51 0 Fermentation and Biochemical Engineering Handbook Often in multiple-effect evaporators the concentration of the liquid being evaporated changes drastically from effect to effect, especially in the latter effects. In such cases, this phenomenon can be used to advantage by staging one or more of the latter effects. Staging is the operation of an effect by maintaining two or more sections in which liquids at different concentra- tions are all being evaporated at the same pressure. The liquid from one stage is fed to the next stage. The heating medium is the same for all stages in a single effect, usually the vapor from the previous effect. Staging can substantially reduce the cost of an evaporator system. The cost is reduced because the wide steps in concentrations from effect to effect permit the stages to operate at intermediate concentrations, which result in both better heat transfer rates and higher temperature differences. 6.0 ENERGY CONSIDERATIONS FOR EVAPORATION SYSTEM DESIGN The single largest variable cost factor in making a separation by evaporation is the cost of energy. Ifcrude oil is the ultimate source of energy, the cost of over $126.67 per m3 ($20 per barrel) is equivalent to more than $3.33 for 1 million kJ. Water has a latent heat of 480 kJkg at 760 mm of mercury, absolute, so the energy required to evaporate 1 kg of water exceeds 0.16 cents. Therefore, the efficient utilization ofenergy is the most important consideration in evaluating which type of evaporation system should be selected. Energy can never be used up; the first law of thermodynamics guarantees its conservation. When normally speaking of “energy use” what is really meant is the lowering ofthe level at which energy is available. Energy has a value that falls sharply with level. Accounting systems need to recognize this fact in order to properly allocate the use of energy level. The best way to conserve energy is not to “use” it in the first place.[22] Of course, this is the goal of every process engineer when he evaluates a process, but once the best system, from an energy point of view, has been selected, the necessary energy should be used to the best advantage. The most efficient use of heat is by the transfer of heat through a heat exchanger with process-oriented heat utilization, or by the generation of steam at sufficient levels to permit it to be used in the process plant directly as heat. When heat is available only at levels too low to permit recovery in the process directly, thermal engine cycles may be used for energy recovery. Heat pumps may also Evaporation 51 I be used to “pump” energy from a lower to a higher level, enabling “waste” heat to be recovered through process utilization. Thermal efficiencies of heat exchangers are high, 90-95%. Thermal efficiencies of thermal engine cycles are low, 10-20%. Heat pumps permit external energy input to be reduced by a factor of 4 to 5; however, the energy required in a heat pump is in the work form, the most expensive energy form. Utility consumption, of course, is one of the major factors which determine operating cost and, hence, the cost of producing the product for which a plant has been designed. In order to select the proper equipment for a specific application it is important to be able to evaluate different alterna- tives, which may result in a reduction of utility usage or enable the use of a less costly utility. For example, the choice of an air-cooled condenser versus a water-cooled condenser can be made only after evaluating both equipment costs and the costs of cooling water and horsepower. When heating with steam, a selection ofthe proper steam pressure level must be made when designing the evaporator. No definite rules for the selection can be established because of changing plant steam balances and availability. However, it is generally more economical to select the lowest available steam pressure level which offers a saturation temperature above the process temperature required. Some evaporator types require relatively low temperature differences. Some products may require low temperature in order to reduce fouling or product degradation. Maximum outlet temperatures for cooling water usually are dictated by the chemistry ofthe cooling water. Most cooling water contains chlorides and carbonates; consequently temperatures at the heat transfer surfaces must not exceed certain values in order to minimize formation of deposits or scale, which reduces heat transfer and leads to excessive corrosion. In addition, velocity restrictions must be imposed and observed to prevent corrosion and fouling as a result of sedimentation and poor venting. Stagnant conditions on the water side must always be avoided. In some plants, water consumption is dictated by thermal pollution restrictions. Unnecessary restraints should not be imposed on the pressure drops permitted across the water side of condensers. All too often, specified design values for pressure drop are too low and much higher values are realized when the unit has been installed and is operating. Not only does this result in more expensive equipment, but frequently the water flow rate is not monitored and cooling water consumption is excessive, increasing operating costs. Because cooling water consumption is governed by factors other than energy conser- vation and because cooling water velocities must be maintained above certain values, tempered water systems can be effectively used at locations where 512 Fermentation and Biochemical Engineering Handbook cooling water temperatures vary with the season of the year. At some locations a 3OoC difference between summer and winter water temperatures is experienced. At such locations a tempered water system may be used in order to reduce both pumping costs and maintenance costs. A tempered water system requires a pump to recycle part of the heated cooling water in order to maintain a constant inlet water temperature. Evaporative-cooled condensers in many applications give greater heat transfer than air-cooled or water-cooled condensers. The evaporative equipment can do this by offering a lower temperature sink. Evaporative- cooled condensers are frequently called wet-surface air-coolers. Perhaps the best description for this type of equipment is a combination shell-and-tube exchanger and cooling tower built into a single package. The tube surfaces are cooled by evaporation of water into air. Air-cooled condensers are especially attractive at locations where water is scarce or expensive to treat. Even when water is plentifbl, air coolers are frequently the more economical alternative. Elimination of the problems associated with the water side of water-cooled equipment, such as fouling, stress-corrosion cracking, and water leaks into the process, is an important advantage of air-cooled equipment. In many cases, carbon steel tubes can be employed in aircooled condensers when more expensive alloy tubes would otherwise have been necessary. The use of air-cooled heat exchangers may eliminate the need for additional investment in plant cooling water facilities. Maintenance costs for air-cooled equipment are about 25% of the maintenance costs for water-cooled equipment. Power requirements for air- coolers can vary throughout the day and the year if the amount of air pumped is controlled. Water rates can be varied to a lesser degree because daily water temperatures are more constant and because water velocities must be kept high to reduce maintenance. The initial investment for an air-cooled condenser is generally higher than that for a water-cooled unit. However, operating costs and maintenance costs are usually considerably less. These factors must be considered when selecting water or air as the cooling medium. Air-cooled condensers employ axial-flow fans to force or induce a flow of ambient air across a bank of externally finned tubes. Finned tubes are used because air is a poor heat transfer fluid, The extended surface enables air to be used economically. Several types of finned-tube construction are avail- able. The most common types are extruded bimetallic finnedtubes and fluted tension-wound finned tubes. The most common fin material is aluminum. Air-cooled heat exchangers generally require more space than other types. However, they can be located in areas that otherwise would not be used (e.g., on top of pipe racks). A forced draft unit has a fan below the tube Evaporation 513 bundle which pushes air across the finned tubes. An induced draft unit has a fan above the tube bundle which pulls air across the finned tubes. Air- cooled condensers are normally controlled by using controllable-pitch fans. Good air distribution is achieved ifat least 40% ofthe face area ofthe bundle is covered with fans. It is most economical to arrange the bundles and select the fan diameters to minimize the number of fans. Controllable-pitch fans permit only the air flow required for heat transfer to be pumped. An important added advantage is the reduction of the power required for operation when ambient air temperature is lower than that used for design. Controllable-pitch fans can result in a 50% reduction in the annual power consumption over fixed-blade fans. There are many ways to waste energy in pumping systems. As energy costs have continued to climb, it has often been found that a complete pumping unit’s initial investment can be less than the equivalent investment value of one electrical horsepower. Calandria circulating pumps require a certain availableNPSH. This is usually obtained by elevating the evaporator, often with a skirt. Quite often the designer establishes the skirt height before he selects the calandria recirculating pump. In the interest of economy he provides a skirt as short as possible, often without realizing that he will be forever paying an energy penalty for a smaller initial capital savings. More efficient pumps often require greater NPSH. Therefore, it is prudent to check the NPSH requirements of pumping applications before establishing skirt heights of evaporator systems. Heat pumps or refrigeration cycles involves the use of external power to “pump” heat from a lower temperature to a higher temperature. The working fluid may be a refrigerant or a process fluid. Heat pumps use energy that often would otherwise be thrown away in the form of waste heat in effluents or stack gases. The external energy input can be reduced by a factor of 4 to 5, depending on the temperature difference and temperature level of the heat pump system. There are several ways to increase the steam economy, or to get more evaporation with less steam input, for certain types of evaporation applica- tions. The use of multiple-effect configurations or compression evaporation can be considered for large flow rates of relatively dilute aqueous solutions. Both multiple-effect and compression evaporation systems require a sizable incremental capital investment over single-effect evaporators, and these systems are larger and more complex than the simpler one-stage evaporators. Like the multiple-effect evaporators described above, compression evapora- tion systems can only be justified by a reduced level of steam consumption. 514 Fermentation and Biochemical Engineering Handbook In a compression evaporation, a part or all of the evaporated vapor is compressed by a compressor to a higher pressure level and then condensed, usually in the heating element, thus providing a large fraction of the heat required for evaporation.[23] Energy economy obtained by multiple-effect evaporation can sometimes be equalled in a singleeffect compression evaporator. Compression can be achieved with mechanical compressors or with steam jet thermo-compressors. To achieve reasonable compressor costs and power requirements, compression evaporators must operate with fairly low temperature differences, usually from 5" to 10°C. This results in a large heat transfer surface, partially offsetting the potential energy economy. When a compression evaporator of any type is designed, the designer must provide adequate heat transfer surface and may decide to provide extra area over that required to anticipate reduced heat transfer should fouling occur. If there is inadequate surface to transfer heat available after compression, the design compression ratio will be exceeded causing a thermo-compressor to break or bacwre or a mechanical compressor to exceed the horsepower provided. Mechanical compression evaporation (Figs. 19 and 20) is generally limited to a single effect. All of the vapor is compressed and condensed, eIiminating the cooling water required for conventional or steam jet thermo- compression evaporators; an advantage when cooling water is costly. Me- chanical compression is ideally suited for locations where power is relatively inexpensive and fuel is expensive. The greatest advantage of mechanical compression is the high energy economy. Compressors may be reciprocating, rotary positive displacement, centrifugal, or axial flow. Single stage positive displacement compressors appear to be better suited to compression evapo- ration because of lower cost and their characteristic fixed capacity, dependent only on speed or discharge pressures. They are, however, limited in developed compression ratios and material of construction. The compressor may be driven with a diesel unit, a steam turbine, a gas turbine or an electric motor. Selecting the compressor drive requires analysis of all factors present at a particular location. One disadvantage of mechanical compression is that most systems require a heat source to initiate evaporation during start-up. Because the vapor is frequently water, which has a low molecular weight and a high specific volume, compressors are usually quite large and costly. Compressors require high purity of the vapor to avoid buildup on the blades of solids that result from evaporation of liquid as the vapor is superheated by compression. Liquids having high boilingpoint elevations are Evaporation 51 5 not usually adaptable to compression evaporation. Mechanical compression evaporation sometimes requires more heat than is available from the com- pressed vapor, so the evaporation rate can be controlled by regulating the makeup steam flow to maintain a constant liquor temperature. Usually, mechanical compression results in slightly higher maintenance costs because of the compressor and its drive. Mechanical compression is best suited for atmospheric or pressure operation, for mildly corrosive vapors, for low boiling-point elevation liquids, low temperature differences across the calandria, and where energy economy is important. Body I1 Mokeup steam -4 Heater 2 41 Condensate II i Circulating pump -, kf E Thick liquor discharge - +Feed Figure 19. Mechanical recompression applied to forced-circulation evaporator. (From Unit Operations of Chemical Engineering by W. L. McCabe and J. D. Smith (2nd. ed., 1967), p. 473. OMcGraw-Hill. Used with permission of McGraw-Hill Book Company) 51 6 Fermentation and Biochemical Engineering Handbook Evaporation 51 7 Steam jet thermo-compressors can be used with either single or multiple-effect evaporators. As a rule-of-thumb, the addition of a thermo- compressor will provide an improved steam economy equivalent to an additional effect, but at a considerably lower cost. Thermo-compressors have low efficiencies which further diminish when the jet is not operated at its design point. Thermo-compressors in a typical operation can entrain one pound of vapor per pound of motive steam. They are available in a wide range of materials of construction, and can have a wide range of design and operating conditions. They should be considered only when high pressure motive steam is available, and when the evaporator can be operated with low pressure steam. Motive steam pressures above 60 psig usually are required to justig using thermo-compressors. Steam condensate from thermo- compressors often is contaminated with trace amounts of product and may have to be treated before being returned to the steam generator. It is relatively easy to design an evaporator using thermo-compression for a given set of operating conditions. However, once the thermo-compres- sor has been designed and fabricated, its performance characteristics are basically fixed. The design of a thermo-compression evaporation system should include an analysis ofthe consequences of changing operating points. The characteristics of a thermo-compressor make it difficult to predict performances at conditions different from the design point, so accurate prediction of evaporator performance at other than design conditions be- comes impossible. Because of the unpredictable performance of thermo-compressors, control of evaporators using them is more difficult than for a conventional system where it is necessary to set only steam and feed rates to maintain a constant evaporation rate. One way to provide flexibility with better operating stability is to use two or more thermo-compressors in parallel. This permits capacity control without loss in energy economy. Thermo-compres- sion evaporators are used for single or double-effect systems where low operating temperatures and improved economy are desired. It costs less to add athermo-compressor instead of an additional effect, and both have about the same effect on energy economy. The temperature differences across the thermocompressor should be below 15OC. This evaporator system is not as flexible as multiple-effect systems because of the unpredictable variation of performance characteristics for the thermo-compressor under changing operating conditions. 518 Fermentation and Biochemical Engineering Handbook 7.0 PROCESS CONTROL SYSTEMS FOR EVAPORATORS From the process viewpoint, the two parameters that should be regulated are the concentration and flow rate of the bottoms product. If the composition of the feed stream is constant, good control of the feed rate and the evaporation rate will give the desired concentrated product at the proper production rate (see Fig. 1). Of course, the method of control can depend upon the evaporator type and method of operation. When evaporation rate is to be maintained at a constant rate, a steam flow controller is generally used. Steam flow control usually is accomplished by throttling the steam which results in a loss of temperature difference. Steam may, therefore, be uncontrolled to achieve maximum capacity. Steam pressure controllers may be used to protect the equipment or to assure substantially constant tempera- tures in the front end of a multistage evaporation system. Constant temperatures in the later effects of the evaporator can be controlled with a pressure controller on the last effect. A control system consists of three parts: a measurement; a control algorithm; and a process actuator. The process actuator (often a control valve) is always a direct user of energy; the measurement may take energy from the process (as in the case of a head-type flow meter); and the control calculation never requires a significant energy supply. However, the correct control calculation is essential for energy-efficient operation of any process. The well-engineered control system depends on the ability to directly measure the parameter that is to be controlled, or to measure another parameter from which the controlled variable can be inferred. In every case, a measurement of the controlled variable is preferred. A survey of the measurements in a major production unit gave the following distribution of process Type of Measurement Percent Flow 34 Temperature and analytical 24 Pressure 22 Liquid level 20 Flow rates are the largest single group of process measurements used for control, and flow is the only process variable for which significant energy may be required by the measuring device. Most flows are measured by orifice meters which are heat-type devices that extract head loss from the pumping Evaporation 51 9 system. The amount of power required by an orifice, nozzle, or venturi tube meter can be significant. There are many flow sensors available and numerous considerations to be evaluated in the design of a flow metering system. The cost of operating each meter should be evaluated and the type selected should have the best balance between operating, maintenance, and capital costs. Although the energy required to operate a process unit can be reduced if the designer becomes sensitive to the hidden cost in each meter installation, the amount of energy required to operate most process meters is small; and the opportunities for significant reductions in energy usage by modifications of flow meters in lines less than 10 inches in diameter is limited. A control system requires a mechanism to change the state of the process when a disturbance causes the control variable to move from the desired value. This control mechanism is most often a control valve although it can be a motor, a set of louvers, an electrical power supply, a fan on an air- cooled condenser, etc. Control which is achieved by changing the area of the valve body opening is a direct energy expense to the operating unit. The control valve is a variable orifice device in which the size of the orifice is adjusted to control a process variable. Consequently, the manufac- turer, type, or even the size of a control valve has no effect on the energy dissipated in the control of a selected stream once the process pressure, line size, and pumps have been selected. This energy-independence of the control valve assures that continuous throttling of the flow stream is required to control a process variable. In those cases where a valve is used for shut-off or ovemde control (not a continuous throttling device), energy savings can be realized by selecting a valve with a minimum pressure loss in the full-open position. Any control system which is properly designed to control the evapora- tion process must maintain both an energy and material balance across the evaporator boundaries. The control system must be able to accommodate some fluctuation in the feed flow rate or composition within a specified range, and still enable the evaporation system to perform the required separation with stable operation. The control system should function to reduce heat input with a reduction in feed rate, or change the evaporation rate as changes in the feed composition occur. The best control system should be used in the design of evaporator systems. Products which are off-specification require additional time, expense, and energy in reprocessing. A properly designed control system can do much to reduce these wastes, and ensure that the evaporation system uses the optimum energy during normal operation. 520 Fermentation and Biochemical Engineering Handbook Product concentration can be controlled by measuring a number of physical properties, most usually specific gravity and boiling point elevation. Control is usually accomplished by controlling the discharge of product from the evaporator. Feed rate and flow rates between effects are then adjusted to maintain constant liquid levels. When this is not possible, product concen- tration may be controlled by throttling the feed. Often there is a considerable time lag before a change in feed rate is reflected by a change in product concentration. Liquid level control in evaporators may be important for product concentration, to prevent scaling, and to maintain heat transfer rates. Level control may also reduce splashing and entrainment. Several methods are used to control the amount of heat removed in the condenser, including controlling the cooling medium flow rate or tempera- ture, changing the amount of surface available for heat transfer, and introducing inert gases into the condensing vapor. The condensate from a condenser is subcooled. Because of the temperature gradient across the condensate liquid film, there is no way to avoid subcooling even when it is not desired. In some cases, the condensate is purposely subcooled several degrees in order to reduce product losses through the vent. For total condensers with essentially isothermal conditions, subcooling results in apressure reduction, unless something is donetoprevent it. This happens because the subcooled liquid has a vapor pressure lower than the operating pressure. The system pressure for a subcooled condensate will be the vapor pressure of a condensate when no inert gases are present. Permitting the system pressure to vary as the degree of subcooling changes is not usually desirable. A constant pressure vent system is normally provided to prevent this pressure kind of variation. Inert gases are introduced or removed as required to ensure that the system pressure drop is reflected only by the friction drop and not by changes in vapor pressure. The inert gases should be introduced downstream of the condenser; introducing inerts upstream of the condenser will reduce the heat transfer rate requiring more heat transfer surface. This approach, however, is sometimes used to control the condenser. When the condenser and condensate tank are closely connected, the condensate tank must be properly sized in order to permit the condensate liquid level to be controlled somewhere in the condensate tank. If the condensate tank is too small, liquid level control can be achieved only by flooding part of the condenser, especially when the condensate is pumped from the tank. Liquid level must be maintained in the condensate tank and not in the condenser. Control of natural circulation calandrias presents some problems not found in other heating elements. When heating with condensing vapors, Evaporation 521 changes in condensing pressure affect four variables: heat transfer coeffi- cients, temperature difference, liquid composition, and circulation rate. The same four variables are affected when throttling the flow of a liquid heating medium. The whole mode of operation is changed when one variable is altered, and it is not always possible to predict from experience which direction the change will take. The liquid level on the shell side of kettle-type re-boilers should be sufficient to ensure that all the tubes are covered with the boiling fluid. Controlling by varying process liquid levels may result in fouling ofthe heat transfer surface as part ofthat surface is deprived of liquid. In addition, the temperature difference may be affected as the hydrostatic head (which affects boiling temperature) is reduced. The temperature gradient across the liquid film in falling-film evapo- rators must be kept relatively low, usually less than 15°C. Excessive temperature difference between the process and utility fluids may result in boiling of the fluid on the heat transfer surface with resulting fouling. Film boiling can also occur with subsequent reduction in the rate of heat transfer. Inert gases are sometimes injected into a falling-film evaporator in order to reduce the partial pressure required to vaporize the volatile component. This technique will often eliminate the need for vacuum operation. Enough inert gas must be injected to achieve the desired results, but too much can produce flooding and entrainment, resulting in poor control. Steam-heated calandrias with process boiling temperature less than 100°C can present control problems, especially at reduced rates and during start-up. In most such cases, low-pressure steam is used for heating. Control is usually achieved by throttling the entering steam in order to reduce the pressure at which it is condensed. At reduced rates this often results in steam pressures less than atmospheric or less than the steam condensate return system pressure. The steam is usually removed through steam traps which require a positive pressure differential to finction. In order for the trap to hnction, steam condensate floods part of the steam chamber until the steam pressure is sufficient to operate the trap. This leads to poor control and all the problems associated with condensate flooding. Calandrias heated with sensible heat from a hot liquid are normally controlled by throttling the liquid flow. Usually, good control may be achieved by controlling the temperature of the heating medium. The best utilization of the available heat transfer surface is achieved by maximizing the temperature difference in the calandria, and this is accomplished by designing for high pumping rates for the heat transfer medium and the process fluid in order to achieve nearly isothermal conditions on both sides of the heating surface. 522 Fermentation and Biochemical Engineering Handbook 8.0 EVAPORATOR PERFORMANCE Energy economy and evaporative capacity are the major measures of evaporator performance. When evaporating water with steam, the economy is nearly always less than 1 .O for singleeffect evaporators, but in multiple- effect evaporators it is considerably greater. Other performance variables to be considered include: product quality, product losses, and decrease in performance as scaling, salting, or fouling occurs. Designers of evaporation systems strive to achieve high heat transfer rates. This can be justified by a costhenefit analysis. High rates of heat transfer in theory must often be proved in practice. Evaporators designed for high rates of heat transfer are generally more affected by traces of scale or non-condensable gases. Product loss requirements may be an important factor in the evaporator design. Provisions to reduce product losses have far less effect on the cost of an evaporator system than does the amount of heat transfer surface. Product losses in evaporator vapor occur as a result of entrainment, splashing, or foaming. Foaming properties of the liquid may at times dictate the selection of evaporator type. Losses from entrainment result from the presence of droplets in the vapor that cannot separate because of the high vapor velocity. Entrainment is thought to be due mainly to the collapse of the liquid film around vapor bubbles. This collapse projects small droplets of liquid into the vapor. The amount of entrainment is a function ofthe size distribution ofthe droplets and the vapor velocity. Bubbles leaving the surface cause droplets of different sizes to be propelled upward. The smaller droplets are caught in the fast moving vapor and are carried upward as entrainment, while the larger ones fall to the surface. The largest size (or mass) drop camed up is dependent on the gas velocity and density, and on other physical properties. At very high gas velocities, large drops produced at the surface are shattered into smaller droplets and all the generated entrainment is carried overhead, flooding the device. This breakup phenomenon occurs when the inertial forces, which cause a pressure or force imbalance, exceed the surface tension forces, which tend to restore a drop of its natural spherical shape. The gas velocity at which this flow crisis develops is the flooding velocity and is given by the following equation: (PL -P& 'Gf E 0'7[ ] Evaporation 523 Equipment containing both gases and liquids in which the gas flows vertically upward will be flooded at velocities exceeding that predicted from the above equation. In practice, most equipment is designed to operate well below the flooding limit. Factors such as disengaging height, convergence effects, and nonuniform gas velocities prevent operation at velocities exceeding roughly half the flooding velocity. The amount of entrainment from an upward flowing gas can be estimated and is a function of gas velocity, gas and liquid densities, and surface tension. Entrainment can be separated from a gas stream with a variety of mechanisms, including gravity, inertial impaction, interception, centrifugal force, and Brownian motion. Separators can be classified according to mechanism, but it is more useful to categorize them by construction type. Separators in common use include: flash tanks, vane impingement separators, wire mesh separators, Karbate strut separators, centrifugal separators, cyclones, and special separator designs, Flash tanks are generally used when the liquid entrainment exceeds 20% ofthe gas flow on a weight basis. Flash tanks may be either vertical or horizontal. Proper sizing of a flash tank should result in a residual entrainment under 3% of the gas rate. Vane impingement, wire mesh, centrifugal, and Karbate strut separators are commercial proprietary design and all compete for similar applications. Performance and cost, however, can vary widely from one type to another. The designer should understand the advantages and disadvantages of each type and the level of separation that each type can achieve. Except for flash tanks and some special separators, the efficiency of all these separators tends to increase with increasing velocity up to a maximum allowable limit. In this region the efficiency seems to depend primarily on gas velocity and particle size, and to be somewhat insensitive to gas and liquid physical properties. Except for the cyclone and some special separators, there is a predictable maximum allowable velocity. The following equation is commonly used: where v, is the maximum gas velocity and F is an experimentally derived constant. Both equations indicate that the term @G)-” is of primary importance in separator sizing. 524 Fermentation and Biochemical Engineering Handbook 9.0 HEAT SENSITIVE PRODUCTS The world value for end products using biological manufacturing methods was approximately $250 billion in 1980.[251 This total value can be broken down as follows: $ Billion $ Billion Food 218.4 Baked goods requiring yeasts 41.4 Butter and cheese 79.2 Alcoholic beverages 90.6 Others 7.2 Drugs Biologicals Antibiotics Hormones 15.2 4.1 7.7 3.4 Others 15.0 Fuel ethanol, amino acids, enzymes 3.2 Miscellaneous 11.8 Total 248.6 248.6 Many of these food and pharmaceutical products are heat sensitive; that is, the finished product may be damaged or destroyed if it is exposed to too great a temperature over an extended period of time. Even common products like tomato catsup and penicillin “spoil” or lose their efficacy when exposed to ambient temperatures for long periods of time. The chemical reactions that limit shelf life are strongly temperature dependent. Some biological products may be handled at elevated temperatures in dilute solutions, but may degrade at the same temperature once the concentration exceeds a certain threshold value. Thus, it is not uncommon for heat sensitive products to be concentrated in two different evaporator types in series-in natural or forced circulation evaporators to perform the bulk of the solvent removal and in a low mean residence time evaporator to finish the concentration step. Because of the recycle and back mixing effects of anatural or forced circulation evaporator, the mean residence time ofthe average molecule can be several hours. Some evaporators operate without recycle; these are called once-through or single Evaporation 525 pass evaporators, and have mean residence times measured in minutes or even seconds. Figure 2 1 represents the “heat history” or the temperature-time rela- tionship as seen by the product, which is to say an average fluid element in the product. After the product reaches the final concentration within the allotted time, it cools to an ambient or final temperature. Because the loss in product quality (degradation, oxidation, polymerization, etc.) is generally due to one or more chemical reactions, this deterioration phenomenon is a rate problem which is determined by chemical reaction kinetics. It is, therefore, possible to evaporate a heat sensitive product at a higher temperature in a short residence time evaporator rather than in a recirculating or other long residence time unit. This could have important positive consequences, including increased energy efficiency and smaller evaporator size. Note that it may make sense to install a low residence time product cooler for the concentrate from a low residence time evaporator, since the product could still be damaged by long residence time cooling. ry a 3 2 n’ a 1 Y c EVAPORATION ac 3 < t TIME E VAPOR At ION Figure 21. Heat history of long (left) and short (right) residence time evaporators. (Luwu Corporation. ) An average residence time expressed as holding volume divided by discharge rate was frequently used in the past for both single-pass and recirculation evaporators. However, statistical analysis of several types of evaporators has revealed that the actual time of replacement of 97% of the 526 Fermentation and Biochemical Engineering Handbook feed in a recirculating evaporator is about 3.2 times the average residence time as defined above. It takes longer to replace a larger percentage of feed. The actual residence time achieved in any evaporator can be calculated from the equation below: where x = fraction of feed removed 0 = time, minutes r = ratio of holding volume to discharge rate, l/min Nearly every supplier of evaporation equipment and systems maintains a pilot plant facility where, for a fee, different evaporation schemes can be set up. Data obtained from several days of testing on small laboratory or pilot plant units can be good predictors of evaporator performance, and these data are very helpful in the scaling-up calculations for production-sized installations. Samplers obtained from the test work can be used to check the mass balances, concentrations, and product quality. Serious operational problems like foaming, plugging, and fouling can occur in even short pilot plant tests and can point to the need for alternative evaporator types or modified designs. 10.0 INSTALLATION OF EVAPORATORS Many details must be considered when installing heat transfer equip- ment. Some of these may seem of minor importance but it is precisely these small details that often lead to poor performance, operational difficulties, and increased maintenance. Vent and drain connections are normally provided, and they should be permanently installed. In some cases, intermittent venting ofnon-condensables may be acceptable. For vertical exchangers with cooling water on the shell side, it is essential to provide means for venting gases that are released as the water is heated. If these gases are not continuously removed, they collect in pockets; the shell side heat transfer is reduced; and corrosive attack may occur in the gaseous region. In vertical units, the gases tend to collect just beneath the top tubesheet and the tubes in this area corrode rapidly when not surrounded by liquid. Corrosion and stress cracking may occur in this area and solids may also buildup on the tubes. Vent connections should be located at positions which enable these gases to be vented and to insure that the tubesheet is swept with water. Evaporation 52 7 Vertical steam-heated exchangers must also be vented to remove carbon dioxide and other gases which can accumulate under the top tubesheet. Corrosion of the shell, tubes, and tubesheet, especially in the area just opposite the steam inlet nozzle, may result ifadequate venting is not provided. Sometimes a continuous steam purge or intermittent venting is recommended. The cooling water side of condensers should operate under a positive pressure whenever possible. Frequently, water-cooled units are mounted high in a structure and the available pressure is barely adequate to deliver water to the user. OAen a vacuum exists and boiling of the water may occur. This causes corrosion and hydraulic problems, both in the exchanger and the outlet water piping. Any control valve should be placed in the outlet piping so that the maximum available pressure is realized at the exchanger. No design should be finalized without considering what the pressure at the outlet of the exchanger will be when operating, and the consequence of a siphon effect in the cooling water line. Generally, U-bend exchangers should not be installed in a vertical position. Vertical U-bends are difficult to vent or drain on the tube side because connections cannot be provided at the U-bend end. Multi-pass exchangers are relatively high and provisions are made to ensure that the tubes operate completely flooded. Sufficient space should be allowed in the equipment layout to remove the tube bundle from the removable bundle units. Consideration should be given to units expected to require periodic maintenance or cleaning. Early recognition should be given to space requirements of air-cooled exchangers. Equipment layouts should be made recognizing that longertube lengths result in more economical heat exchangers. Care should be exercised to avoid forcing the equipment to fit the layout, rather than providing a layout to match the equipment. The manner in which a heat exchanger is piped up can influence its performance. Horizontal units should have inlet and outlet nozzles on the top and bottom of the shell or channel. Nozzles should not be on the horizontal centerline of the unit. In general, fluids should enter the bottom of the exchanger and exit at the top, except when condensation occurs. Units are almost always designed to be counterflow and the piping must reflect this. When cocurrent flow has been specified, it is equally important that it be piped to suit. Ifthe equipment will be used in corrosive service and ifthe components will have a short life expectancy, the design engineer should select units that are easy to repair. Removable bundle units may be required. In addition, the plant equipment layout should be arranged to facilitate removal and repair. 528 Fermentation and Biochemical Engineering Handbook Maintenance costs and production losses can often be reduced by specifying equipment with standard components or designing equipment to be inter- changeable among several different services. By anticipating potential maintenance problems, the designer can avoid high maintenance or cleaning expenses and costly shutdowns. To anticipate maintenance problems, the designer needs to be familiar with the plant location, process flow-sheet, and anticipated plant operation. The designer will need to know, for example, whether the equipment will have to be periodically cleaned and if chemical or mechanical cleaning methods can be used. 11.0 TROUBLESHOOTING EVAPORATION SYSTEMS Occasionally, it becomes necessary to investigate the performance of an evaporator in order to evaluate its performance at other operating conditions or to determine why the system is not performing as expected. Fortunately, most conditions that result in an evaporator not meeting expected performance are easily corrected. Troubleshooting, therefore, often means checking for small details which have agreat effect on the performance of the evaporator system. Of course, it is possible that a type of evaporator has been misapplied, the heat transfer surface that has been provided is not adequate for the intended service, or fouling is occurring. Discrepancies in performance may be caused by deviations in physical properties of fluids, flow rates, inlet temperatures, mechanical construction of the equipment, or by problems caused during the installation of the equipment. The troubleshooter should first check to see that compositions, flows, temperatures, and physical properties agree with those specified for design. He should then examine the equipment drawings to determine if the problem could lie in the manner in which the equipment was constructed, or in the manner in which the equipment has been installed. After these basic items have been reviewed, the checklist below outlines some questions that should be raised. Calandrias: 1. Has the steam side been vented to remove air or other entrapped gases? 2. Has the steam control valve been adequately sized? What is the actual steam pressure in the steam chamber? 3. Has the steam trap been properly selected and sized? Evaporation 529 4. Are the control valve and steam trap hnctioning cor- rectly? 5. Is steam condensate flooding part of the surface? What is the temperature of the steam condensate? Is the conden- sate nozzle large enough? Is steam trap piping adequately sized? 6. Is the process liquid level maintained at the proper place? Are liquid level instruments calibrated? Are instrument leads plugged? 7. Is the liquid holdup adequate to prevent surging? 8. Are process compositions and temperatures equal tothose used for design? Does the process material contain enough volatiles to provide adequate boiling? 9. What is the temperature of the top head for natural circulation calandrias? A temperature higher than the liquid temperature may indicate inadequate circulation for some reason? 10. Is the available steam pressure equal to that used for design? 1 1. Are process nozzles adequate? 12. Is the process side adequately blown down or purged? 13. Were debris and other foreign objects removed from the equipment and piping prior to start-up? How often is the unit cleaned? What is the appearance of the equipment before cleaning? Is the cleaning adequate? 14. If apump is provided, do the pump and the system match? Is the pump cavitating? 15. Has enough back pressure been provided to prevent boiling of the process fluid when the evaporator operation requires this (submerged tube type)? 16. Is entrainment occurring? Are entrainment separators properly sized and installed? Are they plugged? 17. Is dilution important? 18. Are flows adequate to maintain flow regimes used in design? Is the pressure drop out of line? 530 Fermentation and Biochemical Engineering Handbook 19. For falling film units, is the unit plumb? Is there equal liquid distribution to eachtube? Is the inlet channel vented to remove any flashed vapors? Are flows adequate to ensure that a film is formed? Is the outlet flow rate sufficient to prevent the film from breaking? Condensers: 1. Has a constant pressure vent system been provided? Has it been properly installed? Inerts should be injected downstream of the condenser, not upstream? 2. Is the vent system adequate? 3. Are condensate connections properly sized? Is liquid being entrained into the condenser? If horizontal, are the tubes level (or sloped toward the outlet)? 4. Is all piping adequate? 5. Is the water side operating under a vacuum? 6. Are the temperatures and composition equal to those used 7. Is the water flow adequate? Properly vented? 8. Were debris or other foreign objects removed from the equipment and piping prior to start-up? 9. If aircooled, is the inlet piping adequate to effect good distribution? Are fan blades properly pitched? Are motors delivering rated power? Are fan belts slipping? Is there noticeable recirculation of hot exhaust air? Performance testing is an experimental procedure to help understand the performance of an evaporation system. Tests can be performed to identi@ and characterize unsatisfactory performance, and often indicate methods to improve operation. Performance tests may also be required to establish that a new evaporator system has met performance guaranteed by the supplier. Tests may be used to determine evaporator capacity under different operating conditions or to obtain data for designing a new evaporator system. The American Institute of Chemical Engineers has published a procedure entitled Testing Procedure for Evaporators. This procedure covers methods for conducting performance tests and discusses several factors influencing performance and accuracy of test results. for design? Evaporation 531 Tests are conducted to determine capacity, heat transfer rates, steam economy, product losses, and cleaning cycles. Practically all the criteria of evaporator performance are obtained from differences between test measure- ments. Errors can result when measuring flow rates, temperatures and pressures, concentrations, and steam quality. Factors which can have a great effect on performance include dilution, vent losses, heat losses, and physical properties of fluids. Frequent causes for poor performance of an evaporator system include the following: Low Steam Economy: Steam economy with a fixed feed arrangement can be calculated from heat and material balances. Steam economies lower than those calculated during the design ofthe unit may be the result of one or more of the following: 1. Leakage of pump gland seal water 2. Excessive rinsing 3. Excessive venting 4. Flooded barometric condensers 5. Dilution from condensate leakage 1. Salted, scaled, or fouled surfaces 2. Inadequate venting 3. Condensate flooding 4. Inadequate circulation Excessive Entrainment: 1. Air leakage 2. Excessive flashing 3. Sudden pressure changes 4. Inadequate liquid levels 5. Inadequate pressure levels 6. Operation at increased capacity Short Time Between Cleaning Cycles: frequency of cleaning. Short cycles may be caused by: Low Rates of Heat Transfer: Downtime required for cleaning may not agree with the expected 532 Fermentation and Biochemical Engineering Handbook 1. Sudden changes in operating conditions (such as pressure or liquid level) 2. Low vehicles. 3. Introduction of hard water or other contaminants during 4. High temperature differences 5. Improper cleaning procedures cleaning, rinsing, or from seal leaks REFERENCES AND SELECTED READING MATERIAL 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. Perry, R. H., Chilton, C. H., andKirkpatrick, S. D., (eds.), Chemical Engineers' Handbook, Fourth Ed., pp. 1-24, McGraw-Hill, New York (1963) Minton, P. E., Course Director Lecture Outline and Notes from Evapora- tion Technology, p. B-2, The Center for Professional Advancement, East Brunswick, New Jersey, (May 22-24, 1978) Standards of Tubular Exchanger Manufacturer 'sAssociation, Fifth Ed., p. 9, New York (1968) McCabe, L. and Smith, J. C., Unit Operations of Chemical Engineering, Second Ed., pp. 439-440, McGraw-Hill, New York (1967) Kern, D. Q., Process Heat Transfer, pp. 6-7, McGraw-Hill, New York (1954) Bird, R. B., Stewart, W. E., and Lightfoot, E. N., Transport Phenomena, p. 284, John Wiley & Sons, New York (1960) Kern, D. Q., op. cit., p. 845 Gilmour, C. H., A Resume ofExpressions for Heat Transfer CoefJicients, Union Carbide Corporation, South Charleston, West Virginia (Oct. 27, 1959) Kern, D. Q., op. cit. McAdams, W. H., Heat Transmission, McGraw-Hill, New York (1954) Minton, P. E., op. cit. Perry, R. H. and Chilton, C. H., (eds.), Chemical Engineers 'Handbook, Fifth Ed., pp. 11-3 1, McGraw-Hill, New York (1973) Perry, R. H., and Chilton, C. H., ibid., pp. 11-27 Minton, P. E., op. cit., p. B-48 Minton, P. E., ibid., p. B-60 Evaporation 533 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. Sack, M., Falling Film Shell-and-Tube Heat Exchangers, 6 1st National Meeting of American Institute of Chemical Engineers, Houston, Texas (Feb. 22, 1967) Minton, P. E., op. cit., pp. B-65 toB-68 Frees, H. L., Mechanically Agitated Thin-Film Evaporators, from the short course “Evaporation Technology,” p. D-2, The Center for Profes- sional Advancement, East Brunswick, New Jersey (May 2-24, 1978) Frees, H. L., and Glover. W. B., Mechanically Agitated Thin-Film Evaporators, Chem. Eng. Progr., pp. 56-58 (Jan. 1979) Fischer, R., Agitated Thin-Film Evaporators: Part 3-Process Applica- tions, Chem Eng., p. 186 (Sept. 13, 1965) Minton, P. E., op. cit., p. B-118 Minton, P. E., ibid., p. B-91 Minton, P. E., ibid., p. B-111 Minton, P. E., ibid., p. B-149 Opportunities in Biotechnology, 1980 to 1988-1990, T. A. Sheets Company-Management Consultants, Cleveland (1 980) Minton, P. E., op. cit., pp. B-181 toB-184 Testing Procedure for Evaporators, American Institute of Chemical Engineers, New York (1 96 1) Perry, R. H., and Chilton, C. H., op. cit., pp. 11-38 Bird, R. B., Stewart. W. E., and Lightfoot, E. N., Transport Phenomena, John Wiley & Sons, New York. Evaporation: A Prime Target for Industrial Energy Conservation, En- ergy Research and Development Administration, C00/2870-1 UC-95f, Oak Ridge, Tennessee (Feb. 1977) Evaporation Technology, Wnton, P. E., Course Director), The Center for Professional Advancement, East Brunswick, New Jersey, (May 22-24 1978) Freese, H. L., and Glover, W. B., Mechanically Agitated Thin-Film Evaporators, Chem. Eng. Progr. (Jan. 1979) Gilmour, C. H., A Resume ofExpressions for Heat Transfer CoefJicients, Union Carbide Corporation, South Charleston, West Virginia (Oct. 27, 1959) Kent, J. A. (ed.), Riegel ’s Handbook of Industrial Chemistry, Seventh Ed., Van Nostrand Reinhold Company, New York (1 974) Kern, D. Q., Process Heat Transfer, McGraw-Hill, New York. McAdams, W. H., Heat Transmission, McGraw-Hill, New York (1954) McKelvey, J. M., and Sharps, G. V., Fluid Transport in Thin-Film Polymer Processor, Polymer Engineering and Science (July 1979) 534 Fermentation and Biochemical Engineering Handbook 38. 39. 40. 41. 42. 43. 44. 45. 46. Minton, P. E., Lord, R. C., and Slusser, R. P., Design ofHeat Exchangers, Chem. Eng. (Jan. 26, 1970) Mutzenberg, A. B., Parker, N., and Fischer, R., Agitated Thin-Film Evaporators, Chem. Eng. (Sept. 13, 1965) Perlman, D. (ed.), Fermentation Advances, Academic Press, New York (1969) Perry, R. H., and Chilton, C. H. (eds.), Chemical Engineers Handbook, Sixth Edition, McGraw-Hill, New York Solomons, G. L., Materials and Methods of Fermentation, Academic Press, New York (1 969) Standards of Tubular Exchanger Manufacturers Association, Fifth Ed., New York (1968) Upgrading Existing Evaporators to Reduce Energy Consumption, En- ergy Research and Development Administration, C00/2870-2, Oak Ridge, Tennessee (1 977) Underkofler, A. and Hickey, R. J. (eds.), Industrial Fermentation, Vols. I and 11, Chemical Publishing Company, New York (1954) Widmer, F. and Giger, A., Residence Time Control in Thin-Film Evaporators, Chem. and Process Eng., London (Nov. 1970)