ve mentioned words like “separation”, “extraction”, “purification”. Even when we talk about “dissolving away the PMMA cylinders”, we’re talking about separation. In fact, a lot of chemical engineering involves how to get what we want, and eliminate the rest of the stuff. That’s what separations are about.
How do we achieve these reactions and separations? In part, through the transport of material. Molecules A and B cannot react to form C unless they are brought together. And they have to be at the right temperature for them to react—that’s the thermodynamic requirement. We’ll return to these topics later. . .
Attributes of matter
To understand how chemical engineering processes might work, especially separation processes, we need to discuss some attributes of matter. We’ll restrict our discussion to those which are non-biological. They’re the easiest to grasp for those of us who do not have a sophisticated bio-chemical background. What do we mean by attributes? Attributes are characteristics which can distinguish one chemical compound from another, and they can depend on either individual molecules or a homogeneous substance made from them.
Here's an incomplete list.
Properties of a single substance:
a) boiling point
b) freezing point
c) density
d) volatility
e) surface tension
f) viscosity
g) molecular complexity (size, geometry, polarization)
The first two relate to the states or phases of matter. There are three: solid, liquid, and gas. As you probably know, essentially all matter can be made to exist in all three states. It's just a matter of how much heat you add or remove. Water is the easiest to imagine because we use all three forms almost every day—ice, water, and steam. Each substance has its own temperature at which it changes phase. These are the boiling and freezing points. And these points depend on the pressure. At one atmosphere of pressure (760 torr), water freezes and boils at 0(C and 100(C, respectively.
Density is the measure of how massive the atoms are and how closely they are packed in the continuum. In liquid and solid phases, atoms are packed perhaps 2000 times more closely than in gas phases. And, of course, each substance has its own characteristic density under controlled conditions. Density is mass/volume. The density of water is ( = 1gm/cm3 or 1000 kg/m3.
Volatility is a little tricky. Volatility is the tendency for atoms in a liquid (or solid) phase of a substance to evaporate from the liquid and become vapor or gas. Atoms in this gas phase are the things that you can smell. Examples of highly volatile substances are gasoline, fingernail polish remover (acetone), moth balls, or PineSol. A substance never exists just in its liquid (or solid) phase. There are always atoms in the liquid (or solid) phase which escape and enter the gas phase. (Even a metal solid like iron loses atoms to become iron vapor.) Simultaneously, atoms in the gas phase become captured by the liquid and enter the liquid phase. At any temperature there's a balance or equilibrium between those atoms leaving and those atoms entering the liquid phase. The concentration of atoms or molecules in the gas phase at equilibrium is quantified as the substance's vapor pressure. The higher the vapor pressure, the higher concentration of atoms in the gas phase. And the higher the temperature, the higher the vapor pressure. When the vapor pressure reaches that of the atmosphere, the liquid boils. The temperature at which that occurs is called the boiling point. Volatility becomes especially interesting when two or more substances exist in a liquid mixture. Since each substance will likely have different volatilities, the relative concentrations of substances in the gas phase will be different than the relative concentrations in the liquid phase.
Surface tension is a property of a substance in the liquid phase. In a liquid, molecules are attracted to one another. And except for those on the surface, molecules experience this attraction from all directions because they are entirely surrounded. At the surface, however, a different situation exists. Surface molecules are not completely surrounded by other molecules, and they are only attracted inward to the layer of molecules directly below them. This inward net force creates an apparent film or skin on the surface of the liquid. Surface tension is the force needed to stretch this film. It's measured in dynes/cm or N/m. Water has a surface tension of about 74 dynes/cm; mercury has a surface tension of over 700 dynes/cm. Surface tension causes drops of liquid to form into spheres, e.g., mercury on a tabletop. It allows one to overfill a water glass. And it provides the force to allow water to flow up a towel through capillary action.
Viscosity is a measure of how easily a substance can be stirred. Suppose you have cup of water and a cup of honey. If you try to stir each of them with a spoon, one is easy, one is very difficult—especially if the honey is cold. The harder it is to stir, the more viscous it is. It’s an extremely important property in engineering because it determines how easily thing can be mixed and how much energy it takes to deform it. Consider trying to drink a glass of honey through a straw. Viscosity is not just a property of liquids. It is also a property of gases. Well, you say, I can stir air with essentially zero effort—or so it would seem. But one of the parameters that contributes to the force required to stir something is its density. So, to get a better handle on a substance’s viscosity, engineers sometimes use kinematic viscosity which is viscosity divided by density. Using this normalized form of viscosity , air is actually ten times more viscous than water. The viscosity of air is one of the principle components contributing to the drag of an airplane. Viscosity ( is measured in poise or centipoise (1/100th of a poise). A poise = 1gm/ (cm s). Kinematic viscosity is usually indicated by ( = (/(. Some viscosities: (water = 1 centipoise; (glacier = 108 centipose; (earth's interior = 1024 centipoise. (Yes, a glacier is a fluid, but with a very high viscosity. Observe the yearly “dirt” lines on the glacier in the picture. Initially they were straight lines corresponding to the elevation at which new snow begins to fall. As the glacier flows—most rapidly in the center—the dirt lines distort to show that motion.)
Molecular complexity is a property that is responsible for all sorts of behavior in substances: boiling point, viscosity, etc. Molecules can be as simple as two oxygen atoms forming an O2 molecule, or as complicated as a DNA molecule having millions of atom groups in a long chain. Polymers are another class of substances that can have very long chains of atoms strung together. Some of these are so long, that you can see these molecules under a simple microscope. Recall that a single atom has a diameter of approximately 10-10 m. Imagine pouring a pot of peas into a bowl. That’s like pouring water into a bowl. Now imagine pouring a pot of spaghetti into a bowl. That’s like pouring a soup of long-chain polymers into a bowl. Because of their geometries and how they must interact with the neighboring molecules, the behavior of peas vs. spaghetti is different.
And these differences are not just pure geometry. Water H2O has its atoms arranged in a peculiar way. The hydrogen atoms are bonded to the oxygen atom at an angle of 105(. What this means is that, although the water molecule is electrically neutral, that neutrality is not uniformly distributed. In the drawing, the top is more negative because the oxygen atom has valence –2 and the bottom is more positive because the hydrogen atoms have valence +1. This results in a molecule that is a dipole, i.e., one side has a positive charge, the other a negative charge. This dipole will have preferred orientation in its interactions with other dipoles—especially other water molecules.
Molecular complexity, in fact, can explain a considerable amount of a substance’s properties. For example viscosity is a function of two elements: intermolecular forces, and the lengths of the molecules.
The properties we’ve just mentioned exist in a single substance. There are a few more that we should mention that exist in the presence of other substances. Again, a short, incomplete list.
Properties in the presence of another substance:
a) Solubility
b) Chemical reactivity
i) phase change
ii) generation of heat
Solubility is the ability of one substance to dissolve into another. We usually think of solubility in terms of a solid dissolving into a liquid, like sugar into water; but it equally applies to gases into liquids and liquids into liquids. Example of these are CO2 dissolved in water to produce carbonated beverages; or oil not dissolving so readily in vinegar. As we shall see, the property of solubility can be used as the key parameter in some extraction processes. The solubility of oxygen in 100gm of water @ 1 atm is 0.0043gm. 100gm of water can absorb 53gm of ammonia or 36gm of salt under the same conditions. Quite a difference.
Other things can happen. Supply a spark to hydrogen and oxygen gas, and you have water. Two gases combine to form a liquid. Two liquids could combine to form a precipitate—a solid. Almost any phase change is possible in a chemical reaction. And accompanying that reaction could be heat. So taking heat into account is important for the chemical engineer. That’s a part of the science of thermodynamics.
Transport properties:
We've discussed some of the intrinsic characteristics of substances: , volatility, solubility, etc. But, because these substances are made up of atoms having mass, they also carry the properties of momentum and heat. And, of course, the atoms can move with respect to one another. These properties involve transport—the moving of a property from one place to another.
There are three kinds of transport:
1) momentum transfer
2) heat transfer
3) mass transfer
Why do chemical engineers care about these processes? Because they completely determine how chemical engineering is actually accomplished. Moving heat and material from one point to another is essential for creating and controlling chemical reactions.
Momentum transfer is the process by which one part of the substance, say a liquid, is transferred to another part of the substance. The study of momentum transfer is really the study of forces: body forces, pressure forces, friction forces. When you stir a cup of coffee with a thin rod, you move every molecule in the cup. Why? Your rod doesn't touch every molecule in the cup. The answer is that, through viscosity, the momentum of the rod is being transferred to the rest of the fluid. Without momentum transfer it would be almost impossible to uniformly mix cream into coffee or produce an efficient chemical reaction.
Similarly, if your coffee is hot and the cream is cold, after mixing, what do you experience? Little patches of hot and cold in your mouth? No. When hot and cold fluid particles are brought into proximity with one another, heat is transferred by diffusion from the hot one to the cold one. But diffusion is a very slow process and operates best only over short distances. To speed it up, you stir your coffee. Here you are bringing regions of hot and cold together through convection so that diffusion can work more efficiently. Meantime, heat is also being transferred from the coffee through the cup into the atmosphere by conduction.
Finally, what color is your coffee and cream after you've stirred it up? Dark brown with little specks of white? No. (Unless you've used spoiled cream.) it's a uniform light brown. Why? Because your stirring has brought the cream and coffee close together so that molecular diffusion can mix them thoroughly.
How does all this work? The answer is identical for all of them: gradient transport. The rate at which a property can be moved depends on the spatial gradient of property. A gradient is the change over distance. Although the mathematics is a little daunting, we offer here the expressions for transport:
For momentum transfer, it's called Newton's Law:
flux of x-momentum in z direction[pic], vx is velocity in x-direction, ( is density, ( is viscosity.
For heat transfer, it's called Fourier’s Law
heat flux in z-direction [pic]; ( is thermal diffusivity, ( is density, cp is heat capacity, T is thermal energy (heat).
For mass transfer, it's called Fick’s Law
mass flux of A in z-direction [pic]; D is molecular diffusivity of A in B, CA is the concentration of A.
There are many different parameters in these equations but they all have one thing in common. They all have a gradient, i.e., a factor [pic], where ( is a concentration or quantity of a property and z is a direction. In other words, these equations all behave in the same way. The flux or rate of movement of each property depends on the spatial gradient. And the flux is always negative, i.e., the flux travels down the gradient. From hot to cold, or from fast to slow, or from high concentration to low concentration.
To actually “see” gradient transport, explore a couple of our virtual experiments: Heat Conduction and Diffusion Processes. Both can be found at jhu.edu/virtlab/virtlab.html . In these visual experiments, you can set up various diffusion problems and see what happens. Experiment. See if you can get a feel for how gradient transport works. And, although one of these experiments is called “heat conduction”, understand it could just as easily have been called “dye diffusion”. The process is the same: gradient transport. In one case heat is being transported; in the other, dye concentration is being transported.
Transport is a phenomenon important in many engineering disciplines: momentum transfer across airplane wings, dispersion of pollution in the environment, diffusion of oxygen into the blood stream, conduction of heat away from semiconductors. The list is endless. So, although we are talking about transport in the context of chemical engineering problems, it's an extremely important phenomenon in almost all branches of engineering.
Chemical production
Chemical engineers produce chemical compounds and materials. But how? Basically there are two strategies: 1) create the desired compound from raw materials via one or more chemical reactions, or 2) isolate the compound where it exists in combination with other substances. These are the "reactors" and "separation processes" we talked about earlier.
We'll discuss both of these strategies. But we'll concentrate on separation processes, because in order to really discuss reactors we need to have a pretty good grasp of chemistry. We'll not assume that here.
Reactors:
The basics of reactors can be introduced with a single illustration:
This is everything. The box is the reactor—the container where chemical reactions take place. There are inputs and outputs. What we're really interested in is the efficiency of producing "product"—the desired compound. Efficiency usually means cost vs. yield, i.e. how many dollars per kilogram does it cost to create. Sometimes the cost is difficult to calculate. For example, the by-products may be expensive to neutralize and/or dispose of. Or the by-products may have value in and of themselves. And there could be additional costs associated with separating the product from contaminants. All these costs count.
Now, let's talk about the reaction itself. In fact, let's talk about creating a hardened epoxy from two raw materials: A--the epoxy; B—the hardener. Here the reactor will be a piece of wax paper. Problem #1 is to ensure that the correct ratio of A and B are brought together. If the ratios are wrong, the result will be useless. There is a narrow range of ratios that will work, so you must develop a metering system whose fluctuations lie within that range. But simply throwing A and B together is insufficient, because the ratio must be correct everywhere. Problem #2 is how to stir it. What procedure do you use to produce a final homogeneous mixture? If there are regions of too much A and too much B, the reaction will not go to completion. Stirring will work, thanks to viscosity and diffusion.
Suppose you've solved these problems. You can reliably produce a few grams of mixed epoxy that will properly cure. Now I tell you that I need a liter of mixed epoxy at a time. What's your strategy? You say, no problem. I've got the technique. I'll just mix the epoxy in a plastic milk jug, and we're done. So you do that. Unfortunately, seconds after you begin mixing, the mixture becomes scalding hot and the milk jug shatters. What happened?
The curing of epoxy is exothermic—the reaction produces heat. And the speed of all chemical reactions depends on its temperature (an almost universal truth). So, as the curing proceeds, the exothermic reaction causes the mixture to heat up, which in turn causes the reaction to proceed faster. It's a positive feedback system: higher temperature produces a faster reaction which produces a higher temperature.
So why did you have no trouble mixing a few grams on wax paper? There's a twofold reason. First, exothermic reactions generate heat per unit mass. So the mass was a lot less. Second, the mixture was spread out over a comparatively large surface area. So heat could escape more quickly. Transport is important again.
This is a typical chemical engineering problem. A solution to this problem is to not mix the epoxy in batches at all, but rather mix it continuously. Continuous power feeds of A and B into a small mixing nozzle will produce a good result. Mixed epoxy is continuously accessible, and there is never a large volume of mixed epoxy to generate serious heat. This is an apparatus that is used in industry for coating large structures in epoxy.
Sometimes one does not have so much flexibility with a reactor. Take for example a concrete dam (like Boulder Dam). The curing of concrete is an exothermic reaction. When concrete dams are built, they are built with miles of water pipe imbedded in them. The objective is to transfer the heat away from the curing concrete. Nuclear reactors are another example. Except there, the whole purpose of the reactor is to generate heat and put it to use.
Reactors are of two types: batch and continuous. In one case, fixed quantities of materials are mixed together to produce a fixed quantity of product--a single batch. In the other case, materials are introduced at a fixed RATE to produce a continuous output of product. Batch reactors are fairly straightforward because one begins anew for each batch. The trouble is that batch reactors are relatively inefficient, because they must be cleaned and reloaded before a new batch can be started. Batch reactor efficiency depends heavily on "down time". Continuous reactors are efficient because there is no down time, but one must first solve a myriad of problems.
Continuous reactors are presumed to run in '"steady state", i.e., nothing changes in time. But how does one begin a such a reaction? By definition, the startup is not steady state. So that's a problem. Suppose you solve that problem, and you approach steady state production. You can never really achieve steady state production, only approximate it. To remain close, you must monitor and continuously adjust many parameters: input material flow rates, raw material purity. reaction temperature, reactor pressure, energy input. And lots more. So you need a sophisticated feedback control system.
Feedback control systems can be tricky. Let's take an example of the steering wheel of a car as the adjustable parameter, the direction of the wheels as the outcome of the system, and you, as the driver, the feedback control system. Most everyone can drive a car without too much effort, so there appears to be no control problem. But wait. Suppose that it took two seconds for the direction of the wheels to respond to an adjustment of the steering wheel. How well do you think you could steer the car? Even in a straight line. What could happen is that, because of the time lag between adjustments to the steering wheel and turning of the wheels, you would over-adjust the steering wheel. And this would lead to larger and larger swerving of the car. That is, you, as the feedback control system, would become unstable.
The same thing can happen in a reactor: instability. Because of time delays between adjustments and outcomes, feedback control systems in continuous reactors must be quite sophisticated.
So the engineering of reactors is complicated. . . but it's also interesting.
The other major element of chemical engineering is separation processes. We'll spend more time on this topic because there's more that can be learned with limited background in chemistry.
Separation processes:
Remember, chemical engineers produce chemicals in two ways: they create them from raw materials using reactors; or, if the compounds already exist in combination with other substances, they find a way to extract and purify them. The latter method incorporates the field of separation processes.
The idea is a very simple one. Find some characteristic or property of the desired compound that is different from that of the surrounding substances and figure out a way to exploit that difference in a separation method.
Let's talk turkey—at least, turkey gravy. When turkey gravy is first made it contains two principal parts: gravy and fat. What your chemical engineering mother wants to do is separate the gravy from the fat. What does she do? Sometimes two things? One is simply wait. She knows that fat is less dense than the water-based gravy. So given enough time, the fat will rise to the top leaving the gravy relatively pure on the bottom. The second thing she might do is put the combination in the refrigerator. Although the fat is already on the top, she'd like to find a convenient way to mechanically separate the fat from the gravy. She knows that animal fat freezes at a relatively high temperature, maybe 20( C. But to speed up the freezing process she puts it in the refrigerator. The water-based gravy, of course, freezes at 0( C. After a while, she can then simply spoon out the solidified fat to leave just the gravy. That's the gist of separation processes.
A large part of chemical engineering is involved with separating substances from one another. Some of the techniques are extremely simple; others are very complex. But they all depend on exploiting a difference in a material property to effect the separation.
Separation Processes / Unit Operations:
Unit operations? What are they? They represent the one concept that launched the field of chemical engineering. Arthur D. Little in 1915 coined the term 'Unit Operation':
Any chemical process, on whatever scale conducted, may be resolved
into a coordinated series of what may be termed 'Unit Actions', as
pulverizing, mixing, heating, roasting, absorbing, condensing,
lixiviating, electrolyzing and so on. The number of these basic Unit
Operations is not very large and relatively few of them are involved in
any particular process. The complexity of chemical engineering
results from the variety of conditions as to temperature, pressure,
etc., under which the unit actions must be carried out in different
processes and from the limitations as to materials of construction and
design of apparatus imposed by the physical and chemical character
of the reacting substances. (from A.D. Little: Report to the Corporation
of M.I.T., as published in the AIChE Silver Anniversary Volume, p. 7)
The idea was to divide each of the technical processes into steps which could understood in terms of basic principles: physics, chemistry, and mechanics. In taking this approach Little, took chemical engineering from an applied art to a science. Each simple step of a chemical process is now known as a unit operation. And understanding the physical and chemical principles behind each step is the key to becoming a chemical engineer.
We'll begin our discussion with an easy one: evaporation. This process has large-scale applications in chemical production. It is based on differences in volatility. The basic idea is to remove a component from a mixture through vaporization. Typically the mixture consists of a relatively nonvolatile solid or liquid and a volatile liquid. The goal is to isolate and retain the nonvolatile substance. An example is the evaporation of sea water to obtain salt.
Sometimes volatile substances are initially added to a component to give a component a temporary property. Then, when that property is no longer needed the volatiles are evaporated away. Think about fingernail polish or paint.
Distillation is a separation process that's useful in separating components of a liquid mixture that have different boiling points. It's been described as the "work-horse" of chemical engineering because of its widespread use in industry. It's a little more complicated than you might think, but not a lot so. Suppose you have a mixture of water and alcohol (ethanol) which having boiling points of 100°C and 78.5°C, respectively. Naively, you might think that you could put the mixture in a pot at, say 90(C, and wait for the alcohol to boil away. The alcohol vapor could be condensed back into liquid phase and the separation would be completed. It's not quite that easy. Unfortunately, water as well as alcohol enters the vapor phase. Fortunately, since the boiling point of alcohol is lower than that of water, the fraction of alcohol in the vapor phase is larger than that in the liquid phase. So if we now condense this vapor we will have a new mixture which is higher in alcohol concentration than the one we started with. So, we've made progress. The chemical engineering problem is to figure out how many stages of distillation we need to obtain the desired product. We'll go into this problem later in a laboratory experiment.
This is basically how distillation works. It is just a series of vaporization and condensation processes that continues until a desired concentration is reached. One of its biggest applications is in the petroleum industry where crude oil is separated into its man hydrocarbon components: natural gas, gasoline, kerosene, diesel fuel, heating oils, and tars. And, of course, it's used to produce brandy—a high alcohol drink—from wine—a low alcohol drink
Gas Absorption and Gas Desorption (or stripping) constitute another area of separations. As you can guess from the names, both processes involve the transfer of a vapor or gas, and the only difference lies in whether the transfer is to or from a liquid or gas phase.
In gas absorption, a component of a gas mixture is transferred to a liquid. If you have ever seen a fish tank, you have probably seen this process in action. The bubbler in the tank injects a stream of air into the water, and the water absorbs oxygen from the air. This oxygenates the water so that the fish can breathe.
Why does the oxygen decide to move into the water phase? To answer these questions, we turn to the property of solubility. For absorption to work, the gas mixture component that is to be transferred must be soluble in the liquid that is to absorb it. Then, if this condition is met, the component will naturally move from the gas to the liquid phase as long as there is a concentration gradient between the two phases for the transferring component. At some point an equilibrium is reached between the oxygen in the gas phase and the oxygen in the liquid phase. Now there is no more "driving force", and the transfer of oxygen from the gas to liquid phase stops.
Humans are gas absorption/desorption systems. As we breathe, oxygen enters our lungs and is absorbed by the bloodstream. Waste carbon dioxide from tissue metabolism also enters the bloodstream and is desorbed through the lungs. At increased pressures, the solubility of these gases in blood is increased. So, when a SCUBA diver makes a deep dive a higher concentration of these gasses enter the bloodstream than can be handled at the surface. If a SCUBA ascends too quickly, the excess gas will come out of solution as bubbles. This is a serious and sometimes fatal problem.
Gas absorption is used often for environmental reasons. Suppose, for example, you are burning coal with a high sulfur content and you want to control the amount of sulfur (SO2) that reaches the atmosphere. You can insert a "scrubber" which consists of a series of water nozzles spraying water across the smoke stack. The water will absorb the SO2, and thus remove it from the exhaust. Of course, now you have to figure out what to do with sulfur-contaminated water.
Gas desorption or stripping is simply the reverse process of absorption, as it is the transfer of a gas or vapor from a liquid phase to a gas phase. Because of the similarity between processes, the same principles apply to both gas absorption and desorption. A common application of desorption is found in chemical plants. Desorption helps to remove residual amounts of gaseous contaminants in process water before it leaves the plant. For instance, gaseous ammonia can be removed by a desorption process between the water stream and an air stream. This helps chemical plants meet the strict environmental requirements for water purity.
Extraction or solvent extraction is another type of separation process, and it includes a number of techniques. All of the extraction techniques share the characteristic of using one liquid to remove a component from another liquid or solid. The liquid which will acquire this component is called the solvent.
In liquid-liquid extraction, a component dissolved in one liquid mixture is transferred to another liquid. There are two main requirements for this process. First, the substance that is moving between the two liquids must be soluble in both. Second, the two liquids between which the substance is transferred must be immiscible. A common example of immiscible liquids is oil and water. Because they are immiscible in each other, they form two distinct layers when placed in the same container. If the two liquids were not immiscible, then there would not be two distinct phases between which the third substance could be transferred, and the whole process would make no sense. A useful thing about immiscibility is that we do not have to worry about separating the two liquids after combining them for the extraction to take place. They will naturally separate from each other.
Liquid-liquid extraction occurs in a manner similar to that of gas absorption. The liquid containing the component to be transferred is brought into contact with the liquid solvent. The component then moves to the liquid solvent phase, and this movement can continue until equilibrium is reached. Engineers have given fancy names to the two liquid phases that are the result of extraction. The liquid that originally contained the component is called the raffinate, and the liquid solvent that extracts the component is appropriately called the extract.
Extraction processes are used in the production of penicillin. First, an extraction step separates the penicillin from other liquid materials in the fermenting vessel. Then a second extraction process moves the penicillin into an aqueous solution so that it can be “dried” and made into its final form for packaging.
In another type of extraction technique, a liquid solvent removes a component from a solid. This is called leaching. And this is a technique you use every day (if you drink coffee! A liquid solvent (water) extracts the coffee flavor and color that we are familiar with from the solid phase coffee beans. This is possible because the compounds that give coffee its flavor and color are soluble in water. Contact between the water and ground up beans allows the water to extract the compounds.
More separation processes: filtration. This one's pretty obvious. Filtration is the process of removing a solid from a liquid/solid or gas/solid mixture. It is very likely that you have already observed several filtration processes. For instance, air conditioners usually have some sort of foam pad that filters dust and other solid particles from the air stream. Draining cooked spaghetti with a colander could even be considered a filtration process. A solid (the spaghetti) is separated from a liquid (the water) by means of a filter (the colander.)
Filtration is different than the other types of separation processes we have talked about because it separates materials based on physical properties. Previously, chemical properties like boiling points and solubilities were important and determined whether or not the separation could occur. With filtration, physical properties like size and shape are what counts. One of the real challenges in filtration is keeping the filter from clogging and keeping the filter characteristics from changing due to loading.
Chromatography is another type of separation process used by chemical engineers. Its primary advantages are that it is typically a gentle process and does not pose the threat of hurting or destroying the substances that are being separated. This feature is particularly useful in the biotechnology field in isolating delicate biological substances such as proteins for use in the production of therapeutic drugs. Because many biological substances are easily harmed by the high temperatures or harsh chemicals involved in other types of separation processes, the gentler chromatography process is a much-needed alternative.
There are many types of chromatography, including thin-layer, gas, and wet-column chromatography. We'll only look at the simplest variety of this type of process – paper chromatography.
Paper chromatography uses three basic elements: a solvent, a solute (a mixture of components that are to be separated), and a strip of absorbent paper (to provide a surface on which the separation can take place.) In a typical paper chromatography experiment, a small amount of solute is deposited near one end of a strip of the absorbent paper. The paper is then placed in a container holding a small amount of solvent, and the separation process begins. The solvent is drawn up the surface of the paper by capillary action. After travelling a short distance alone, the solvent reaches the point where the solute was deposited on the paper. At this point, the solute dissolves in the solvent, and the entire mixture continues to travel up the paper.
The actual separation between the components of the solute occurs as the mixture travels up the paper. The distance that a component will travel up the paper with respect to the solvent is based on the degree of “affinity”, or attraction, a component has for the paper and the solvent. As the component travels up the paper with the solvent, there is an ongoing competition between the solvent and the paper for the component. Whether the component will remain in the solvent or attach to the paper depends on this competition.
If the component has a stronger affinity for the paper, it will prefer to attach to the paper rather than remain dissolved in the solvent. In this case, the component will not travel very far up the paper. Conversely, if component has a stronger affinity for the solvent, it will prefer to remain in solution with the solvent rather than attach to the paper. Thus, the solvent will be able to carry the component farther up the paper. Or, more appropriately, we can look at it in terms of flow rates. Components having a greater affinity for the solvent will have higher flow rates since they prefer to stay in the mobile solvent phase. Components having a greater affinity for the paper will have slower flow rates since they prefer to attach to the stationary paper phase. Each component in the original solute mixture has at least slightly different affinities for the paper. Thus over the distance traveled by the solvent, the components will be physically spread out. And each component can be separately rinsed from its position on the paper.
All solvents and papers are not the same. Each combination can yield different component flow rates. You will get a better feel for this separation process when you carry out a paper chromatography laboratory assignment.
That's about it for the major unit operations found in the chemical engineering world. There are a few others: centrifugation to accelerate separation based on density differences; refrigeration to exploit differences in freezing points; electrophoretic separations which exploit ionic differences.
There is one common denominator to all these processes: differences. They all rely on one or more properties that are different between the component of interest and everything else. However, it's one thing to know how to separate the components, it's another to do it efficiently. Again, the engineer must keep in mind the universal engineering parameter $$$.
OK. Now we know how, in principle, to isolate a chemical or component. Now the question becomes "How do we actually carry it out?" The answer: in steps. . .
Single-stage processes
We concentrate now on how separation techniques are carried out at large scale. In our discussion above, we talked about making coffee and separating grease from gravy—all small-quantity processes. But, in real chemical plants where these processes are employed, it is more likely that we would be dealing with flowing streams of materials in large quantities. Still these larger scale processes would be analyzed as though they were divided into individual stages. So, what exactly is a stage and what happens there?
Basically, a stage is an area where phases are brought into close contact so that a component can be transferred. Usually, this contact is in the form of counter-current flow between two streams. This means that the stream of one phase is flowing in the direction opposite to the stream of the other phase. While the streams are in contact, a component can move from its original phase to the other phase. Ideally, this redistribution continues in the stage until both phases have reached equilibrium concentrations. (In reality, equilibrium conditions are seldom reached, and this causes process inefficiency.) The phases then leave the stage in separate streams, and each carries a new component concentration.
To analyze this process, we need to look at the Newton's Law of chemical engineering: mass balance. What goes in must come out. This law has nothing to do with Newton, but I use the expression here to suggest that mass balance to the chemical engineer is just as important as F=ma is to the mechanical engineer. Using the simple principle of mass and/or energy balance, a chemical engineer can analyze almost any process. To illustrate this, let's analyze what goes on in one counter-current flow stage. Here's a simple picture.
[pic]
L and V designate the mass flow rates of the two streams. L always represents one phase, while V represents the other. The number subscripts relate what stage a stream is coming from. (Although we are only looking at one stage, the streams are labeled as if a second stage existed. This helps us get accustomed to the idea of how the labeling system works, and it gives us four clearly identified streams.)
With all of the streams associated with the stage accounted for and labeled, we are ready to write a mass balance. Our first assumption is that the stage is at steady-state. This means that there is no accumulation of mass in the stage. Therefore, the general equation for a mass balance is:
IN = OUT
Easy enough! Now let’s write this equation in terms of our labeled streams. All of the streams pointing to the stage go on the “IN” side of the equation, and the streams pointing away from the box go on the “OUT” side of the equation. This gives us
L0 + V2 = L1 + V1 = M
The M is there simply to represent the total mass entering or leaving the stage.
Let's say we want to use this stage to remove dye from oil in a liquid-liquid extraction process. We will want to transport this dye to a second fluid, water, which is immiscible in oil. So we have three components – dye, oil, and water . We'll label them as follows:
A = dye
B = oil
C = water
The amount of each of these substances in each stream is what we want to determine. We can represent these amounts in terms of mass fractions and label them in the following manner:
xAO = mass fraction of A in stream LO
yA1 = mass fraction of A in stream V1, and so forth…
If we need to represent the total mass of a component in a stream rather than its mass fraction, all we have to do is multiply the total mass flow rate of the stream by the mass fraction of the component. For example,
L0 xA0 = mass of component A in stream L0
The equation we wrote earlier is called on overall mass balance.
L0 + V2 = L1 + V1 = M
It shows the total mass going into and out of the stage. Usually, we also need to write individual mass balances for each of the components, as well. This is done in the same way as it was for the overall mass balance. Let’s begin with the dye (component A). The basic equation is the same.
A in = A out
Now, we rewrite the equation in terms of the individual streams. The definitions established earlier should be helpful.
L0 xA0 + V2 yA2 = L1 xA1 + V1 yA1 = M xAM
Let's interpret the first term. L0 is the mass flow rate of the incoming water stream. xA0 is the percentage of A (dye) in that stream. So, L0 xA0 is the mass flow of A in the incoming L stream. The inflows are in streams L0 and V2; the outflows are in streams L1 and V1. The input mass flow of A must equal the output mass flow of A.
Following the same procedure, the mass balance for the water (component C) is
L0 xC0 + V2 yC2 = L1 xC1 + V1 yC1 = M xCM
A mass balance could also be written for the oil (component B). However, the equation would not be independent of the three we have already written for the stage since we also know that the sum of the mass fractions in each stream equals unity,
i.e., xA + xB + xC = 1.
OK. Now let's apply this to our problem—dye-contaminated oil. The phase represented by V is oil (or component B) contaminated with dye (component A.) L represents the second phase and consists of water (component C.) It is used to extract the dye from the oil. When V comes in contact with L, the dye redistributes itself between the two phases. Also, and very importantly, L and V are immiscible, making them two distinct liquid phases. The picture below clarifies what each stream looks like in this scenario.
[pic]
Let's analyze this a component at a time. First, the oil mass flow rate. Well, we really don't know what the oil mass flow rate is, but we do know the following:
Oil mass flow rate = V( = V(1 –yA) = constant.
That is, the rate at which oil is entering, V(, whatever that value is, equals the rate at which oil is leaving V(. We can write the same thing about the water mass flow rate:
Water mass flow = L( = L(1 – xA) = constant.
Let’s use component A as our analyzed component. From above, the mass balance for component A is
L0 xA0 + V2 yA2 = L1 xA1 + V1 yA1
Now, we’re going to employ a trick and multiply each term in the equation by 1. Of course we'll do it in a special way:
[pic]
Let's look at the first term. Part of that term is L0 (1 – xA0) = L( = constant. That's the mass flow rate of the L stream minus the mass of dye, i.e., the mass flow rate of just the water. Using the same reasoning for the other three terms we get a mass balance equation for the dye as:
[pic]
To make this equation useful, we need one other assumption or piece of information. And that is, what is the equilibrium concentration of dye when both the oil and water phases are in close contact. The concentrations are assumed to follow Henry's Law that says that the concentrations will come into equilibrium in a fixed ratio. In our case we can write this as yA1 = H xA1, where H will depend on the chemistry of all three components, A, B, and C.
We now have all the tools to solve a problem. There is a crucial assumption, however. And that is that all concentrations come into equilibrium according to Henry's Law before components leave the stage. Just because we can solve the mathematical problem doesn't necessarily mean that we can actually match the mathematical results in a real stage.
OK. A problem. Suppose we have 100kg/hr of dye-contaminated oil (1% by weight) that we mix with 100kg/hr of water to reduce the dye concentration in the oil. What is the resulting dye concentration in the oil after passing through one mixing stage? Assume that Henry's constant H = 4. Here's a graphic of the problem.
The solution. Start with the dye mass flow balance and plug in the numbers:
[pic]
Then, using Henry’s Law for the relationship between xA1 and yA1 as
xA1 = (1/4) yA1 = 0.25 yA1
the dye mass balance equation is reduced to
[pic]
This yields a result of yA1 = .008. So we reduced the dye concentration from 0.01 to 0.008. That's a reduction of 20%. Not a lot, but it's a start.
Here's a few problems you can try yourself: 1) What would be the resulting dye concentration in the oil if the water flow rate were 200kg/hr? 2) What water flow rate would be necessary to reduce the dye concentration in the oil to 0.001?
We've just analyzed a single stage separator. Maybe we need better dye removal. What can we do? What about a multistage countercurrent process?
[pic]
Notice that the initial dye-containing oil stream is not met by clean water, but rather by dye-contaminated water from previous stages. The analysis of this system is a bit more tedious than that of a single stage, but the principle is the same. Here, one might ask the question "How many stages do we need to bring the dye concentration down to, say, 0.0001?"
You should leave this single stage process example with two thoughts. First, with very simple concepts, .e.g, mass balance, you can analyze what appear to be quite complex processes. And second, the process we analyzed here is universal for all countercurrent processes. That is, we analyzed a particular unit operation that can be used in a number of different applications.
As we mentioned earlier, just because we can do the arithmetic to get an answer doesn't mean that we can create the apparatus to carry it out. That's where chemical engineering comes in. For example, in the problem we studied, how big a mixing container should one use. The larger the container, the longer the residence time, the greater the opportunity for the dye to come into equilibrium in the two phases. But size costs $$$. Chemical engineers must find the solution.
In this section, we have opened only the tiniest window into the field of chemical engineering. But one can now imagine what a chemical engineer does. He produces things. What things? Almost everything. How? Using a very large bag of tricks, techniques and technologies. Sometimes he creates the desired product; sometimes he isolates the desired product from other contaminants. What does he need to know? Everything—chemistry, physics, mathematics, transport, thermodynamics, gene engineering, biology, materials. What does he need to be concerned with? Environmental pollution, safety, and $$$. No wonder they call the chemical engineer a universal engineer.
-----------------------
raw materials
energy
product + contaminants
energy
byproducts
catalyst
catalyst
Stage 1
L( = 100kg/hr
xA0 = 0 (pure water)
[pic]
[pic]
reactor
Oxygen
Hydrogen
Hydrogen
V( = 99kg/hr
yA1 = ?
L( = 100kg/hr
xA1 = ?
V( = 100(1-0.01) = 99kg/hr
yA2 = 0.01
................
................
In order to avoid copyright disputes, this page is only a partial summary.
To fulfill the demand for quickly locating and searching documents.
It is intelligent file search solution for home and business.
Related searches
- chemical engineering conversion sheet
- introduction to engineering course
- chemical engineering unit conversion sheet
- introduction to engineering design pdf
- introduction to electrical engineering pdf
- introduction to engineering online free
- chemical processes in the body
- chemical engineering phd jobs
- chemical engineering professional society
- the chemical processes of cellular respiration
- chemical engineering programs ranking
- top chemical engineering phd programs